Regulations Requiring Onboard Diagnostic
Systems on 2010 and Later Heavy-Duty
Engines Used in Highway Vehicles Over
14,000 Pounds; Revisions to Onboard
Diagnostic Requirements for Diesel
Highway Vehicles Under 14,000 Pounds
Draft Technical Support Document
&EPA
United States
Environmental Protection
Agency
Office of Transportation and Air Quality
EPA420-D-06-006
December 2006
-------
Regulations Requiring Onboard Diagnostic
Systems on 2010 and Later Heavy-Duty
Engines Used in Highway Vehicles Over
14,000 Pounds; Revisions to Onboard
Diagnostic Requirements for Diesel
Highway Vehicles Under 14,000 Pounds
Technical Support Document
Docket ID Number: EPA-HQ-OAR-2005-0047
Compliance and Innovative Strategies Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
v>EPA
United States EPA420-D-06-006
Environmental Protection December 2006
Agency
-------
DRAFT Technical Support Document; HDOBD NPRM
Table of Contents
List of Tables 2
1. Introduction 3
2. Technological Feasibility 3
2.1. Update on Oxygen Sensor Development for HD OBD 3
2.1.1. Current Technology 3
2.1.2. Heavy-duty Air-fuel Ratio (APR) Measurement for 2010 Technology Engines 6
2.2. Update on ZrO2 NOx Sensor Development 7
2.2.1. Current Technology 7
2.2.2. Future Improvements 10
2.2.3. Heavy-duty NOx Detect!on for 2010 Technology Engines 10
2.3. Fuel Injection Timing Monitor 11
3. Costs 13
3.1. Cost Analysis for Engines Used in Over 14,000 Pound Applications 14
3.1.1. Variable Costs 14
3.1.2. Fixed Costs 16
3.1.3. Total Costs 35
3.2. Cost Analysis for Under 14,000 Pound Applications 36
3.3. Updated 2007/2010 HD Highway Costs Including OBD 42
References 46
-------
DRAFT Technical Support Document; HDOBD NPRM
List of Tables
Table 1. Estimated OBD Hardware Costs for Diesel and Gasoline Engines 14
Table 2. Total OBD Variable Costs for Diesel and Gasoline Engines Used in Vehicles Over
14,000 Pounds 16
Table 3. R&D Costs for OBD Algorithm Development and Application - Diesel Engines for
Over 14,000 Pound Applications 18
Table 4. R&D Costs for OBD Algorithm Development and Application - Gasoline Engines
for Over 14,000 Pound Applications 21
Table 5. OBD R&D Test Cell Costs - Diesel Engines for Over 14,000 Pound Applications
23
Table 6. OBD R&D Test Cell Costs - Gasoline Engines for Over 14,000 Pound
Applications 24
Table 7. OBD R&D Test Cell Demand per Manufacturer - Diesel Engines for Over 14,000
Pound Applications 25
Table 8. OBD R&D Test Cell Demand per Manufacturer - Gasoline Engines for Over
14,000 Pound Applications 25
Table 9. Summary of OBD R&D Costs - Diesel and Gasoline Engines for Over 14,000
Pound Applications 26
Table 10. Cost for OBD Certification Demonstration Limit Parts - Diesel Engines for Over
14,000 Pound Applications 28
Table 11. OBD Certification and Production Evaluation Testing Costs - Diesel Engines for
Over 14,000 Pound Applications 29
Table 12. OBD Certification and Production Evaluation Testing Costs - Gasoline Engines
for Over 14,000 Pound Applications 30
Table 13. Total OBD Fixed Costs - Diesel and Gasoline Engines for Over 14,000 Pound
Applications 34
Table 14. Total Estimated OBD Costs - Diesel and Gasoline Engines for Over 14,000 Pound
Applications 35
Table 15. R&D Costs for OBD Algorithm Development and Application - 38
Table 16. OBD R&D Test Cell Costs - Diesel Applications Under 14,000 Pounds 39
Table 17. Cost for OBD Certification Demonstration Limit Parts - Under 14,000 Pound
Diesel Applications 40
Table 18. OBD Certification and Production Evaluation Testing Costs - Diesel Applications
Under 14,000 Pounds 41
Table 19. Total Estimated OBD Costs - Diesel Applications Under 14,000 Pounds 42
Table 20. Costs of the 2007/2010 Heavy-duty Highway Program* 43
Table 21. Updated 2007/2010 Program Costs Including New OBD-Related Costs 44
Table 22. Producer Price Index Data for Motor Vehicle Exhaust System Parts* 45
-------
DRAFT Technical Support Document; HDOBD NPRM
1. Introduction
This document contains technical details in support of the proposed requirements for
onboard diagnostic (OBD) systems on highway applications over 14,000 pounds, and the
proposed revisions to existing OBD requirements on highway diesel applications under
14,000 pounds. The details of these proposed requirements are not covered in this document
and can be found in the preamble to the proposed regulations contained in the docket for the
rule.l Comments on the technical details presented in this document are welcomed. Details
regarding how to comment on this document can be found in the preamble to the proposed
regulations in both the ADDRESSES and SUPPLEMENTARY INFORMATION sections.
The details presented in this document support statements in the technological feasibility
and costs sections of the preamble for this rule (sections III and VI, respectively). As such,
this document is broken into two sections: technological feasibility and costs. Note that the
bulk of our technological feasibility arguments are presented in section III of the preamble.
Only the very detailed information behind some of our findings are contained in this
technical support document. By contrast, the preamble to the rule contains only a brief
summary of our cost estimates while the details behind our cost estimates are presented here.
2. Technological Feasibility
2.1. Update on Oxygen Sensor Development for HD OBD
2.1.1. Current Technology
a. Manufacturers
Zirconium Oxide oxygen sensors have been developed to measure modal O2
concentration in spark ignition and lean burn engines. There are many manufacturers of
these devices.
b. Measurement Principle
There are two typical O2 sensor designs. The first is the lambda sensor. This sensor
consists of a main body that is a U-shaped tube of Zirconia electrolyte. Zirconia is a well
known ionic oxygen conductor at high temperatures. Pt electrodes were applied to both sides
of the zirconia tube. The inner electrode is open to the atmosphere and the outer side is open
to the exhaust gas. An example of the design of a convention lambda sensor can be seen in
Figure 1. The electromotive force of the cell is governed by the Nernst equation and can be
described as follows:
-------
DRAFT Technical Support Document; HDOBD NPRM
E =
Where R is the ideal gas constant, T is absolute temperature, F is the Faraday constant,
PO2(ref)is the partial pressure of the reference gas and PO2(test)is the partial pressure of the
sample gas.
The partial pressure of oxygen in the air, PO2(ref) is almost constant and E depends on the
partial pressure of the exhaust gas, PC^test). In a lean environment, PC^test) is close to PO2(ref)
and E approaches 0V. Under rich conditions, PO2(test) is negligible and E approaches IV. At
the stoichiometric point, E is about 0.5V. The equilibrium pressure of oxygen abruptly
changes near the stoichiometric point and a step change in E is evident at this point.2'3'4
The second type of oxygen sensor design is the UEGO (Universal air to fuel ration
Exhaust Gas Oxygen) sensor. The UEGO sensor can detect a wide range of A/F ratios,
making it possible to control an engine in a very lean or very rich fuel mixture state. These
sensors are usually considered wide range A/F ratio sensors.
The UEGO sensor is amperometric while the lambda sensor is potentiometric. The
sensor measures current which is proportional to the partial pressure of oxygen in lean
environments and the partial pressure of CO, H2, and hydrocarbons (CmHn) in rich
environments. This then provides quantitative information on the A/F ratio. As the A/F ratio
increases in a lean environment, excess oxygen in the exhaust increases. As the A/F ratio
Pt
Exhaust rl
Air p|-'$ Electrode
Zirconia
Electro lyte
Figure 1. Design of a conventional lambda sensor.
-------
DRAFT Technical Support Document; HDOBD NPRM
decreases in a rich environment, the partial pressure of CO, H2 and CmHn increases in the
exhaust due to oxygen deficiency. The concentration of these gases approach zero at the
stoichiometric point with the exhaust being composed of mostly H^O and CC>2 due to
complete combustion.
There are several other types of UEGO sensors, however their basic operating principles
are not greatly different. The most common commercial sensor used today is based on
zirconia (ZrO2) partly or fully stabilized with ytteria (Y2O3). Figure 2 shows a cross-
sectional view of the sensor element. Most sensor designs consist of pumping and
potentiometric cells with slight variations in structure. Some sensors have adopted an air-
biased pumping cell without the potentiometric cell because it automatically reverses current
flow between the rich and lean.
5678
The potentiometric cell decides whether the exhaust is lean or rich and applies the
appropriate pumping voltage to the pumping cell depending on the signal. The presence of
oxygen vacancies in the material makes the mobility of the oxygen ion 02" possible. The
resulting conductivity is very low at room temperatures, but reaches values of a wet
electrolyte when the sensor is heated up to < 600°C. An oxygen sensor can be constructed if
the solid electrolyte is provided with porous electrodes separating two gas chambers. At
higher cell temperatures the solid electrolyte conducts oxygen ions, thus an oxygen
concentration difference between the two chambers results in a voltage signal. The half cell
reactions are as follows:
4e'
2O
2-
2-
4e
•2O2
O2(ref)
Just like the lambda sensor, this voltage signal is described in a very good agreement by
the Nernst equation:
£ =
Pumping
Cell
Gap
Potentiometric
Cell
Air
Heater
Figure 2. Cross-sectional view of oxygen sensing element.
-------
DRAFT Technical Support Document; HDOBD NPRM
Each part in the sensing element functions as follows:
By pumping oxygen, the pumping cell controls the partial oxygen pressure in the
detecting cavity, which is surrounded by the inner electrode of the pumping cell and the
potentiometric cell. The potentiometric cell, made from the oxygen galvanic cell element,
works as a conventional oxygen sensor, but without any reference oxygen from atmosphere.
By supplying a very small constant pumping current to the potentiometric cell, oxygen is
pumped to the reference oxygen cavity (Air in figure 2) from the detecting cavity, resulting
in a constant self-generated oxygen partial pressure in the cavity. The pumping current is
controlled by using a feedback circuit to maintain the potentiometric cell voltage.
The pumping cell current is the sensor output and the UEGO sensor shows the pump
current respectively proportional to the oxygen amount in the exhaust gas in the lean side and
to the oxygen amount required for the complete combustion of combustible gas in the
exhaust gas in the rich side. Hence, this value corresponds to the air/fuel ratio.8
c. Durability
Durability data for diesel applications is limited. NGK has reported data for 4,700 hours
of testing (135,000 mile equivalent) on a 2.5 L diesel engine. This data showed that the aged
sensor accuracy was equivalent to the accuracy of the fresh sensor for APR determination.9
These types of sensors have been used in OBD II spark ignition applications for years
and have proven to be durable. This durability should transfer directly over to lean burn
applications.
APBF-DEC aging to 4,000 hours has been completed and data analysis is in process.
Investigation into failure and degradation process is ongoing.
10
2.1.2. Heavy-duty Air-fuel Ratio (APR) Measurement for 2010 Technology Engines
a. Usage
It is anticipated that both wide-range APR sensors and conventional oxygen sensors will
be used by the heavy-duty engine manufacturers to optimize their emission control
technologies as well as to satisfy many of the proposed heavy-duty OBD monitoring
requirements such as, fuel system, catalyst monitoring, and EGR system monitoring. Since
these sensors can be a critical component of a vehicle's fuel and emission control system, the
proper performance needs to be assured in order to maintain low emissions. Therefore, any
malfunction that adversely affects performance must be detected by the OBD system. This
can be achieved through monitoring of the sensor output voltage, resistance, impedance,
response rate, and any other characteristic of the oxygen sensor that can effect emissions
and/or other diagnostics.
-------
DRAFT Technical Support Document; HDOBD NPRM
2.2. Update on ZrO2 NOx Sensor Development
2.2.1. Current Technology
a. Manufacturers
Zirconium Oxide NOX sensors have been developed to measure modal NOX emissions
from lean burn engines. Currently there are three companies that are selling these devices.
They are as follows: NGK Automotive Ceramics, lonotec, and Ceramatec.
b. Measurement Principle
Typical NOX sensor design consists of two internal cavities and three oxygen pumping
cells designed to measure both oxygen (air to fuel ratio measurement) and NOX
concentrations. The most common commercial sensor used today is based on zirconia (ZrO2)
partly or fully stabilized with ytteria (Y2O3). The presence of oxygen vacancies in the
material makes the mobility of the oxygen ion O2" possible. The resulting conductivity is
very low at room temperatures, but reaches values of a wet electrolyte when the sensor is
heated up to < 600°C. An oxygen sensor can be constructed if the solid electrolyte is
provided with porous electrodes separating two gas chambers. At higher cell temperatures
the solid electrolyte conducts oxygen ions, thus an oxygen concentration difference between
the two chambers results in a voltage signal. The half cell reactions are as follows:
4e + O2(test) -» 2O22'
2O22' -» 4e" + O2(ref)
Firs
it d
(lP
f
H
t internal cavit
>
iffusion path
ty
{
First pumping
9l9ctrode(-) Secor
First pumping / f}, /
\ 9lectrodei + i | \\£jf I
I \)\ }J
\ \ / / /
" 1
ZT
I
eater
5
^nn
id diffusion path
Second internal cavity
£ / Second pumping 9l9ctrod9f-'i
» 1 /
i j r.fensLiring 9l9ctrod9
i SJ -1
' 1 ! '
1
— (vq) —
nv./SJ'
1
|
1
^y
Sfl
~\
\
s
v£3r
Air Diet — j
3
x -h
\ ^
R9f9renc9 9lectrcd9
•
400 mV
J ^ -7rO9 1
^-Zr02-2
^- ZrO2-3
. — ZrO2-4
~— ZrO2-5
^ZrO2-6
' 400 mV
Figure 3. Cross-sectional view of NOX sensing element.2
-------
DRAFT Technical Support Document; HDOBD NPRM
This voltage signal is described in a very good agreement by the Nernst equation:
E = \
Where R is the ideal gas constant, T is absolute temperature, F is the Faraday constant,
PO2(ref)is the partial pressure of the reference gas and PO2(test)is the partial pressure of the
sample gas.
In general, the measurement concept consists of:
1) Lowering the oxygen concentration of a measuring gas to a predetermined level in the
first internal cavity, in which NOX does not decompose, and
2) Further lowering the oxygen concentration of the measuring gas to a predetermined
level in the second internal cavity, in which NOX decomposes on a measuring
electrode and the oxygen generated is detected as a sensor signal.
Figure 3 shows a cross-sectional view of the NOX sensor element. Each part in the
sensing element functions as follows:
First Internal Cavity
The first internal cavity connects a measuring gas stream through the first diffusion path
under a predetermined diffusion resistance. There is an oxygen pumping cell and an oxygen
sensing cell inside the first internal cavity.
The first oxygen pumping cell consists of a pair of first pumping (+) and (-) electrodes
on the ZrC>2-l layer, in order to lower the oxygen concentration to a predetermined level.
The first pumping electrode (+) is platinum and the (-) electrode is a platinum/gold alloy to
reduce NOX reduction catalytic activity.
The oxygen sensing cell consists of the first pumping (-) electrode in the first internal
cavity and a reference electrode in an air duct. This allows monitoring of the oxygen
concentration in the first internal cavity by generated electromotive force and feedback to the
first oxygen pumping cell.
Second Internal Cavity
The second internal cavity connects to the first internal cavity through the second
diffusion path under a predetermined diffusion resistance. There are two different oxygen
pumping cells and an oxygen sensing cell inside the second internal cavity.
The second oxygen pumping cell consists of the second pumping (-) electrode in the
second internal cavity and the first pumping (+) electrode on the ZrCVl layer, in order to
further lower the oxygen concentration to a predetermined level. The second pumping (-)
electrode is also made of a platinum/gold alloy.
-------
DRAFT Technical Support Document; HDOBD NPRM
The oxygen sensing cell consists of the second pumping (-) electrode and the reference
electrode in the air duct to monitor the oxygen concentration in the second internal cavity by
generated electromotive force and feedback to the second oxygen pumping cell.
The NOX sensing cell consists of a measuring electrode in the second internal cavity and
the reference electrode in the air duct. The measuring electrode is rhodium and has a NOX
reduction catalytic activity. Therefore, NOX decomposes on the measuring electrode and the
oxygen generated is detected as an oxygen pumping current in the NOX sensing cell. The
sensor signal is in proportion to the NOX concentration in the measuring gas.u
c. Measurement Range
ZrC>2 NOX sensors are currently available in the 0 - 500 ppm, 0 - 1500 ppm, and 0 -
2000 ppm range. Reported accuracy is in the ± 10% range for readings in the 100 to 2000
ppm range and ±10 ppm for readings in the 0 to 100 ppm range.
d. Interference
ZrC>2 NOX sensor interference has been limited to ammonia (NH3). Sensitivity to NH3
has been shown to be up to 65% of the amount of NH3 present in the sample gas. This NH3
is converted to NOX in the internal cavities of the sensor and then measured.12 This
phenomenon may only plague urea SCR applications, where over dosing of urea could lead
to NH3 slip. In addition, urea SCR feedback control studies have shown that the NH3
interference signal is discernable from the NOX signal and can, in effect, allow the design of a
better feedback control loop than a NOX sensor that doesn't have any NH3 cross-sensitivity.
The signal conditioning method developed, resulted in a linear output for both NH3 and NOX
from the NOX sensor downstream of the catalyst.12
e. Durability
Durability data for diesel applications is limited. NGK has reported data for 1000 hours
of testing (60,000 mile equivalent) on a 2.5 L diesel engine. This data showed that the aged
sensor achieved ± 20 ppm (or ± 7% measurement error) NOX accuracy for a 300 ppm NOX
sample on a 0 to 2000 ppm range sensor. This is almost equivalent to the accuracy of the
fresh sensor in this concentration range.13
Twenty-five NGK NOX sensors in the 0 to 2000 ppm range are currently undergoing
6,000 hours of aging on a 12 L Caterpillar C-12 engine. Five of these sensors are in the
engine out location, 10 are located downstream of the DPF and upstream of the SCR catalyst,
and 10 are located downstream of the clean-up catalyst. NOX sensors are compared every
1,000 hours and are independently calibrated every 2,000 hours. Currently, data has been
reported through 2,000 hours of aging.
Typical sensor NOX exposure varies by location. On average, the 15 sensors located
upstream of the SCR catalyst were exposed to NOX concentrations in the 100 to 600 ppm
-------
DRAFT Technical Support Document; HDOBD NPRM
range. This is close to the expected range of engine out exhaust emissions for a 2010 engine,
but the range maximum is on the low side. The 10 sensors located downstream of the
cleanup catalyst were exposed to NOX concentrations in the 10 to 200 ppm range. Of the pre-
catalyst sensors, 12 degraded by 3 to 4%, while the remaining three degraded by 5 to 7%. Of
the post-catalyst sensors, 8 had minimal degradation, one failed completely, and one
degraded 30%. For those sensors that degraded a similar amount, degradation was linear.
Overall relative error ranged from 4% at engine-out concentrations to 12% at lower
concentrations.
Aging to 4,000 hours has been completed and data analysis is in process. Investigation
into failure and degradation process is ongoing.
14
2.2.2. Future Improvements
As with any maturing technologies, it is expected that improvements will be made to
sensor accuracy and durability in the near future. Requests by engine manufacturers have
been made to instrument manufacturers to develop sensors that have improved accuracy in
the 0 to 100 ppm range. Instrument manufacturers are complying with these requests and it
is expected that NOX sensors in the 0 to 100 ppm range with ± 5 ppm accuracy will be
available by the middle of 2006.
2.2.3. Heavy-duty NOx Detection for 2010 Technology Engines
a. Future NOx Emission Levels
It is expected thatNOx concentrations downstream of an emission control system on an
engine meeting the 2010 NOX standard will be in the 0 to 50 ppm range, on average,
depending on engine speed, load, and the state of the emission control system (ECS).
As an example, a 5.9 L Cummins ISB meeting the 2010 NOX standard for the FTP (0.13
g/hp-hr) and SET (0.12 g/hp-hr) using a NOX adsorber based ECS will have average NOX
emissions ranging from 0 to 60 ppm.15 Data from the APBF-DEC Heavy-Duty NOX
Adsorber/DPF Project: Heavy Duty Linehaul Platform reported NOX emissions downstream
of the ECS in the range of 0 to 200 ppm for an engine emitting NOX in the range of 0.05 to
0.5 g/hp-hr NOX over 2000 hours.16 It is important to note that the average NOX emissions
are less than 40 ppm for this engine and ECS. Therefore it is important to note that NOX
spikes larger than the average will have to be dealt with accordingly by the OBD system.
b. Current NOx Sensor Detection Limits
Current NOX sensors have a stated accuracy of ± 10 ppm in the zero to 100 ppm range
for a 0 to 2000 ppm range. Accuracies for some sensors have been reported as high as ± 30
ppm. With this in mind, current NOX sensor technology should be able detect NOX emissions
that exceed the standard by 2 to 3 times the 2010 limit.
10
-------
DRAFT Technical Support Document; HDOBD NPRM
c. Future NOx Sensor Detection Limits
If NOX sensor manufacturers are able to develop the proposed 0 to 100 ppm range sensor
with ± 5 ppm accuracy, it should be possible to accurately measure emissions increases as
low as 1.5 times the 2010 NOX emission standard. With sensor development underway, this
sensor should be available by early to mid 2006 for evaluation.
2.3. Fuel Injection Timing Monitor
It should be possible to monitor fuel injection timing by monitoring the crankshaft speed
fluctuation and, most notably, the time at which such fluctuation begins, ends, or reaches a
peak. The OBD system could compare the time to the commanded fuel injection timing
point and verify that the crankcase fluctuation occurred within an acceptable time delay
relative to the commanded fuel injection. If the system was working improperly and actual
fuel injection was delayed relative to when it was commanded, the corresponding crankshaft
speed fluctuation would also be delayed and would result in a longer than acceptable time
period between commanded fuel injection timing and crankshaft speed fluctuation.
Such a method has been described as follows in "Controls for Modern Diesel Engines,"
found at www.dieselnet.com.
In fact, some experiments were conducted at the Bendix Diesel Engine
Controls in which a signal was obtained and digitized to analyze the impulsive
flywheel motion that results from the torque development. Figure 5 shows the
results of this experiment which was conducted on a 4-cylinder Volkswagen
diesel engine. While the general observation is that in an engine the flywheel
is rotating at a steady speed, it is in fact rotating in a pulsating pattern as
shown in Figure 5. By referencing the trace in Figure 5, control engineers at
Bendix were able to infer injection timing and fueling for each cylinder.
Analysis of such trace can yield information regarding when the piston began
its downward acceleration. From this determination, an injection timing is
inferred by referencing the start of piston acceleration to a set top-dead-
center reference. Comparative analysis is then conducted by the electronic
control unit to determine the injection timing for each individual cylinder. In
injection systems where individual cylinder control of the fuel injection is
available, adjustments can be made to equalize the effective injection timing
in all cylinders. Likewise, the rate and amount of acceleration of each
flywheel impulse can be used to infer the fueling in each cylinder. Once again,
the electronic control unit is capable to adjust the cylinder-to-cylinder fueling
rate for smoother engine operation... [Emphasis added]
11
-------
DRAFT Technical Support Document; HDOBD NPRM
L'fr-J
Crank Angle
Figure 5. Torque Pulses Development in a 4-Cylinder Diesel Engine
The emphasized text suggests that, in the opinion of the author, such a torque pulse
monitor could be used to determine when injection had occurred and, therefore, if injection
had occurred at the desired timing. The author also suggests that the technique could be used
to determine if the desired fuel quantity had actually been injected. The torque pulses could
be determined using the crankshaft position sensor—that exists on the engine for proper
engine control absent OBD requirements—that also would be used for engine misfire
detection.
12
-------
DRAFT Technical Support Document; HDOBD NPRM
3. Costs
This section provides the details behind the cost analysis done in support of our proposed
Overl4,000 pound OBD program and our proposed changes to the existing Underl4,000
pound diesel OBD requirements. Details associated with the proposed requirements and
proposed changes to existing requirements can be found in the preamble to the rulemaking
and are not presented here. As a result, there may be details within this report that can be
understood only by reading the associated preamble for the proposed rulemaking.
This analysis breaks estimated costs into two primary categories: variable costs and fixed
costs. Variable costs are those costs associated with any new hardware required to meet the
proposed requirements, the associated assembly time to install that hardware, and any
increased warranty costs associated with the new hardware. Variable costs are additionally
marked up to account for both manufacturer and dealer overhead and carrying costs. The
manufacturer's carrying cost was estimated to be four percent of the direct costs to account
for the capital cost of the extra inventory and the incremental costs of insurance, handling,
and storage. The dealer's carrying cost was estimated to be three percent of their direct costs
to account for the cost of capital tied up in inventory. We adopted this same approach to
markups in the heavy-duty 2007/2010 rule and our more recent Nonroad Tier 4 rule based on
industry input.17
Fixed costs considered here are those for research and development (R&D), certification,
and production evaluation testing. The fixed costs for engine R&D are estimated to be
incurred over the four-year period preceding introduction of the engine. The fixed costs for
certification include costs associated with demonstration testing of OBD parent engines
including the "limit" parts used to demonstrate detection of malfunctions at or near the
applicable OBD thresholds. The demonstration testing costs are estimated to be incurred one
year preceding introduction of the engine while the production evaluation testing is estimated
to occur in the same year as introduction. Importantly, none of the fixed costs estimated here
consider the recent California Air Resources Board approved requirements for over 14,000
pound OBD.18
We present all of these costs in the year during which we estimate they will be incurred
by manufacturers over the 30 year time period following publication of the final rule. We
then calculate a 30 year net present value of those cost streams using both a three percent and
a seven percent discount rate to reflect the time value of money at both ends of the most
likely range.
We present all costs in 2004 dollars. We refer to both near term costs and long term costs.
The near term costs represent those costs when warranty costs are estimated to be the highest.
The long term costs consider the effects of a reduction in warranty costs. For warranty costs,
we have estimated a three percent near term rate for warranty claims and a one percent long
term rate for warranty claims.
13
-------
DRAFT Technical Support Document; HDOBD NPRM
3.1. Cost Analysis for Engines Used in Over 14,000 Pound Applications
3.1.1. Variable Costs
The variable costs we have estimated represent those costs associated with various
sensors that we believe would have to be added to the engine to provide the required OBD
monitoring capability. Our cost estimates are summarized in Table 1.
Table 1. Estimated OBD Hardware Costs for Diesel and Gasoline Engines
Used in Vehicles Over 14,000 Pounds
201 0-201 2 Model Year
New Hardware
ECU upgrade
Purge solenoid for evap leak check
Pressure sensor for evap leak check
Subtotal
Assembly labor (hours)
Assembly labor cost
Assembly labor overhead at 40%
Cost to Mfr
Warranty cost - near term at 3% claim rate
Mfr. Carrying cost at 4%
Cost to Buyer - near term
2013+ Model Year
New Hardware
MIL and wiring
Subtotal (2010+2013)
Assembly labor (hours)
Assembly labor cost
Assembly labor overhead at 40%
Cost to Mfr
Warranty cost - long term at 1% claim rate
Mfr carrying cost at 4%
Cost to Buyer - long term
Diesel
$ 30
-
-
$ 30
0.10
$ 3
$ 1
$ 34
$ 4
$ 1
$ 39
$ 10
$ 40
0.20
$ 6
$ 2
$ 48
$0
£
$ 2
$ 52
Gasoline
$ 10
$ 10
$ 10
$ 30
0.30
$ 9
$ 4
$ 43
$ 4
$ 2
$ 48
$ 10
$ 40
0.40
$ 12
$ 5
$ 57
$0
£
$ 2
$ 61
For the 2010 model year, we believe that both diesel and gasoline engines would have to
upgrade their engine control computers, or engine control units, to accommodate the
increased computing capacity required for the proposed OBD. We have estimated this cost
at $30 per engine for diesel engines and $10 for gasoline engines, inclusive of supplier
markup. We have estimated a different cost because we believe that the gasoline engines are
using computers similar, if not in fact identical to, their under 14,000 pound counterparts.
Therefore, those computer upgrades should cost little, if anything. For diesel engines, we
believe that the OBD requirements will result in a more substantial upgrade to existing
computers. Also for the 2010 model year, we believe that gasoline engines would have to
add both a purge solenoid and a pressure sensor for the evaporative system monitoring
requirement. We have estimated the cost of both of these items at $10 a piece inclusive of
supplier markup. We believe that the other sensors needed by the OBD system on both
14
-------
DRAFT Technical Support Document; HDOBD NPRM
diesel and gasoline engines will already be on the engines for either emissions control and/or
protection of the engine (e.g., temperature sensors used to protect against condensation
formation caused by overcooling of the EGR gases—engine protection—can also be used to
monitor the effectiveness of the EGR cooler—OBD). The result is a manufacturer cost
subtotal of $30 for both diesel and gasoline engines in the 2010 model year. Note that we
have not included costs for a malfunction indicator light (MIL) and associated wiring in the
2010 timeframe since we are not requiring a dedicated MIL until the 2013 model year.
We have estimated that adding these sensors and actuators will require increased
assembly time. We have estimated these times at one-tenth of an hour for diesel engines and
one-third of an hour for gasoline engines (i.e., six minutes for each newly added part). We
have estimated a labor rate of $30 per hour for this assembly along with overhead at 40
percent. This results in an estimated cost to the manufacturer of $34 and $43 for diesel and
gasoline engines, respectively, in the 2010 model year.
We have included a warranty cost recovery estimating a three percent warranty claim rate
in the near term. We have also included a four percent manufacturer carrying cost to cover
increased insurance and inventory costs incurred by the manufacturer.19 Including these
costs results in an end cost to the buyer of roughly $40 and $50 for diesel and gasoline
engines, respectively, in the 2010 model year.
For the 2013 model year, we have included costs associated with the dedicated MIL and
its wiring. These costs were estimated at $10 per engine inclusive of supplier markup.
Following the same process for assembly costs (another one-tenth of an hour per engine),
warranty costs (one percent claim rate for the long term), and carrying costs, we have
estimated the long term hardware cost to the buyer at roughly $50 and $60 for diesel and
gasoline engines, respectively.
To determine the fleetwide estimated hardware costs, or total variable costs, we looked at
the projected over 14,000 pound sales data from our 2004 model year certification database
which showed projected US sales less projected California sales of 614,500 for diesel
engines and 39,400 for gasoline engines. In the 2010 through 2012 model years, we
estimated 50 percent of engines would comply with the proposed OBD requirements based
on our proposed phase-in schedule. For model years 2013 and later, we will have 100
percent compliance. Applying the estimated hardware costs presented in Table 1 to the
appropriate projected sales in each model year through 2035, estimating a two percent
growth in sales based on 2004 sales, results in a 30 year net present value (NPV) cost of $620
million and $47 million for diesel and gasoline engines, respectively, using a three percent
discount rate. These costs, including a NPV at a seven percent rate, are shown in Table 2.
15
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 2. Total OBD Variable Costs for Diesel and Gasoline Engines Used in Vehicles Over 14,000
Pounds
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Diesel
Projected Sales
639,103
651,393
663,684
675,974
688,265
700,555
712,846
725,136
737,426
749,717
762,007
774,298
786,588
798,879
811,169
823,459
835,750
848,040
860,331
872,621
884,912
897,202
909,493
921,783
934,073
946,364
958,654
970,945
983,235
995,526
$/engine
$
$
$
$
$ 39
$ 39
$ 39
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
$ 52
% complying
0%
0%
0%
0%
50%
50%
50%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
Diesel Subtotal
$
$
$
$
$ 13,531,000
$ 13,772,000
$ 14,014,000
$ 37,588,000
$ 38,225,000
$ 38,862,000
$ 39,499,000
$ 40,136,000
$ 40,774,000
$ 41,411,000
$ 42,048,000
$ 42,685,000
$ 43,322,000
$ 43,959,000
$ 44,596,000
$ 45,233,000
$ 45,870,000
$ 46,507,000
$ 47,144,000
$ 47,782,000
$ 48,419,000
$ 49,056,000
$ 49,693,000
$ 50,330,000
$ 50,967,000
$ 51,604,000
$ 619,863,000
$ 327,800,000
Gasoline
Projected Sales
40,976
41,764
42,552
43,340
44,128
44,916
45,704
46,492
47,280
48,068
48,856
49,644
50,432
51,220
52,008
52,796
53,584
54,372
55,160
55,948
56,736
57,524
58,312
59,100
59,888
60,676
61,464
62,252
63,040
63,828
$/engine
$
$
$
$
$ 48
$ 48
$ 48
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
% complying
0%
0%
0%
0%
50%
50%
50%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
Gasoline Subtotal
$
$
$
$
$ 1,060,000
$ 1,079,000
$ 1,098,000
$ 2,816,000
$ 2,864,000
$ 2,912,000
$ 2,959,000
$ 3,007,000
$ 3,055,000
$ 3,102,000
$ 3,150,000
$ 3,198,000
$ 3,246,000
$ 3,293,000
$ 3,341,000
$ 3,389,000
$ 3,437,000
$ 3,484,000
$ 3,532,000
$ 3,580,000
$ 3,628,000
$ 3,675,000
$ 3,723,000
$ 3,771,000
$ 3,818,000
$ 3,866,000
$ 46,559,000
$ 24,653,000
Total
$
$
$
$
$ 14,591,000
$ 14,851,000
$ 15,112,000
$ 40,404,000
$ 41,089,000
$ 41,774,000
$ 42,458,000
$ 43,143,000
$ 43,829,000
$ 44,513,000
$ 45,198,000
$ 45,883,000
$ 46,568,000
$ 47,252,000
$ 47,937,000
$ 48,622,000
$ 49,307,000
$ 49,991,000
$ 50,676,000
$ 51,362,000
$ 52,047,000
$ 52,731,000
$ 53,416,000
$ 54,101,000
$ 54,785,000
$ 55,470,000
$ 666,422,000
$ 352,453,000
3.1.2. Fixed Costs
We have estimated fixed costs for research and development (R&D), certification, and
production evaluation testing. The R&D costs include the costs to develop the computer
algorithms required to diagnose engine and emission control systems, and the costs for
applying the developed algorithms to each engine family and to each variant within each
engine family. The certification costs include the costs associated with testing of durability
data vehicles (i.e., the OBD parent engines), the costs associated with generating the "limit"
parts that are required to demonstrate OBD detection at or near the applicable emissions
thresholds, and the costs associated with generating the necessary certification
documentation. Production evaluation testing costs consist of the costs associated with the
three different elements of production evaluation testing.
a. Research & Development Costs
We have broken the estimated R&D costs into three separate categories. The first of
these is the cost for developing computer controlled diagnostic algorithms. These costs are
estimated to be incurred once per manufacturer since once an algorithm is developed, it can,
practically speaking, be used over and over again with only minor changes, if any, to
improve upon the original. The second R&D cost is that for applying the manufacturer's
developed algorithm to each of its engine families. Each engine family may have a different
number of cylinders or different emissions control architecture (e.g., different combinations
16
-------
DRAFT Technical Support Document; HDOBD NPRM
of aftertreatment devices) and the algorithm may have to be adapted for each of these engine
families. Consequently, this cost is estimated to be incurred once for each of the engine
families expected to be sold. The third R&D cost is that for applying the algorithm that has
been adapted for each engine family to every variant within each engine family. Variants
within engine families have different horsepower and/or torque characteristics and, therefore,
the adapted algorithm would have to be fine tuned to each of the engine family's variants.
These costs are estimated to be incurred once for each of the remaining variants within each
family (i.e., one variant will use the adapted algorithm while the remaining variants will
require further fine tuning).
We have estimated separate development and separate application costs for the different
types of monitors—system monitors, rationality monitors, and comprehensive component
monitors. System monitors are generally the most difficult monitors and for the most part are
those monitors for which an emissions threshold exists. Nonetheless, most system monitors
are not correlated to an emissions threshold and are, instead, functional monitors that can
detect a malfunctioning component prior to emissions exceeding the applicable thresholds.
For such monitors, manufacturers generally forego the more costly emissions correlation
work and rely on the functional check alone which saves both time and money.
We have estimated that an engineer and a technician would be involved in most of the
development work since much of the work will entail testing on an engine test bed. We have
estimated that an engineer costs $100,000 a year while a technician costs $60,000 a year, and
that they each work 48 forty hour weeks per year. Table 3 shows these R&D costs for diesel
engines. The total costs shown represent industry totals for ten manufacturers.
17
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 3. R&D Costs for OBD Algorithm Development and Application - Diesel Engines for Over 14,000
Pound Applications
A. Algorithm Development Costs
System Threshold Monitors
Engineer $
Technician $
Subtotal
System Functional Monitors
Engineer $
Technician $
Subtotal
CCM Rationality Monitors
Engineer $
Technician $
Subtotal
CCM Continuity Monitors
Engineer $
Technician $
Subtotal
Total
weeks/monitor
30
15
20
5
15
1
2
Cost/monitor
$ 63,000
$ 19,000
$ 82,000
$ 42,000
$ 6,000
$ 48,000
$ 31,000
$ 1,000
$ 32,000
$ 4,000
$
$ 4,000
# of monitors
13
37
50
80
Total/Mfr
$ 1,066,000
$ 1,776,000
$ 1,600,000
$ 320,000
$ 4,762,000
Total
$ 10,660,000
$ 17,760,000
$ 16,000,000
$ 3,200,000
$ 47,620,000
B. Application Costs to each Family
System Threshold Monitors
Engineer $
Technician $
Subtotal
System Functional Monitors
Engineer $
Technician $
Subtotal
CCM Rationality Monitors
Engineer $
Technician $
Subtotal
Total
weeks/monitor
5
10
5
10
3
1
Cost/monitor
$ 10,000
$ 13,000
$ 23,000
$ 10,000
$ 13,000
$ 23,000
$ 6,000
$ 1,000
$ 7,000
# of monitors
13
37
50
Total/Family
$ 299,000
$ 851,000
$ 350,000
$ 1,500,000
#families/mfr
6.5
6.5
6.5
Total/Mfr
$ 1,944,000
$ 5,532,000
$ 2,275,000
$ 9,751,000
Total
$ 19,440,000
$ 55,320,000
$ 22,750,000
$ 97,510,000
C. Application Costs to remaining Variants
Total
Total/Variant
$ 375,000
# variants/family
4
#families/mfr
6.5
Total/Mfr
$ 9,750,000
Total
$ 97,500,000
For diesel engines, using industry input and our own engineering analysis, we have
estimated that there will be roughly 50 system monitors. Of these, we treated 13 as threshold
monitors with the remainders being functional monitors.21 Based on industry input, we have
also estimated that there will be an additional 50 rationality monitors and 80 circuit
continuity monitors.
a The 13 threshold monitors for diesel engines, based on our engineering judgment, would be: fuel system
pressure high; fuel system injection timing too advanced; fuel system injection timing too retarded; EGR low
flow; EGR slow response; EGR low cooling; variable valve timing (WT) above target; WT below target;
WT slow response; NMHC catalyst conversion; NOx catalyst system conversion; NOx catalyst system
reductant delivery; NOx adsorber performance; DPF filtering performance; DFP NMHC conversion; NOx
sensor slow response; and, NOx sensor offset. We have estimated 50 percent of engines to be SCR equipped
with 50 percent being NOx adsorber equipped. Similarly, we have estimated 50 percent to be EGR equipped
with 50 percent being VVT equipped. Using these factors on the list of threshold monitors results in 12.5
monitors for the "average" diesel engine which we have rounded to 13.
18
-------
DRAFT Technical Support Document; HDOBD NPRM
We believe that algorithm development will be more resource intensive than will
algorithm application (on a per monitor basis). For algorithm development of system
threshold monitors, we have estimated 30 engineer-weeks of development per monitor and
15 technician-weeks per monitor while for system functional monitors we have estimated 20
and 5 weeks of development per monitor, respectively. For rationality monitors, we have
estimated 15 engineer-weeks and only one technician-week since determining the proper
rationality—the engineer's job—can be difficult but testing and verifying that it works—the
technician's job—should not be difficult. For circuit continuity monitors, we have estimated
only two engineer-weeks and no technician weeks since these monitors are relatively straight
forward (open circuit/short circuit).
Multiplying by the engineer and technician labor rates and the number of monitors results
in total costs of $48 million which will be incurred during the four year period leading up to
implementation (i.e., during the years 2006 through 2009). These costs are shown in Table
3A.
For algorithm application to each engine family, we have estimated that the majority of
the work will entail testing and, therefore, it will be done by the technician. For system
threshold monitors and functional monitors, we have estimated five engineer-weeks and 10
technician weeks. For rationality monitors, we have estimated three engineer-weeks and one
technician-week because adapting these algorithms should be more straight forward than
adapting system monitors. For circuit continuity monitors, we have estimated no costs for
applying algorithms since these should be directly applicable to any engine.
These algorithm application costs will be incurred on each engine family. Our 2004
model year database shows a total of 65 diesel engine families meant for over 14,000 pound
vehicles. The database also shows 10 heavy-duty diesel engine manufacturers for an average
of 6.5 engine families per manufacturer. Multiplying the estimated weeks by the appropriate
engineering and technician labor rates, the number of monitors, the number of engine
families per manufacturer, and the number of manufacturers results in total costs of $98
million dollars. These costs are shown in Table 3B. These costs will be incurred on some
engine families during the four years leading up to the 2010 model year (i.e., one engine
family per manufacturer) and on the remaining families during the four years leading up to
the 2013 model year.
To estimate the costs for fine tuning the adapted algorithm to the remaining variants
within each engine family, we have considered this to take roughly one-quarter the effort
required for the initial engine family application. Therefore, the $375,000 cost per variant is
estimated as one-quarter of the $1.5 million per family cost to apply the algorithm to the
engine family. The variant based application costs are estimated to be incurred by those
remaining variants within the engine family (i.e., these costs are not incurred on the variant
for which the initial application work was done). Based on input from industry, we have
estimated that there is an average of five variants per engine family. As a result, the variant
application cost will be incurred on four variants per engine family. Multiplying the cost per
variant by the number of remaining variants, the average number of engine families per
manufacturer and again by the number of manufacturers results in another $98 million
19
-------
DRAFT Technical Support Document; HDOBD NPRM
dollars in total costs. These costs are shown in Table 3C. These costs will be incurred on
some engine families during the four years leading up to the 2010 model year (i.e., four
variants within one engine family per manufacturer) and on the variants of the remaining
families during the four years leading up to the 2013 model year.
We have used this same process for estimating the R&D costs for gasoline engines which
are shown in Table 4. We have used many of the same estimates for gasoline engines as for
diesel engines with the exception that we have estimated only eight system threshold
monitors for gasoline engines.b As shown in Table 4A, we have estimated that the algorithm
development costs for gasoline engines will be zero since the manufacturers of gasoline
engines (only Ford and General Motors have certified gasoline engines for over 14,000
pound vehicles) have been complying with OBD requirements for over 10 years on their
under 14,000 pound vehicles. We believe that the algorithms used in under 14,000 pound
vehicles will be directly applicable to over 14,000 pound vehicles with only some adapting of
those algorithms. The costs for adapting the existing algorithms to each engine family are
shown in Table 4B where we have estimated the costs at $4.5 million. Note that our 2004
model year certification database shows two over 14,000 pound engine families certified by
General Motors and none certified by Ford. We have estimated that Ford will certify an
engine family in future model years and, therefore, have estimated an average of 1.5 engine
families per manufacturer. Table 4C shows the costs for applying algorithms to each
remaining variant within the engine family. Again, we have estimated this cost at one-
quarter the cost of first adapting an algorithm to the engine family. These efforts are
estimated to result in another $4.5 million. All of these gasoline engine costs will be incurred
in a manner analogous to that described above for diesel engines.
b The eight threshold monitors for gasoline engines, based on our engineering judgement, would be: fuel
system too rich; fuel system too lean; multiple cylinder random misfire; secondary air system low flow; catalyst
conversion; EGR low flow; variable valve timing (WT) above target; WT below target; WT slow response;
and primary exhaust gas sensor slow response. As with diesel engines, we have estimated 50 percent to be
EGR equipped with 50 percent being WT equipped.
20
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 4. R&D Costs for OBD Algorithm Development and Application - Gasoline Engines for Over
14,000 Pound Applications
A. Algorithm Development Costs
System Threshold Monitors
Engineer $
Technician $
Subtotal
System Functional Monitors
Engineer $
Technician $
Subtotal
CCM Rationality Monitors
Engineer $
Technician $
Subtotal
CCM Continuity Monitors
Engineer $
Technician $
Subtotal
Total
weeks/monitor
30
15
20
5
15
1
2
-
Cost/monitor
$ 63,000
$ 19,000
$ 82,000
$ 42,000
$ 6,000
$ 48,000
$ 31,000
$ 1,000
$ 32,000
$ 4,000
$
$ 4,000
# of monitors
-
-
-
-
Total/Mfr
$
$
$
$
$
Total
$
$
$
$
$
B. Application Costs to each Family
System Threshold Monitors
Engineer $
Technician $
Subtotal
System Functional Monitors
Engineer $
Technician $
Subtotal
CCM Rationality Monitors
Engineer $
Technician $
Subtotal
Total
weeks/monitor
5
10
5
10
3
1
Cost/monitor
$ 10,000
$ 13,000
$ 23,000
$ 10,000
$ 13,000
$ 23,000
$ 6,000
$ 1,000
$ 7,000
# of monitors
8
42
50
Total/Family
$ 184,000
$ 966,000
$ 350,000
$ 1,500,000
#families/mfr
1.5
1.5
1.5
Total/Mfr
$ 276,000
$ 1,449,000
$ 525,000
$ 2,250,000
Total
$ 552,000
$ 2,898,000
$ 1,050,000
$ 4,500,000
1C. Application Costs to remaining Variants
[Total
Total/Variant
$ 375,000
# variants/family
4
#families/mfr
1.5
Total/Mfr
$ 2,250,000
Total
$ 4,500,000
Closely associated with the costs shown in Table 3 and Table 4 would be costs associated
with operating and maintaining the test cells required for testing and evaluating the OBD
systems and associated algorithms. To determine these costs we projected that two types of
test cell work would be done. The first would be actual emissions testing using a certified
emissions test cell. The other would be performance and/or endurance testing done in a
development test cell where OBD monitors could be evaluated against functional criteria
rather than emissions criteria and where operating hours can be amassed far more cost
efficiently than by using a certified emissions test cell. The costs associated with these
different test cells were estimated at $700 per hour for an emissions test cell and $100 per
hour for an endurance test cell. We also estimated that 90 percent of the test cell time for
OBD development work would be done in an endurance test cell with the remaining 10
percent being done in an emissions test cell.
Table 5 shows the test cell costs we have estimated for diesel engines. Note that these
costs represent the costs associated with operating existing test cells for the sake of meeting
the proposed OBD requirements. We are not projecting that any new test cells would have to
be built. As shown in Table 5, we have estimated the test cell demand for algorithm
development of a system threshold monitor at three weeks. Algorithm development of a
system functional monitor was estimated to require two weeks of test cell time while a
rationality monitor was estimated at one week. We have estimated no test cell demand for
21
-------
DRAFT Technical Support Document; HDOBD NPRM
circuit continuity monitors. We have used the same base estimates for the test cell demand
associated with applying algorithms to individual engine families except that we have
estimated the demand to be only 30 percent of that required for algorithm development. The
same is true for applying engine family algorithms to individual variants except here we have
estimated the demand to be only 10 percent of that required for initial algorithm development.
Table 5 shows how these costs are incurred in preparation for compliance in the 2010
model year and the 2013 model year. As stated above, 90 percent of the test cell demand—
i.e., the total test weeks—would be met using an endurance test cell at $100 per hour while
the remaining 10 percent of the demand would be met using an emissions test cell at $700
per hour. Note that there would be no test cell demand for algorithm development beyond
that incurred for 2010 since the same algorithms would be used for 2010 and later model
years. Table 5A shows an estimated cost for test cell operation of $1.8 million per
manufacturer or $18 million for the industry in preparation for the 2010 model year. These
costs would be incurred over the four year period leading up to the 2010 model year. For the
2013 model year when 100 percent compliance is required, the cost is estimated at $4 million
per manufacturer or $40 million total to be spread over the four year period leading up to the
2013 model year. The 2013 costs are shown in Table 5B.
Table 6A and Table 6B show the analogous information for gasoline engines complying
in the 2010 and 2013 model years, respectively. The table shows that we have estimated no
costs—development or test cell—for developing monitoring algorithms for gasoline engines
since the same algorithms as are used on under 14,000 pound vehicles can be used for over
14,000 pound vehicles. The test cell costs for gasoline engines are estimated at $1.4 million
for 2010 model year compliance and $700 thousand for 2013 model year compliance. As
with the diesel costs, these costs are expected to be incurred over the four year period leading
up to the first year of compliance.
Table 7 and Table 8 summarize the estimated test cell demand per manufacturer for
meeting the 2010 and the 2013 requirements. These summaries estimate that testing is
conducted during 48 weeks in a given year.
22
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 5. OBD R&D Test Cell Costs - Diesel Engines for Over 14,000 Pound Applications
A. R&D Test Cell Costs - Diesel
Monitor Algorithms
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family variant
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Total R&D Test Cell Costs
$ per year for 4 years
Cost for 2010
test wks
3
2
1
30%
0.9
0.6
0.3
10%
0.3
0.2
0.1
# of monitors
13
37
50
# of monitors
13
37
50
# of monitors
13
37
50
#families/mfr
1.0
1.0
1.0
#families/mfr
1.0
1.0
1.0
additional variants
4.0
4.0
4.0
total test wks
39.0
74.0
50.0
total test wks
11.7
22.2
15.0
total test wks
15.6
29.6
20.0
Costs/mfr
$ 250,000
$ 474,000
$ 320,000
$ 1,044,000
$ 261,000
Costs/mfr
$ 75,000
$ 142,000
$ 96,000
$ 313,000
$ 78,250
Costs/mfr
$ 100,000
$ 189,000
$ 128,000
$ 417,000
$ 104,250
$ 1,774,000
$ 443,500
#mfrs
10
10
10
#mfrs
10
10
10
#mfrs
10
10
10
Total
$ 2,500,000
$ 4,740,000
$ 3,200,000
$ 10,440,000
$ 2,610,000
Total
$ 750,000
$ 1,420,000
$ 960,000
$ 3,130,000
$ 782,500
Total
$ 1,000,000
$ 1,890,000
$ 1,280,000
$ 4,170,000
$ 1,042,500
$ 17,740,000
$ 4,435,000
B. R&D Test Cell Costs - Diesel
Monitor Algorithms
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family variant
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Total R&D Test Cell Costs
$ per year for 4 years
Costs for 20 13
test wks
3
2
1
30%
0.9
0.6
0.3
10%
0.3
0.2
0.1
# of monitors
# of monitors
13
37
50
# of monitors
13
37
50
#families/mfr
5.5
5.5
5.5
#families/mfr
5.5
5.5
5.5
additional variants
4.0
4.0
4.0
total test wks
total test wks
64.4
122.1
82.5
total test wks
85.8
162.8
110.0
Costs/mfr
$
$
$
$
$
Costs/mfr
$ 412,000
$ 781,000
$ 528,000
$ 1,721,000
$ 430,250
Costs/mfr
$ 549,000
$ 1,042,000
$ 704,000
$ 2,295,000
$ 4,016,000
$ 1 ,004,000
#mfrs
10
10
10
#mfrs
10
10
10
#mfrs
10
10
10
Total
$
$
$
$
$
Total
$ 4,120,000
$ 7,810,000
$ 5,280,000
$ 17,210,000
$ 4,302,500
Total
$ 5,490,000
$ 10,420,000
$ 7,040,000
$ 22,950,000
$ 40,160,000
$ 10,040,000
23
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 6. OBD R&D Test Cell Costs - Gasoline Engines for Over 14,000 Pound Applications
A. R&D Test Cell Costs - Gasoline
Monitor Algorithms
System monitor- threshold
System monitor -functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family
factor
System monitor- threshold
System monitor -functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family variant
factor
System monitor- threshold
System monitor -functional
Rationality monitor
Subtotal
$ per year for 4 years
Total R&D Test Cell Costs
$ per year for 4 years
Cost for 20 10
test wks
3
2
1
30%
0.9
0.6
0.3
10%
0.3
0.2
0.1
# of monitors
-
-
-
# of monitors
8
42
50
# of monitors
8
42
50
# families/ mfr
1.0
1.0
1.0
# families/mfr
1.0
1.0
1.0
additional variants
4.0
4.0
4.0
total test wks
-
-
-
total test wks
7.2
25.2
15.0
total test wks
9.6
33.6
20.0
Costs/ mfr
$
$
$
$
$
Costs/mfr
$ 46,000
$ 161,000
$ 96,000
$ 303,000
$ 75,750
Costs/mfr
$ 61,000
$ 215,000
$ 128,000
$ 404,000
$ 101,000
$ 707,000
$ 176,750
# mfrs
2
2
2
#mfrs
2
2
2
#mfrs
2
2
2
Total
$
$
$
$
$
Total
$ 92,000
$ 322,000
$ 192,000
$ 606,000
$ 151,500
Total
$ 122,000
$ 430,000
$ 256,000
$ 808,000
$ 202,000
$ 1,414,000
$ 353,500
B. R&D Test Cell Costs - Gasoline
Monitor Algorithms
System monitor- threshold
System monitor -functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family
factor
System monitor- threshold
System monitor -functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family variant
factor
System monitor- threshold
System monitor -functional
Rationality monitor
Subtotal
$ per year for 4 years
Total R&D Test Cell Costs
$ per year for 4 years
Costs for 20 13
test wks
3
2
1
30%
0.9
0.6
0.3
10%
0.3
0.2
0.1
# of monitors
-
-
-
# of monitors
8
42
50
# of monitors
8
42
50
# families/mfr
0.5
0.5
0.5
# families/mfr
0.5
0.5
0.5
additional variants
4.0
4.0
4.0
total test wks
-
-
-
total test wks
3.6
12.6
7.5
total test wks
4.8
16.8
10.0
Costs/mfr
$
$
$
$
$
Costs/mfr
$ 23,000
$ 81,000
$ 48,000
$ 152,000
$ 38,000
Costs/mfr
$ 31,000
$ 108,000
$ 64,000
$ 203,000
$ 50,750
$ 355,000
$ 88,750
#mfrs
2
2
2
#mfrs
2
2
2
# mfrs
2
2
2
Total
$
$
$
$
$
Total
$ 46,000
$ 162,000
$ 96,000
$ 304,000
$ 76,000
Total
$ 62,000
$ 216,000
$ 128,000
$ 406,000
$ 101,500
$ 710,000
$ 177,500
24
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 7. OBD R&D Test Cell Demand per Manufacturer - Diesel Engines for Over 14,000 Pound
Applications
A. R&D Test Cell Demand - Diesel
Monitor Algorithms
Monitor Application to each engine family
Monitor Application to each engine family variant
Total
Cells needed per mfr
Cells needed per mfr per each of 4 years
For 2010
Total test wks
163.0
48.9
65.2
277.1
CVS cell test wks
16.3
4.9
6.5
27.7
0.6
0.1
Endurance cell test wks
146.7
44.0
58.7
249.4
5.2
1.3
B. R&D Test Cell Demand - Diesel
Monitor Algorithms
Monitor Application to each engine family
Monitor Application to each engine family variant
Total
Cells needed per mfr
Cells needed per mfr per each of 4 years
For 201 3
Total test wks
-
269.0
358.6
627.6
CVS cell test wks
-
26.9
35.9
62.8
1.3
0.3
Endurance cell test wks
-
242.1
322.7
564.8
11.8
2.9
Table 8. OBD R&D Test Cell Demand per Manufacturer - Gasoline Engines for Over 14,000 Pound
Applications
A. R&D Test Cell Demand - Gasoline
Monitor Algorithms
Monitor Application to each engine family
Monitor Application to each engine family variant
Total
Cells needed per mfr
Cells needed per mfr per each of 4 years
For 2010
Total test wks
-
47.4
63.2
110.6
CVS cell test wks
-
4.7
6.3
11.1
0.2
0.1
Endurance cell test wks
-
42.7
56.9
99.5
2.1
0.5
B. R&D Test Cell Demand - Gasoline
Monitor Algorithms
Monitor Application to each engine family
Monitor Application to each engine family variant
Total
Cells needed per mfr
Cells needed per mfr per each of 4 years
For 201 3
Total test wks
-
23.7
31.6
55.3
CVS cell test wks
-
2.4
3.2
5.5
0.1
0.0
Endurance cell test wks
-
21.3
28.4
49.8
1.0
0.3
These R&D costs—algorithm development, algorithm application, and test cell—are
summarized in Table 9 for both diesel and gasoline engines. The net present value of the
estimated R&D costs through 2035 is $273 million using a three percent discount rate.
25
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 9. Summary of OBD R&D Costs - Diesel and Gasoline Engines for Over 14,000 Pound Applications
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Diesel
R&D-Algorithms
$ 11,905,000
$ 11,905,000
$ 11,905,000
$ 11,905,000
$
$
$
-
$
$
$
$
-
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$ 44,252,000
$ 40,325,000
R&D-Application
$ 7,500,000
$ 7,500,000
$ 7,500,000
$ 48,750,000
$ 41,250,000
$ 41,250,000
$ 41 ,250,000
$
$
$
$
$
$
$
$
$
$
$
-
$
$
$
-
$
$
$
$
-
$
$
$ 168,197,000
$ 139,459,000
R&D-Test Cell
$ 4,435,000
$ 4,435,000
$ 4,435,000
$ 14,475,000
$ 10,040,000
$ 10,040,000
$ 10,040,000
$
$
$
$
$
-
$
$
$
$
-
-
$
$
$
-
$
$
$
$
-
$
$
$ 50,638,000
$ 42,783,000
Subtotal R&D
$ 23,840,000
$ 23,840,000
$ 23,840,000
$ 75,130,000
$ 51,290,000
$ 51,290,000
$ 51,290,000
$
$
$
$
$
-
$
$
$
$
-
-
$
$
$
-
$
$
$
$
-
$
$
$ 263,087,000
$ 222,567,000
Gasoline
R&D-Algorithms
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
R&D-Application
$ 1 ,500,000
$ 1 ,500,000
$ 1 ,500,000
$ 2,250,000
$ 750,000
$ 750,000
$ 750,000
$ 8,127,000
$ 7,155,000
R&D-Test Cell
$ 354,000
$ 354,000
$ 354,000
$ 532,000
$ 178,000
$ 178,000
$ 178,000
$ 1,921,000
$ 1,691,000
Subtotal R&D
$ 1 ,854,000
$ 1 ,854,000
$ 1 ,854,000
$ 2,782,000
$ 928,000
$ 928,000
$ 928,000
$
$
$
$
$
-
$
$
$
$
-
-
$
$
$
$
$
$
$
$
$
$
$
$ 10,048,000
$ 8,846,000
Total R&D
$ 25,694,000
$ 25,694,000
$ 25,694,000
$ 77,912,000
$ 52,218,000
$ 52,218,000
$ 52,218,000
$
$
$
$
$
-
$
$
$
$
-
-
$
$
$
$
$
$
$
$
$
$
$
$ 273,136,000
$ 231,412,000
26
-------
DRAFT Technical Support Document; HDOBD NPRM
b. Certification and Production Evaluation Testing Costs
As noted above, the certification costs include the costs associated with testing of durability
data vehicles (i.e., the OBD parent engines), the costs associated with generating the "limit" parts
that are required to demonstrate OBD detection at or near the applicable emissions thresholds,
and the costs associated with generating the necessary certification documentation.
Cost of OBD Limit Parts
We look first at the costs associated with generating limit parts for certification
demonstration testing. These are the parts used to demonstrate OBD detection at or near the
applicable emissions thresholds. Such parts can be very difficult to generate because of the
difficulties associated with deteriorating parts just the right amount—not so much that the
thresholds are grossly exceeded thereby making the demonstration test somewhat meaningless
and not so little that emissions remain well below the thresholds.
Table 10 shows the costs we have estimated for the limit parts needed for diesel engine
demonstration testing. To arrive at these costs, we estimated the part costs of aftertreatment
devices based on our 2007/2010 highway heavy-duty rule and our recent nonroad Tier 4 rule.
However, since those costs represented costs of new parts being mass produced, we doubled the
costs here to represent the higher costs associated with orders to suppliers consisting of only one
or two parts. Fuel system costs were estimated to include costs for injectors, pressure regulators,
etc. The exhaust gas sensor costs estimate NOx sensors and estimate that these are ordered (and
costed) in sets of two. We estimated the costs for a typical light-heavy, medium-heavy, and
heavy-heavy engine assuming 6, 8, and 14 liter displacements, respectively. We sales weighted
these costs using the projected sales data from our 2004 model year certification database
excluding California sales and excluding those engines certified for use in vehicles under 14,000
pounds. We have estimated that two parts would be needed to account for possible errors and/or
the need for parts to demonstrate both a high and a low failure (e.g., EGR flow high/EGR flow
low). For variable valve timing (VVT) costs, we have estimated these based on input from
industry and not based on our prior analyses which did not consider costs for VVT systems. As
shown in Table 10, multiplying through and including the percent of engines we expect will need
the particular limit parts, results in limit parts cost of $19,400 for each diesel engine undergoing
demonstration testing.
27
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 10. Cost for OBD Certification Demonstration Limit Parts - Diesel Engines for Over 14,000 Pound
Applications
Diesel Engines
Displacement
2004 Projected Sales less CA sales
NOx Adsorber
SCR
DPF
Fuel system
Exhaust gas sensors
Turbo
EGR System
VVT
Total for Limit Parts
Light-heavy
14-19.5K
6
21,695
$ 1,500
$ 1,500
$ 2,500
$ 1,250
$ 200
$ 560
$ 370
$ 1,500
Medium-heavy
8
361,393
$ 2,000
$ 2,000
$ 3,200
$ 1,250
$ 200
$ 570
$ 440
$ 1,500
Heavy-heavy
14
231,434
$ 3,300
$ 3,300
$ 5,600
$ 1,500
$ 200
$ 630
$ 660
$ 1,500
Sales Weighted
614,522
$ 2,500
$ 2,500
$ 4,100
$ 1,300
$ 200
$ 600
$ 500
$ 1,500
Parts needed
(incl errors)
2
2
2
2
2
2
2
2
Percent
needing part
50%
50%
100%
100%
100%
100%
50%
50%
Fleet
weighted
$ 2,500
$ 2,500
$ 8,200
$ 2,600
$ 400
$ 1,200
$ 500
$ 1,500
$ 19,400
We have not estimated costs associated with generating limit parts for gasoline engines
because we do not expect that over 14,000 pound engines will be used for certification
demonstration. Instead, we expect that manufacturers will demonstrate their OBD systems using
an engine or vehicle in the under 14,000 pound range and then provide documentation in their
certification package showing how their over 14,000 pound engine is represented by the under
14,000 pound demonstration as allowed by the proposed program. While this may also be the
case for some diesel engine manufacturers, we have chosen to be conservative in our estimates
by assuming that all diesel demonstrations will be over 14,000 pounds.
We have estimated that these costs for limit parts will be incurred every three years going
forward. In 2010, one engine family per manufacturer will have to be demonstrated and in 2013
we expect another two engine families per manufacturer to undergo demonstration testing (for
diesels). We would then expect engine families to be carried-over for three years at which time
another three engines would be demonstrated, etc. This is an over simplification of the carry-
over provisions of our certification program, but it serves our purpose here and does not under
estimate the costs but rather impacts only when those costs are incurred. We use this same
simplifying assumption throughout our analysis of certification and production evaluation testing
costs as is shown in Table 11 which shows all our estimated certification and production
evaluation testing costs for diesel engines and Table 12 which shows the analogous costs for
gasoline engines.
28
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 11. OBD Certification and Production Evaluation Testing Costs - Diesel Engines for Over 14,000 Pound Applications
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Demonstration Testing Related
#of
parent
test
engines
-
-
-
10
-
-
20
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
Costs for Limit
Parts
$
$
$
$ 194,000
$
$
$ 388,000
$
$
$ 582,000
$
$
$ 582,000
$
$
$ 582,000
$
$
$ 582,000
$
$
$ 582,000
$
$
$ 582,000
$
$
$ 582,000
$
$
$ 2,848,000
$ 1,611,000
DDV Testing
Costs
$
$
$
$ 728,000
$
$
$ 1 ,456,000
$
$
$ 2,184,000
$
$
$ 2,184,000
$
$
$ 2,184,000
$
$
$ 2,184,000
$
$
$ 2,184,000
$
$
$ 2,184,000
$
$
$ 2,184,000
$
$
$ 10,687,000
$ 6,046,000
Total DDV
Costs
$
$
$
$ 922,000
$
$
$ 1 ,844,000
$
$
$ 2,766,000
$
$
$ 2,766,000
$
$
$ 2,766,000
$
$
$ 2,766,000
$
$
$ 2,766,000
$
$
$ 2,766,000
$
$
$ 2,766,000
$
$
$ 13,535,000
$ 7,657,000
Certification Documentation Related
#of
parent
families
-
-
-
10
-
-
20
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
#
remaining
families
-
-
-
-
-
-
45
-
-
35
-
-
35
-
-
35
-
-
35
-
-
35
-
-
35
-
-
35
-
-
Cert
Documentation
Costs
$
$
$
$ 50,000
$
$
$ 213,000
$
$
$ 238,000
$
$
$ 238,000
$
$
$ 238,000
$
$
$ 238,000
$
$
$ 238,000
$
$
$ 238,000
$
$
$ 238,000
$
$
$ 1,183,000
$ 670,000
Production Evaluation Testing Related
PE Testing - Scan Tool
# of engine
families for
testing
-
-
-
-
10
-
-
55
-
-
65
-
-
65
-
-
65
-
-
65
-
-
65
-
-
65
-
-
65
-
PE Costs
$
$
$
$
$ 21 ,000
$
$
$ 115,000
$
$
$ 135,000
$
$
$ 135,000
$
$
$ 135,000
$
$
$ 135,000
$
$
$ 135,000
$
$
$ 135,000
$
$
$ 135,000
$
$ 640,000
$ 347,000
PE Testing - Monitors
# of OBD
Groups
tested
-
-
-
-
10
-
-
20
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
-
30
-
PE Costs (incl
vehicle rental)
$
$
$
$
$ 193,000
$
$
$ 255,000
$
$
$ 318,000
$
$
$ 318,000
$
$
$ 318,000
$
$
$ 318,000
$
$
$ 318,000
$
$
$ 318,000
$
$
$ 318,000
$
$ 1,620,000
$ 910,000
PE Testing - Ratios
#of
monitoring
groups
tested
-
-
-
-
30
30
30
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
PE Costs
$
$
$
$
$ 7,000
$ 7,000
$ 7,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 14,000
$ 205,000
$ 112,000
PE Costs -
Total
$
$
$
$
$ 221 ,000
$ 7,000
$ 7,000
$ 384,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 14,000
$ 467,000
$ 14,000
$ 2,465,000
$ 1,369,000
Total
Certification &
PE Testing
Costs
$
$
$
$ 972,000
$ 221,000
$ 7,000
$ 2,064,000
$ 384,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 17,182,000
$ 9,697,000
29
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 12. OBD Certification and Production Evaluation Testing Costs - Gasoline Engines for Over 14,000 Pound Applications
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Demonstration Testing Related
#of
parent
test
engines
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Costs for
Limit Parts
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
DDV Testing
Costs
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
Total DDV
Costs
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
Certification Documentation Related
# of parent
families
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
# remaining
families
-
-
-
-
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
Cert
Documentation
Costs
$
$
$
$
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 8,000
$
$
$ 39,000
$ 22,000
Production Evaluation Testing Related
PE Testing - Scan Tool
# of engine
families for
testing
-
-
-
-
2
-
-
1
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
PE Costs
$
$
$
$
$ 4,000
$
$
$ 2,000
$
$
$ 6,000
$
$
$ 6,000
$
$
$ 6,000
$
$
$ 6,000
$
$
$ 6,000
$
$
$ 6,000
$
$
$ 6,000
$
$ 29,000
$ 16,000
PE Testing - Monitors
# of OBD
Groups
tested
-
-
-
-
2
-
-
1
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
-
3
-
PE Costs (incl
vehicle rental)
$
$
$
$
$ 39,000
$
$
$ 19,000
$
$
$ 45,000
$
$
$ 45,000
$
$
$ 45,000
$
$
$ 45,000
$
$
$ 45,000
$
$
$ 45,000
$
$
$ 45,000
$
$ 226,000
$ 127,000
PE Testing - Ratios
#of
monitoring
groups
tested
-
-
-
-
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
PE Costs
$
$
$
$
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1,000
$ 1 ,000
$ 1 ,000
$ 1 ,000
$ 1,000
$ 1,000
$ 1 ,000
$ 1,000
$ 16,000
$ 9,000
PE Costs -
Total
$
$
$
$
$ 44,000
$ 1 ,000
$ 1 ,000
$ 22,000
$ 1,000
$ 1 ,000
$ 52,000
$ 1,000
$ 1,000
$ 52,000
$ 1 ,000
$ 1 ,000
$ 52,000
$ 1,000
$ 1 ,000
$ 52,000
$ 1,000
$ 1,000
$ 52,000
$ 1 ,000
$ 1 ,000
$ 52,000
$ 1 ,000
$ 1 ,000
$ 52,000
$ 1 ,000
$ 270,000
$ 152,000
Total
Certification &
PE Testing Costs
$
$
$
$
$ 44,000
$ 1 ,000
$ 9,000
$ 22,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 9,000
$ 52,000
$ 1 ,000
$ 309,000
$ 174,000
30
-------
DRAFT Technical Support Document; HDOBD NPRM
Focusing first on Table 11, the limit parts costs are first incurred in 2009 in advance of the
2010 model year. The limit parts cost estimate shown in Table 10 ($19,400 per engine) is
incurred on one engine family for each of 10 engine manufacturers for a total cost that year of
$194,000. This process is carried forward every three years as discussed above. As noted, for
gasoline engines, Table 12 shows no limit parts costs or demonstration testing costs.
OBD Certification Demonstration Testing Costs
For costs associated with the actual demonstration testing of OBD parent engines (diesel
only), we have estimated that two OBD threshold monitors can be demonstrated during a given
day of testing in an emissions test cell. With our estimate of 13 threshold monitors per engine,
this means 13 days of testing in an emissions test cell that costs $700 dollars per hour or $5,600
per day to operate. The OBD parent engine, or durability data vehicle (DDV), demonstration
testing costs were then calculated by multiplying the test days per engine (13) by the dollars per
day ($5,600) and again by the number of demonstration engines being demonstrated for the
given model year. The result in 2009 is $728,000 for all 10 engine manufacturers which is
incurred one year in advance of implementation because they are certification costs. These costs
change depending on the number of engine families undergoing demonstration testing.
OBD Certification Documentation Costs
For certification documentation costs, we have estimated that a certification documentation
package for an OBD parent engine would cost $5,000 while it would cost $2,500 for a non-OBD
parent engine (i.e., an OBD child rating). We consider this to be a conservative estimate since
most child ratings would very likely incur no costs since it would be part of an OBD group
represented by the OBD parent engine and should, therefore, require no further certification
documentation. Our certification database for the 2004 model year showed 65 diesel engine
families and three gasoline engine families in the over 14,000 pound range. Multiplying the
expected number of OBD parent engines and child engines being certified for each given year by
the estimated costs to generate the certification documentation packages results in the costs
shown in Table 11 and Table 12.
OBD Production Evaluation Testing Costs
Also shown are costs for production evaluation (PE) testing. The required production
evaluation testing consists of three elements. The first of these is testing to ensure that
engines/vehicles comply with the standardization requirements of the OBD rule. This is done by
connecting a scan tool to a production vehicle to ensure that the onboard systems communicate
properly to an off board device (e.g., a scan tool). We would expect this testing to be done as
vehicles roll off the vehicle assembly line. The second element of PE testing is testing to ensure
that the OBD monitors are functioning properly. This is done by implanting or simulating
malfunctions and determining whether or not the OBD monitors run and detects them. This
testing does not involve any actual emissions testing. We would expect this testing to be done on
one to three production vehicles but required test beyond one vehicle could be done on
production engines rather than production vehicles. The third element of PE testing is testing to
-------
DRAFT Technical Support Document; HDOBD NPRM
ensure that OBD monitors are running and making diagnostic decisions with sufficient frequency
in the real world. This is done by scanning the stored OBD information contained in actual in-
use vehicles and noting the performance ratios for various non-continuous monitors. Since the
production evaluation testing is a post-certification requirement, the costs would be incurred
either as new engines/vehicles are rolling off the assembly line or during the six to 12 months
following introduction into commerce.
OBD Production Evaluation Testing Costs - Standardization Requirements
To estimate the PE testing costs for verifying the standardization requirements, we have
conservatively estimated that the actual test would take four hours and that for each engine
family sold the maximum of 10 vehicles would be tested. We have also conservatively estimated
that the testing would be done by an engineer at $100,000 per year rather than the more likely
choice of a technician at $60,000 per year. Multiplying the number of engine families by the
number of vehicles tested per family, the hours per test, and the engineer's cost per hour results
in the yearly estimated costs. This cost—shown as "PE testing - scan tool" in the tables—is
estimated at $21,000 for diesel engines in 2010 and $4,000 for gasoline engines in 2010. These
costs would be incurred on newly introduced OBD-compliant engine families. Therefore, we
have estimated costs for testing the engine families from which the OBD parent engine has been
chosen. We have also included costs for future model years assuming that most engines undergo
enough changes over a three year period to nullify the ability to carry-over from a prior year's
certification. When that occurs, we would expect the PE scan tool testing to be done.
OBD Production Evaluation Testing Costs - Monitor Verification
To estimate the PE testing costs for verifying monitors, we have first been conservative by
estimating that each manufacturer would conduct the testing for each of three OBD groups. This
overestimates these costs because some manufacturers will only have to conduct the testing on
one, and others on two, OBD groups because they do not sell enough different engine families to
require testing of three. We have also estimated that, as allowed by the proposed rule, the first
OBD group tested would have to be tested using a production vehicle while the remaining OBD
groups tested would use a production engine. We have estimated the time required to conduct
the testing at three weeks and that the testing would be done by an engineer costing $100,000 per
year. We have also estimated that it would cost $10,000 to rent or otherwise acquire a vehicle
for testing while acquiring an engine would not cost the engine manufacturer anything. Lastly,
we have estimated travel costs at $3,000 dollars for testing done on a production vehicle while
travel costs for testing on production engines would be zero. The certification and production
engine testing cost tables show—in the columns under "PE testing - monitors"—the number of
OBD groups undergoing this testing in given years. The 10 shown for 2010 represent one engine
tested from each OBD compliant engine family by each of 10 manufacturers. In 2013, we
require all engine families to comply but only up to two new engine families must undergo
certification demonstration testing and, consequently, PE testing for monitors. For simplicity, as
stated elsewhere, we have estimated three new parent engines per manufacturer undergo
certification demonstration testing every three years and, consequently, they undergo PE testing
for monitors.
32
-------
DRAFT Technical Support Document; HDOBD NPRM
OBD Production Evaluation Testing Costs - Performance Ratios
To estimate the PE testing costs for evaluating in-use performance ratios, we have first
conservatively estimated that every OBD monitoring group would have to test the maximum of
15 vehicles. An OBD monitoring group is defined first by emissions control architecture (i.e.,
combination of EGR, turbo, and aftertreatment devices) and secondly by application type (i.e.,
line haul, urban delivery, other). We have estimated that each manufacturer would have two
emissions control architectures and engines sold into each of the three application types. As a
result, there would be six monitoring groups per each of 10 different manufacturers for 60
monitoring groups being tested. This is true except for the 2010 to 2012 model years when,
since only one engine family is compliant, we have assumed only one emissions control
architecture and, therefore, only three OBD monitoring groups for each of 10 manufacturers for
30 total. We have also estimated that the test itself—simply connecting a scan tool and
downloading the performance ratio data—would take half an hour to complete by a technician
costing $60,000 per year. We have been conservative in our estimate by including costs for this
testing in every year even though we would expect that data could be carried over from one year
to the next once we are sure that monitors are indeed running at sufficient frequency in-use.
Table 13 shows the cost streams presented above for all fixed costs. The fixed costs
consist of R&D, certification, and production evaluation testing costs. Also shown are the 30
year net present values at a three percent discount rate which are $280 million for diesel, $10
million for gasoline and $291 million for the entire industry. The total fixed costs are also shown
on a per engine basis using the projected sales shown in Table 2.
33
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 13. Total OBD Fixed Costs - Diesel and Gasoline Engines for Over 14,000 Pound Applications
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Diesel
R&D
$ 23,840,000
$ 23,840,000
$ 23,840,000
$ 75,130,000
$ 51,290,000
$ 51,290,000
$ 51,290,000
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$ 263,087,000
$ 222,567,000
Cert/PE Testing
$
$
$
$ 972,000
$ 221,000
$ 7,000
$ 2,064,000
$ 384,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 17,182,000
$ 9,697,000
Subtotal
$ 23,840,000
$ 23,840,000
$ 23,840,000
$ 76,102,000
$ 51,511,000
$ 51,297,000
$ 53,354,000
$ 384,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 280,270,000
$ 232,263,000
Projected Sales
639,103
651,393
663,684
675,974
688,265
700,555
712,846
725,136
737,426
749,717
762,007
774,298
786,588
798,879
811,169
823,459
835,750
848,040
860,331
872,621
884,912
897,202
909,493
921,783
934,073
946,364
958,654
970,945
983,235
995,526
$/engine
$ 37
$ 37
$ 36
$ 113
$ 75
$ 73
$ 75
$ 1
$ 0
$ 4
$ 1
$ 0
$ 4
$ 1
$ 0
$ 4
$ 1
$ 0
$ 4
$ 1
$ 0
$ 3
$ 1
$ 0
$ 3
$ 0
$ 0
$ 3
$ 0
$ 0
Gasoline
R&D
$ 1,854,000
$ 1,854,000
$ 1,854,000
$ 2,782,000
$ 928,000
$ 928,000
$ 928,000
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$ 10,048,000
$ 8,846,000
Cert/PE Testing
$
$
$
$
$ 44,000
$ 1,000
$ 9,000
$ 22,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 309,000
$ 174,000
Subtotal
$ 1,854,000
$ 1,854,000
$ 1,854,000
$ 2,782,000
$ 972,000
$ 929,000
$ 937,000
$ 22,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 10,358,000
$ 9,020,000
Projected Sales
40,976
41,764
42,552
43,340
44,128
44,916
45,704
46,492
47,280
48,068
48,856
49,644
50,432
51,220
52,008
52,796
53,584
54,372
55,160
55,948
56,736
57,524
58,312
59,100
59,888
60,676
61,464
62,252
63,040
63,828
$/engine
$ 45
$ 44
$ 44
$ 64
$ 22
$ 21
$ 21
$ 0
$ 0
$ 0
$ 1
$ 0
$ 0
$ 1
$ 0
$ 0
$ 1
$ 0
$ 0
$ 1
$ 0
$ 0
$ 1
$ 0
$ 0
$ 1
$ 0
$ 0
$ 1
$ 0
Total
$ 25,694,000
$ 25,694,000
$ 25,694,000
$ 78,884,000
$ 52,483,000
$ 52,226,000
$ 54,291,000
$ 406,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 3,027,000
$ 519,000
$ 15,000
$ 290,627,000
$ 241,283,000
34
-------
DRAFT Technical Support Document; HDOBD NPRM
3.1.3. Total Costs
Combining the variable cost streams shown in Table 2 and the fixed costs streams shown in
Table 13 results in the total estimated costs for the over 14,000 pound proposed OBD
requirements. The results are shown in Table 14. As shown, the 30 year net present value at a
three percent discount rate is estimated at just under $1 billion with the bulk of those costs being
for new hardware in the form of more powerful engine and emissions control system computers.
Note that the per engine costs shown in Table 14 use the engine sales estimates shown in Table 2
without accounting for any phase-in (i.e., the costs have been divided by the total new engine
sales rather than dividing by the fraction of new engine sales that are compliant).
Table 14. Total Estimated OBD Costs - Diesel and Gasoline Engines for Over 14,000 Pound Applications
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Diesel
Variable
$
$
$
$
$ 13,531,000
$ 13,772,000
$ 14,014,000
$ 37,588,000
$ 38,225,000
$ 38,862,000
$ 39,499,000
$ 40,136,000
$ 40,774,000
$ 41,411,000
$ 42,048,000
$ 42,685,000
$ 43,322,000
$ 43,959,000
$ 44,596,000
$ 45,233,000
$ 45,870,000
$ 46,507,000
$ 47,144,000
$ 47,782,000
$ 48,419,000
$ 49,056,000
$ 49,693,000
$ 50,330,000
$ 50,967,000
$ 51,604,000
$ 619,863,000
$ 327,800,000
Fixed
$ 23,840,000
$ 23,840,000
$ 23,840,000
$ 76,102,000
$ 51,511,000
$ 51,297,000
$ 53,354,000
$ 384,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 3,018,000
$ 467,000
$ 14,000
$ 280,270,000
$ 232,263,000
Subtotal
$ 23,840,000
$ 23,840,000
$ 23,840,000
$ 76,102,000
$ 65,042,000
$ 65,069,000
$ 67,368,000
$ 37,972,000
$ 38,239,000
$ 41,880,000
$ 39,966,000
$ 40,150,000
$ 43,792,000
$ 41,878,000
$ 42,062,000
$ 45,703,000
$ 43,789,000
$ 43,973,000
$ 47,614,000
$ 45,700,000
$ 45,884,000
$ 49,525,000
$ 47,611,000
$ 47,796,000
$ 51,437,000
$ 49,523,000
$ 49,707,000
$ 53,348,000
$ 51,434,000
$ 51,618,000
$ 900,133,000
$ 560,063,000
Gasoline
Variable
$
$
$
$
$ 1,060,000
$ 1,079,000
$ 1,098,000
$ 2,816,000
$ 2,864,000
$ 2,912,000
$ 2,959,000
$ 3,007,000
$ 3,055,000
$ 3,102,000
$ 3,150,000
$ 3,198,000
$ 3,246,000
$ 3,293,000
$ 3,341,000
$ 3,389,000
$ 3,437,000
$ 3,484,000
$ 3,532,000
$ 3,580,000
$ 3,628,000
$ 3,675,000
$ 3,723,000
$ 3,771,000
$ 3,818,000
$ 3,866,000
$ 46,559,000
$ 24,653,000
Fixed
$ 1,854,000
$ 1,854,000
$ 1,854,000
$ 2,782,000
$ 972,000
$ 929,000
$ 937,000
$ 22,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 9,000
$ 52,000
$ 1,000
$ 10,358,000
$ 9,020,000
Subtotal
$ 1,854,000
$ 1,854,000
$ 1,854,000
$ 2,782,000
$ 2,032,000
$ 2,008,000
$ 2,035,000
$ 2,838,000
$ 2,865,000
$ 2,921,000
$ 3,011,000
$ 3,008,000
$ 3,064,000
$ 3,154,000
$ 3,151,000
$ 3,207,000
$ 3,298,000
$ 3,294,000
$ 3,350,000
$ 3,441,000
$ 3,438,000
$ 3,493,000
$ 3,584,000
$ 3,581,000
$ 3,637,000
$ 3,727,000
$ 3,724,000
$ 3,780,000
$ 3,870,000
$ 3,867,000
$ 56,916,000
$ 33,673,000
Total
$ 25,694,000
$ 25,694,000
$ 25,694,000
$ 78,884,000
$ 67,074,000
$ 67,077,000
$ 69,403,000
$ 40,810,000
$ 41,104,000
$ 44,801,000
$ 42,977,000
$ 43,158,000
$ 46,856,000
$ 45,032,000
$ 45,213,000
$ 48,910,000
$ 47,087,000
$ 47,267,000
$ 50,964,000
$ 49,141,000
$ 49,322,000
$ 53,018,000
$ 51,195,000
$ 51,377,000
$ 55,074,000
$ 53,250,000
$ 53,431,000
$ 57,128,000
$ 55,304,000
$ 55,485,000
$ 957,049,000
$ 593,736,000
Table 15 shows these costs on a per engine basis by combining the per engine costs shown
in Table 2 and Table 13.
35
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 15. Total Estimated OBD Costs per Engine for Over 14,000 Pound Applications
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
Total $/engine
Diesel
$ 37
$ 37
$ 36
$ 113
$ 114
$ 113
$ 114
$ 52
$ 52
$ 56
$ 52
$ 52
$ 56
$ 52
$ 52
$ 56
$ 52
$ 52
$ 55
$ 52
$ 52
$ 55
$ 52
$ 52
$ 55
$ 52
$ 52
$ 55
$ 52
$ 52
Gasoline
$ 45
$ 44
$ 44
$ 64
$ 70
$ 69
$ 69
$ 61
$ 61
$ 61
$ 62
$ 61
$ 61
$ 62
$ 61
$ 61
$ 62
$ 61
$ 61
$ 62
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
$ 61
3.2. Cost Analysis for Under 14,000 Pound Applications
We have used the same approach as described above for estimating costs associated with the
under 14,000 pound OBD requirements. Since we have had OBD requirements for many years
on such vehicles and engines the costs described here are incremental to past requirements. For
hardware costs, we anticipate no new costs since all sensors and actuators should already be
present and the computers should already be capable of handling the demands of OBD. We have
estimated some new R&D costs to develop the DPF monitor since our current DPF monitoring
requirement is to detect only a catastrophic failure while the proposed requirement would be
more difficult. This requirement would begin in the 2010 model year and the R&D associated
with it would be incurred over the four year period leading up to 2010.
We have estimated that nine manufacturers would be making diesels in the under 14,000
pound market. This estimates that four of the light-duty manufacturers will be selling diesels in
the 2010 timeframe. We have also used the same engineering and testing related costs for the
under 14,000 pound requirements as used above for the over 14,000 pound requirements. This is
being conservative since most testing related costs, especially official emissions testing in a
36
-------
DRAFT Technical Support Document; HDOBD NPRM
certification test cell, is generally less costly on a chassis dynamometer than on an engine
dynamometer. We have also been conservative by developing R&D costs for three
manufacturers of engines in the 8,500 to 14,000 pound range despite the fact that they each sell
into the over 14,000 pound range and, presumably, their R&D efforts there would suffice for
much of their R&D needs in the under 14,000 pound range.
The analogous tables to those presented above are presented here. Table 16 shows the R&D
costs for OBD algorithm development and application. We have estimated costs for two new
threshold monitors for the new DPF monitoring requirement, for one new threshold monitor for
the new NMHC catalyst monitoring requirement, for four and a half (on average) functional
monitors associated with DPF and NMHC catalyst monitoring, and for nine continuity monitors
associated with DPF and NMHC catalyst monitoring. We have also estimated costs for two
engine families per manufacturer with two variants each. The total costs are estimated at $8
million to be spread over the four year period prior to the 2010 implementation date for the new
monitoring requirements.
37
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 16. R&D Costs for OBD Algorithm Development and Application -
Diesel Applications Under 14,000 Pounds
A. Algorithm Development Costs
System Threshold Monitors
Engineer $
Technician $
Subtotal
System Functional Monitors
Engineer $
Technician $
Subtotal
CCM Rationality Monitors
Engineer $
Technician $
Subtotal
CCM Continuity Monitors
Engineer $
Technician $
Subtotal
Total
weeks/monitor
30
15
20
5
15
1
2
-
Cost/monitor
$ 63,000
$ 19,000
$ 82,000
$ 42,000
$ 6,000
$ 48,000
$ 31,000
$ 1,000
$ 32,000
$ 4,000
$
$ 4,000
# of monitors
3.0
4.5
4.5
9.0
Total/Mfr
$ 246,000
$ 216,000
$ 144,000
$ 36,000
$ 642,000
Total
$ 984,000
$ 864,000
$ 576,000
$ 144,000
$ 2,568,000
B. Application Costs to each Family
System Threshold Monitors
Engineer $
Technician $
Subtotal
System Functional Monitors
Engineer $
Technician $
Subtotal
CCM Rationality Monitors
Engineer $
Technician $
Subtotal
Total
weeks/monitor
5
10
5
10
3
1
Cost/monitor
$ 10,000
$ 13,000
$ 23,000
$ 10,000
$ 13,000
$ 23,000
$ 6,000
$ 1,000
$ 7,000
# of monitors
3.0
4.5
4.5
Total/Family
$ 69,000
$ 104,000
$ 32,000
$ 205,000
#families/mfr
2
2
2
Total/Mfr
$ 138,000
$ 208,000
$ 64,000
$ 410,000
Total
$ 1,242,000
$ 1,872,000
$ 576,000
$ 3,690,000
C. Application Costs to remaining Variants
Total
Total/Variant
$ 51,000
# variants/family
2.0
#families/mfr
2
Total/Mfr
$ 204,000
Total
$ 1,836,000
The R&D testing costs associated with the R&D effort that we have estimated are shown in
Table 17. These costs are estimated at $1.7 million to be spread over the four year period prior
to the 2010 implementation date, and just over $800,000 to be spread over the four year period
prior to the 2013 implementation date, for the new monitoring requirements.
38
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 17. OBD R&D Test Cell Costs - Diesel Applications Under 14,000 Pounds
A. R&D Test Cell Costs - Diesel
Monitor Algorithms
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Cost for 2007
test wks
3
2
1
30%
0.9
0.6
0.3
Monitor Application to each engine family variant
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Total R&D Test Cell Costs
$ per year for 4 years
10%
0.3
0.2
0.1
# of monitors
1.0
4.5
4.5
# of monitors
1.0
4.5
4.5
# of monitors
1.0
4.5
4.5
#families/mfr
2.0
2.0
2.0
#families/mfr
2.0
2.0
2.0
additional variants
1.0
1.0
1.0
total test wks
3.0
9.0
4.5
total test wks
1.8
5.4
2.7
total test wks
0.6
1.8
0.9
Costs/mfr
$ 19,000
$ 58,000
$ 29,000
$ 106,000
$ 26,500
Costs/mfr
$ 12,000
$ 35,000
$ 17,000
$ 64,000
$ 16,000
Costs/mfr
$ 4,000
$ 12,000
$ 6,000
$ 22,000
$ 5,500
$ 192,000
$ 48,000
#mfrs
9
9
9
#mfrs
9
9
9
#mfrs
9
9
9
Total
$ 171,000
$ 522,000
$ 261,000
$ 954,000
$ 238,500
Total
$ 108,000
$ 315,000
$ 153,000
$ 576,000
$ 144,000
Total
$ 36,000
$ 108,000
$ 54,000
$ 198,000
$ 49,500
$ 1,728,000
$ 432,000
B. R&D Test Cell Costs - Diesel
Monitor Algorithms
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Monitor Application to each engine family
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Costs for 2010
test wks
3.0
2.0
1.0
30%
0.9
0.6
0.3
Monitor Application to each engine family variant
factor
System monitor - threshold
System monitor - functional
Rationality monitor
Subtotal
$ per year for 4 years
Total R&D Test Cell Costs
$ per year for 4 years
10%
0.3
0.2
0.1
# of monitors
2.0
# of monitors
2.0
-
-
# of monitors
2.0
#families/mfr
2.0
2.0
2.0
#families/mfr
2.0
2.0
2.0
additional variants
4.0
4.0
4.0
total test wks
6.0
total test wks
3.6
-
-
total test wks
4.8
Costs/mfr
$ 38,000
$
$
$ 38,000
$ 9,500
Costs/mfr
$ 23,000
$
$
$ 23,000
$ 5,750
Costs/mfr
$ 31,000
$
$
$ 31,000
$ 92,000
$ 23,000
#mfrs
9
9
9
#mfrs
9
9
9
#mfrs
9
9
9
Total
$ 342,000
$
$
$ 342,000
$ 85,500
Total
$ 207,000
$
$
$ 207,000
$ 51,750
Total
$ 279,000
$
$
$ 279,000
$ 828,000
$ 207,000
39
-------
DRAFT Technical Support Document; HDOBD NPRM
For certification costs, we have first estimated costs for limit parts for certification
demonstration. These costs are shown in Table 18 as $4,600 per vehicle. The projected sales
numbers shown in the table are based loosely on the 2004 certification database and engineering
judgement.
Table 18. Cost for OBD Certification Demonstration Limit Parts - Under 14,000 Pound Diesel Applications
Diesel Engines
Displacement
2010 Projected Sales less CA sales
NOx Adsorber
SCR
DPF
Total for Limit Parts
LDD
2.5
100,000
$
$
$ 1,250
8.5-14K
6
470,000
$
$
$ 2,500
Sales Weighted
570,000
$
$
$ 2,281
Parts needed (incl
errors)
2
2
2
Percent
needing part
50%
50%
1 00%
Fleet Weighted
$
$
$ 4,600
$ 4,600
Table 19 shows the estimated costs for demonstration testing. Note that we have not
estimated costs for certification documentation since all of the under 14,000 pound diesel
applications are already generating and submitting OBD certification documentation. We have
also estimated no costs for production evaluation testing since we do not have requirements for
such testing in our under 14,000 pound OBD program. We have estimated costs for a total of 18
engine families with only one per manufacturer being demonstrated every three years, on
average. The 30 year net present value costs for certification demonstration testing are estimated
at $3 million and $2 million at a three percent and a seven percent discount rate, respectively.
The total costs for under 14,000 pound diesel applications are shown in Table 20. The per
vehicle numbers assume a two percent sales growth rate using the projected sales number shown
in Table 18, and entries of $0 represent costs less than $1 per vehicle. The 30 year net present
value of total costs are estimated at $13 million and $11 million at a three percent and a seven
percent discount rate, respectively. Importantly, these costs represent the incremental costs of
the proposed additional OBD requirements, as compared to our current OBD requirements, for
under 14,000 pound applications and do not represent the total costs for under 14,000 pound
OBD.
40
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 19. OBD Certification and Production Evaluation Testing Costs - Diesel Applications Under 14,000 Pounds
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Demonstration Testing Related
# of parent test
engines
-
-
9
6
-
-
6
6
-
6
-
6
-
6
-
6
6
-
Costs for Limit
Parts
$
$
$
$ 41,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 28,000
$
$
$ 173,000
$ 107,000
DDV Testing
Costs
$
$
$
$ 655,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 437,000
$
$
$ 2,709,000
$ 1,689,000
Total DDV
Costs
$
$
$
$ 696,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 2,882,000
$ 1,797,000
Production Evaluation Testing Related
PE Testing - Scan Tool
# of engine
families for
testing
-
-
-
-
-
-
-
-
-
-
-
-
-
-
PE Costs
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
PE Testing - Monitors
# of OBD
Groups tested
-
-
-
-
-
-
-
-
-
-
-
-
-
-
PE Costs
(incl vehicle
rental)
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
PE Testing - Ratios
# of monitoring
groups tested
-
-
-
-
-
-
-
-
-
-
-
-
-
-
PE Costs
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
PE Costs -
Total
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
Total
Certification &
PE Testing
Costs
$
$
$
$ 696,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 2,882,000
$ 1,797,000
41
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 20. Total Estimated OBD Costs - Diesel Applications Under 14,000 Pounds
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
CY
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
R&D
$ 2,455,500
$ 2,455,500
$ 2,455,500
$ 2,662,500
$ 207,000
$ 207,000
$ 207,000
$
$
$
$
$
$
$
$
-
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$ 9,831,000
$ 8,890,000
Cert/PE Testing
$
$
$
$ 696,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 2,882,000
$ 1,797,000
Hardware
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
-
-
-
-
Total
$ 2,455,500
$ 2,455,500
$ 2,455,500
$ 3,358,500
$ 207,000
$ 207,000
$ 672,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 465,000
$
$
$ 12,714,000
$ 10,686,000
Projected
Sales
570,000
581,400
592,800
604,200
615,600
627,000
638,400
649,800
661,200
672,600
684,000
695,400
706,800
718,200
729,600
741 ,000
752,400
763,800
775,200
786,600
798,000
809,400
820,800
832,200
843,600
855,000
866,400
877,800
889,200
900,600
$/vehicle
$ 4
$ 4
$ 4
$ 6
$ 0
$ 0
$ 1
-
$
$ 1
$
$
$ 1
$
$
$-\
I
$
$
$ 1
$
$
$ 1
$
$
$ 1
$
$
$ 1
$
$
3.3. Updated 2007/2010 HD Highway Costs Including OBD
Table 21 shows the cost estimates for the 2007/2010 heavy-duty highway program. As
shown, the 30 year net present value cost at a three percent discount rate was estimated at $70
billion with $25 billion of that being engine related costs.
42
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 21. Costs of the 2007/2010 Heavy-duty Highway Program*
(All Costs in SMillions; 1999 Dollars)
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
Calendar
Year
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Diesel Engines
HD2007 FRM
$ (80)
$ 1 ,266
$ 1 ,321
$ 1 ,072
$ 1 ,520
$ 1 ,225
$ 1,133
$ 1,157
$ 1,180
$ 1,141
$ 1,156
$ 1,159
$ 1,182
$ 1 ,205
$ 1 ,226
$ 1 ,247
$ 1 ,268
$ 1 ,288
$ 1 ,307
$ 1 ,326
$ 1 ,344
$ 1 ,362
$ 1 ,380
$ 1 ,398
$ 1,415
$ 1 ,432
$ 1 ,450
$ 1 ,467
$ 1 ,484
$ 1 ,500
$ 23,721
$ 14,369
Gasoline
Vehicles &
Engines HD2007
FRM
$
$
$ 46
$ 80
$ 81
$ 82
$ 83
$ 78
$ 79
$ 80
$ 82
$ 83
$ 84
$ 85
$ 86
$ 87
$ 89
$ 90
$ 91
$ 92
$ 93
$ 94
$ 95
$ 97
$ 98
$ 99
$ 100
$ 101
$ 102
$ 104
$ 1,514
$ 877
Diesel Fuel
$ 880
$ 1 ,786
$ 1 ,809
$ 1 ,904
$ 2,014
$ 2,128
$ 2,160
$ 2,192
$ 2,225
$ 2,258
$ 2,292
$ 2,327
$ 2,362
$ 2,397
$ 2,433
$ 2,469
$ 2,506
$ 2,544
$ 2,582
$ 2,621
$ 2,660
$ 2,700
$ 2,741
$ 2,782
$ 2,824
$ 2,866
$ 2,909
$ 2,953
$ 2,997
$ 3,042
$ 45,191
$ 26,957
Total Costs -
Engines, Fuel
$ 799
$ 3,052
$ 3,177
$ 3,056
$ 3,615
$ 3,434
$ 3,376
$ 3,427
$ 3,484
$ 3,480
$ 3,530
$ 3,568
$ 3,628
$ 3,687
$ 3,746
$ 3,804
$ 3,863
$ 3,921
$ 3,980
$ 4,039
$ 4,098
$ 4,157
$ 4,217
$ 4,276
$ 4,337
$ 4,397
$ 4,459
$ 4,521
$ 4,583
$ 4,646
$ 70,427
$ 42,203
EPA420-R-00-026; Table V.D-1 & Appendix VI-B; December 2000.
As shown in Table 14 (OBD costs for over 14,000 pounds) and Table 20 (OBD costs for
under 14,000 pounds), the 2007/2010 program costs far outweigh the OBD related costs of $1
billion and $13 million, respectively. The updated 2007/2010 program costs are shown in Table
22. Note that the 2007/2010 program costs were generated using 1999 dollars. Normally, we
would adjust 1999 dollars to 2004 dollars to make all costs consistent. However, we consulted
the Producer Price Index (PPI) for "Motor vehicle parts manufacturing-new exhaust system
parts" developed by the Bureau of Labor Statistics and found that the annual PPI adjustment for
such parts had actually decreased from 1999 to 2004.20 The PPI data are shown in Table 23.
This suggests that the cost to produce exhaust system parts has decreased since 1999 (note that
the preliminary data for 2005 suggest that the PPI adjustment for 2005 will be roughly equal to
that for 1999). For clarity, rather than adjusting downward the 2007/2010 program costs from
1999 dollars, or adjusting upward the OBD costs from 2004 dollars, we have chosen to present
the 2007/2010 costs as they were presented in that final rule alongside the OBD costs as
43
-------
DRAFT Technical Support Document; HDOBD NPRM
presented in sections 2 and 3 of this report. In short, the costs shown in Table 22 ignore the PPI
effect because it is, essentially, negligible over the timeframe of consideration.
Table 22. Updated 2007/2010 Program Costs Including New OBD-Related Costs
(All costs in SMillions)
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NPV@
NPV@
Calendar
Year
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
3%
7%
Diesel Engines
HD2007 FRM
$ (80)
$ 1 ,266
$ 1,321
$ 1 ,072
$ 1 ,520
$ 1 ,225
$ 1,133
$ 1,157
$ 1,180
$ 1,141
$ 1,156
$ 1,159
$ 1,182
$ 1 ,205
$ 1 ,226
$ 1 ,247
$ 1 ,268
$ 1 ,288
$ 1 ,307
$ 1 ,326
$ 1 ,344
$ 1 ,362
$ 1 ,380
$ 1 ,398
$ 1,415
$ 1 ,432
$ 1 ,450
$ 1 ,467
$ 1 ,484
$ 1,500
$ 23,721
$ 14,369
Diesel Engines
>14KOBD
$ 23.8
$ 23.8
$ 23.8
$ 76.1
$ 65.0
$ 65.1
$ 67.4
$ 38.0
$ 38.2
$ 41.9
$ 40.0
$ 40.2
$ 43.8
$ 41.9
$ 42.1
$ 45.7
$ 43.8
$ 44.0
$ 47.6
$ 45.7
$ 45.9
$ 49.5
$ 47.6
$ 47.8
$ 51.4
$ 49.5
$ 49.7
$ 53.3
$ 51.4
$ 51.6
$ 900
$ 560
Diesel
Applications
<14KOBD*
$ 2.5
$ 2.5
$ 2.5
$ 3.4
$ 0.2
$ 0.2
$ 0.7
$
$
$ 0.5
$
$
$ 0.5
$
$
$ 0.5
$
$
$ 0.5
$
$
$ 0.5
$
$
$ 0.5
$
$
$ 0.5
$
$
$ 13
$ 11
Gasoline
Vehicles &
Engines
HD2007 FRM
$
$
$ 46
$ 80
$ 81
$ 82
$ 83
$ 78
$ 79
$ 80
$ 82
$ 83
$ 84
$ 85
$ 86
$ 87
$ 89
$ 90
$ 91
$ 92
$ 93
$ 94
$ 95
$ 97
$ 98
$ 99
$ 100
$ 101
$ 102
$ 104
$ 1,514
$ 877
Gasoline
Engines
>14KOBD
$ 1.9
$ 1.9
$ 1.9
$ 2.8
$ 2.0
$ 2.0
$ 2.0
$ 2.8
$ 2.9
$ 2.9
$ 3.0
$ 3.0
$ 3.1
$ 3.2
$ 3.2
$ 3.2
$ 3.3
$ 3.3
$ 3.4
$ 3.4
$ 3.4
$ 3.5
$ 3.6
$ 3.6
$ 3.6
$ 3.7
$ 3.7
$ 3.8
$ 3.9
$ 3.9
$ 57
$ 34
Diesel Fuel
$ 880
$ 1 ,786
$ 1 ,809
$ 1 ,904
$ 2,014
$ 2,128
$ 2,160
$ 2,192
$ 2,225
$ 2,258
$ 2,292
$ 2,327
$ 2,362
$ 2,397
$ 2,433
$ 2,469
$ 2,506
$ 2,544
$ 2,582
$ 2,621
$ 2,660
$ 2,700
$ 2,741
$ 2,782
$ 2,824
$ 2,866
$ 2,909
$ 2,953
$ 2,997
$ 3,042
$ 45,191
$ 26,957
Total Costs -
Engines, OBD, Fuel
$ 828
$ 3,080
$ 3,204
$ 3,138
$ 3,682
$ 3,502
$ 3,446
$ 3,468
$ 3,525
$ 3,524
$ 3,573
$ 3,612
$ 3,675
$ 3,732
$ 3,790
$ 3,852
$ 3,910
$ 3,969
$ 4,031
$ 4,088
$ 4,146
$ 4,209
$ 4,267
$ 4,328
$ 4,393
$ 4,450
$ 4,512
$ 4,579
$ 4,638
$ 4,701
$ 71 ,395
$ 42,807
* Note that the 2007/2010 final rule did not apply to <8,500 pound applications.
44
-------
DRAFT Technical Support Document; HDOBD NPRM
Table 23. Producer Price Index Data for Motor Vehicle Exhaust System Parts*
Series Id: PCU3363993363993
Industry: All other motor vehicle parts mfg
Product: Exhaust system parts, new
Base Date: 8812
Year
1995
1996
1997
1998
1999
2000
2001
2002
Jan
111.9
116
117.5
119.8
119
118.6
117.2
119.9
2003 | 116.1
2004 | 115.9
2005 | 116.4
Feb
111.8
116.5
119.1
119.7
119
118.6
123.6
119.6
116.1
116.2
116.4
Mar
113.5
116.5
119.1
119.7
119
118.6
123.5
119.6
116.1
116.2
116.4
Apr
113.5
116.5
118.6
119.7
119
118.4
122.7
116.5
115.9
116.1
116.4
May
113.7
116.4
118.6
119.7
119
118.1
122.7
116.1
115.9
116.1
118.2
Jun
113.7
117.1
118.6
119.2
118.9
118.2
122.7
116.1
115.9
116.4
118.2
Jul
113.7
117.1
118.4
119
118.9
118.2
121
116.1
115.9
116.4
116.4(P)
Aug
113.7
117.1
118.4
119
118.9
118.3
120.1
115.7
115.9
116.4
116.4(P)
Sep
113.5
117.1
118.3
119
118.8
118.3
120.1
115.7
115.9
116.4
118. 2(P)
Oct
113.5
117.1
118.2
119
118.8
117.1
119.9
115.9
115.9
116.4
118. 2(P)
Nov
116
117.5
118.1
119
118.8
117.1
119.9
116.1
115.9
116.4
Dec
116
117.5
118.1
118.7
118.6
117.2
119.9
116.1
115.9
116.4
Annual
113.7
116.9
118.4
119.3
118.9
118.1
121.1
116.9
116
116.3
|P : Preliminary. All indexes are subject to revision four months after original publication.
See www.bls.gov/ppi.
45
-------
DRAFT Technical Support Document; HDOBD NPRM
References
1 Docket ID Number EPA-HQ-OAR-2005-0047 which can be found at www.regulations.gov.
2 Dueker, H.; Friese, K.; Haecker, D. Ceramic Aspects of the Bosch Lambda-Sensor. Soc.
Automot. Eng. Tech. Pap. Ser. 1975, No. 750223.
3 Hamann, E.; Manger, H.; Steinke, L. Lambda-Sensor with Y2O3-Stabilized ZrtVCeramic for
Application in Automotive Emission Control Systems. Soc. Automot. Eng. Tech. Pap. Ser. 1977,
No. 770401.
4 Wiedenmann, H.; Raff, L.; Noack, R. Heated Zirconia Oxygen Sensor for Stoichiometric and
Lean Air-Fuel Ratios. Soc. Automot. Eng. Tech. Pap. Ser. 1984, No. 840141.
5 Soejima, S.; Mase, S. Multi-Layered Zirconia Oxygen Sensor for Lean Burn Engine
Application. Soc. Automot. Eng. Tech. Pap. Ser. 1985, No. 850378.
6 Suzuki, S.; Sasayama, T.; Miki, M.; Ohsuga, M.; Tanaka, S.; Ueno, S.; Ichikawa, N. Air-Fuel
Ratio Sensor for Rich, Stoichiometric and Lean Ranges. Soc. Automot. Eng. Tech. Pap. Ser.
1986, No. 860408.
7 Ueno, S.; Ichikawa, N.; Suzuki, S.; Terakado, K. Wide-Range Air-Fuel Ratio Sensor. Soc.
Automot. Eng. Tech. Pap. Ser. 1986, No. 860409.
8 Yamada, T.; Hayakawa, N.; Kami, Y.; Kawai, T. Universal Air-Fuel Ratio Heated Exhaust Gas
Oxygen Sensor and Further Applications. Soc. Automot. Eng. Tech. Pap. Ser. 1992, No. 920234.
9 Walsh, R. Technical Trends, Usage and Performance of Smart NOX Sensors. Presented at EPA
MECA Meeting, Ann Arbor, MI, November 16, 2004.
10 Orban, J.; Wendt, D. Long-Term Aging of NOX Sensors in Heavy-Duty Engine Exhaust.
Presented at the 10th Diesel Engine Emissions Reduction Conference, Coronado, CA, September
1, 2004.
11 Kato, N.; Hamada, Y.; Kurachi, H. Performance of Thick Film NOX sensor on Diesel and
Gasoline Engines. Soc. Automot. Eng. Tech. Pap. Ser. 1997, No. 970858.
12 Schar, C.; Onder, C.; Geering, H. Control of a Urea SCR Catalytic Converter System for a
Mobile Heavy Duty Diesel Engine. Soc. Automot. Eng. Tech. Pap. Ser. 2003, No. 2003-01-0776.
13 Kato, N.; Kokune, N.; Lemire, B.; Walde, T. Long Term Stable NOX Sensor with Integrated
In-Connector Control Electronics. Soc. Automot. Eng. Tech. Pap. Ser. 1999, No. 1999-01-0202.
14 Orban, J.; Wendt, D. Long-Term Aging of NOX Sensors in Heavy-Duty Engine Exhaust.
Presented at the 10th Diesel Engine Emissions Reduction Conference, Coronado, CA, September
1, 2004.
15 Schenk, C.; McDonald, J.; Laroo, C. High-Efficiency NOX and PM Exhaust Emission Control
for Heavy-Duty On-Highway Diesel Engines - Part Two. SAE Tech. Pap. Ser. 2001, No. 2001-
01-3619.
16 May, M.; Adelman, B. APBF-DEC Heavy Duty NOX Adsorber/DPF Project: Heavy-Duty
Linehaul Platform Final Project Meeting. Presented at the Heavy-Duty Final Project Meeting,
Rosemont, IL, April 27, 2004.
17 66 FR 5002 and 69 FR 38958, respectively.
18 Board approved as of July 21, 2005 (see 13 CCR 1971.1).
46
-------
DRAFT Technical Support Document; HDOBD NPRM
19 "Final Regulatory Analysis: Control of Emissions from Nonroad Diesel Engines," EPA420-R-
04-007, May 2004.
on
See www.bls.gov/ppi: All other motor vehicle parts mfg; Exhaust system parts, new; series ID
PCU3363993363993; Base date 8812.
47
------- |