944 OBD Project (On-Board Diagnostics)
#481
Drifting
Looking forward to seeing how things are progressing. Dynoed my 944 2.8 L stroker engine. Maximum HP was 149 @ 5500 rpm with A/F ratio 12.6. Using Rogue NA Tune with 2.8 L Rogue chip. Would have been nice to boost HP with spark advance. Installed new Lucas low impedance fuel injectors, 30 lb/hr, fuel pressure @ 20 psig. The Rogue chip must have increased the pulse width time for old style Bosch 24/26 lb/hr fuel injectors with the fuel rail pressure @ 36 psig.
#482
Addict
Rennlist
Lifetime Member
Rennlist
Lifetime Member
I updated to v0.25.270 yesterday and the 951 started and ran fine. I do wonder about the reading of the Barometric Air Pressure. Is that currently functional? It seems to be fixed at 14.5 since my weather station reported 30.05 in Hg today, thus my local pressure in the Seattle area is 14.76 psi (30.05/2.03602).
#484
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
I have been working diligently on MAF, especially in the last week. Despite some significant setbacks, everything is coming together nicely. I keep saying this, but I hope to be road testing on my 944 NA this or next week. Right now, I am waiting on some parts that will let me finish supporting the MAF's built in IAT (Intake Air Temperature) sensor. The BAP (Barometric Air Pressure) sensor is nearly fully implemented into the MAF code module. The code that back-calculates the MAF airflow in terms of volume is also implemented as is the timing system to continuously update the emulated AFM signal. Reading out MAF airflow in terms of density has been in the code for a while now, but I need to generate a new calibration table using the updated air-flow bench software. Once I have all of that, I can generate the calibration table for the AFM which is the last of the core missing pieces. None of this is difficult, I just need to keep hammering on it until it all works together.
-Joe
-Joe
__________________
http://www.ftech9.com/
http://www.ftech9.com/
The following 4 users liked this post by Ftech9:
#486
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
MAF Project Update
I was able to generate a calibration table for the MAF sensors built-in IAT sensor, which was the major goal for last week.
I am now working on the next phase which gets us very close to something tangible. For development, I have special MAF housing in series with the AFM so I can verify the calculated AFM signal matches the real AFM signal while driving under various conditions. This means I need to connect the MAF airflow signal to the OBD+ "MAP" connector and its IAT signal to the "AUX" connector. The airflow signal is easy, it is just 0V to 5V. However, to connect the IAT signal, I needed to design a small circuit board to drive it properly. When this circuit board arrives early next week, I can build the wiring harness and electrically connect everything to my 944 NA.
For the rest of this week, I will be setting up the flow-bench to generate calibration tables for both the AFM and development MAF sensor. Once that is done the next thing to do is update the firmware/software to display the AFM in terms of airflow volume (m3/s or CFM) instead of just raw voltage. As a bonus, I will also be able to calculate and display AFM airflow in terms of density (g/s or lb/min). In theory, all of these values from the AFM and MAF sensor should match reasonably well. If they do, I can start running the engine solely off the MAF sensor.
- Joe
I was able to generate a calibration table for the MAF sensors built-in IAT sensor, which was the major goal for last week.
I am now working on the next phase which gets us very close to something tangible. For development, I have special MAF housing in series with the AFM so I can verify the calculated AFM signal matches the real AFM signal while driving under various conditions. This means I need to connect the MAF airflow signal to the OBD+ "MAP" connector and its IAT signal to the "AUX" connector. The airflow signal is easy, it is just 0V to 5V. However, to connect the IAT signal, I needed to design a small circuit board to drive it properly. When this circuit board arrives early next week, I can build the wiring harness and electrically connect everything to my 944 NA.
For the rest of this week, I will be setting up the flow-bench to generate calibration tables for both the AFM and development MAF sensor. Once that is done the next thing to do is update the firmware/software to display the AFM in terms of airflow volume (m3/s or CFM) instead of just raw voltage. As a bonus, I will also be able to calculate and display AFM airflow in terms of density (g/s or lb/min). In theory, all of these values from the AFM and MAF sensor should match reasonably well. If they do, I can start running the engine solely off the MAF sensor.
- Joe
Last edited by Ftech9; 10-12-2022 at 08:55 PM.
The following 8 users liked this post by Ftech9:
fejjj (10-14-2022),
J1NX3D (10-12-2022),
Millermatic (10-13-2022),
riouxc (10-14-2022),
spencang (10-13-2022),
and 3 others liked this post.
#487
Addict
Rennlist
Lifetime Member
Rennlist
Lifetime Member
I've been using the OBD+ to diagnose an issue I'm having with the idle speed on my 951; I'm getting some weird results. I recently replaced the NBO2 sensor and now the engine idles in the 1200-1300 rpm range when warm. OBD+ indicates the ICV % Open is ~ 87.7, which seems high as I would think the ICV would move towards the closed position to slow the idle speed, but it doesn't respond at idle (it does respond towards closed then the throttle is not at idle). The other issue is the Throttle Switch Mode; with the throttle closed the mode is "Full Load," which also does not make sense. Advancing the throttle off the idle position changes the Mode to "Above Idle." Can someone with OBD+ installed confirm that the Mode should be "Idle" (or whatever the correct indication is) when the throttle is closed?
So, the DME apparently thinks the throttle is in the "Full Load" position (greater than 65 deg throttle position per Workshop Manual). In Full Load mode the DME ignores the input from the O2 sensor, and it would also not control the ICV to stabilize the idle speed.
Update: The correct mode for idle is "Idle." I just confirmed this by hooking up my laptop to the DME and without the engine running and the ignition ON, "Idle" was displayed for the mode. Moving the throttle to any other position changed the mode to "Full Load." However (!), turning the ignition OFF and ON again the idle position mode was now "Full Load" and the off-idle position was "Above Idle." I tried numerous times to get it back to "Idle" and "Full Load" but could not repeat the initial test.
This weird behavior explains another issue I've observed: when the engine is at idle and warmed up the NBO2 Digital signal should switch between RICH and LEAN, and the NBO2 voltage should increase or decrease accordingly as the DME dithers the fuel mixture about the stoichiometric line in response to the NBO2 signal. In my case, with my old NBO2 sensor this behavior never happened at idle, or at any other rpm, which is what prompted me to change the NBO2 sensor.
Now, with the new NBO2 sensor the digital and voltage output are as before at idle: Digital is always RICH and the voltage is high. But just increasing the throttle position ever so slightly to off-idle, the Digital and Voltage signals respond as expected.
So, typing all of this may have convinced me that my Throttle Switch may be faulty (last changed in 2002 from the records I have), or maybe an issue with the KLR. Wow, prices on the TPS certainly have changed. In 2002 it was $70 and now the Bosch version is $320 and genuine Porsche is $500+.
Any thoughts on this behavior are appreciated.
So, the DME apparently thinks the throttle is in the "Full Load" position (greater than 65 deg throttle position per Workshop Manual). In Full Load mode the DME ignores the input from the O2 sensor, and it would also not control the ICV to stabilize the idle speed.
Update: The correct mode for idle is "Idle." I just confirmed this by hooking up my laptop to the DME and without the engine running and the ignition ON, "Idle" was displayed for the mode. Moving the throttle to any other position changed the mode to "Full Load." However (!), turning the ignition OFF and ON again the idle position mode was now "Full Load" and the off-idle position was "Above Idle." I tried numerous times to get it back to "Idle" and "Full Load" but could not repeat the initial test.
This weird behavior explains another issue I've observed: when the engine is at idle and warmed up the NBO2 Digital signal should switch between RICH and LEAN, and the NBO2 voltage should increase or decrease accordingly as the DME dithers the fuel mixture about the stoichiometric line in response to the NBO2 signal. In my case, with my old NBO2 sensor this behavior never happened at idle, or at any other rpm, which is what prompted me to change the NBO2 sensor.
Now, with the new NBO2 sensor the digital and voltage output are as before at idle: Digital is always RICH and the voltage is high. But just increasing the throttle position ever so slightly to off-idle, the Digital and Voltage signals respond as expected.
So, typing all of this may have convinced me that my Throttle Switch may be faulty (last changed in 2002 from the records I have), or maybe an issue with the KLR. Wow, prices on the TPS certainly have changed. In 2002 it was $70 and now the Bosch version is $320 and genuine Porsche is $500+.
Any thoughts on this behavior are appreciated.
Last edited by Tom Pultz; 10-23-2022 at 06:05 PM. Reason: Add information
#488
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
I've been using the OBD+ to diagnose an issue I'm having with the idle speed on my 951; I'm getting some weird results. I recently replaced the NBO2 sensor and now the engine idles in the 1200-1300 rpm range when warm. OBD+ indicates the ICV % Open is ~ 87.7, which seems high as I would think the ICV would move towards the closed position to slow the idle speed, but it doesn't respond at idle (it does respond towards closed then the throttle is not at idle). The other issue is the Throttle Switch Mode; with the throttle closed the mode is "Full Load," which also does not make sense. Advancing the throttle off the idle position changes the Mode to "Above Idle." Can someone with OBD+ installed confirm that the Mode should be "Idle" (or whatever the correct indication is) when the throttle is closed?
So, the DME apparently thinks the throttle is in the "Full Load" position (greater than 65 deg throttle position per Workshop Manual). In Full Load mode the DME ignores the input from the O2 sensor, and it would also not control the ICV to stabilize the idle speed.
Update: The correct mode for idle is "Idle." I just confirmed this by hooking up my laptop to the DME and without the engine running and the ignition ON, "Idle" was displayed for the mode. Moving the throttle to any other position changed the mode to "Full Load." However (!), turning the ignition OFF and ON again the idle position mode was now "Full Load" and the off-idle position was "Above Idle." I tried numerous times to get it back to "Idle" and "Full Load" but could not repeat the initial test.
This weird behavior explains another issue I've observed: when the engine is at idle and warmed up the NBO2 Digital signal should switch between RICH and LEAN, and the NBO2 voltage should increase or decrease accordingly as the DME dithers the fuel mixture about the stoichiometric line in response to the NBO2 signal. In my case, with my old NBO2 sensor this behavior never happened at idle, or at any other rpm, which is what prompted me to change the NBO2 sensor.
Now, with the new NBO2 sensor the digital and voltage output are as before at idle: Digital is always RICH and the voltage is high. But just increasing the throttle position ever so slightly to off-idle, the Digital and Voltage signals respond as expected.
So, typing all of this may have convinced me that my Throttle Switch may be faulty (last changed in 2002 from the records I have), or maybe an issue with the KLR. Wow, prices on the TPS certainly have changed. In 2002 it was $70 and now the Bosch version is $320 and genuine Porsche is $500+.
Any thoughts on this behavior are appreciated.
So, the DME apparently thinks the throttle is in the "Full Load" position (greater than 65 deg throttle position per Workshop Manual). In Full Load mode the DME ignores the input from the O2 sensor, and it would also not control the ICV to stabilize the idle speed.
Update: The correct mode for idle is "Idle." I just confirmed this by hooking up my laptop to the DME and without the engine running and the ignition ON, "Idle" was displayed for the mode. Moving the throttle to any other position changed the mode to "Full Load." However (!), turning the ignition OFF and ON again the idle position mode was now "Full Load" and the off-idle position was "Above Idle." I tried numerous times to get it back to "Idle" and "Full Load" but could not repeat the initial test.
This weird behavior explains another issue I've observed: when the engine is at idle and warmed up the NBO2 Digital signal should switch between RICH and LEAN, and the NBO2 voltage should increase or decrease accordingly as the DME dithers the fuel mixture about the stoichiometric line in response to the NBO2 signal. In my case, with my old NBO2 sensor this behavior never happened at idle, or at any other rpm, which is what prompted me to change the NBO2 sensor.
Now, with the new NBO2 sensor the digital and voltage output are as before at idle: Digital is always RICH and the voltage is high. But just increasing the throttle position ever so slightly to off-idle, the Digital and Voltage signals respond as expected.
So, typing all of this may have convinced me that my Throttle Switch may be faulty (last changed in 2002 from the records I have), or maybe an issue with the KLR. Wow, prices on the TPS certainly have changed. In 2002 it was $70 and now the Bosch version is $320 and genuine Porsche is $500+.
Any thoughts on this behavior are appreciated.
The 944 Turbo has a combination TS (Throttle Switch) and TPS (Throttle Position Sensor). The TS is connected directly to the DME and provides the “Idel” and “Off Idel” signal. The TPS is connected to the KLR which provides the DME with a WOT (Wide Open Throttle) signal when it senses the throttle fully open. For the DME to enter “Full Load” mode, it must sense both “Off Idel” and “WOT” signals.
From what you describe, it is likely the idle contacts are not electrically making a connection when throttle is closed. If you have a DMM (Digital Multi-Meter) with a continuity function, I recommend verifying the TPS is defective by unplugging it and checking pins 4 and 6 on the TPS/TS side. These pins should have continuity (i.e., electrically shorted) when the throttle is closed.
Below is a link to a detailed write up of the 951 TPS and how to perform these measurements. He also shows you how top “pop” the housing off and potentially fix this issue.
951 TPS Repair : ArnnWorx Specialty Tools
-Joe Baumbach
Last edited by Ftech9; 10-23-2022 at 08:34 PM.
The following users liked this post:
riouxc (10-23-2022)
#489
Rennlist Member
You also make sure that when it's on idle, you hear the switch "click". Just play with it when everything is off. If not, you can adjust it.
http://www.clarks-garage.com/shop-manual/fuel-06.htm
http://www.clarks-garage.com/shop-manual/fuel-06.htm
#490
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
MAF Support Update
A quick update on the MAF project...
I finished building the IAT adapter board and wiring harness, I expect to start bench testing early this week after some software updates. Below is a picture of the development wiring harness connected to an OBD+ Sport DME and one of our direct-fit MAF sensors. This is just for develop so I can read the AFM and MAF signals simultaneously. Once I have a high level of confidence, I will remove the AFM and run only on the MAF sensor using the factory AFM connector via an adapter.
The small green circuit board adapts the IAT's variable resistance to a voltage that can be measured by the OBD+ Module. This is essentially the same circuit used in the factory Bosch DME as the analog front end to its ADC (Analog to Digital Converter). It also scales the voltage to a usable range and has a basic over-voltage protection feature.
I finished building the IAT adapter board and wiring harness, I expect to start bench testing early this week after some software updates. Below is a picture of the development wiring harness connected to an OBD+ Sport DME and one of our direct-fit MAF sensors. This is just for develop so I can read the AFM and MAF signals simultaneously. Once I have a high level of confidence, I will remove the AFM and run only on the MAF sensor using the factory AFM connector via an adapter.
The small green circuit board adapts the IAT's variable resistance to a voltage that can be measured by the OBD+ Module. This is essentially the same circuit used in the factory Bosch DME as the analog front end to its ADC (Analog to Digital Converter). It also scales the voltage to a usable range and has a basic over-voltage protection feature.
The following 6 users liked this post by Ftech9:
J1NX3D (10-24-2022),
Millermatic (10-25-2022),
riouxc (10-24-2022),
spencang (10-24-2022),
Tom Pultz (10-24-2022),
and 1 others liked this post.
#491
Addict
Rennlist
Lifetime Member
Rennlist
Lifetime Member
You also make sure that when it's on idle, you hear the switch "click". Just play with it when everything is off. If not, you can adjust it.
http://www.clarks-garage.com/shop-manual/fuel-06.htm
http://www.clarks-garage.com/shop-manual/fuel-06.htm
And thanks Joe for your input. However, I'm still confused how the OBD+ can indicate Full Load with the throttle at idle. Even if the closed throttle position microswitch inside the combo throttle switch is defective, how does the KLR get a WOT reading from the TPS part of the combo switch when the throttle is closed? Is the potentiometer section also messed up, so the KLR thinks the resistance is at or above the WOT setting?
#492
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
Thanks. Per the ArnnWorx link from Joe, even if you hear the click the switch can still be bad. I ordered a new Bosch 0 280 120 400 switch from an eBay seller ($255). Hard to believe these switches cost so much when you see what's inside.
And thanks Joe for your input. However, I'm still confused how the OBD+ can indicate Full Load with the throttle at idle. Even if the closed throttle position microswitch inside the combo throttle switch is defective, how does the KLR get a WOT reading from the TPS part of the combo switch when the throttle is closed? Is the potentiometer section also messed up, so the KLR thinks the resistance is at or above the WOT setting?
And thanks Joe for your input. However, I'm still confused how the OBD+ can indicate Full Load with the throttle at idle. Even if the closed throttle position microswitch inside the combo throttle switch is defective, how does the KLR get a WOT reading from the TPS part of the combo switch when the throttle is closed? Is the potentiometer section also messed up, so the KLR thinks the resistance is at or above the WOT setting?
-Joe
#493
Addict
Rennlist
Lifetime Member
Rennlist
Lifetime Member
FWIW, I did the "click test" today for the combo TS/TPS switch. I didn't hear a "click" when I slowly opened the throttle, but I did hear a "click" when I slowly closed the throttle just prior to being fully closed. I could try disconnecting the KLR and see what Mode I get in the idle position, but I don't think that will prove anything since with the KLR disconnected it should be impossible to see Full Load for the Mode (unless the OBD+ SW is faulty). I need to change the combo switch first and THEN if I'm still getting Full Load at idle that might point to a bad KLR.
#494
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Thread Starter
FWIW, I did the "click test" today for the combo TS/TPS switch. I didn't hear a "click" when I slowly opened the throttle, but I did hear a "click" when I slowly closed the throttle just prior to being fully closed. I could try disconnecting the KLR and see what Mode I get in the idle position, but I don't think that will prove anything since with the KLR disconnected it should be impossible to see Full Load for the Mode (unless the OBD+ SW is faulty). I need to change the combo switch first and THEN if I'm still getting Full Load at idle that might point to a bad KLR.
Sorry, I have been going over different fault paths in my head and have not done a good job communicating. It seems unlikely both the idle contacts and potentiometer would fail simultaneously. The test for disconnecting the KLR would be a step towards determining if this is the case and also verify the OBD+ is functioning correctly. Let me know if you would like to continue diagnosing this issue or wait until the replacement TPS/TS arrives. If you wish to continue diagnostics, I am happy to come up with a test plan to determine the root cause of this issue.
-Joe
Last edited by Ftech9; 10-24-2022 at 11:34 PM.
#495
Addict
Rennlist
Lifetime Member
Rennlist
Lifetime Member
Hi Tom,
Sorry, I have been going over different fault paths in my head and have not done a good job communicating. It seems unlikely both the idle contacts and potentiometer would fail simultaneously. The test for disconnecting the KLR would be a step towards determining if this is the case and also verify the OBD+ is functioning correctly. Let me know if you would like to continue diagnosing this issue or wait until the replacement TPS/TS arrives. If you wish to continue diagnostics, I am happy to come up with a test plan to determine the root cause of this issue.
-Joe
Sorry, I have been going over different fault paths in my head and have not done a good job communicating. It seems unlikely both the idle contacts and potentiometer would fail simultaneously. The test for disconnecting the KLR would be a step towards determining if this is the case and also verify the OBD+ is functioning correctly. Let me know if you would like to continue diagnosing this issue or wait until the replacement TPS/TS arrives. If you wish to continue diagnostics, I am happy to come up with a test plan to determine the root cause of this issue.
-Joe