Engine/Transmission (1998.5 - 2002) Inside the ECU

[Gary - K7GLD]

Neither, your truck has too much base timing. As does mine and several others on here. Search my post history, I installed a pigtail between the IAT sensor and its connector so I could splice a switch and potentiometer into the circuit without damaging the stock harness. When it gets cold and starts acting up I flip the switch and pull out timing by altering the IAT signal with the potentiometer. Switch off = stock, Switch on = potentiometer in parallel w/ IAT sensor. Works great. Start another thread or send me a PM if you need more info.

Thanks for the helpful info - I know a fair number of others have the same issue, especially noticeable and annoying when towing or otherwise loaded. Yeah, I'll start another thread on the issue, but also wonder if you know of others who had the same problem, and totally eliminated it by your method?

 

[ERuhl]

Yeah, I'll start another thread on the issue, but also wonder if you know of others who had the same problem, and totally eliminated it by your method?

Yes, search for posts by user "timbo" as well. He and I sorted this out together.

But, what I had been referring to, was which analog input pin on the CPU chip itself each input was connected to. I have since figured that out, with the exception of the coolant temp sensor. Most of my spare time these days has been spent picking through the program code to see what makes it tick.

I wasn't sure which pins you were referring to, glad it was still of some use. There's a similar pinout diagram for the PCM, let me know if you need those posted as well.

Have you scanned the data yet to identify any likely map regions? I found a tool that graphically displays the data in a binary file (2D and 3D) which can be handy for quickly identifying map regions. It was made for an 8 bit data file, but perhaps something similar exists for this 32 bit file? I run an emulator in two of my vehicles (one w/ GM TBI, the other a Nissan Z32) and the ability to make changes on the fly has been quite rewarding. I'd love to be able to tweak this one as well. Best of luck with it, I'm anxiously following your progress

 

[Jdonoghue]

Have you scanned the data yet to identify any likely map regions? I found a tool that graphically displays the data in a binary file (2D and 3D) which can be handy for quickly identifying map regions. It was made for an 8 bit data file, but perhaps something similar exists for this 32 bit file? I run an emulator in two of my vehicles (one w/ GM TBI, the other a Nissan Z32) and the ability to make changes on the fly has been quite rewarding. I'd love to be able to tweak this one as well. Best of luck with it, I'm anxiously following your progress

I identified some areas that looked interesting. I spent some time this past weekend going over the disassembled code, and I was able to identify most of the maps by which sensor inputs are used to index into them. I think that I have found the place where it sends the data to the VP44. The subroutine that generates the data that is sent has an area that looks like it is used for the 3-cylinder idle, and the 'bitmap' can be changed to disable cylinders (for diagnostic purposes?).



Tonight I am going to gather my notes and post an update. I've found a lot of interesting stuff!

 

[Spunkys Ram]

THIS THREAD ROCKS!!! I, as most other people here really don't have a clue what does what in the ECM/PCM, but I have already marked your ECM page as my top favorite and will be following closly along with any changes/info made. I know I probably can't be much/if any help 'reading' code but let me know if there is anything I can do for you, research, test, etc.

It's really great that someone is taking the time and SHARING the info with us 'other' people. THANKS AGIAN!!!!!

 

[Jdonoghue]

vp44

Well, I think I found the place where the data is sent to the vp44. But since I can find no reference information, all I can do is speculate at this point.



There appears to be 4 data items that get sent to the vp44. Each is a 2-byte value. These get sent in a single CAN message (max. message length of a CAN message is 8 bytes). As of right now, I do not know what the PGN or address is - but this is what I think the data consists of, so far:



Bytes 1-2: Range: 0x0000 - 0x0ff7 (Fuel rate)

Bytes 3-4: Range: 0x0000 - 0x0582 (base timing adjustment?? seems to be set to 0, from a constant in the calibration table

bytes 5-6: Range: 0x1de2 - 0xfd2a (timing command)

bytes 7-8: Range: 0x0000 - 0x658e (I do not know what this could be)



Well, at least that's what I think those bytes mean. I am probably going to have to get a CAN interface for the PC, and set up a sniffing operation.



It looks like this information is not 'pushed' to the vp44, but the vp44 sends a request message to the ECU, and the ECU responds with this information.



Also, I read somewhere that the vp44 can send fuel temperature back to the ECU.

 

[Gary - K7GLD]

Since the VP-44 is capable of dual injection events, perhaps bytes 7 and 8 control that function when it's used? At least, That's how some of the add-on boxes get improved performance from the VP-44...

 

[Jdonoghue]

More info

The 'fuel injection pump sync signal' is apparently a signal FROM the computer TO the vp44, when the engine is at TDC. It is used by the vp44 to compensate for mechanical inaccuracies in timing.



On to more interesting things:



There are 3 data buses in/out of the ECU:

The CAN bus, which goes to the vp44, and the 'J1939' or 'Cummins Bus' connector that hangs out by the lift pump.

The CCD bus, which connects all things in the truck like the PCM, dashboard, overhead console, etc.

The bus that's labelled 'SCI' or 'ISO9141'.



What is necessary is to figure out which of these can be used to reflash the ECU. Smarty plugs into the OBDII connector, so it must either use the CCD bus or the SCI. Interesting to note that there are 2 separate SCI pins on the diesel - one goes to the PCM, the other to the Cummins ECU.



What would be even more interesting, is if there are functions available to read/write memory while the ECU is in operation. This would make it possible to adjust things 'on the fly' before committing them to a permanent update.

 

[mhenon]

Smarty plugs into the OBDII connector, so it must either use the CCD bus or the SCI. Interesting to note that there are 2 separate SCI pins on the diesel - one goes to the PCM, the other... .

Umm... Marco... now take it easy. The cat isn't completely out of the bag... yet!

Mike

 

[Michael Hughes]

Also read somewhere that the vp44 can send fuel temperature back to the ECU.

I can tell you that with a SnapOn scanner, the fuel temperature is one of the values you can display.

 

[mhenon]

I can tell you that with a SnapOn scanner, the fuel temperature is one of the values you can display.

Didn't know that. My Scangauge doesn't read fuel temp but would love to have that feature.

Mike

 

[rweis]

ANY idea where the fuel temp is read?



I have been working fuel temperatures for about a year now. I have a sensor in the input fuel flow line, I had a sensor on the output fuel flow back to the tank, and I had a sensor on the ECB (Electronic Bay Cover).



Those gave me good data relationships on what seems to raise the input fuel temperature based on a rpm and load relationship. Then based on that data I installed fuel coolers, and replaced and rerouted the return fuel line to the tank vent line.



I would really like to be able to read an internal fuel temperature off the bus. I wonder if the bus fuel temperature is before or after the intermediate vane pump or before or after the rotary high presure pump.



Cranking the pressure up to injection pressures seems to input about 10* to the input fuel temp. There is a 1 to 1 relationship of rpm to increases in output fuel temp. I bet Gary (and lost the bet) that the VP44 would not know if there was a load or not (why would it care? it pumps the same amount of fuel regardless), but more load also increases output fuel temp.



If anyone in Florida has a scanner that reads VP44 fuel temp I would be happy to drive to you to be able to use it one afternoon to see how it plays with the other 3 fuel temperature readings.



Bob Weis

 

[mhenon]

I've just sent an e-mail off to ScanGauge. They're doing free up-grades to previously purchased units and I asked them about reading fuel temps.

There's software for my VW Jetta (Vag-Com) that also reads fuel temps but again the ScanGauge doesn't display this. Maybe we can get them to add this feature if enough interest is shown.

Mike

 

[Gary - K7GLD]

ANY idea where the fuel temp is read?



I have been working fuel temperatures for about a year now. I have a sensor in the input fuel flow line, I had a sensor on the output fuel flow back to the tank, and I had a sensor on the ECB (Electronic Bay Cover).



Those gave me good data relationships on what seems to raise the input fuel temperature based on a rpm and load relationship. Then based on that data I installed fuel coolers, and replaced and rerouted the return fuel line to the tank vent line.



I would really like to be able to read an internal fuel temperature off the bus. I wonder if the bus fuel temperature is before or after the intermediate vane pump or before or after the rotary high presure pump.



Cranking the pressure up to injection pressures seems to input about 10* to the input fuel temp. There is a 1 to 1 relationship of rpm to increases in output fuel temp. I bet Gary (and lost the bet) that the VP44 would not know if there was a load or not (why would it care? it pumps the same amount of fuel regardless), but more load also increases output fuel temp.



If anyone in Florida has a scanner that reads VP44 fuel temp I would be happy to drive to you to be able to use it one afternoon to see how it plays with the other 3 fuel temperature readings.



Bob Weis

Bob, after eliminating the several internal VP-44 sensors that don't APPEAR to be temperature related, I'd bet that the temp reading for fuel comes from a sensor source mounted directly on the VP-44 electronics bay board itself - after all, THAT temp location is probably the most important...

 

[mhenon]

The info's here: VP44 Electronic Inj Pump

"A fuel temperature sensor is located inside the VP44 pump to compensate for changes in the temperature of the fuel. The pump also contains... ... "

The info must be going to the OBDII connector as some of the scanners can read it. I would think any scanner "could" pick up the voltage signal and translate that to a temp reading. Just has to be designed into the scanner.

Mike

 

[Gary - K7GLD]

The info's here: VP44 Electronic Inj Pump

"A fuel temperature sensor is located inside the VP44 pump to compensate for changes in the temperature of the fuel. The pump also contains... ... "

The info must be going to the OBDII connector as some of the scanners can read it. I would think any scanner "could" pick up the voltage signal and translate that to a temp reading. Just has to be designed into the scanner.

Mike

YUP - but I think Bob's question is exactly WHERE - at what POINT - is that VP-44 internal fuel temp read and reported - there are multiple points inside the VP-44 where temperature rises are introduced, so a temperature reading at the fuel input point usually differs substantially from temperatures at other points in the fuel flow.



As I recall, there are only *3* possible temp sensing points, the board itself - another is at the point where the VP-44 control board has an attached internal sensor that picks up timing impulses from the rotating pump plunger assembly - and the only other one is the leads to the fuel solenoids that control and meters fuel delivery to the injectors. The temp sensor COULD be included in one of those, but I doubt it, since each of those 2 solenoid leads only has 2 leads (as I recall) - not enough for the additional temp sensing function in addition to solenoid activation...

 

[Jdonoghue]

I believe the temperature sensor is right on the circuit board. Somewhere else I read that the sensor in the pump that is connected by the thin flexible wires contains 2 coils and nothing else.

 

[Gary - K7GLD]

I believe the temperature sensor is right on the circuit board. Somewhere else I read that the sensor in the pump that is connected by the thin flexible wires contains 2 coils and nothing else.

2 pairs of wires run partially externally on the VP-44, down to the distribution area where the injector lines connect - those pairs each connect to an internal fuel delivery/control solenoid. One of those wire pairs is also the one famous as the external connecting point for add-on power boxes such as the Edge Comp.



I believe the actual internal sensor you refer to is likely a simple inductive coil rotation sensor pickup used to sense internal position and timing of the rotary pump plunger assembly...



But I too feel the actual VP-44 fuel temp sensor is most likely mounted directly to the internal main circuit board.

 

[Wiredawg]

Have you seen the Cummins ECM trouble shooting manual? It has detailed block diagrams of the ECM, operating processes and trouble shooting for all the components controlled by it. A Cummins mechnic recommend it to me.



If interested, I can try to scan and post it.



Wiredawg

 

[Jdonoghue]

Have you seen the Cummins ECM trouble shooting manual? It has detailed block diagrams of the ECM, operating processes and trouble shooting for all the components controlled by it. A Cummins mechnic recommend it to me.



If interested, I can try to scan and post it.



Wiredawg

No, I have not seen it. That would be a very valuable thing to have - I wasn't aware such a thing existed. I would be very interested in seeing what's in there.

 

[rweis]

I would think the fuel temp sensor is somewhere in the high pressure section and somewhere around the solenoid since the timing and volume delivery of the fuel to the injector distribution section occurs there.



WE are interested in fuel temp to the PSG (Pump Supply Group) electronics in our attempts to keep the electronics cool. However, it is after the pressure increase in the high pressure section that the temperature rise happens I would think and that would affect how dense the fuel is and how long the VP44 solenoid maintains the high pressure charge to the distribution section before collapsing the pressure charge.



I would be interested in the Cummins ECM troubleshooting manual.



Knowledge IS power!



GREAT discussion!



Bob Weis