99 Isuzu 3.5 anyone know about these ECUs?

[CyrodiilSavior]

To start, I'm new to this board so hopefully my project I'm embarking on will be welcome here and there will be folks here both interested in helping as well as knowledgeable enough to help me in the areas I'm lacking in.

As my day job I work as a software engineer. I primarily work in languages such as puppet, go-lang and Ruby and focus heavily in a Linux infrastructure-oriented development environment. I also know C/C++ and have spent a lot of time toying around with the arduino platform. This doesn't necessarily I"m ready to take on everything this project will most likely present me with but it shows I'm no stranger to computers and how they work. I also spent a few years as a professional automotive tech, I have been working on cars basically my whole life, am good with a wrench and am not afraid to get my hands dirty.

Now about this project....
I am a huge Isuzu enthusiast (yes we exist) I grew up in my dad's truck and now have the pleasure of owning my own. They're ultra reliable, easy to work on, and can take on just about anything off the road on the trails. This isn't to say they don't have their issues. Probably the #1 problem found on the late model Isuzu V6s (named 6VD1 3.2 and 6VE1 3.5) is their tendency to trigger CEL for a P0401 which is EGR flow insufficient. This is often times fixed by replacing the GM EGR valve, cleaning the intake side of the tube as it sometimes can get clogged with carbon or sometimes both. I want to try to establish a 3rd way of fixing this problem. And that being to delete the EGR system all-together. The benefits to this would include both lowing air intake temperatures and would lessen the amount of carbon and other junk being recirculated into the engine.

You can get an EGR delete plate for a pretty good price as this specific valve was very common among late 90s GM cars/trucks but doing this isn't enough to remove the problem as unplugging the EGR valve will throw a code of its own. This means I will have to get personal with my PCM. The PCM used on the 98/99 troopers can be found under the radio behind some trim. I originally was under the impression that despite all the delco electronics throughout the truck the ECU was not one of them. This is due to a complete lack of information on the PCM, and there being no established tools that can interface/program this PCM except for the TECH 2. This most likely means I will have to reverse engineer it myself. I've been doing a lot of reading on the topic and I think the best way to go about solving this problem would be to disassemble the PCM, locate the EEPROM that contains the map for the EGR system. import this EEPROM onto a computer, remap the EGR to be completely flat and therefore accept all ranges of flow when it runs its EGR test. And delete the DTC directly related to unplugging the EGR valve from the system. Simple enough right? LOL

I pulled my PCM the other day to start studying the internals and figuring out how it works (this PCM not be touched. I will be pulling a computer from a donor truck to experiment with and remove IC chips) What I found was somewhat surprising...

I see a whole lot of delco parts on this computer. Now whether or not this is a PCM manufactured by Delco for Isuzu still remains to be seem, all I know is its made in Singapore, and contains a few Delco chips. Perhaps someone can chime in on this.

Next I started to pull numbers and began identifying chips. For simplicity I have colorcoded what I have.

I have missed a few chips so I'll have to pull it apart again and get the ones I missed... But as it stands now What heres what I've found (start at top move clockwise):

LIGHT GREEN: Delco 27833
LIGHT ORANGE: Don't have numbers for these. need to pull apart again
RED: AMD Am29F010 this is some sort of flash memory, holds 1 Megabit of information, found a pdf on it here: http://noel.feld.cvut.cz/hw/amd/16736f1.pdf
BLUE: Some sort of Motorola chip, has masket E87J http://matthieu.benoit.free.fr/cross/co ... askset.htm
GREEN: DELCO 66285 can't find much more info on this. Just a bunch for sale
ORANGE: DELCO 40854 even less info on this one comared to the last.
BLUE: Another moto chip. masket E57S http://matthieu.benoit.free.fr/cross/co ... askset.htm
PURPLE: THAILAND 50610 given its structure I believe its some sort of EPROM. Can't get much more info online about it. No pinouts etc.
LIGHT BLUE: DELCO 51862 once again just a bunch for sale online, noticed they came in a few other ECUs saw one in a saturn and mitsubishi, weird.
YELLOW: Don't have clear numbers on these 2 have to take it apart again
PURPLE: Same as the other THAILAND 50610
PINK: M83AD can't find anything on this
LIGHT PURPLE: 16132083 don't know the manufacturer for this one. found a couple for sale online

So at this point I think my best leads are the AMD flash memory and the 2 Motorola processors. If anybody can help get me more info about anything or what I should do next anything is appreciated!

Thanks!
Garrett

[Charlescrown]

I have a 2016 MUX and am well aware of the EGR problem. Just blanking it won't work with the CEL coming on. You can turn them off in the program but not sure if that's a good idea. I operated one with my Tech2 and found at idle when I closed it off the diesel knock got very loud so figured they do more than reduce NOx but also provide a smooth quiet idle.

[CyrodiilSavior]

Its important to keep in mind, I'm working on a 3.5l 6VE1 which is a naturally aspirated Gasoline motor.
I'm totally jealous of your 2016 MUX with the 4jj1 diesel, but totally different setup compared to mine as far as I know

[delcowizzid]

Is the thing with push on it a removable memcal with an eprom in it

[CyrodiilSavior]

Here's what's under the plastic cover. Some sort of IO. Not sure what this does, is it a GM specific thing? I saw a couple others online with this port.

[delcowizzid]

If the diagnostic plug has pins 4 5 6 9 13 and 16 populated it probably talks 8192 baud aldl .the am29f010 should hold the whole calibration if you can get it off the board without cooking it you should be able to put it in a soic to dip adaptor and read it with a chip programmer.its probably possible to read it with an aldl cable in the diagnostic port if we knew the memory mapping and required commands

[CyrodiilSavior]

So I looked through the FSM, I didn't see anything that specifies the protocol used but given some online information I found + the fact that the factory programming /dealer scan tool is a TECH 2. I'm going with yes it uses a JPW J1850 protocol.

In terms of next steps to be taking I planned on grabbing a donor PCM from the junkyard to experiment on. Once I pull the chip off the board my next step is reading/programming it... The only thing I can find with clear support for this chip is http://www.ezoflash.com

Does anyone have experience with this reader/software? It seems open and has ability to both read and flash calibrations to the chip in question.

Would anyone recommend using something different?

[j_ds_au]

The logo on the violet (you've called it light purple) chip is Allegro, who do interfacing stuff.

Also note you've misinterpreted delcowizzid's post. 8192 baud ALDL is not J1850.

Joe.

PS. This 2015 thread seems to give lots of clues on a similar ECU :
http://www.gearhead-efi.com/Fuel-Inject ... 196397-yet

[delcowizzid]

So I found out what the thing is that says push on it its the knock filter I just stumbled apon the answer in my travels lol

[Maxzillian]

I thought about starting my own thread due to the age of this one, but... best not to clutter things up.

So I have a 2000 Vehicross which uses this PCM. In pursuit of being able to pull the bin off of it I procured the same series of PCM from a 98 Rodeo and socketed the flash memory so I could read out the binary (attached). This a 1Mbit binary file that is pretty similar to the Holden VX/VY pcms, but not quite the same. It also appears to have an EEPROM to store things like the VIN because I could not find the string in the flash dump.

The OBD plug is wired with a VPW and ALDL interface as well as a "PCM diagnostics enable" wire. Curiously I can't get so much as a peep out of the ALDL interface and that's with the PCM diagnostic wire open, grounded, 10k resistor, etc. Presently I have a script running that's hammering it with every command from 0x00 to 0xFFFFF, but I have doubts it'll yield anything.

Meanwhile the VPW interface has been promising. After doing some more script hammering I managed to figure out that the Seed/Key algorithm is 0x1C and when manually using the key in PCM hammer I did manage to unlock the PCM. However, it presently defaults to loading a P01 kernel and the PCM is rejecting the default addresses used. Just a general properties query yields:

98 Rodeo:
Calibration ID: 9393339
Serial: CKMA7828007
Broadcast code: DDLY
00 Vehicross:
Calibration ID: 9375599
Serial: CXWP0539326
Broadcast code: CXWP

Both respond to a request for OSID, but with what appears to be a NACK response; I don't presently have that exact response available to share.

So I'm kinda scratching my head at where to go from here. It seems like PCM Hammer would be the way to go, but I'm a novice at this so I have some questions. I assume the purpose of the kernel upload is to push some code to the PCM "ram" space that will serve as a download service to dump the binary? This, so far, has been the norm for any Delco VPW PCM?

Is it unusual for a PCM to be wired for ALDL, but not respond to anything? I'd be keen to blame my DIY interface, but I've also looked at the signal on my oscilloscope and I've seen no evidence of the PCM doing anything other than pulling the line low for about a millisecond around 3-5 minutes after my last attempt to send a command. This seems to suggest there's something there, but so far it hasn't responded to any of the typical ALDL commands and my script is presently about 5% done with no responses... that said, I have about another 1.5-2 months to go before it has a chance of sending every possible 3 byte combination.

What scares me is that even once I simplify pulling a binary off the PCM... there's still presently no definitions for it; certainly no shortage of work to be done.