Computer Problems with Manual Trans Swap

[TrickyT]

USA spec 1FZ-FE engines with OBD-II use the J1850-VPW protocol. Also important is what OBD standard they follow (see OBD-II PIDs - Wikipedia). 1FZ-FE's sold in North America follow the "OBD-II as defined by CARB" standard. The 89661-60281 is a Euro spec ECU that will almost certainly follow one of the European OBD standards.

 

[burbuja96]

Is there a way to check the protocol with the actual ECU?
Thanks

 

[TrickyT]

Is there a way to check the protocol with the actual ECU?
Thanks

Yes. Power it up in a vehicle, plug in an OBD-II scanner, and send it PID 011C and then see what it reports. The number you get back will indicate the row in the Wikipedia table I linked to in my post above.

EDIT: Engine does not have to be running, but ignition must on.

 

[burbuja96]

Got it, I will do that and report back. Thanks for your help.

 

[landtank]

J1850 VPW, there is like 5 flavors of protocols.

 

[GLTHFJ60]

Changing out the ECM is very easy. 30 minute job max. Swapped in a 89661-60340 this weekend. Tachometer no longer works as expected, but the truck runs as well as it did before.

We'll see this week or next what codes it throws and whether or not US readers can read codes from it:

 

[GLTHFJ60]

As expected the tachometer doesn't work, but the CEL is off and didn't come back on. It comes on with the ignition as expected, but not when the engine is running.

Unfortunately the local inspection shop said that they couldn't connect to the ECU, claiming some sort of electrical problem, so the truck still won't pass inspection

This computer fix would work well for a 96 and below truck, but not for a 97

 

[burbuja96]

Crap, mine is a 97 too. Why would a 96 work but not a 97?

 

[GLTHFJ60]

Because to pass 1997 inspection (US OBD2) the truck has to be connected to the inspection station system in order for it to register and pass. It's not enough for the light just to be off. Pre-OBD2 (aka 1996 and older) the light just has to be off.

@landtank and @TrickyT were right.

 

[TrickyT]

Bummer on not being able to pass inspection.

So it sounds like the 89661-60340 doesn't implement OBD-II at all, or at least not J1850 VPW which was one of the primary protocols in use when these vehicles were made. Prior to OBD-II, Toyota had some proprietary interface they called "multiplex OBD" (M-OBD) that used the same 16-pin connector but totally different signalling. You can read about it here: What is data stream M-OBD of Asian Toyota vehicle? So it could be that the 89661-60340 uses M-OBD. Unfortunately that won't help you because M-OBD was never an accepted standard in the USA.

 

[TrickyT]

Because to pass 1997 inspection (US OBD2) the truck has to be connected to the inspection station system in order for it to register and pass. It's not enough for the light just to be off. Pre-OBD2 (aka 1996 and older) the light just has to be off.

@landtank and @TrickyT were right.

So what are your options now, Johnny? You've obviously got a lot of time and money invested in doing the manual trans swap. I sure hope you're not just limited to towing this rig to off road parks to wheel it.

 

[GLTHFJ60]

I was told by a couple of people that 89661-60340 would work, but clearly it does not, not at least for a 97, due to inspection requirements. This is the achilles heel of the swap. No clue how anyone else has legally completed a H151 conversion on a 1997 US spec FZJ80, but from what I've read and experienced first-hand, it's not possible. I'm investigating options now. I've got no ideas other than to buy a 96 or older FZJ80 and move everything transmission related over to the older truck.

I'm trying to identify how to determine which protocol this ECM utilizes as a way to move forward, but even if I knew, it wouldn't change the fact that the truck won't pass inspection as it sits.

This is really depressing.

Let's be clear for future readers. This ECU *should* work for a pre-OBD2 vehicle (96 and older), with the A343, because the CEL comes on with the ignition and stays off when running. I *believe* that the only check for the pre-OBD2 trucks is making sure that the CEL is off when running AND that plugging into the inspector's computer is NOT part of the inspection procedure. It would be ideal for someone with a 1996 FZJ80 to verify.

*sigh*

 

[sbman]

I've read most of this thread, but not all of it. Excuse me if this has already been tried.

My basic plan with this when I do the swap on my '97 here in CA, (I've got the whole kit in the garage) is to wire up the shift indicator so that the ECU think's it is in 'N' all the time, dummy up loads on any solenoids it expects to see and work around starter/shift/key interlocks so that things work like normal. Has this approach already been tried/failed?

 

[GLTHFJ60]

I wired the PRND indicator so it does show in neutral all the time (unless in reverse). Doesn't help.

If you can figure out how to wire up dummy loads to simulate the A343, please, let us know. I doubt you'll be able to design a circuit to emulate the transmission well enough that the ECM doesn't realize it's missing. From what I've read, no one has tried it.

The starter/shift/key interlock is easy to work around. Simply disconnect the shift interlock computer and the shift lockout is removed without any CEL from what I can tell.

 

[TrickyT]

...I'm trying to identify how to determine which protocol this ECM utilizes as a way to move forward, but even if I knew, it wouldn't change the fact that the truck won't pass inspection as it sits.

I think the only way to know for sure will be by finding literature about the model vehicle this ECM was used on and then hopefully find a reference to the diagnostic port and protocol. There are some other links you can follow from the link I provided in post #150, including this one: Scanner - Scantool - OBDI - OBDII - OBD2 - Automotive training - Cartool - Vocational - Trainer which is an ad for a diagnostic scanner that claims to work on "M-OBD (car uses K-line for markets as [sic] Asian, Africa, middle East and Australia)." Farther down on that page it shows that scanner connected to the DLC3 connector using pin #13 for the signal. Pin #13 is a make/model specific pin in the OBD-II standard. U.S. spec 80's use pin #2 for the J1850-VPW signal. The wiring harness on a U.S. 80 does not have any connection to pin #13, so it may be that by combining the 89661-60340 ECM with the U.S. harness you've routed an incompatible ECM signal to pin #2 and this is what led to the "some sort of electrical problem" reported by the inspection shop.

...This is really depressing.

I can only imagine. A real bummer for sure. I'm for you.

 

[GLTHFJ60]

Thanks for trying to help explain Tom.

I know which models this ECU was listed in, all "General Market". It's still a mystery to me how to see which OBD protocol all of these models adhered to.

Model Dates: Matching Models:
01/1995-12/1997 FZJ80L-GCMEKV
11/1997-11/2006 FZJ80L-GNMEK
01/1995-12/1997 FZJ80L-GNMEKV
11/1997-11/2006 FZJ80L-GNMRK

 

[KliersLC]

I think the only way to know for sure will be by finding literature about the model vehicle this ECM was used on and then hopefully find a reference to the diagnostic port and protocol. There are some other links you can follow from the link I provided in post #150, including this one: Scanner - Scantool - OBDI - OBDII - OBD2 - Automotive training - Cartool - Vocational - Trainer which is an ad for a diagnostic scanner that claims to work on "M-OBD (car uses K-line for markets as [sic] Asian, Africa, middle East and Australia)." Farther down on that page it shows that scanner connected to the DLC3 connector using pin #13 for the signal. Pin #13 is a make/model specific pin in the OBD-II standard. U.S. spec 80's use pin #2 for the J1850-VPW signal. The wiring harness on a U.S. 80 does not have any connection to pin #13, so it may be that by combining the 89661-60340 ECM with the U.S. harness you've routed an incompatible ECM signal to pin #2 and this is what led to the "some sort of electrical problem" reported by the inspection shop.



I can only imagine. A real bummer for sure. I'm for you.

To be honest, most of this post is going over my head, but it sounds like you are saying that the M/T ECU might be sending the signal to the wrong pin on the OBD connector. If that is the case, can you swap the pins/wires on the OBD connector? That might not solve the issue if the information transmitted is not compatible, but it might at least get the information to a point where it can be accessed....

Another thought after reading this thread--with the expertise we have on the board and a couple extra A/T ECUs laying around, could we open one up and do some wizardry to the inside to help out?

Or, is there a way to re-flash the ecu to solve the issues?

 

[GLTHFJ60]

OBD2 is a standard that is implemented by many protocols. TrickyT is saying that the US has one protocol (J1850-VPW) and that other countries use other protocols. Even if we swapped pins, the protocol would still be wrong.

No way to re-flash the ECM that any business has figured out.

I'm not going to open up an ECM we paid to have shipped from another country, one that I can't easily replace. By all means, buy one, open it up and start soldering

 

[KliersLC]

I meant your old ECU. It isn't doing anything for you currently...


If you could modify it internally to not throw those codes, you could reinstall it and have proper communication with the OBD scanner. The how of that is way beyond me though.

One question though--does the MIL come on immediately when starting the truck? Is it possible to clear the codes right before the inspection or is NC one of the places where the ECU has to have cycled through all of the diagnostics before it will pass?

 

[GLTHFJ60]

Gotcha, you mean the old ECM. That's not a bad question thinking about it that way. I doubt that the transmission portion is separate from the engine portion though.

NC is one of those places where the ECM has cycled through all of those diagnostics. They'll be able to see that the ECM is in "learning mode" and won't pass you.