What ever they did, it looks like they didn't even touch the capacitors. Tedious, but you might want to take a magnifying glass and scan both sides of the board carefully for any signs of rework.

If no rework is found, this might be a classic case of ESD damage.

 

I'm ****ing devastated guys. What do you think the next step I should do is? I'm utterly lost.

I did my best to avoid ESD damage. I've worked with computers and built my own computers for my entire life, my dad taught me when I was a kid. I only have hardwood floors in my house, no carpets. I only touched the board from the edges or on the plastic mounts. I didn't mess around with it. I just took pictures of it, then buttoned it back up. It's unknown what the vendor did while they had possession.

My eye is untrained, but I can't spot any sign of defect. I spent about 10 minutes examining both sides closely. No burn marks, no scrapes, nothing out of place as far as I can tell.

I think the following pictures say it all. Those are the original Nichron caps still in place behind a warranty sticker for replaced capacitors from SIA Electronics.

 

As long as you touched something metal first(like the case) you are fine.

To get esd damaged you have to touch the processor pins and have static on you too, at least from my experience.

It can happen but it sounds like you were careful. 🤓

​​​​​
PS, don't be afraid of the circuit board. Just stay or sit on one place(so you're not picking up static from your shoes) when handling it and make sure you touch something bare metal first. Then you can touch the board all over. Touching something bare metal does a good job of grounding out static from you

One more thing, if you don't have an antistatic mat you can lay the board on a piece of cardboard or a ceramic tile.

 

This is a tough one, you can't charge someone for replacing them if they weren't replaced, well I guess you can, and you did get charged... It sounds like they have some kind of struggle with these 98-00 ecu's. Your transparency is valuable for anyone else reading this, though this is very bad for you obviously however. So any kind of theoretical speculation of what happened or who did what is mute at this point because this shows the work wasn't done. So if an entity is willing to be inaccurate about work performed, what else are they hiding. I've worked on computers too, circuit boards some, you can handle them like you did all day long I've handled many without issue. Its when you start touching the boards with instruments is where you put them at most risk. I don't think you did any of that from what you are stating, you would have no reason to.

 

aww man this was a painful read

yea if there's no visible signs of damage my only assumption would be an internal short somewhere

also idk if i missed an initial post before this one but what made you think there was ECU damage in the first place?

 

Thanks for the comments guys. I appreciate it. I'll of course continue testing and triple checking my work, I'm going to re-assemble everything and try starting the car again.

Yeah I've written a lot sorry lol. here is the "why".

https://www.clublexus.com/forums/ls-...l#post11792113

I want to make it clear I don't blame Yamae or hold him accountable in any way. He didn't tell me to do this, I did it on my own. I would actually really appreciate his expertise and help in this manner, I come on a bended knee asking for council.

At least there is good news. My car has now been sitting for 5! weeks. No oil leaks, not one. I'm not sure how long that will continue, but goddamn are the mechanicals of my car still in stellar working order. Never once not started, never once thrown a check engine light in 50,000 miles. However, there is a fuel leak because the engine has been flooded now from me starting it so many times. Gas is running out of the exhaust valves and pooling in the exhaust. And leaking out of the flange between the manifold and the cats. So the first leak my car develops is a gas leak out of the exhaust 🤣, and, gives me a hint at to where a suspected exhaust leak I've had is!

 

Before you re-install the ECU, inspect the connector pins and contacts in the harness for damage. Once installed, I would still check for an immobilizer issue. Read thru pages DI-621 thru DI-638 in the '98 FSM. I don't know it it would be of any help, but there's a way to check for immobilizer DTC's without the Lexus hand-held tester.

I did come across an immobilizer troubleshooting tree from the '97 FSM that mentioned a crank but no start condition. I'll try to find it again. TBH, I've never actually witnessed how a car acts if there's an immobilizer problem; crank no start, no crank, no start, etc.

 

What sort of damage should I be inspecting for? I remember feeling one or two of the wires were "loose" in their plugs, but only a few thousands of movement, they were still attached. Would that be enough to cause continuity issues? I did my best not to yank on the wires to pull out the plugs. For most of them I used a flathead screwdriver to pry the connector out once I realized how snug they were. I will go through the wires on the harness to inspect them more closely this evening.

I took a cursory look at the immobilizer. I don't think there is an issue? The security light is in a stable blinking state, and it correctly shuts off completely when I insert the key. For example, a DTC of the immobilizer says there should be a 4.5 second pause in the blinking light to signify there being a problem. Are there any other "features" of the car which are locked behind it, that if I have access to, should eliminate the immobilizer?

 

For starters (no pun intended), check for bent pins. It's also common to do a "wiggle test" once the harness is plugged in. This will sometimes expose connections/plug issues.

Right now, it's a guessing game. I know it's frustrating, but I think the best course of action is focus on the most likely problem; the only thing that changed before and after, is the ECU. Because the symptoms appear to be immobilizer related, you'd want to absolutely eliminate that possibility. This is from a '97 FSM; no mention of the Security light (entire document attached). I don't know how this differs from the '98 model year (probably does).



The '98 FSM describes how to check for DTC's on page DI-623 (attached). Also attached is a list of possible trouble codes. If this checks out, then you can confidently move on to to other things.

Also. make sure your battery is fully charged.

 

+1, cool.
​​​​​​
Like mentioned above, it's all guessing at this point.

I've repaired 1,000's of circuit boards over the years and other than checking the basics like good battery and tracing one's steps like mentioned above, having a known good Ecm for further testing is what's needed/wanted.

How you get there I don't know because the ecm was already worked on so it's automatically suspect.

You did mention the car sat for a few weeks. I have seen cases where a board has caps that were in bad shape and in a short time of not running the board, the caps failed completely and killed the board from working properly. For this to happen would mean the caps were already bad and just sitting not being used was the final nail in the coffin. An ESR meter is needed.

However, this would've been caught by a repair shop because the capacitors ESR would be elevated and they would've caught that, hopefully.

Like mentioned above, it's all a guessing game now. Lol

PS, fine members above gave you the steps to rule out other stuff before condemning the ecm, such as immobilizer, etc.. This will give you a high degree of certainty whether or not your Ecm is bad. Not 100% of course but the best you can do. 🤓

 

Thanks Margate330. I don't trust what the shop did at all, if they lied about replacing the capacitors, they didn't check anything else. Having a known good ECM is not an option here. The immobilizer is a chip on the ECM. However it would offer some interesting options for testing as I could see if error codes change or who knows what else. But not for now.

P.S I am going to keep replies as short as I can now - this thread is so long!

So here is the documentation talking about the security light. The key piece of information I didn't read correctly is that these patterns only show up when you have the wire plugged in? I'm not actually sure if it only happens when performing the test, or if it always will display it if there is a problem.

I just have on interem question Paulo - Two methods of doing this. One using the scanner, which I don't think my ELM327 scanner can do? The other using SST

https://www.cbsracingshop.com/en/098...eck-wire-no2-2

The SST is just a merged pair of wires? How should I best go about fabricating one of these on my own. Is it as simple as any random wire to bridge the two terminals?

 

If you look at page DI-626, this shows the immobilizer DTC's. These are manufacturer-specific codes; most consumer types of scan tools will not read these codes.

Special Service Tool (SST) 09843-18040 is a paired wire jumper wire. But any single length of wire will work to jump Terminals 4 and 13 of the DLC3 (OBDII Data Plug). The limitations of this jumper method is that it will only tell you that trouble codes exist. If codes are stored, you have to work through Page DI-629 to find the cause.

In your case, if a code is stored (Code 99) indicating the immobilizer is at fault, you can stop and pretty much assume that the ECU is the problem; it the only thing that was physically changed. Optionally, you can go to Page DI-629 and continue to troubleshoot.

 

You don’t need any special service tools or wire a simple large paper clip works just fine.
As long as you insert into the correct terminals of course…
Just unfold the paper clip and bend it into a U shape.

 

Sorry for the delay - life caught up with me. I got some testing done this afternoon though.

I spent about 30 minutes reading and re-reading all the relevant instructions. Fun fact I don't have any paperclips, but I have I think 18 gauge single conductor electrical wire. I snipped the ends so they would fit in the terminals snugly and I am able to insert as much copper wire as the metal contacts on the male end of the OBD plug. I checked 5 times I am plugging into the correct holes. I made sure the taper and clip are on the correct orientation as the diagram I counted holes multiple times. I inserted wires before turning the key to ON, and then I re-ran the test in the correct procedure inserting them after.

But at the end of the day, every time I did this was the same result. The security light never turns on again for any flashes for any reason after inserting the key. I still have the same 4 check engine codes and VSC error as before. Am I correct in that this test will not show those and is only checking immobilizer codes?

Oh I almost forgot, I also checked the board again for another 10 or 15 minutes visually before closing it back up. For whatever good that is. This includes the plugs, and, includes the cables on the car's harness too. Nothing is loose or out of place as far as I can tell. The plugs click cleanly into their sockets, all of them.

 

When using the jumper wire, what codes did you get? Did the MIL/CEL flash a Code 12 (one blink - pause - 12 blinks)? Did you get any other codes?

When you state that you still get P0120, P1120, P1126, P1656, was this with your USB dongle, yes?