2002 2.7L starts then dies immediately

Working on a 2002 Tacoma Pre Runner Crew Cab with the 2.7L 3RZ engine. A few months ago we replaced the head gasket and installed ARP head studs, and it has run perfectly since then. Wednesday, he drove the truck to work and it was fine. However when he got off of work, the truck just fired and died. No engine codes. If you crank it after it dies without cycling the key all the way off, it will not fire at all, but if you cycle the key all the way off, it will repeat the process of firing once and dyeing. I went and tried some diagnosis yesterday.

There are no engine codes.

He replaced the fuel pump yesterday, and with the key on, you can hear fuel flowing through the fuel system under the hood.

The MAF is clean.

The coils fire.

I have valid readings from all of the sensors on my scan tool. Coolant temp, RPM, TPS, Injector duty cycle, ignition advance, air temp, airflow in lbs, etc..

I installed a noid light on the #1 injector. When the key is cycled all the way off and then started, the injector fires once, and then never again. You can crank forever and it never fires again, so this is the obvious issue. It is not putting fuel into the engine, but why?

The Cam sensor had a broken edge where the o-ring goes when we changed the head gasket, and since there is no option to see the crank and cam signals on my scan tool, we opted to replace the cam sensor since it seems like the ECU is not getting a cam sync to fire the injectors. Same issue.

We have tried firing it with the MAF unplugged, TPS unplugged, Engine Coolant Temp sensor unplugged, all 1 at a time, and same result each time. The only codes given is for the component we had unplugged during the test.

The engine gives an RPM signal to the ECU, and I believe that is from the crank sensor, so that indicates it is working.

It is acting like a security issue, but his truck does not have chipped keys, and he has disconnected the battery for an hour with the same result. I even locked the doors and unlocked them with the key and same result. Later found out he doesn't even have a key fob to unlock the doors remotely so always has to use the key anyway.

This really makes no sense to me. If the injector drivers in the ECU were bad, it wouldn't fire the injectors at all, but always fires them 1 time when the key has been cycled all the way off, and if not cycled off, they will not fire at all no matter how long you crank the engine. I am at a loss at this point as it is not making any sense.

I hope someone has some insight to this issue.

 

I would try checking all your power and grounds to the ecm. A faulty power circuit to a computer can do some weird stuff

 

I did make sure all of the grounds were tight. That was one of my first thoughts was that a bolt holding a ground vibrated loose, but no, all are good. Fuses are all good. The voltage to the ECU is fine, the scan tool gives the actual reading the ECU sees in one of the PID's. Im seriously leaning towards leaking caps in the ECU, but hoping it is something else. Ive had to replace caps in lots of Toyota ECU's, but all were early to mid 90's, never had to touch any in ECU's this new.

 

Attached is the wiring diagram.

The noid light issue is a difficult one to figure, could be a clue, could be rabbit hole, unfortunately you wont know until it is fixed. You would think that is there was a bad PCM ground that it wouldn't fire at all. I would verify all the grounds as mentioned above. As an afterthought, the CEL does light up right?

 

Yes, the CEL comes on with the key on as it should. As mentioned, it does not throw any codes at all which is frustrating. You would think if there was a required sensor/signal input missing to the ECU, that it would throw a CEL and tell us, but nope. And the ECU is capable of seeing issues. if you unplug the MAF and try to start it, you get codes for the MAF and air temp sensor, If the TPS is unplugged, we get a code for that. Same for Coolant temp as well.

One thing I did not try that I wish I would have is trying to start it with the cam sensor unplugged to verify it still showed RPM. Which for certain would mean the crank sensor is good. I do not see how it could calculate RPM from the cam sensor, but who knows. Toyota does crazy things sometimes. This is why I love my stand alones. I can literally scope out every signal for every input and see what is wonky or missing. I do have Techsys on my laptop. I may take it over there and see if I can figure anything out. I have never really used Techsys before so not familiar with it.

 

Cam operates at 1/2 crank-speed so it is possible. I feel your frustration… it could be worse, one of these times it just might start, then what? No definitive diagnosis and the thought that it just might do this again in an hour.

 

If you've verified all the inputs and outputs and know it's not an anti-theft issue then it could be the ecm. I've had two vehicles in my career so far that were caused by faulty Ecm's. One was an intermittent no start where a 2012 equinox would start fine with no check engine lights then randomly it wouldn't crank or do anything but kill the instrument cluster as if it was cranking. Still with no codes. Release the key and try again and it would start just fine. It would happen between 1/4 and 1/50 times of consecutive starts. I was an apprentice at the time too and freaking out believing there was no way it could be the ecm because there are no other issues, even when I was testing with a breakout box at the ecm and watching the scan tool data telling me that all modules were reporting to the ecm the crank request and that the ecm was cranking but it wasn't outputting to the starter.

Second one was on a 2018 Colorado with a manual trans which would have no back up camera when cold and the battery would die after a few days. Grounds and all were fine, terminal tension okay, camera, reverse switch and battery both replaced over 7 times when checking the repair history of the vehicle. Even got TAC ( the gm technical assistance centre) involved and we agreed that we don't really have much else to check because the other module reports to the ecm then sends the camera on signal but we don't get a camera.

Hopefully that's not it but after hearing for most of my junior mechanic years that it's rarely the ecm and if your diag leads to the ecm you probably did something wrong. I still kind of agree but after those two years I don't reject my diag now, just double check it before I order one

 

Well, I am thinking the ECM is bad, but am hoping otherwise. I also do not know if a reprogram is needed if he buys an ECM from a junkyard. Before i let him buy an ECM, I will pull his apart to inspect for leaking capacitors and replace if there are. Ive done it many times before, just never anything with an ECU this new. Seems to only ever be issues with 90's vehicles. Typically, unless something was done to fry the ecu, the ecu is almost never the issue. Almost. I am just hoping there is something about this truck I am not aware of that could be the issue aside from the ECM. I have built Megasquirt ECM's and built custom harnesses etc.. and typically work on Supras, Celicas, Corollas, etc... I had ECU issues many times on my old GS300, but I took that thing to 440k miles before selling it and the 2jzge still ran perfect. This Tacoma is driving me nuts. Even when its been the ECU in the past, there have been fairly obvious issues of it. This time, not so much. I really wish I had the wiring diagram for the ECM/Engine harness. Would make things a lot easier.

 

Maybe you can get a couple days Tis or shopkey subscription to look at wiring diagrams? I myself am not sure if those vintage Ecm's need to be programmed either and am definitely not a fan of going parts cannon on things

 

I got to thinking about your problem and this came to mind… I personally haven't used this particular unit but similar ones a while ago… for the price and if you can either prove or disprove the CKP and CMP signals, for $60, I would rip my pocket to get the money out to pay for this… I can see more than $60 in frustration.

$55 Mini Handheld Digital Oscilloscope, Support Persistence, PWM, 14 Type Measurement, 40MSa/s Sampling Rate, 18MHz Bandwidth, Signal Generator Frequency 500Khz, 2.4” LCD Display

https://www.amazon.com/Oscilloscope...5145436&mcid=a22055a67f1e36dab29fec177503485c

 

Ok. It runs now.

We were admitting defeat that it must be the ECU. He pulled the ECU and brought it to me last night. I pulled it apart and it is MINT inside. No leaking on any of the elctrolytics. All components look perfect. I made note of the values of the electrolytics and the numbers from the transistors etc... He also disconnected his battery last night and put it on his trickle charger as it was getting a bit low after all of the cranking we did during triage. Anyhow, I was looking up the procedure to program the VIN using Techstream as we found a closish junkyard that showed they had the ECU.

I found mention of a procedure to force the ECU to "Handshake" with the imobilizer/security module if it has one. You use a paper clip to jump pins 4 and 13 of the ODB2 connector under the dash, then turn the key to "ON" for 30min, then turn it off and remove the paper clip. I dropped his ECU off to him this morning. He plugged it in, hooked his battery up and tried starting it for giggles before trying the "Handshake" procedure. It started up and ran without issue. We literally changed nothing and only disconnected the ECU and battery overnight.

I am certain it was some sort of "security" issue and the power loss overnight cleared it, whatever it was. I am betting he at minimum has a key-less entry module in the truck even though he does not have key fobs, and likely has a security module of some sort that just happened to "hiccup". It makes no sense. He has shut it off and started it back up multiple times without issue today, so this crazy issue has "resolved itself".

I wish I had more info in case this happens to anyone else going forward. I have recommend he find another ECU for a backup in case this ever happens again.


TLDR = We disconnected the ECU and battery last night, and connected them back up this morning and the issue is resolved!

 

I looked up the owners manuals for an 01, 02 and 03 just to verify if there is a factory anti theft and there is no mention of one so if it is there it is aftermarket. There is a post on the 1st gen forum where a guy had an aftermarket immobilizer and if I am understanding his post the immobilizer was activated by turning on the headlights after he turned the ignition off, crazy stuff. 99% of the time when an alarm/immobilizer has been installed there is strange wiring under the D/S knee panel… it is pretty obvious when you see it. I sure hope it is fixed… like I said the worst case scenario is it runs with no definitive “fix”.

There have been times where I have wired in a few 12v 194 sockets and bulbs to circuits that I suspect are causing the vehicle to stall, fuel pump, power to PCM etc, and drive around until it dies if the light goes out well at least you have somewhere to start looking, if they all stay lite, then it is something else. Kind of archaic diagnostics but what other choice do you have, the parts cannon?