Anyone with TPI(TBI too?) and a "noid" light?

[dyeager535]

Wondering if any of you would be so kind as to unplug one of your injectors when the engine is running, and hook up a noid light to the injector connector. Then hook the noid light up to your battery terminals, comparing how much light is given off.

I'm not saying I got my no start issue resolved, but throwing a different ECM does now allow the engine to start after being shut off, more than once a day.

I have 12V on both the injector connector leads (as there should be) but when I disconnect the injector connector and test mine, the bulb is quite dim. If I had to guess, I'd say it LOOKS like it's getting 6V or so at best. It pulses as it should, but the whole thing looks weak. Nice bright bulb at the battery.

Trying to make sure the setup isn't somehow killing the ECM's, as I'm on my third one now, (luckily none new) and would really hate to have this one die another similar death.

If anyone with TPI (I'll listen to TBI folks as well though) would be kind enough to try this, it will tell me if I need to work on my grounds or not. Hoping not, but better to do it now than later.

Thanks!

 

[4X4HIGH]

I thought i remembered reading that there is only supposed to be between 3-5 volts at the injector wires.

 

[Russell]

Dorian, my truck is much the same, very little light put out when I use a marker light bulb as a noid light.

That said, my TPI runs like crap right now too, lol

 

[dyeager535]

Maybe on some other setups, but since one half of the injector lead comes directly from the fuse panel, it's 12V.

I'm going to try and test with my digital meter, but I'm sure response isn't going to be fast enough. I guess if I can see the peak voltage it will do its job, but otherwise the light is the only thing I've got to go off of.

It gets the 12V constantly, so the only possible reason for low voltage (if indeed that is true) has to be a bad ground, but I know the wiring is good, and don't want to dig the grounds out again if I don't have to. I spent a lot of time making sure they were all good on install, shouldn't be problematic now. But I'm also not going to take chances.

Russ, thanks for the test. I figure one more person with a good running setup with the same results will settle this for me. It seems like the injector drivers slowly get overloaded and eventually get to the point where they are good for only one start a day.

 

[HarryH3]

I don't think that a noid light will blink very bright because the duty cycle is so short. The filament barely starts to get hot before the circuit is open again.

Does your engine start easier when the ambient temp is warm? If so, is the "Crank" wire to the ECM wired to get voltage from the starter when the engine is cranking over? That is what helps the ECM decide when to apply "cold start" mode.

 

[dyeager535]

Mine doesn't use a crank signal from the starter, that's handled from the ignition module.

Starts a bit easier, but at the same time I'm dealing with gas that's been in the rig for about a year now. Runs fine (so it seems) but cranking times seem to have increased a bit. Especially since I had modded the PROM to fire the engine at about 4 distributor pulses, not the 8 or 16 GM had originally. Not to mention I'm running '89 code, so no cold start injector, just increased injector flow at cold temps. Confirmed the temp sensor is working.

Anyways, that's kind of what I was thinking with the bulb. Not even sure an analog meter could respond close to fast enough, but I don't have one anyways. Which is why it would reassure me that everything is "normal" if someone else with a good running setup confirms the bulb test.

To revisit, whole issue began when the truck randomly wouldn't start when cold. Could never get it to be consistent, so couldn't do much diagnosis. Finally after swapping in a third (two used, one new Delco) ignition module, the problem became 100% consistent...if the truck was started and turned off, it would not start again that day. No matter if running 2 seconds or 10 minutes. Checked injector pulse, and there was no injector pulse when it was cranking. Swapped ECM's (two "known good") and same problem. This one I didn't even know if it was good, but the price was right, so tried it. Now it starts as often as I want it to.

Only thing I've never been able to test is if the ECM is actually seeing that "crank" signal when it was doing it's thing. My scanner has NEVER recorded cranking RPM, but shows running RPM no problem. So impossible to tell if it's a scanner or ignition problem. Following the GM diagnostics, that's one of the first things to look at, but if I follow either branch of the flow chart from there, it still lead back to the ECM. I could power the "crank" ignition wire as the test said, and I could hear the all the different components working, so I know that given a signal, the ECM DOES see it.

With this third ECM having "fixed" the problem, my only concern is that something else (like bad ground or injectors) is the real root of this problem, not the ECM. At least if someone elses setup has the same dim light, then that would rule out a ground issue. My injectors spec out ok, (not great, but within tolerance) but they are too small and I've already got brand new replacements for them, so that's next step to eliminate one known and potential problem.

 

[HarryH3]

Sounds like you've got the bases covered. Have fun changing out those injectors. THAT was a job that made me decide that TBI is a much more desireable system, at least from a maintenance viewpoint.

 

[dyeager535]

Ehh, already been in there once, setup hasn't been together for too many years, so besides the one bolt I crossthreaded (shh, those bottom ones on the runners are a pain sometimes!) so I should be able to make short work of it. Not even sure I have to pull the runners. I'll try without, but have a feeling it will.

Course, the rain has started again, was enjoying the clear warm-ish weather we were having for working on it!

 

[HarryH3]

You have to pull the runners, again. The injectors won't come up out of the holes far enough to remove otherwise. The fuel rails hit the runners first. It sucks to have to replace all of those gaskets every time you have to take the top end apart.

 

[dyeager535]

Last time it was so shortly after the build I just reused them. Don't tell anyone, it couldn't possibly work.

I've gotta steal the injector o-rings too, but they are new so I'm not feeling too bad. Going to use the dial calipers on them to double check their dimensions, but the Ford o-rings are a bit different and I'm not going to chance it.

 

[southernspeed]

Course, the rain has started again, was enjoying the clear warm-ish weather we were having for working on it!

Where are you at now then Dorian that's so warm and wet? You hussling out in Malasia or something???

 

[dyeager535]

Warm's relative around here. It was below freezing for quite some time, but not now.

I'm back home now Mark, (You too I see from your other post!) just outside of Seattle.

 

[1979jimmy350]

have you checked the pick up coil by chance? One on the verge of going out could cause somproblmes like this, Thoese TPI computers must be pretty sensative, im on my second one, now im just waiting to gater parts to convert to a newer PCM so i can controal electronic trannys also with my TPI

 

[southernspeed]

Warm's relative around here. It was below freezing for quite some time, but not now.

I'm back home now Mark, (You too I see from your other post!) just outside of Seattle.

Seattle? Hell....that's almost Scotland ain't it?!!

 

[Shiprekd]

Sounds to me like there's a short or something awry in the injector harness. I peeked in on chevy thunder and here's a direct copy/paste:

"Disconnect on of the injectors from the harness, and with a injector tester lamp (noid lamp), or a test lamp that one end is connected to a hot (12V), probe the green/blk or blue/black wire and with the engine cranking, the lamp should blink. If it blinking, the problem lies in the fuel delivery system, that will be discussed a little later. If the light is steady, the problem is in the injector drive circuits, the green/blk or blue/black wires that are shorted somewhere to ground. You'll have to pull the terminal to the ecm and with all the injectors disconnected, hook up your continuity tester to ground and probe these wires, the light should not come on. If it does, find and repair the shorted wires. If the lamp does not come on, check the resistance of each injector, with your ohmmeter, the resistance should be more than 10 ohms( typically around 15-16 ohms). If the injectors check out okay, the ecm is faulty. If not okay, replace the any injectors that measure less than 15 ohms. Now, if you don't have a blinking light, plug in the injectors back in, then you'll need to turn the ignition key on and probe the injector harness terminals to ground with your test lamp, the light should be on at both terminals, this is due to the fact the injectors are wired in parallel. If there is not a light at both terminals, the problem is in the injector harness. If the light comes on one wire, the problem is in the harness to the injectors.
Now, if the light comes on in both terminals, reconnect your test light to injector harness, and disconnect the 4 way distributor connector. Momentarily touch the connector with the purple/white wire with a test light hooked to 12V, if the injector light blinks the problem is a bad ignition module. If there is no blinking light, disconnect the tps and repeat the test. If the light blinks, replace tps. No blinking light means you'll need to make sure the fuel pump runs for its two second priming period. If the fuel pump checks out okay, reconnect all your injectors, turn the ignition off, disconnect your ecm, turn the ignition key back on, and probe terminal D15 with the test light connected to a hot power source, if the light comes on, you'll need to wires to injectors (green/blk or blue/black) being shorted to a voltage source. If they are okay, the ecm is at fault."

 

[dyeager535]

All wires test fine. All ECM grounds are grounded, no injector related wires *in the entire harness* are shorted to ground, OR shorted to another voltage source. All injector wires have power, all injector leads pulse as they should.

With the other ECM's the injector pulse was so weak you could barely see it, new one is brighter, but still only appears to be 6 volts or so. As Harry mentioned, THAT may be because of the duty cycle, but I'm not having luck with anyone that has a properly running TPI setup testing theirs for me to make sure. I need to get an analog meter and see if it is any better.

I know these ECM's are "notorious" for having weak injector drivers, but two ECM's failing back to back is suspicious enough for me to be completely sure I've solved the problem.

 

[Shiprekd]

I'm running a TPI in mine. Matter of fact, I'll never go back to Carb or TBI. The tpi noid never really is "bright"; At least mine isn't during daylight anyway..lol - Personally, I would poke around looking at the resistance in the wires on the injector circuit. You obviously already know they should be dead on zero.

As far as grounds are concerned, even the battery ground could cause it although highly unlikely. I had trouble about a week after I had mine on the road. seems one of the grounds jiggle loose where I had it bolted to the engine/trans. Caused the ecm to throw a O2 sensor code.

I pieced my set-up together. I run an 86' lower intake for older heads and a speed density system on top except for the TPS, which I use an older style so I can adjust it. I rewired an '86 harness for the 727 ecm and had the memcal burned to my specs.