[L78toLT1]

Hey guys, please allow me to apologize in advance for the lengthy post. This dissertation is both a rant about the unbelievable amount of bad juju I am experiencing in trying to get my new motor up and running, and a solicitation for suggestions on what could be causing my problem.
I’ll start with the problem: I have no fuel injection on the right bank. I can completely unplug the connector to the r/h fuel rail with the car running, and it makes no difference at all. The left side appears to be good. The motor actually sounds like a reasonably smooth, decent running four cylinder.
Here’s some background info: I melted an exhaust valve at Road Atlanta last fall and trashed the short block and one head on my original motor; some shrapnel got into the blower and nicked up the rotors pretty good also. I replaced all the damaged parts and sent the blower to Superchargers online to be refurbished. I also sent them the injectors to check and clean. I need to update my profile, but the new motor is basically the same as before, except for ported CID heads from LME and a slightly less aggressive CamMotion cam.
Once I finally sourced all my parts, I reassembled the motor and started remote tuning with Ted Jannetty, I should mention that when I first fired the motor, I had no fuel injection on the right bank, but disconnecting & reconnecting the fuel rail harness cleared things up. Ted and I were on about the third logging/tuning iteration when the motor started missing. I diagnosed a bad injector on cylinder #7. Given that I had to pull the damn blower back off to replace the injector, I decided to replace all eight. I suspect the bad one was in the cylinder with the dropped valve, but I don’t know for sure, there was no visible damage to that injector, or any of the others. John at Fuel Injector Connection verified that #7 was bad and sold me a full set. When I reassembled everything with the new injectors, it fired up and ran fine. This was the weekend, so I couldn’t resume logging & tuning until Ted was back at work. I decided to put a few gentle, no boost miles on the motor over the weekend. I put around 150 miles on the motor when I noticed steam coming from the exhaust. I determined it was coming from cylinder #2. I pulled the blower again, along with the head and found that my new head was leaking coolant into the #2 exhaust port, you could see the void in the casting. It apparently took several heat cycles to open up and start leaking as the motor had around 350 miles at this point. I returned the head to LME and they replaced it in a very timely manner. I also respectfully requested reimbursement for the extra gaskets and consumables, which included those dreaded one-time-use HPFP tubes that I had to buy. I’ll admit to reusing these tubes in the past, but this would be the third time for this set, so I replaced them; I did not want to pull the blower yet again to replace these tubes. Nothing but crickets from LME on this, which surprised and disappointed me. It’s not so much the money, I think it came to around $300, but the principal of the thing. I didn’t ask for labor and offered to provide receipts for the tubes gaskets, coolant, etc. do you guys think I was being unreasonable?
Anyway, installed the new head, jumped in, and fired her up. Running on four cylinders again, so I did my usual trick of disconnecting the connectors, multiple times, to no avail. I assumed the problem was with the big 12 pin rail connector, so I ordered new engine side rail harnesses, left and right. At this point, I wanted everything under the blower to be brand new, so I won't have to pull it off again. When I got everything back together this time, I was afraid to push the start button. I honestly thought I may need to go into therapy if it was still running on four cylinders. Since I’m writing this rambling narrative, you probably already know the answer, right bank was dead as ever.
Now that I have extra rail harnesses, it at least makes troubleshooting easier. I plugged the old r/h harness in and checked the power and signal wires; I have 4.8v on the power wire but no signal. I verified this with a test light. I know my signal wire test methodology is good as I performed the same test on the left side as a check and the damn light flashed away. My next step is to pull the connector from the ECM and verify signal wire connectivity from the injector plug to the ECM. I’m assuming the harness will check good, I have never disturbed that end of the harness and visually, it looks good as new. I should also mention that, per Ted’s recommendation, I verified all grounds that were disturbed over my five-year course of modding. I need to consult my service manual to determine the locations and check all other relevant grounds, just in case.
If the signal wires check good, and I’ll be shocked if they don’t, I’m thinking the ECM itself must be the problem. As I mentioned, the harness looks good, there are no melted wires adjacent to the headers, and no damaged insulation anywhere that I can see. Also, considering the number of wires in that harness, it seems almost statistically impossible for the four right bank signal wires to be damaged when all other systems on the car work fine.
Is anyone familiar enough with the inner workings of the ECM to know if it would be possible for the right bank injector signal wires to be dead while the left work seemingly perfectly? If it’s not the harness, which I will know as soon as I get a minute to check it out, what could it be other than the ECM? If anyone has any idea as to what could be causing this problem, please let me know, I’m at my wits end with this thing.
One final note: Even though I knew it wasn’t the problem, Early on in my troubleshooting, I checked the compression on the right bank and all cylinders checked good. I didn’t do a leak down. With good compression, it would at least run regardless of how badly it was leaking, not to mention, there was no reason to suspect anything mechanical.

[KingLT1]

It sounds like a ground issue to me.

[cjperformance]

I dont know details on how inj signal works, but for basic troubleshooting...Can you backpin the harness at the ECU and check the signal right out of the source (ECU)? I was taught to take a straight pin and slide it in from the back of the connector so i could get an oscilloscope lead or other meter on signal wires. Not sure if the ecu harness is backpinnable. Only had it off one time to install my flex fuel kit and i dont recall the construction other than some levers to release. You dont want to pierce wires or damage waterproofing though....but if all else fails... probably more desirable than getting a new ecu and find out you didnt need it.

That would validate or invalidate the wiring harness.

[L78toLT1]

Originally Posted by cjperformance

I dont know details on how inj signal works, but for basic troubleshooting...Can you backpin the harness at the ECU and check the signal right out of the source (ECU)? I was taught to take a straight pin and slide it in from the back of the connector so i could get an oscilloscope lead or other meter on signal wires. Not sure if the ecu harness is backpinnable. Only had it off one time to install my flex fuel kit and i dont recall the construction other than some levers to release. You dont want to pierce wires or damage waterproofing though....but if all else fails... probably more desirable than getting a new ecu and find out you didnt need it.

That would validate or invalidate the wiring harness.

There are two wires on each injector. The power wire is hot all the time. The ECM grounds the signal wire at the appropriate time to complete 5he circuit and fire the injector. If the harness checks good from the injector to the to the ECM, then I don't see how it could be anything other than the ECM, although, if the ECM connector can be back pinned, it won't hurt to try.

[Joshinator99]

Originally Posted by L78toLT1

There are two wires on each injector. The power wire is hot all the time. The ECM grounds the signal wire at the appropriate time to complete 5he circuit and fire the injector. If the harness checks good from the injector to the to the ECM, then I don't see how it could be anything other than the ECM, although, if the ECM connector can be back pinned, it won't hurt to try.

If it’s the ECM then that’s a pretty straightforward fix thankfully. Might be the first case I’ve heard of one actually dying?

[L78toLT1]

Originally Posted by Joshinator99

If it’s the ECM then that’s a pretty straightforward fix thankfully. Might be the first case I’ve heard of one actually dying?

I agree, in fact I just pulled the ECM out last night. Apparently several vendors can supply a plug and play unit if you provide your vin and ECM part #.
From what I've read, the car may or may not start when the new unit is installed; the crank reluctor may need to be relearned, which requires GM software. I'll also need to re-liscense with HPTuners & load Ted's latest tune as the plug and play unit will have my original stock tune loaded. With my luck the damn car won't crank and I'll have to flat bed it somewhere to do the crank relearm process.
If my harnesses checks good from ECM connector pin to injector plug, I'm going to try the ECM. It's around $200.

[arpad_m]

Man, you can't seem to get a break with this car :/ Hope the ECM was the culprit indeed.

[arpad_m]

Jim, here's the only schematic I found that indicates a distinction between the two injector banks and not individual injectors, check connectors X160 vs X161. If it isn't that, then I don't see how a single bad ground wire could cause the problem, my next step would be the ECM.

[L78toLT1]

Originally Posted by arpad_m

Jim, here's the only schematic I found that indicates a distinction between the two injector banks and not individual injectors, check connectors X160 vs X161. If it isn't that, then I don't see how a single bad ground wire could cause the problem, my next step would be the ECM.

Thanks, Arpad. I believe you are correct. The way I see it, if I probe pin #'s 45, 46, 47 & 48 on the ECM, they should all be grounded. Pins 65, 66, 67 & 68 would show open until they are grounded by the ECM to fire the injector. Do you agree with my interpretation? If I'm correct, I couldn't tell anything by just probing the signal pins with the engine off, they would all indicate an open circuit. I would have to pin ECM connector or probe the pin on the ECM while cranking the engine to see if the ECM is triggering the signal wires. Pinning the connector would probably be best as I could actually start the engine.
If I don't get a signal, the problem would have to be with the ECM, unless the ECM itself has more than one ground to the engine/body, and the left and right injector circuits use different grounds. Can you determine if the ECM has multiple grounds? There are a ton of grounds on these cars. I know from pulling my engine (multiple times) that there are three ground straps on the back of the heads, one on the right side of the block, and another on the front of the motor in the vicinity of the AC compressor. I'm certain there are many more on the body.

[arpad_m]

I'm not sure about your interpretation, because the diagram indicates open ground by default on 45-48 (bank 1) and 49-52 (bank 0) and an activation switch inside the ECM.

The ECM has several grounds indeed, but they are not specific to injector banks, let alone individual injectors, so that can be ruled out.

How would catch injector pulses with anything less than an oscilloscope when these only last a few milliseconds even at WOT and in fact barely longer than 1ms at idle? I'm no mechanic, but since each injector pulse will only take a small percentage of total elapsed time, I'd think one needs a proper tool to see if the ECM actually sent the signal to fire an injector, you can't observe these fast and short voltage changes or power bursts with normal garage tools.

[L78toLT1]

Originally Posted by arpad_m

I'm not sure about your interpretation, because the diagram indicates open ground by default on 45-48 (bank 1) and 49-52 (bank 0) and an activation switch inside the ECM.

The ECM has several grounds indeed, but they are not specific to injector banks, let alone individual injectors, so that can be ruled out.

How would catch injector pulses with anything less than an oscilloscope when these only last a few milliseconds even at WOT and in fact barely longer than 1ms at idle? I'm no mechanic, but since each injector pulse will only take a small percentage of total elapsed time, I'd think one needs a proper tool to see if the ECM actually sent the signal to fire an injector, you can't observe these fast and short voltage changes or power bursts with normal garage tools.

Believe it or not, just a couple of milliseconds is enough to check with a trouble light. I verified this at the injector on the left side. Since I had an extra fuel rail harness, I was able to do this without removing the blower. I hooked the light to the positive battery post on the fuse box and probed the signal wire on the injector harness while my cranked the motor over. The light flashed away on the left bank.
You state that the ECM has several grounds, but none are specific to the injector circuits, so that can be ruled out. For talking purposes, lets say that the ECM has five grounds, are you saying that if any one, or more of these break down, it could not cause my problem? Just trying to understand, I know your knowledge of all things related to circuits and electronics far surpasses mine!

[arpad_m]

Heh, very gallant of you, but I don't know much at all. However, and color me surprised, I've now learned about noid lights from this article, so I'm a bit less clueless than I was yesterday, yay (although the same article also says an oscilloscope is the best injector troubleshooting tool)

About the grounds question, I just don't see how any common/shared ground that is not specific to an injector bank could cause a whole bank, but nothing else, to fail. Did you check X161 thoroughly? I can't yet dismiss the idea that it isn't seated properly or has some damage, it would be such an obvious explanation.

Unfortunately I'm not familiar with ECM internals, maybe it does have bank separated grounding on its circuit board, but at that point your only realistic choice is ECM replacement anyway.

[L78toLT1]

Originally Posted by arpad_m

Heh, very gallant of you, but I don't know much at all. However, and color me surprised, I've now learned about noid lights from this article, so I'm a bit less clueless than I was yesterday, yay (although the same article also says an oscilloscope is the best injector troubleshooting tool)

About the grounds question, I just don't see how any common/shared ground that is not specific to an injector bank could cause a whole bank, but nothing else, to fail. Did you check X161 thoroughly? I can't yet dismiss the idea that it isn't seated properly or has some damage, it would be such an obvious explanation.

Unfortunately I'm not familiar with ECM internals, maybe it does have bank separated grounding on its circuit board, but at that point your only realistic choice is ECM replacement anyway.

Hey Arpad, I spent some quality time with the Service Manual and did some more troubleshooting yesterday. Here’s what I found:


· As I mentioned earlier, all injectors have power.



· All injector signal wires have continuity from the X1 ECM connector to the injector i.e. the entire run, upstream and downstream of connector X161. I’m positive there is no signal; I validated my technique on the left bank using a noid test light. It flashed away on the left side.



· Pin number 72 on ECM connectors X2 and X3 is grounded to the back side of the driver’s side head, I believe this is G103 on the wiring diagram. Both ground paths are good. There are no other external ECM grounds.


So….I am confident that I don’t have a harness problem or a ground problem, which makes me think ECM. The only other possibility is maybe a sensor issue or sensor harness issue. While I don’t know the details, I am sure that the ECM uses the input from several sensors when firing the injectors, however I can’t envision how a sensor issue could affect only one bank. Take the cam or crankshaft position sensors for example, there are only one of each.



Unless you, or anyone else, can think of something else to check, I’m going to roll the dice on that $220 plug-and-play ECM. It will come loaded with my stock tune so I’ll have to license with HPTuners, reload my latest tune, etc. but that shouldn’t be a big deal.

[Trochoidal]

Originally Posted by L78toLT1

Hey Arpad, I spent some quality time with the Service Manual and did some more troubleshooting yesterday. Here’s what I found:


· As I mentioned earlier, all injectors have power.



· All injector signal wires have continuity from the X1 ECM connector to the injector i.e. the entire run, upstream and downstream of connector X161. I’m positive there is no signal; I validated my technique on the left bank using a noid test light. It flashed away on the left side.



· Pin number 72 on ECM connectors X2 and X3 is grounded to the back side of the driver’s side head, I believe this is G103 on the wiring diagram. Both ground paths are good. There are no other external ECM grounds.


So….I am confident that I don’t have a harness problem or a ground problem, which makes me think ECM. The only other possibility is maybe a sensor issue or sensor harness issue. While I don’t know the details, I am sure that the ECM uses the input from several sensors when firing the injectors, however I can’t envision how a sensor issue could affect only one bank. Take the cam or crankshaft position sensors for example, there are only one of each.



Unless you, or anyone else, can think of something else to check, I’m going to roll the dice on that $220 plug-and-play ECM. It will come loaded with my stock tune so I’ll have to license with HPTuners, reload my latest tune, etc. but that shouldn’t be a big deal.

I see pin 71 & 72 on X2 as unoccupied and pin 73 as ground for LT1 (2017 manual) page 11-668.

I see pin 72 on X3 as DFI High supply and pin 73 as ground for LT1 (same manual) page 11-680.

This may not be relevant to your setup, but I wanted to take a look to be a 2nd/3rd set of eyes.

Although the noid test did show no signal for pins 45-52.