Just one bank of injectors working
Just one bank of injectors working
Author
Discussion

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
Am having problems with injectors. Have installed a Megasquirt to run fuel and spark. But only one injector bank seems to be working properly. Have posted same question on Megasquirt forum, but thought I'd try here in case someone knows.

Have connected up the Megasquirt with an adaptor cable to link the ECU into the existing loom and run new cables where necessary.

The car now idles, but only one bank of injectors is working (normally). The other bank appears to not be working at all (of if so, is squirting very little fuel)

I have tried changing all the settings everywhere to no avail. One set of injectors don't fire with group fire or bank fire configuration.

The ECU appears to be outputting a signal (can get LEDs to flash on bench), and it seems the injectors on the car are receiving a signal, as shown by alternating voltage on DMM, but one bank doesn't seem to fire. The bad bank has come good twice for unexplained reasons, then died again.

Have spent entire weekend scratching head - can't understand why there is a signal on the injectors but they won't fire (spark is definitely there)- tried switching the injector bank outputs from the Megasquirt and the bad bank ran perfectly, so definitely not a problem with the injectors themselves. Could the injector signal be too weak?

Thoughts are:

1. Wrong setting in software - think I've tried everything though
2. Bad trigger signal? - can log trigger input, but display is not as good as seen in documentation. Would this prevent injection events.
3. Dodgy injector transistor inside ECU - can run an LED, but struggling with real injector.
4. Why is the fault intermittent - have checked all wiring good and solder joints fine.

Any ideas?

spend

12,581 posts

282 months

Monday 11th October 2010
quotequote all
Check there is +12v on each bank from the fuel relay feed first, there is a splice in the loom that can 'break' particularly if you have been manhandling the original loom.

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
Thanks Spend.

12V is there - been checked.

The injector signal seems to be there too. Don't have an oscilloscope, but a multimeter shows a varying signal which looks the same as being received on the good bank. (MS shows the same pulse widths for each bank too). And as described - if I connect the bad bank to the known good output, the injectors fire fine - and yet both banks seem to be receiving the same signal but only one side chooses to work.

MikeTVR

167 posts

314 months

Monday 11th October 2010
quotequote all
Check that the 12V remains solid at the injector even when the engine is running. If there is a high resistance in the 12V feed the voltage will appear OK under 'no load' conditions.
Mike

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
MikeTVR said:
Check that the 12V remains solid at the injector even when the engine is running. If there is a high resistance in the 12V feed the voltage will appear OK under 'no load' conditions.
Mike
Tried that too, but will check again. When injection bank signal is swapped they fire, so implies that the wiring for signal and ground to the injectors is fine?

Meatball

1,638 posts

241 months

Monday 11th October 2010
quotequote all
Teetertank said:
Am having problems with injectors. Have installed a Megasquirt to run fuel and spark. But only one injector bank seems to be working properly. Have posted same question on Megasquirt forum, but thought I'd try here in case someone knows.

Have connected up the Megasquirt with an adaptor cable to link the ECU into the existing loom and run new cables where necessary.

The car now idles, but only one bank of injectors is working (normally). The other bank appears to not be working at all (of if so, is squirting very little fuel)

I have tried changing all the settings everywhere to no avail. One set of injectors don't fire with group fire or bank fire configuration.

The ECU appears to be outputting a signal (can get LEDs to flash on bench), and it seems the injectors on the car are receiving a signal, as shown by alternating voltage on DMM, but one bank doesn't seem to fire. The bad bank has come good twice for unexplained reasons, then died again.

Have spent entire weekend scratching head - can't understand why there is a signal on the injectors but they won't fire (spark is definitely there)- tried switching the injector bank outputs from the Megasquirt and the bad bank ran perfectly, so definitely not a problem with the injectors themselves. Could the injector signal be too weak?

Thoughts are:

1. Wrong setting in software - think I've tried everything though
2. Bad trigger signal? - can log trigger input, but display is not as good as seen in documentation. Would this prevent injection events.
3. Dodgy injector transistor inside ECU - can run an LED, but struggling with real injector.
4. Why is the fault intermittent - have checked all wiring good and solder joints fine.

Any ideas?
Its a software setting issue, i had the exact same thing last week with my megasquirt.
I was getting no pulse signal on the driver side bank.
I can send you a copy of my tuner studio settings if you want to cross check it.

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
Thanks Ben,

Reckon I've tried every setting and combination possible in Tunerstudio.

But a comparison with yours would give peace of mind. Will send mail.

I said before that the bad bank is receiving a signal - but is it possible that the signal is coming from the good bank, as they are all connected?

blitzracing

6,422 posts

251 months

Monday 11th October 2010
quotequote all
Can you quantify how the injectors are fires on the MS setup? The 14CUX puts fixed 12 volts on one side of the injectors, and then grounds the otherside when it wants the injectors to fire. In this case you would see fixed 12 volts on one side, and switching from 12 volts to .7 (or so) volts on the other as the injector is earthed when its fired. It would be quite obvious if you have a high resistance or poor switching transistor somewhere as the voltages would not swing through this range, so its important to know these values. If the voltage is moving significantly on the 12 volt side, then you have a high resistance in the supply to the injectors.

spend

12,581 posts

282 months

Monday 11th October 2010
quotequote all
Teetertank said:
When injection bank signal is swapped they fire, so implies that the wiring for signal and ground to the injectors is fine?
What exactly do you mean by that?

daxtojeiro

743 posts

277 months

Monday 11th October 2010
quotequote all
Hi,
send me the msq file and I will ensure its setup correctly. Also send me a log of the engine running. Where did you get the MS from, really this is an issue the guy you purchased it from should help you with. If its one of mine then it would have all been setup correctly for your application with a base map, so it wouldnt be a configuration issue.
cheers,
Phil

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
blitzracing said:
Can you quantify how the injectors are fires on the MS setup? The 14CUX puts fixed 12 volts on one side of the injectors, and then grounds the otherside when it wants the injectors to fire. In this case you would see fixed 12 volts on one side, and switching from 12 volts to .7 (or so) volts on the other as the injector is earthed when its fired. It would be quite obvious if you have a high resistance or poor switching transistor somewhere as the voltages would not swing through this range, so its important to know these values. If the voltage is moving significantly on the 12 volt side, then you have a high resistance in the supply to the injectors.
The injectors on the MS are fired in a similar way to how you describe - i.e. switched ground. The voltages observed are about the same as you quote.

Resistance in cables has been checked, and no high values found.

The 12V supply isn't sagging either.

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
spend said:
What exactly do you mean by that?
The MS has four injector outputs (MS pins 32, 33, 34, 35) these are connected to two pins on the Lucas loom (14CUX 11 & 12). Lucas pin 11 supplies bank 1, pin 12 bank 2.

So I have:
MS pin 32 & 33 connected to 14CUX loom pin 11 (bank 1)
MS pin 34 & 35 connected to 14CUX loom pin 12 (bank 2)

To answer you question, if I swap the above connections at the output from the ECU, so that the MS output for bank 1 is now wired to bank 2 - the other set of injectors fire up fine, and the previously good side goes bad. This implies that there is no dodgy connections or high resistance in the cabling feeding them - have measured too).

Yet, I can measure the same signals that Blitz talks of on both injector banks, even though only one side will fire.

Hope that makes sense.


Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
daxtojeiro said:
Hi,
send me the msq file and I will ensure its setup correctly. Also send me a log of the engine running. Where did you get the MS from, really this is an issue the guy you purchased it from should help you with. If its one of mine then it would have all been setup correctly for your application with a base map, so it wouldnt be a configuration issue.
cheers,
Phil
Thanks Phil, files sent.

MS purchased used, but has the base map you installed when modding the board.

daxtojeiro

743 posts

277 months

Monday 11th October 2010
quotequote all
Teetertank said:
spend said:
What exactly do you mean by that?
The MS has four injector outputs (MS pins 32, 33, 34, 35) these are connected to two pins on the Lucas loom (14CUX 11 & 12). Lucas pin 11 supplies bank 1, pin 12 bank 2.

So I have:
MS pin 32 & 33 connected to 14CUX loom pin 11 (bank 1)
MS pin 34 & 35 connected to 14CUX loom pin 12 (bank 2)

To answer you question, if I swap the above connections at the output from the ECU, so that the MS output for bank 1 is now wired to bank 2 - the other set of injectors fire up fine, and the previously good side goes bad. This implies that there is no dodgy connections or high resistance in the cabling feeding them - have measured too).

Yet, I can measure the same signals that Blitz talks of on both injector banks, even though only one side will fire.

Hope that makes sense.
Errm, Lucas pin 11 and pin 12 are not for the injectors, pins 11 and 13 are wink
Phil

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
daxtojeiro said:
Errm, Lucas pin 11 and pin 12 are not for the injectors, pins 11 and 13 are wink
Phil
Well spotted. Yes, should have typed 13. Unfortunately the MS is wired up to 11 & 13, so not the easy fix it could have been.

spend

12,581 posts

282 months

Monday 11th October 2010
quotequote all
Sounds like its in your MS rather than the car/wiring, so Phil should be able to sort you out.

Teetertank

Original Poster:

358 posts

218 months

Monday 11th October 2010
quotequote all
spend said:
Sounds like its in your MS rather than the car/wiring, so Phil should be able to sort you out.
Agree. Yet both injection outputs fire the LEDs on the stimulator perfectly when off the car banghead
If it wasn't for that, then I'd say for sure the problem is inside the MS.

Does anyone know enough about electronics to say if a MOSFET could be fit enough to light an LED but not real life injectors?


daxtojeiro

743 posts

277 months

Monday 11th October 2010
quotequote all
Hi,
the MS has flyback transistors that switch in during the spikes off the coil as the injectors are switched off. The LEDs on the stim wouldnt test these. Im afraid it looks like a poorly built MS, much better to buy one from a reputable place with a warranty smile
cheers,
Phil