1987 3.2 Carrera Backdate project - Sladeys hotrod

[impmad2000]

...actually, pink, blue and is that a dark green ?

[Gary71]

Yes, let's have all the traces. I'm trying to guess which is which and failing!

I'm sure we have TPS, lambda, and maybe MAP in there somewhere!

[sladey]

OK blue=voltage
Red = RPM
Green = Load
Black - Engine Temp
Yellowy orange = Air Temp
Purple = Oxygen raw
Orange = pulsewidth
Pink = Advance
Grey = Idle duty


Gary = yes I'm using the new screened cable from the connection in the engine (which comes from the crank sensor) to the ECU - that's what I installed on Saturday

Not sure where the voltage is measured

[sladey]

I did a cut-out power run from 547 onwards, and the same thing again from about 552 onwards, then around 557.5 you see the first of those things where I back off slightly at around 3500 revs and it's just cuts out temporarily - that happened 5 or 6 times after that.

[sladey]

And I haven't shown voltage on the scale - but when it's flatlining it's around 13.6 - when it spikes it maxes out at around 15

[Gary71]

Assuming the voltage is simply measured at the board I would expect this to be stable. My megasquirt ECU kept cutting out and resetting due to voltage drop which Tim eventually traced to dodgy wiring (my work!)

I can't explain why it would occur at a certain rpm other than a mechanical resonance or alternator noise causing a reset.

[Gary71]

Could it be the voltage regulator on the alternator? You could experiment with running it from a dedicated battery (short term) for a stable supply.

[impmad2000]

In the diagnostics, is there a reset count and brown out count. If the voltage actually crashes to zero, and I'm not sure about the scale of the graph, then the problem is more to do with power supplies than crank trigger. It doesn't mean that noise or high voltage spikes from the crank sensor aren't causing the problem.
Engine temp and advance also drop too ? But I suspect this is just the effects of the power supply dropping out.
If the power supply is dropping like that, then you should see resets and brown outs in the diagnostics.
I don't know what the input stage of the Canems is like. Perhaps it can't cope with high voltage spikes, or something has been damaged to cause a susceptibility to this. Do Canems publish circuit schematics ?

[Jonny Hart]

So I was right, the problem you've got isn't the the problem you think you've got!

I agree, this looks like a power issue. As Tim says, if ECU loses power and then resets, it should store that history somewhere in the log.

With regard to the 12v feed, in your original post about this you talked about splicing into the loom. I'd start there. Check the grounding too.

[sladey]

Yes there is a reset count and a brown out count - I need to check those with the engine running and then create the fault - and see if the counts increase.

It's a bit late for that tonight so might try it on the way to work in the morning

Dunno on the schematics. I'm going to email David at Canems if work is less manic tomorrow

[sladey]

So I was right, the problem you've got isn't the the problem you think you've got!

I agree, this looks like a power issue. As Tim says, if ECU loses power and then resets, it should store that history somewhere in the log.

With regard to the 12v feed, in your original post about this you talked about splicing into the loom. I'd start there. Check the grounding too.

I didn't really splice into the loom - I replaced one shielded wire with another. There was a connection that joins three lots of shielding - I connected the shielding from my new cable onto that.

I will be able to check on the power loss thing but want to be able to do it with and without recreating the problem.

Are you suggesting trying a clean power supply to eliminate problems with the power supply?

[Gary71]

Are you suggesting trying a clean power supply to eliminate problems with the power supply?

You could experiment with running it from a dedicated battery (short term) for a stable supply.

Yup. But also making sure the connections are clean all the way to the board and the wires are broken internally.

[sladey]

Might try that with Ian's help at the week-end

[Gary71]

Sounds like a plan. At least if you have a stable source it gives you a better chance of working through any remaining issues .Best of luck

[corner shop]

Hi Mark - my motec install exhibited similar symptoms in Neil Bainbridge's hands. He couldn't get it to rev past 5500 initially. He ended up fitting a clewett crankfire kit - something to do with the number of teeth on the flywheel (129 standard, motec prefers 60). Not saying it's your problem, just saying

http://www.clewett.com/products/adapters/adapter.htm