Peugeot 207 Vti crank no start ongoing issue

[MikeT]

Thanks for this Mike. Right at the end the variables are missing, do you know if they were still showing the same values ?

I noticed that too and suspect there's no variables because we re-read the codes before trying to start the engine.

[ekjdm14]

I can't help but notice that one of the original variables was engine rpm "0". Are you (or rather, is the ECU) seeing a signal from the CPS on cranking?

I think in the absence of any other info other than the dodgy CTS supply voltage at the moment, I'd go back to first principles and check for a spark and injector pulses on cranking as its a complete non-start.

Even a completely messed up CTS reading should still allow the car to run even if it were a pain to start, so I suspect there's either something else at play causing the no-start or, as you mention, the ECU itself

[MikeT]

I'm thinking the opposite - that at an assumed -40c coolant temperature, it's injecting so much fuel (for an ambient +20c) it's flooding the plugs as wheeler says, and that's why we heard it try to fire but not catch.

As I said, I've got the Peugeot diagrams but not sure of your cars particulars to accurately find the pertinent ones, but if I have guessed correct, we want to test pin 32 for signal and pin 14 for ground.

[max1]

Yes the ECU will read the temp sensor on crank/ignition, that's normal to calculate correct fuel mixture, no?

I have had confirmation from a few sources that a fault with the ECT or it's wiring can cause a non start in some circumstances, but yeah there could well be something else coincidental causing it. However, the fact remains I shouldn't have 3.3v where the 5v should be & this does APPEAR to cause the ECU to not see the sensor. Specs for this car state I should see between a 4.8v & 5.2v reference at the ECT sensor. Pins 32 & 14 are indeed the pins I have been testing.

Ill do some more wiring integrity tests tomorrow & take it from there.

[MikeT]

I'm free tomorrow if you could use a spare pair of hands.

[max1]

I can't help but notice that one of the original variables was engine rpm "0". Are you (or rather, is the ECU) seeing a signal from the CPS on cranking?

I think in the absence of any other info other than the dodgy CTS supply voltage at the moment, I'd go back to first principles and check for a spark and injector pulses on cranking as its a complete non-start.

Even a completely messed up CTS reading should still allow the car to run even if it were a pain to start, so I suspect there's either something else at play causing the no-start or, as you mention, the ECU itself

Sorry, i didn't read this right.

The CPS signal is not something we have looked at.

[max1]

Power for ECU now tested & is good.

Disconnected middle ECU connector, with ignition on probed ECT Vref & ground directly from ECU still measuring 3.3v. Disconnecting this connector disconnects all the sensors.

Connected back up & probed the two 3 wire sensors on the top right of the engine block, I don't know what these are, getting 5v reference from these ok. Can't access any other sensors as they are under the car.

Thanks again Mike for coming to help out yesterday.

Have ordered a new sensor just to be 100% sure the one from my old engine that I put in is not the problem, 3.3v remains a mystery though.

[MikeT]

I can't help but notice that one of the original variables was engine rpm "0". Are you (or rather, is the ECU) seeing a signal from the CPS on cranking?

I think in the absence of any other info other than the dodgy CTS supply voltage at the moment, I'd go back to first principles and check for a spark and injector pulses on cranking as its a complete non-start.

Even a completely messed up CTS reading should still allow the car to run even if it were a pain to start, so I suspect there's either something else at play causing the no-start or, as you mention, the ECU itself

Sorry, i didn't read this right.

The CPS signal is not something we have looked at.

I too must apologise for misreading ekjdm so my earlier follow-up post was completely inappropriate.

How can we check there is a CPS signal getting to the ECU during cranking? Non-destructive back-probing seems impossible. Can it be done with diagbox? Regarding the wiring diagram, the CPS is referred to as "Engine Speed Sensor" (1313) in this case?

Seems even spark-plug removal requires a special tool, none of my thin-walled, deep-sockets could fit in the recess and reach the plug. Something thinner and longer is needed. Are they 12mm hex?

BTW max, I've found the official battery disconnect procedure for your car and have copied it (MHTML file). Thought I could send it to you via PM but it won't accept the file's extension even when zipped.

And just to confirm, what were your findings when you tested the ECT's resistance?

While I have access to all these wonderful resources, I can't find voltage and/or resistance values anywhere.

[ekjdm14]

Not a problem lads, it's a reasonable oversight considering how similar the two abbreviations are. What a pain on the spark plug front as that'd at least tell you if there was a spark. I don't suppose it's possible to backprobe the coil pack LT connector and see if it's being triggered at least is it? One of the main reasons I don't like working on more modern cars as much, everything seems such a ballache to get at.

I *think* you should be able to see the engine's cranking speed shown in live data, although I'm not entirely sure what menu it'd be under. (based on what I've heard on diagnosis of HDi models at least, where cranking speed was cross-referenced with rail pressure). Just thought it worth ruling out anything simple before chasing tails around this low ECT reference voltage, even if that does seem a likely culprit.

[MikeT]

I've since read engine speed can be viewed live through Diagbox when cranking so that's handy (and easier than probing).

To recap, or at least they way I understand it, the ECU is putting out a 3.3v Vref instead of a stated 5v to the ECT.
Diagbox only shows the ECT circuit (which includes the ECU) as being at fault, though max also measured 3.3v at another (as yet unidentified) sensor and yet 5v is seen on another.
Is that a defnitive diagnostic (the ECU is at fault) or is it possible another circuit is pulling down the Vref voltage somewhere? We did measure 3.3v at the ECU pins with the relevant connector removed and two others connected to allow for power etc but I don't know if that's the correct way to measure it or if other tests can be done.

[ekjdm14]

That does sound to me like there's possibly something pulling the voltage down, my best guess at the moment would be (after confirming through Diagbox that the ECU is seeing the engine being cranked) to start unplugging & measuring voltages at all the sensors until you either find out how many are on 3.3v OR (hopefully) you unplug one particular connector and the 3.3's magically shoot up to 5.0v giving a clue as to where the issue may lie. Don't know where you guys stand on that plan but I'd say it's got to be worth a shot before condemning the ECU or it's supply.

[max1]

Thanks for the replies

That ECU connector that we unplugged to test the ECU pins directly was the connector that had all the other sensors on it as well - pretty sure anyhow & would need to confirm this somehow? So if this is the case this would have eliminated any other sensor pulling the voltage down.

[MikeT]

That does sound to me like there's possibly something pulling the voltage down, my best guess at the moment would be (after confirming through Diagbox that the ECU is seeing the engine being cranked) to start unplugging & measuring voltages at all the sensors until you either find out how many are on 3.3v OR (hopefully) you unplug one particular connector and the 3.3's magically shoot up to 5.0v giving a clue as to where the issue may lie. Don't know where you guys stand on that plan but I'd say it's got to be worth a shot before condemning the ECU or it's supply.

I'm all for it! Just watched an online class on auto-diagnosis that says exactly this ^^^ (they had 1.7v Vref on a couple of sensors and 5v on others following a non-start after an engine swap) and showed next steps testing of all sensors/circuits before blaming the ECU.

[MikeT]

Oh btw, just for fun, who'd like to speculate on this one?

This non-start vehicle used as the training class vehicle, after the fix eventually started and ran sweet but had a crank sensor fault logged in diagnostic which caused much confusion for the master tech. He looked at live data and saw an engine speed of 0 - with the engine running! He cleared the codes, but every time the engine was restarted, the fault code returned.

[ekjdm14]

Engine was able to start using signal from the cam sensor? Only other thing I can think is the crank sensor had a dual output and one disagreed with the other...