Starter Motor Interfering With Engine Management

[jackherer 543]

Four years ago, almost to the day, my Mi16 broke down and I couldn't get it running again for a long time, despite having a lot of spare parts and other running Mi16s to swap parts from etc.

 

I tried everything, even a different loom and different fuel as I was getting desperate. In the end the starter motor burnt out, I assumed it was caused by turning the engine over repeatedly to try to get it running but once I fitted a new starter motor the engine unexpectedly ran fine again. I never really understood how the starter could prevent the engine from running yet still turn it over at a good speed, I can only assume it was generating a lot of electrical noise perhaps?

 

Anyway, a few days ago I was looking at Ian's (Maturin23) Mi16 and I soon realised it was behaving just as mine had, it was turning over at a good speed, had compression, was timed correctly, had a strong spark and (a bit too much) fuel was being injected but wasn't firing at all apart from the odd stutter and backfire.

 

It seemed to be over fuelling a bit as the plugs were fouling up rapidly so I checked the obvious suspects first, i.e. the AFM and temp sensor but it was the same with known good replacements and the connectors and wiring checked out fine.

 

Having seen the same symptoms before I had to try another starter and (after fitting one that just clicked ) it fired up and ran well. There were some other issues but once I put the new starter on the over fuelling was gone never to return.

 

This is the old thread when it happened to mine: http://forum.205gtidrivers.com/index.php?s...c=63132&hl=

[maturin23 10]

I have to thank jackherer Holmes for solving the unsolvable

 

There's no way me or any garage would have considered changing the starter - it was turning over the engine fine and looked fairly new and shiny. I'd probably have paid a few hundred quid and be no further on.

 

To be honest it doesn't make much sense why it would affect the fueling - but the problem was immediately sorted. Does anyone have any thoughts why?

 

It took a little longer to find the frayed coil live feed that was also intermittantly cutting the ignition completely (but never on the driveway!) but he found that too in the end.

[steve@cornwall 100]

I ran a Bedford CF camper years ago that developed the same symptoms (ok it was an old, old carb only engine) turned out it had a relay on the starter that prevented all the available battery power from going to the starter and not leaving enough for a strong spark. I would guess that a starter with slightly decaying windings could well draw more than its share of available current? . The Cf would bump start in a few short feet, but even with the battery fully charged, would not fire off the starter.

 

If your vehicle will bump, but not start on the key I guess its a good starting point to change the starter, not something I'd have thought of without your post, even withmy previous problem!

[jackherer 543]

I am pretty certain it wasn't drawing too much current, the voltage drop wasn't excessive and the spark seemed pretty strong. I can only imagine it was generating a lot of electrical noise which was interfering with the signal from either the AFM, ECU temp sensor or the crank sensor.

 

Also in my experience the Mi16 ECU will still fire at very low voltage, sometimes the battery can be barely strong enough to turn the engine but it will still catch and fire after just a couple of revolutions. Modern cars seem much worse, my mates Alfa 166 V6 wont do anything unless the battery is fully charged, it even seems to hold the valves open somehow, when the battery gets too low it actually spins faster with little or no compression, it was quite worrying the first time it happened as I assumed the timing belt had snapped.

 

Ian told me he did get a tow start at one point but it took a long time to fire, presumably because it was flooded from over fuelling while trying the starter beforehand.

[Anthony 1,003]

Top problem solving there, as I (like most people I suspect) would have never associated a starter with causing issues like that!

 

Reminds me of a oddball problem I had with my 205 1.9 Mi16 back in the day, that stumped me and baffled people on here from what I recall. The car ran and performed very well, but then one day, started periodically flooding itself when trying to start it, and killing dizzy caps at an alarming and ruinously expensive rate (about every 100-200 miles!) with what looked to be arcing damage inside . I changed everything that anyone thought could have been related, and even checked things like cam runout to see if that was behind the cap failures, but the breakdowns persisted and I pretty much gave up on the car, never owning another Mi since.

 

That eventually turned out to be an oddball alternator fault, which I never did truely get my head around (it charged at a steady 14.2v and appeared to work fine) but I'm wondering if that was a similar noise related issue that was upsetting the ECU, causing the spark timing and injection pulses to go awry.

 

The same engine and componants are still running to this day in a friend of mines car, and never again did it overfuel/flood or kill another dizzy cap once the alternator had been replaced. Strange but true....

[jackherer 543]

That is interesting, I remember discussing that dizzy cap issue with you at one point but I didn't realise it was over fuelling too.

 

The more I think about it the more I suspect it is the crank sensor signal that is being interfered with but even if you use an oscilloscope to check it and see some noise you still wouldn't suspect the starter, or in your case the alternator. If I'd seen noise on a 'scope I think I would probably have tried changing the shielded wire between the CPS and ECU which would have masked the symptoms at best depending on the quality of wire I used

 

Ian, have you driven it much? Is it still running right?

[maturin23 10]

I've done my shamefully short commute in the rain three times, plus a 15 minute detour.

 

No problems restarting, no missing, pulling very well and no more 'hunting' on idle. Smells a lot less like an Esso station in the work carpark after I arrive too.

 

Feels absolutely spot on.

[hcmini1989 16]

Got a very similar problem at the minute .We have an early evo in at work and its missing every say 10 clyles of the engine really weird i have checked evey conceivable part on it cant find anything ,i have allso changed all sensors,ecu ,injectors

 

I have had a wideband wired in and every time it starts to miss it sends the fueling all over the shop like its getting noise in one of the signals ,And when the fan kicks in it goes mental pops and bangs etc

 

Does anyone know how to check if theres noise in any of the sensors with out using an osciloscope?

[jackherer 543]

Excellent.

 

pulling very well

 

It is definitely a good one, as good as standard Mi16s get. It sounded pretty good when you drove off the other day too

[jackherer 543]

Does anyone know how to check if theres noise in any of the sensors with out using an osciloscope?

 

I don't think there is any other way, although you can get very cheap 'scopes that connect to PCs these days. But I'm not sure how much help that would be in actually tracking down the source anyway.

[hcmini1989 16]

I don't think there is any other way, although you can get very cheap 'scopes that connect to PCs these days. But I'm not sure how much help that would be in actually tracking down the source anyway.

Will have a gander on flebay .I have only seen the 5v ones .Im sure there out there its finding where the noise is coming from though .All though as you say would probably be easier to fit/change the sheild on the culprit sensor

[GLPoomobile 958]

That's one hell of a stroke of luck having Keiron (sp?) on hand, and that he'd experienced the problem before! Something like that is a real interest killer, and as you've all said, it's highly likely you'd have never got to the bottom of the problem.

 

It makes me wonder how many others have had this problem without knowing. These symptoms seem to crop up quite regularly in member's topics, so I reckon this will be a turning point in the forum troubleshooting process - i.e. if you have a turning starter, fuel and spark, but the engine won't fire, first check will now be to see if it'll bump start. I'll certainly be filing this one away in the old grey matter

[maturin23 10]

Getting Kieran to look everything over was a true stroke of genius on my behalf

 

With the risk of sounding like an arse kisser he really does seem to have an extra bit of his brain dedicated to fixing 205s.

These cars are getting on now, regular garage mechanics just don't have the specific experience of their quirks - fortunately Kieran does.

[pug_ham 244 3 Cars]

Nice detective work Kieran, certainly thinking outside the box on that one.

 

It must've been givng off quite a lot of noise to get past the CPS shielding & cause interference though imo unless the sheilding was damaged where it runs near the starter or maybe just due to the close proximity of the CPS plug.

 

Graham.

[jackherer 543]

I've been trying to research this further as I would like to understand it properly and I found a very interesting article:

 

http://www.motor.com/article_pdf_download....article_ID=1007

 

Our third example (Fig. 6) is a 1995 Volvo 850 with an intermittent no-start problem. Occasionally during the no-start, if the ignition key is left in the ON position, the vehicle will burn up the ignition coil. When the coil has melted, the intermittent no-start turns into a permanent no-start. The 850 is then towed to a repair facility. The module and ignition coil are replaced and the vehicle is sent back out on the road, only to be towed back in to the shop at a later date. The ignition coil is melting due to the module not turning off the ignition coil driver. This allows the current to continue to flow, causing the ignition coil to overheat and melt.

 

To find the cause of this problem, an oscilloscope was connected to several circuits of the engine control system. The yellow trace was connected to the wire that powers the ignition module, the green trace was connected to the ignition module ground circuit, the blue trace was connected to the ignition coil control circuit and the orange trace was connected to the injector control circuit. The data was collected under cranking conditions.

 

Notice that all of the signals have heavy electrical noise riding on them until just after 4.1 seconds of time (bottom of chart). The waveforms to the right of this point show no electrical noise at all. The 4.1-second mark is where the starter motor disengaged and stopped turning. This electrical noise is so severe that it has affected the PCM's internal clock, allowing the signals to become out-of-sync. The PCM output to the ignition module then locked the coil driver circuit high. In turn, this caused the module transistor to stay on, allowing the current to create enough heat to melt the ignition coil. The repair for this no-start 850 was to replace the starter motor.

 

Different symptoms but the same cause I am almost certain.

 

I need to buy a decent 'scope...

 

Edit: Another similar one - http://www.honestjohn.co.uk/forum/post/index.htm?t=66222

 

The starter was creating a magnetic field and stopping the fly wheel sensor from collecting the information from the crank sensor. Hence not firing on the key.

 

Edit2: It seems to affect a variety of vehicles - http://www.carelect.demon.co.uk/rrdisco2.html

 

We have had reports from several sources now that Discovery TDi 5 from 1998 to date are suffering starting problems which may be related to electrical noise radiating from the starter motor cable and interfering with the vehicle immobiliser. The fix appears to be fitting new starter motor even though it appears ok and cranks at full speed.

 

If you have any experience of this or can throw any light on this problem please email me from this page.

 

SYMPTOMS

 

Engine cranks AT FULL SPEED but does not start. There may have been a history of intermittent difficult starting.

 

DIAGNOSTIC CHECKLIST

 

1) Confirm starter crank speed above 300RPM, if necessary fit new battery.

2) Confirm vehicle will start easily from push start or tow (DO NOT PUSH OR TOW AUTOMATICS!). if yes, try replacing starter motor with new.

Edited August 12, 2010 by jackherer

[Rob_the_Sparky 9]

Wow - I'm impressed you found that one...

 

Old serial motors can certainly draw quite a bit more current than good ones. I guess you could be right, it is just the starter. I can't help but ponder whether it could be due to a not ideal 0V connection. My guess is that the ECU 0V is from the body work but the sensors get 0V from the engine block. The starter is going to get through an awful lot of amps so will cause a significant voltage drop across any 12V or 0V connections. The 12V connection is dedicated to the starter but the 0V is not so this coudl combine with the higher than normal currents to screw up the sensor readings.

 

Having said all electrical noise issues an absolute bar steward to diagnose, even in the world of micro electronics, so treat the above as only a guess. As soon as you use a scope (or anything else) to test for noise you change the impedance of the thing you are testing, changing the noise!

 

Anyway, congratulations on finding it, I'll store that one away in the back of the head for later use!!

 

Rob

 

P.S. Could it be that the conversion looms for the Mi16 aren't the best leaving these engines more liable to a noise problem? Getting a harness to work is a different thing from getting the signal quality right...particularly the shielding of cables.

[maturin23 10]

Good work Kieran.

 

Just to add a little more detail the starter was a nearly-new one from a diesel 309. All the earths/battery leads have been recently replaced.

[jackherer 543]

Rob, I certainly wouldn't entirely rule out a 0V connection issue, I was reading about a similar problem that was solved by removing the gearbox and cleaning the bellhousing > engine block mating faces!

 

If that was the cause for mine and Ian's cars it could only be the starter motor to gearbox mating faces or bolts I guess. As Ian says, his grounds are spot on and I replaced *everything* on my car four years ago. However when it happened to mine the starter eventually burnt out quite badly so I do think it was faulty in some underlying way.

 

I don't think it is due to the looms being more susceptible to noise because they've been converted for 205s, Ian has a 405 loom and my car had a BX loom which is wired up very differently and earthed in different places. I also changed my M1.3 loom for an ML4.1 (both BX) to rule out any wiring issues.

[roote 0]

i will also be keeping this one in the back of my head as ive just solved my problems today and been doing a lot of research on here for various solutions, and as its quite a good artical will it be put in to the arcives ?

[Kitch 4]

Interesting reading. I've got a similar problem on the BX. It'll crank over happily, it's got fuel, it's got a spark (if I pull a plug out and hold to the engine while cranking) and I'm pretty sure it's got air

 

Won't start though. It cranks over, sometimes the speed of which accelerates as the engine tries to catch. When you give up, you get a couple of back puffs from the air filter.

 

The relays are fine, the resistances at the ECU plug are fine (except the CTS one which wouldn't read) and many of the components are new. The coil is only 18months or so old, yet when checking the resistances listed on CAPS I should have had 0.8ohms across pins 2-3 and 6500 across 2-4. What I actually had was 1.2 across 2-3 and zero across 2-4. Now theres a chance I'm doing it wrong, but if thats the right readouts, theres a chance my coil's died. Now did it die because I've got a cheap s*it one, or has my starter now killed two on the trot? If they're being fried because the starter is scrambling the signal from the amp, maybe the spark is strong enough to jump from plug to engine when out for testing, but struggles once wound back in under compression?

 

Really interesting thread, thanks for pointing it out.

 

Cheers

[jackherer 543]

Won't start though. It cranks over, sometimes the speed of which accelerates as the engine tries to catch. When you give up, you get a couple of back puffs from the air filter.

 

That does sound like this issue...

 

However:

 

the resistances at the ECU plug are fine (except the CTS one which wouldn't read)

 

That will be causing your ECU to massively overfuel which has similar symptoms. Do you get a sensible reading on the CTS itself? If so check the pins in the blue plug are secure then follow the wiring back from there.

 

maybe the spark is strong enough to jump from plug to engine when out for testing, but struggles once wound back in under compression

 

I think the starter only causes interference when it is working against compression, when the plugs are out the starter spins easily and doesn't make as much noise.

[kyepan 291]

this is really interesting, thanks for sharing guys!

[C_W 3 1 Cars]

Very bizarre. I suppose bump starting it is one way to check if it's the starter causing the problem? Ie if it bump starts easily then it could be that.

 

I think the injection system will run on down to around 9v so definitely never too much drain, if it was that the car would not turn over anyway.

[jackherer 543]

Bump starting should work but the plugs end up pretty fouled due to unburnt fuel so it might take a few goes before it fires. I mention Ian getting his one tow started in post #4 which worked but took a while.

 

You're right about the injection system working on low voltage, I have had the battery so low it is barely able to turn over but then it fires a cylinder which speeds it up a bit and then its off!

[Kitch 4]

Well it got much more interesting today. I borrowed a strobe gun from work and hooked it up to no1 lead to check if the plug is actually sparking in the head, not just in air. It was, happy days.

Hooked up to no2, and the bloody thing fired up! The only thing I did differently was to clip a strobe gun to the lead!!

 

So it's nice that it started, but it's also a PITA because I can't find the fault if it's running! It'll restart all day long when warm, it's when it's cold that it plays up so I'll wait until tomorrow and see what's what. All I've actually changed is the fuel pressure regulator since this issue came about, but since then I've fully charged the battery and re-done a few block connections. I wonder if my original problem was being masked by a slightly less than perfect battery?

 

Or I wonder if I do have some electronic interferance and the strobe gun somehow affected it for the good?!