Two years of intermittent hard starts, please help!

[eacmen]

Well it's not so much that it controls the starter relay, but it effectively *is* the starter relay. To bypass it completely, you'd have to install a relay in its place (and if you still want a clutch interlock, you'd need a second relay to work based off the clutch switch)

Thanks Terra.

My only experience is on the ms43 EWS. On that one we just jumped two of the heavy gauge wires together and disabled ews in the ECU. Been rock solid for months now.

I suspected the M3 EWS was more complicated but didn’t find any good wiring diagrams.

Any refs you can share?

[EthanolTurbo]

Has anyone here tried replacing the oring on the driver side fuel transfer tube? I was having hard start issues, idle issues, loss of power and the car would outright die sometimes. My friend pointed out that the o-ring was expanding and causing a "leak." Well I replaced it today and no more hard starts or issues. Just a thought.

[terra]

You still need the EWS module when doing EWS delete since it controls the starter relay (?).

The EWs delete just means that the ECU and EWS don’t need to be paired.

There are two wires you need to jump in order to fully delete the EWS I think.

Well it's not so much that it controls the starter relay, but it effectively *is* the starter relay. To bypass it completely, you'd have to install a relay in its place (and if you still want a clutch interlock, you'd need a second relay to work based off the clutch switch)

[M3_CJ]

Want to chime in here, also still experiencing this issue still, although it hasn't gotten any worse. I've not been driving mine lately as life got busy (currently at around 68k miles). Have not done anything other than an oil change so far.

I'll keep subscribed to this thread and wish you the best of luck!

[ATB88]

You still need the EWS module when doing EWS delete since it controls the starter relay (?).

The EWs delete just means that the ECU and EWS don’t need to be paired.

There are two wires you need to jump in order to fully delete the EWS I think.

Ah okay, that makes sense. I guess I'll try that if I can't get any other ideas :/

[eacmen]

Hmmm tried to do this with the ECUWorx tool, but not sure if it actually worked properly. Pulled my partial binary with BMWFlash, loaded it into the ECUWorx tool, checked the "EWS Delete + EWS Tampering DTC Delete" box, applied and wrote the changes to a new file, and flashed the newly modified partial back to my ECU.

Car starts fine, but, then, if I unplug the EWS module, it won't start at all. So that means I didn't code it out?

You still need the EWS module when doing EWS delete since it controls the starter relay (?).

The EWs delete just means that the ECU and EWS don’t need to be paired.

There are two wires you need to jump in order to fully delete the EWS I think.

[ATB88]

My gut says electrical gremlin, especially given how intermittent it is.

My buddy's 330 zhp has issue current where it just won't start, replaced the starter, everything else, still starting intermittently. Live data for the dme/ews show ignition, ews unlock, clutch switch registers the pedal press and you can hear the relay click. Sometimes it starts, sometimes it doesn't.

I don't know how but we figured out when it doesn't start if you turn the key off, hold the key on start and put pressure on the starter interlock fuse it starts right up. Fuses are fine, harness is fine, pins aren't bent, continuity is fine, everything checks out.

Maybe code out the ews and clutch switch and see if anything changes. Definitely helps to narrow the window anyway.

Sent from my SM-G965U using Tapatalk

Hmmm tried to do this with the ECUWorx tool, but not sure if it actually worked properly. Pulled my partial binary with BMWFlash, loaded it into the ECUWorx tool, checked the "EWS Delete + EWS Tampering DTC Delete" box, applied and wrote the changes to a new file, and flashed the newly modified partial back to my ECU.

Car starts fine, but, then, if I unplug the EWS module, it won't start at all. So that means I didn't code it out?

[ATB88]

My gut says electrical gremlin, especially given how intermittent it is.

My buddy's 330 zhp has issue current where it just won't start, replaced the starter, everything else, still starting intermittently. Live data for the dme/ews show ignition, ews unlock, clutch switch registers the pedal press and you can hear the relay click. Sometimes it starts, sometimes it doesn't.

I don't know how but we figured out when it doesn't start if you turn the key off, hold the key on start and put pressure on the starter interlock fuse it starts right up. Fuses are fine, harness is fine, pins aren't bent, continuity is fine, everything checks out.

Maybe code out the ews and clutch switch and see if anything changes. Definitely helps to narrow the window anyway.

Sent from my SM-G965U using Tapatalk

Yeah, not a bad suggestion! I had thought about coding out the EWS at some point to see if that changes anything. I guess it's time.

[exodus454]

My gut says electrical gremlin, especially given how intermittent it is.

My buddy's 330 zhp has issue current where it just won't start, replaced the starter, everything else, still starting intermittently. Live data for the dme/ews show ignition, ews unlock, clutch switch registers the pedal press and you can hear the relay click. Sometimes it starts, sometimes it doesn't.

I don't know how but we figured out when it doesn't start if you turn the key off, hold the key on start and put pressure on the starter interlock fuse it starts right up. Fuses are fine, harness is fine, pins aren't bent, continuity is fine, everything checks out.

Maybe code out the ews and clutch switch and see if anything changes. Definitely helps to narrow the window anyway.

Sent from my SM-G965U using Tapatalk

[ATB88]

I HAVE NOW BEEN CHASING THIS PROBLEM FOR NEARLY TWO WHOLE YEARS :'<

Would like to revive this thread to get a fresh set of eyes on it. Edited first post to reflect additional replacements made. In the last year have replaced a few more things: Alt, starter, battery. Absolutely no change in behaviour. Will be installing a new ignition switch later today because at this point, why the f not??

Also, the issue has worsened a bit: every once in a while (not often), it will actually refuse to start. If I stop cranking and try again, it then starts right up.

For an engine that runs as well as mine does, and has been as well taken care of as mine is, this hard start thing really is a total mystery, but it really truly is killing me inside. I would have no problem taking it to a pro to diagnose it, but only if I knew they were going to get creative and go the extra mile in trying to figure it out -- i'd brought it to a couple places but they ended just saying "well, you replaced everything I would have tried, so, idk"..

Absolutely willing to entertain any ideas from anyone who has read through what I've tried!

As we saw in the videos I posted earlier in the thread, fuel pressure looks fine during the (barely) stumbley starts that I was able to get on camera. This makes me wonder if it's some sort of a weird electrical gremlin, and, if so, how to go about chasing/figuring that out?

Here's a current theory (a total reach, but a theory nonetheless): this all started after I did some VANOS/re-timing work (and yes, VANOS pass test with flying colours). Could be a coincidence, could be related. One possibility: I somehow manhandled the cam impulse sending wheel (at the back of the cam that give magnetic signals to the cam sensors) and warped a finger or something, and when that finger is in a certain position at startup it causes weird behaviour? Again, a total reach, but seriously, what else could it be at this point??

[yhp2009]

Would be interested to see the starter replaced as well. Seems like the first time hard starts is a more common issue among owners lately

[mmobmw]

Excellent. Thank you for sharing! I'll try it next time I have the intake off (probably next week). At least if I do end up selling this thing I'll have be able to advertise exquisite maintenance lol.

Good luck and I’m sorry you replaced so many things without success, but at least your engine is in mint condition now!

[ATB88]

Yes, mine sounded just like yours and I didn’t notice it until it regularly would not start on the first try. And then it got more sluggish over several months, then wouldn’t start at all. I think the hard starts are symptoms of an aged starter. Mine was 17 years old when I replaced it with the same Valeo unit you mentioned.

Excellent. Thank you for sharing! I'll try it next time I have the intake off (probably next week). At least if I do end up selling this thing I'll have be able to advertise exquisite maintenance lol.

[mmobmw]

This is interesting, and possibly promising.

May I ask, did your hard/sluggish starts sound like mine do in the youtube videos I posted on the previous page? A couple of people have told me it doesn't sound like a starter problem, but, if yours sounded similar in behavior before your starter completely gave up then I'd be definitely be willing to look into the starter! $160 for a valeo unit from FCP... not tooo bad

Yes, mine sounded just like yours and I didn’t notice it until it regularly would not start on the first try. And then it got more sluggish over several months, then wouldn’t start at all. I think the hard starts are symptoms of an aged starter. Mine was 17 years old when I replaced it with the same Valeo unit you mentioned.

[ATB88]

Just out of curiosity have you gotten a smoke test done? It's possible there's a leak somewhere that's throwing things off on the intake or exhaust side.
If the CSL software runs rich out of the box I'd expect to see negative fuel trims.

If you reset all the adaptations does the problem still occur?

If you look at the air/water/oil temps when the car cold, do they all match ambient temps? Does the baro and MAP match ambient? What's your RF at idle?

Good questions. Yes, I've had a few smoke tests done, no leaks found. Also switched to a CSL intake recently, so, I would suspect that if there was a leak in the intake system before, switching to CSL would have at least changed behavior. Pretty much all air/pcv hoses are new (less than year old), except for the sucking jet pump lines (going from air rail to brake booster). I have a new one that I'll be installing next time I get in there.

Fuel trims: so I currently get between -2 to +1 on bank 2, +1 to 5 on bank 1 (delta between banks is consistent with what I was seeing before switching to CSL, was 5-7ish bank2 and 7-11ish bank1, it's just shifted richer now because CSL runs rich out of the box)

I'll take a look at all those sensor temps next time I drive the car cold! Haven't gotten testo set up to log the MAP sensor yet -- but this behavior existed long before I switched from MAF to MAP, so I would assume the MAP's okay. Still plan to take a look at it though.

What do you mean by "RF" at idle?