E92 M3 - Metallic clunk / grind on cold start (With 3 Videos + Image of Noise)

[bradleyland]

Originally Posted by ///M Power-Belgium

1- When replacing the Vanos Gears the clunck noise disappears .
2- The Vanos Gears are replaced by exactly the same Vanos Gears as when the car left out of the Factory .
3- So when thinking logic , pull out the problem and put back in the same problem .
4- And so why was the noise gone now when replacing with exactly the same Vanos Gears ?
5- And 4 was my question ???

Valid question, but a question no one can answer until we know the specific mechanism that causes the noise.

We have one case of replacement resulting from a PuMA case being filed, but others have been through similar channels and nothing has happened. This sounds, to me, like appeasement. Amazing that BMW would go through the expense, but we don't know the full circumstance surrounding the replacement.

We also have no further information from that case where the gears were actually replaced. There was no indication that BMW found anything wrong with the Vanos gears that were removed. We could assert that BMW examined the gears, found nothing wrong, and is no longer approving replacements because of what they learned in that case. That's one of the functions of PuMA; to centralize knowledge and make sure responses are uniform across the service network.

Eliminating the noise is an outcome, but it's looking right past my assertion: that the noise may not indicate a problem at all.

[Beemdog]

I've heard a similar short-burst noise on Nissan 3.5L V-6s right after start-up which was attributed to timing chain tensioners (yes, this engine used chains and not belts). It also tended to occur when the car sat for a couple days. As with the M3, people drove the car with this noise for 100k miles with no other effects.

[anom3]

hi bradley.

good to see you back in this thread.

although we stand on different sides of this issue, there is no doubt you know what your talking about and i appreciate your input as im sure many others do.

although i am of the belief as well that this is vanos related (the puma case by m3po on here, sticking my head under the hood while gf starts the car and __CLEARLY__ hearing it from the drivers side vanos area, the 0.5-1.0 second delay between engine ignite and noise occurrence playing right into the vanos initialization the ecu does) its very valid to point out that yes, we do in fact only have 1 confirmed case of this being vanos related (m3po's case)...

i took your advice that you gave me about 1-2 months ago (its here... quite a few pages back)... instead of jumping at the vanos gears i wanted to have a 'professional' investigate it and find a cause on their own... i have been at the mechanics about 5-6 times over the last few months as i have been doing other upgrades / mods / small fixes. i have left the car there over night twice, and had 'longer' sessions where it sat cold for 5-6 hours.

not once did it do it there. not kidding. i was almost starting to think its related to their 'parking' procedure as they usually turn off the lights (to "0") and leave the key in the ignition. but it doesnt affect it when i do that on my own here... still get it... also thought maybe its temperature related as their bay is a bit warmer then the underground... but on the other hand over the 5-6 times i have tried to reproduce it there the temps have fluctuated quite a bit from ~20C start of summer, to 35C+ in the peak of summer, to 15-18C a week or so ago when i had my akrapovic installed... so i dont think 'ambient' temperature has a huge effect on it. it just wont do it when it sits at their shop.

so im doing as you suggested... not going to replace the vanos gears 'just cause'... that being said, i AM doing the rod bearings over the next 2-4 weeks and am going to have them change a few additional lower cost items when they do those... mainly the engine mounts, oil pump chain tensioner (it has one too btw) as well as have them "flush" the oil "lines" to make sure things are as clean as possible as well as having them have a look at the vanos chain as best they can from the bottom with the pan off. the cost of the above is fairly minor considering the front end and pan will be coming off anyways.

Originally Posted by bradleyland

Valid question, but a question no one can answer until we know the specific mechanism that causes the noise.

We have one case of replacement resulting from a PuMA case being filed, but others have been through similar channels and nothing has happened. This sounds, to me, like appeasement. Amazing that BMW would go through the expense, but we don't know the full circumstance surrounding the replacement.

We also have no further information from that case where the gears were actually replaced. There was no indication that BMW found anything wrong with the Vanos gears that were removed. We could assert that BMW examined the gears, found nothing wrong, and is no longer approving replacements because of what they learned in that case. That's one of the functions of PuMA; to centralize knowledge and make sure responses are uniform across the service network.

Eliminating the noise is an outcome, but it's looking right past my assertion: that the noise may not indicate a problem at all.

[bradleyland]

Really frustrating that it's not at least more consistent. It'd be nice to have the noise reproduced in the presence of a tech who is intimately familiar with the S65. I'd also really love to know/hear what the techs found in an examination of the Vanos gears pulled off m3po's car. Chances of BMW releasing that information is near zero though :\

[lockeed]

OP, so far, did you do a cold start while listening from under the engine compartment?

[anom3]

negative.

i have done numerous ones with my head under the hood, most notably with the air intake off.

i know noises can reverberate and all, but everything points to the drivers side vanos / drivers side valve cover.

Originally Posted by lockeed

OP, so far, did you do a cold start while listening from under the engine compartment?

[anom3]

ive mentioned this in a previous post but...

my mechanic was more then willing to pull the puma case for me so we can have a look.

i gave him the # on m3po's invoice (listed as puma #) and when they entered it, it did not exist.

unsure if this is related to us being in europe and m3po being in na.

Originally Posted by bradleyland

Really frustrating that it's not at least more consistent. It'd be nice to have the noise reproduced in the presence of a tech who is intimately familiar with the S65. I'd also really love to know/hear what the techs found in an examination of the Vanos gears pulled off m3po's car. Chances of BMW releasing that information is near zero though :\

[lockeed]

Originally Posted by anom3

negative.

i have done numerous ones with my head under the hood, most notably with the air intake off.

i know noises can reverberate and all, but everything points to the drivers side vanos / drivers side valve cover.

OK. I'd get my ears closer to the starter (prime suspect), engine mounts, exhaust hangers... etc... Just to be 100% sure before you start going "internal".

I might be wrong but from the sound I hear on your video, it doesn't seam internal to me...

[lockeed]

link not working...

[FrankieP]

I guess the use of "normal" on my previous post has started a new discussion.

When VANOS gears, cams, lifters have been sitting on the top of an engine for some time without use, it is "normal" for these parts to not be fully immersed in oil. When metal parts slap, rub, and grind together without proper lubrication some clanking will be heard during cold starts on any engine.

So yeah, normal. No failure, no issue. Metal is hard, once oiled, no noise.

[anom3]

just fyi, i know what a vanos sounds like after a day or two has passed and the oil had a chance to drain down.

i get that 'noise' as well. it sounds much different then this.

the non-oiled vanos has a very industrial growl to it.

this is much different, its quite "clack clack clack".

i know... describing sounds using words is like trying to explain the color red to a blind man...

but in my opinion, this is not the typical dry vanos growl that i have heard not just in my own car, but in other bmw's... most recently my friends 550i.

Originally Posted by FrankieP

I guess the use of "normal" on my previous post has started a new discussion.

When VANOS gears, cams, lifters have been sitting on the top of an engine for some time without use, it is "normal" for these parts to not be fully immersed in oil. When metal parts slap, rub, and grind together without proper lubrication some clanking will be heard during cold starts on any engine.

So yeah, normal. No failure, no issue. Metal is hard, once oiled, no noise.

[bradleyland]

Yeah, I agree that it's different than the traditional Vanos noise. It sounds very much like lifter knock to me, which doesn't sound like Vanos growl.

[anom3]

so i have a potential solution... its probably not what most of you will want to hear, and as far as how this affects DCT vehicles, i am unsure... but it appears to work in my 6MT.

so i noticed whenever im just idling at a stop light or whatever, and i throw in the clutch (clutch all the way down) the car will bump up the idle about 200 rpm... as if its getting ready for a gear to be engaged and some load to come. makes sense.

a few nights ago i was thinking, since MT cars are in essence started with the clutch in (clutch down, start button, ignition) maybe the car is giving it this 200 rpm right after start and this has something to do with the noise.

i then recalled that the CAS module has 3 settings that can be changed.

a) clutch down required for start
b) break down required for start
c) neutral / park required for start

so i coded my car over to in essence make it like an 'automatic'... to start my car now instead of applying the clutch, i need to apply the break and press the start button. i know, slightly dangerous... but im no dummy and wont make the mistake of starting the car 'in gear'. and i dont plan on letting anyone else drive my car.

since i coded this over (2 days ago), i have not heard the start noise once. bare in mind, in the past i have had it not make the noise a few cold starts in a row. but i have a feeling its not going to come back.

also worth mentioning, the 3rd setting, park / neutral, doesnt work... i tried to code it, but the car wont start. im guessing its actually "park" not park (as in dct) / neutral (as in 6mt) as ncs dummy (for those that know) lists it as. or maybe the dct has a neutral "gear" that i am not aware of? i recall my old old (way back, 1994 era) civic had one.

gf and i have a nice 400km drive coming up today, will post weather the noise is back or not.

---

last but not least, i had bpmtune code me a tune with idle raised to 950rpm... thought maybe shuting the motor down with a bit more load and oil spraying around would help. it didn't. i have since come back to the stock idle, which after things are all warmed up appears to be about 680rpm (according to my perf steering wheel).

---

[///M Power-Belgium]

Originally Posted by anom3

so i have a potential solution... its probably not what most of you will want to hear, and as far as how this affects DCT vehicles, i am unsure... but it appears to work in my 6MT.

so i noticed whenever im just idling at a stop light or whatever, and i throw in the clutch (clutch all the way down) the car will bump up the idle about 200 rpm... as if its getting ready for a gear to be engaged and some load to come. makes sense.

a few nights ago i was thinking, since MT cars are in essence started with the clutch in (clutch down, start button, ignition) maybe the car is giving it this 200 rpm right after start and this has something to do with the noise.

i then recalled that the CAS module has 3 settings that can be changed.

a) clutch down required for start
b) break down required for start
c) neutral / park required for start

so i coded my car over to in essence make it like an 'automatic'... to start my car now instead of applying the clutch, i need to apply the break and press the start button. i know, slightly dangerous... but im no dummy and wont make the mistake of starting the car 'in gear'. and i dont plan on letting anyone else drive my car.

since i coded this over (2 days ago), i have not heard the start noise once. bare in mind, in the past i have had it not make the noise a few cold starts in a row. but i have a feeling its not going to come back.

also worth mentioning, the 3rd setting, park / neutral, doesnt work... i tried to code it, but the car wont start. im guessing its actually "park" not park (as in dct) / neutral (as in 6mt) as ncs dummy (for those that know) lists it as. or maybe the dct has a neutral "gear" that i am not aware of? i recall my old old (way back, 1994 era) civic had one.

gf and i have a nice 400km drive coming up today, will post weather the noise is back or not.

---

last but not least, i had bpmtune code me a tune with idle raised to 950rpm... thought maybe shuting the motor down with a bit more load and oil spraying around would help. it didn't. i have since come back to the stock idle, which after things are all warmed up appears to be about 680rpm (according to my perf steering wheel).

---

Need to say the coding is a good point of view , actualy you start your car now such as a DCT car, in other words a 6MT with a DCT start simulation !

DCT simulation which means pushing the brake pedal during start procedure , and exactly this creates the DCT brake procedure to avoid the clunck noise with DCT cars !

So actualy you are doing now the brake procedure , and need to say it still works for me 100 % of the time , every single time that i'm using the brake procedure ( about 30sec. pumping the brake pedal ) before pushing the start button I've never had the clunck noise again !

And of course I was testing to start the car without the brake procedure and the result was....the clunck noise was back !

BTW...You was mentioning " the DCT has a neutral "gear" that i am not aware of " Think that also is correct because when shut down engine with DCT in N (neutral) the DCT sets it in park !

Saw also you are doing a longer ride from 400km , when I drove longer trips I don't hear the clunck anymore for about 10-15 days , that's why I was saying on here our cars needs to be driven more , and so it's actualy a bad moment for you after the longer ride to test the clunck noise !

[bradleyland]

Originally Posted by anom3

so i noticed whenever im just idling at a stop light or whatever, and i throw in the clutch (clutch all the way down) the car will bump up the idle about 200 rpm... as if its getting ready for a gear to be engaged and some load to come. makes sense.

The throttle actually doesn't change position when you depress the clutch. Even when in neutral, a manual transmission puts some parasitic load on the engine. When the clutch is engaged, this parasitic load reduces engine RPM slightly.

Your ECU has a target idle RPM range (which is different for startup and normal operation) and a pre-set throttle angle for idle speed. If the RPM is within the specified range, the throttle stays at the preset angle. If it deviates from this range, throttle will be applied.

Just clarifying that. Anecdotally, I've noticed that the brake depress thing does appear to make the sound go away. I think it's vacuum related. Pressing the brake pedal creates an open circuit in the vacuum system, which will affect startup vacuum pressure.

[bmwnukid]

Originally Posted by anom3

I know this has been discussed a few times before, but I wonder if anyone had found a solution to this.

I have uploaded 2 videos.

I understand that a slightly noisy vanos for a second or so on cold start is normal as oil pressure needs to build up.

I am not talking about that, I am talking about the noise at 0:03 on the first video and at 0:04 on the 2nd.

Anyone have an idea of what this is?

A) ONLY happens on cold start. But even if it has sat for as little as 4 hours. Also worth mentioning, even if I "cold start" the car, wait 2 seconds, turn it off, and back on again, the noise will not happen again until next cold start (a few hours later).

B) Happens on EVERY cold start.

C) NEVER happens otherwise.

D) Car runs perfectly otherwise.

E) E92 M3, production 2007.07.

F - Update) Car has bpmsport.com tune with cold start delete enabled. So this is not related to the secondary air pump as this was happening before and after. Car also has 240E installed, previously had 50E. Noise was / is present on both.

G - Update #2) Updated a new video on post #16 of the thread ( http://www.m3post.com/forums/showpos...1&postcount=16 ) with a VERY clear sound of this grinding at 0:49.

H - Update #3) I isolated the Start -> Noise sequence using a spectrum analyzer tool for the PC. The image is attached as post #20 ( http://www.m3post.com/forums/showpos...5&postcount=20 ).

I - Update #4) *NOISE BACK, NOT THIS* Intresting development. See post #37 ( http://www.m3post.com/forums/showpos...4&postcount=37 )... Perhaps power related?!?

J - Update #5) Check valves on order, install early next week. Will report if it helps or not.

K - Update #6) Check valves changed, unfortunatly due to time, only 3 of 4 (last one gonna be done next early next week). Also, new development, engine temperature HAS an influence on if / how loud the noise is ( http://www.m3post.com/forums/showpos...7&postcount=69 ). Waiting 12+ hours as I write this update before testing if the check valves help or not.

L - Update #7) Previous "run cycle" temperature has a clear effect on the loudness of the noise. If on previous start / stop the motor / oiil doesnt warm up, the next time the noise is expected (6+ hours later), the noise is almost not there. See post video in post #84 ( http://www.m3post.com/forums/showpos...5&postcount=84 ).

M - Update #8) I was unable to determine exactly where the noise was coming from when I listened for it with the intake manifold off + my head under the hood. It *SOUNDED* like it was coming from under the drivers side valve cover area.

O - Update #9) The following were replaced: check valves and chain tensioners. The starter was pulled, cleaned and greased. The noise still persists. In my case, I can say for sure, the 3 listed are NOT responsible for the noise.

P - Update #10) Changed over to 0W40. Did not fix it, noise still there but noticeably better. 1/4 the time the noise doesnt happen, and after 20+ cold starts (a couple weeks worth at least) i have yet to hear it long and loud like numerous times before. This points me to this being somewhere within the oil system.

Latest video, intake manifold removed, recorded directly above motor (under the hood):

My car does the same thing and the dealership gave me what I thought was an off the wall answer. I was told that upon startup of my 2009 M3 DCT that the drive shaft was aligning itself with the transmission........................not buying that one. The dealership put in a request to PUMA and thier response was that if the car isn't broken that an unusual noise doesn't constitute something wrong and therefore they had nothing to go on. Oh this after the guys put seventy-five miles on my car and failed to correctly replace an oil line that was the initial cause to the dealership. Did I mention that the car did not make the noise until after the dealership pulled the transmission to install the oil cooler lines that were not properly repaired the first time. Anyway the has to go if someone is interested because I can stand to look at it and I don't care for the dealership's poor performance. The car sits in the garage under a car cover and will sit there and rot before I give it away or trade in in.

[///M Power-Belgium]

Originally Posted by bmwnukid

My car does the same thing and the dealership gave me what I thought was an off the wall answer. I was told that upon startup of my 2009 M3 DCT that the drive shaft was aligning itself with the transmission........................not buying that one. The dealership put in a request to PUMA and thier response was that if the car isn't broken that an unusual noise doesn't constitute something wrong and therefore they had nothing to go on. Oh this after the guys put seventy-five miles on my car and failed to correctly replace an oil line that was the initial cause to the dealership. Did I mention that the car did not make the noise until after the dealership pulled the transmission to install the oil cooler lines that were not properly repaired the first time. Anyway the has to go if someone is interested because I can stand to look at it and I don't care for the dealership's poor performance. The car sits in the garage under a car cover and will sit there and rot before I give it away or trade in in.

Wow what a story !

[s85e90]

Still normal IMO. You're gonna throw all this money to fix a problem that doesn't exist. Just deal with it.

[anom3]

just fyi, coding the break instead of clutch was free, i did it myself with my own cable that i already had... and the bpmtune with raised rpm was fairly inexpensive...

so very little money has been thrown around thus far...

that being said 10000km with this issue (bought my m3 used, who knows how long it was happening before i took possession), and no issues.

and for what its worth, 3rd day with the break instead of clutch and it didnt make the noise when i fired it up 30 mins ago.

Originally Posted by s85e90

Still normal IMO. You're gonna throw all this money to fix a problem that doesn't exist. Just deal with it.

[///M Power-Belgium]

Originally Posted by anom3

just fyi, coding the break instead of clutch was free, i did it myself with my own cable that i already had... and the bpmtune with raised rpm was fairly inexpensive...

so very little money has been thrown around thus far...

that being said 10000km with this issue (bought my m3 used, who knows how long it was happening before i took possession), and no issues.

and for what its worth, 3rd day with the break instead of clutch and it didnt make the noise when i fired it up 30 mins ago.

It's normal if you don't have the noise now,because yesterday you drove 400km !
If you realy want to know if the coding with the brake works for you ...don't drive her for about 10-12 days and fire here up ...only at that moment you will know if it works or not !
Anyway let's hope

[s85e90]

Originally Posted by anom3

just fyi, coding the break instead of clutch was free, i did it myself with my own cable that i already had... and the bpmtune with raised rpm was fairly inexpensive...

so very little money has been thrown around thus far...

that being said 10000km with this issue (bought my m3 used, who knows how long it was happening before i took possession), and no issues.

and for what its worth, 3rd day with the break instead of clutch and it didnt make the noise when i fired it up 30 mins ago.

Yes but how much have you spent already chasing this? Just deal with it

[anom3]

a drop in the bucket is how much.

this IS me dealing with it, enjoy the show

Originally Posted by s85e90

Yes but how much have you spent already chasing this? Just deal with it