S38 Valve problem UPDATE 2/12/14 & solution

[M5BB]

Here's a good one for all the S38 guru's out there.

I built this S38 in 2008.
Pretty high performance. See info with signature.
800 miles into the rebuild it spit out a valve adjustment puck.
Made a hole in the cam box.
http://www.mye28.com/viewtopic.php?t=60420
You can read about details there.
.
Before reinstalling cam box I had to polish hole for bucket as fit was a little tight. Discussed with Paul Burke at time.
Probably caused by heat from welding even though I used the best guy I could find at the time.

26500 miles later a new issue has manifested itself.
Started car up last Sunday morning and similar noise to the puck coming out of the bucket. Shut off in 5 seconds or so.
Garage is heated and probably about 60 degrees that morning.
Came home later and removed valve cover. No puck out of place. Whoa!
Started examining and rotating pucks and buckets very carefully and turned over motor with starter in short bursts so I could try to isolate the noise. Deffinately in the valve train. Decided to do a leak down and check clearances of each puck to see if any big discrepancy. None found.
While using special tool to depress buckets I noticed that #5 exhaust right was slow to come back up.
Push bucket down and let go of tool and bucket slowly rises.
First thought was broken valve spring. Thought about that for a while and next day I called Paul Burke to see what he thought.
He said he has practically never seen a valve spring break in an S38. More discussion and possibility of bent valve or bad valve guide or maybe a valve seat fell out.

Leak down test showed #5 was leaking about 30% and I think I could feel and hear air coming out the exhaust.
#5 exhaust (R) is the same cylinder where the puck was spit out 26000 miles and over 4 years ago.
#6 was also leaking but not as much. The other 4 cylinders were 10%,5%, 2%,2%.
The really weird thing is the car was driven into the garage about a week before and shut off running like it normally does. Then on startup this noise shows up?
Borrowed a bore scope from a friend that has a shop and checked all cylinders and particularly #5. Bore scope was not that good but thought I could see some shiny spots in valve pocket cut outs in piston. Matched to valve that was sticking.
Removed exhaust cam and tested valve up and down with bucket out and I could push valve down and count to 5 or so before it got all the way up. Springs seemed fine.
So as cam went over this bucket and pushed the valve down it would stick and then slam back up into cam after lift area had passed. Hence the slamming sound.
More discussion with Paul and we think the valve guide has failed.
Head will be removed tomorrow. Yuck.
To be continued..................

2/10/14
Ok, head removed, problem discovered.
It looks like the valve guide clearances were too tight.
Head is at machinist and he is removing all the valves and checking particularly the exhaust valve guide for sticking.
BMW spec says maximum wear would be .031.
Machinist says he used .015 for clearance.
Paul Burke says it should probably be .020
The exhaust side is hotter of course and particularly cyl 4-5 are the hottest because of their location and cooling available.
Intake max wear is .025 so .015 is probably ok.

So here are some photos.

cyl 5 right valve pocket shows where valve hit.


cyl 5 valves. right valve was hit by piston


Valve stem of sticking valve. May be deposit problem on stem.


cyl 1 shows intake valve was leaking. Deposits on valve. Leak down did not show this.


THIS IS THE HEAD FROM A SPARE ENGINE- NOT MY ENGINE
It was known to have a bad head. Well I guess, as this is what I just found. It has 125k on it when car was totaled.


My engine has not been babied.
It is driven hard and shifted at 7000 at least a few times every time I drive it. Hence no carbon build up on the intake valves except the one that was leaking.
It has been on the track at Barber Motorsports for 40-50 laps in one day and Mid Ohio for the same.
It looks like these problems are from improper machining.
I thought I had a good machinist but I do not have the ability to check all his measurements.
Machinists with the knowledge to set these motors properly are getting quite hard to find so please be careful and check on your machinist before you give him your cyl head.


.

[Da_Hose]

Dammit .... I hate to hear stories like this.

I hope it goes well.

Head work on an S38. What a friggin nightmare.

Jose

[S38Mike]

Here's a good one for all the S38 guru's out there.

I built this S38 in 2008.
Pretty high performance. See info with signature.
800 miles into the rebuild it spit out a valve adjustment puck.
Made a hole in the cam box.
http://www.mye28.com/viewtopic.php?t=60420
You can read about details there.
.
Before reinstalling cam box I had to polish hole for bucket as fit was a little tight. Discussed with Paul Burke at time.
Probably caused by heat from welding even though I used the best guy I could find at the time.

26500 miles later a new issue has manifested itself.
Started car up last Sunday morning and similar noise to the puck coming out of the bucket. Shut off in 5 seconds or so.
Garage is heated and probably about 60 degrees that morning.
Came home later and removed valve cover. No puck out of place. Whoa!
Started examining and rotating pucks and buckets very carefully and turned over motor with starter in short bursts so I could try to isolate the noise. Deffinately in the valve train. Decided to do a leak down and check clearances of each puck to see if any big discrepancy. None found.
While using special tool to depress buckets I noticed that #5 exhaust right was slow to come back up.
Push bucket down and let go of tool and bucket slowly rises.
First thought was broken valve spring. Thought about that for a while and next day I called Paul Burke to see what he thought.
He said he has practically never seen a valve spring break in an S38. More discussion and possibility of bent valve or bad valve guide or maybe a valve seat fell out.

Leak down test showed #5 was leaking about 30% and I think I could feel and hear air coming out the exhaust.
#5 exhaust (R) is the same cylinder where the puck was spit out 26000 miles and over 4 years ago.
#6 was also leaking but not as much. The other 4 cylinders were 10%,5%, 2%,2%.
The really weird thing is the car was driven into the garage about a week before and shut off running like it normally does. Then on startup this noise shows up?
Borrowed a bore scope from a friend that has a shop and checked all cylinders and particularly #5. Bore scope was not that good but thought I could see some shiny spots in valve pocket cut outs in piston. Matched to valve that was sticking.
Removed exhaust cam and tested valve up and down with bucket out and I could push valve down and count to 5 or so before it got all the way up. Springs seemed fine.
So as cam went over this bucket and pushed the valve down it would stick and then slam back up into cam after lift area had passed. Hence the slamming sound.
More discussion with Paul and we think the valve guide has failed.
Head will be removed tomorrow. Yuck.
To be continued..................


.

Ouch!! Hey look at the bright side, you caught it before the piston and valve got together and turned the engine into an inline 5. Make sure to clean off the piston top and check it for cracks. Best of luck, hope it turns out ok.

[Devinder]

What is the theory of valve guide failure that would cause this? Just asking because I would have jumped to bent valve first. How do guides get tight over time?

[Mario L.]

Thought I'd share my experiences with this on the s14. I've never seen it on an S38 . . but the mechanics and the design are the same.

This can happen with just a very light over rev where one or two valves make piston contact. The shaft of the valve gets lightly bent as it's open, and this causes a restriction as the shaft has to move through the guide. This will increase the clearance to the shim and cam, and may be enough to cause the shim to pop out. Note that the bend in the valve stem is so minor that the valve still seats adequately and it will not be noticed during operation, or possibly even at idle other than a bit more valve clearance noise.

When this engine was rebuilt in 2008, what was replaced?
Guides? Valves? Valve springs?


Were other cams installed, something more aggressive?


Mario L.
VSR1.com

[GaAlpinaOwner]

Yikes Gary sorry to hear this. Let me know if you want my favorite local mechanic (Mike Spooner) to look at it.

David

[M5BB]

Thought I'd share my experiences with this on the s14. I've never seen it on an S38 . . but the mechanics and the design are the same.

This can happen with just a very light over rev where one or two valves make piston contact. The shaft of the valve gets lightly bent as it's open, and this causes a restriction as the shaft has to move through the guide. This will increase the clearance to the shim and cam, and may be enough to cause the shim to pop out. Note that the bend in the valve stem is so minor that the valve still seats adequately and it will not be noticed during operation, or possibly even at idle other than a bit more valve clearance noise.

When this engine was rebuilt in 2008, what was replaced?
Guides? Valves? Valve springs?


Were other cams installed, something more aggressive?


Mario L.
VSR1.com

Hey Mario,
Thanks for the input.
Yes, I'm running B36 cams and CP 11:1 pistons.
Pretty large valve eyebrows for valves.
On the rebuild all valve guides were replaced.
Also about 6 or 7 exhaust valves and a couple intakes.
Springs were checked and used again. 125k on them at the time.
All new main and rod bearings. Stock sizes as there was not enough wear to go to a plus size.
When this ( the shim spit out) happened the engine only had about 800 miles on it so doubt I had an over rev situation as it was just getting broken in.
See my addition to my OP.
I think the valve guide tolerance was too small and the fact that the car had been out overnight and it was 24 degrees when I started it. Thick oil, contracted valve and guide that was sticking allowed the thick oil to pick up the shim as the pressure on the shim and bucket was not maintained because the valve stuck in the guide.
This is the first time after 26000 miles that I've ever been able to see a possible cause for the shim to get spit out???
Think that is a reasonable cause?

Did you get the cats for sale on ebay?
Haven't checked.
Gary

[mooseheadm5]

Hmm, exhaust valve fouling. How are your AFRs? M5s will destroy valves if run lean.

[M5BB]

Hmm, exhaust valve fouling. How are your AFRs? M5s will destroy valves if run lean.

Hey Paul.
I do have a Innovate AFR in my gauge set in the car.
I do keep and eye on it.

See what you think of these?




CYL 1-2


CYL 3-4


CYL 5-6

[Mario L.]

Gary,

S38 B3.6 uses the BMW Motorsport valve springs with the yellow red stripes. To the best of my knowledge the S38 B3.5 used the lighter duty valve spring, looking at ETK the part number is not given for the E28 M5, but is for the E24 M6 as the early spring.
11 34 1 312 704 early S38 B3.5
11 34 1 315 912 Yellow/Red MS Spring B3.6
We do know that the MS Spring does a better job of controlling valve float when used with the EVO cams in an S14, and are essential for higher RPM use. I have not compared cam profiles of the EVO S14 to a B3.6 cam, but suspect they are extremely similar . . . there has to be a reason BMW went to the stiffer spring with the B3.6 cam.

After all your miles on this engine, I can't agree with the oil viscosity valve stem clearance thought . . spring pressures will overcome all of that. Your oil would have to be the thickness of molasses . . and I suspect you were using a synthetic . .
My guess without seeing and handling parts
cam follower bucket and shim radial clearance a bit worn combined with a slight past overrev . .

Mario L.

[Mario L.]

I just looked at your pictures . . .

lots of carbon, did a loose piece get under a valve and hold it open?

Mario

[M5BB]

I just looked at your pictures . . .

lots of carbon, did a loose piece get under a valve and hold it open?

Mario

Mario, either I didn't explain very well or you are not reading what I wrote.

The engine spit out a puck 800 miles after a rebuild.
I had to leave the car out overnight and it was 24 degrees when I started it up. It had Castrol 20-50 in it. Not synthetic
A new bucket and shim were installed after repair.
A valve guide sticking certainly could have caused this.
Now 26000 miles later the same valve was sticking so much that it would not return to the cam quickly enough to not make a terrible noise.
The car lives in a climate controlled garage and is not driven much in the Winter and if it is it's just local area drives where it would not have a chance to get cold over night.
The other S38 engine I just tore down looked very similar to my engine as far as carbon and valve condition. It had 125k on it. This does not seem to me to be lots of carbon.
All the intakes valves heads (not the combustion face)were very clean except the one valve in cyl #1 that was leaking.
That's the one in the photo.

What valve guide clearance would you have a machinist run on the exhaust valves for an S38 B35?

Yes, I understand about the valve springs. I've had a long conversation with Paul Burke and he says he has probably never seen an S38 spring failure. Yes the spring rate may be higher for the B36 springs probably should have gone there.

Some of this look it has could be the 10% ethanol gas?

[Mario L.]

Got it, sorry . . . too many engines, too many cars and too much time from when I originally looked at this thread.

You asked ... "What valve guide clearance would you have a machinist run on the exhaust valves for an S38 B35? "

We run them a bit looser that the .015 mm your tech indicated, closer to the limit of BMW spec. But note that we change that depending on engine application and use . . street or race, and what we expect the life of the engine is between rebuilds.

Your #1 cylinder intake valve that was leaking, you saw no leakage during the leakdown test, ??? so that to me would indicate that the deposits on the valve are from oil leakage past the guide seal ..this will also cause what you saw on the valve seat . . as the oil leaking past the guide will not burn off on a cold intake seat.

I have had no issues with ethanol fuel in any BMW M engine, even race track engines. We also make certain that fuel tanks are full during off season storage with Startron preservative and new fuel goes in before a track session.
Different scenario for marine and open to atmosphere fuel systems!!!

Mario

[M5BB]

Consensus on valve problem is that clearances for exhaust valves were too tight.
Guide clearance was set at .015.
BMW specs say wear limit is .031.

Because of greater heat generated in the exhaust valves the clearance should be around .020.

So since they were tight out of the rebuild it spit out the valve adjustment shim at 800 miles on a cold morning when car was left outside overnight. Cold parts contracted.
Up until now we had never been able to determine a reason the shim was spit out.

Now 26000 miles later it was cool in the garage but not below 50 and the stem was sticking in the guide. Different expansion rates.

I have no way to measure this spec so I'm afraid this was an error by the machinist.

Ironically the valve was not bent even though it hit the piston top.
Remember the valve was sticking not stuck so when engine started the valve was being pushed back up by the piston and with nothing to bang into it made it back to it's seat and by that time the piston was not longer in contact.