-Marlin-

^ That's a great idea. Good tip

Euler

The head bolts from the dealer are $15/ea x 24 bolts. Labor should be about 4-6 hours if done one at a time. Twice that plus a pile of gaskets if they take the heads off.

JimmyR

FYI from one of the UK sites:
http://www.pistonheads.com/gassing/t...ues+%21%21%21+

The bloke posting is an indie specialising in AMG and has fitted a few superchargers to C63/E63s so would have plenty of experience with the M156 motor. The coolant leak issue appears to be new to him. Note that he diagnosed the fault as due to several loose, as opposed to broken, head bolts.

Ingenieur

It did not show a broken bolt?
only 2 'loose' ones?
perhaps stretched?

Euler

Ok, I just got off the phone with the owner of Buckhead imports in Atlanta. He has seen dozens of failures related to head bolts, cam shafts, cam buckets, valve springs, etc. in the affected VIN range for these engines (all models). He is adament that at some point it will happen to all of these motors and so far he sees it in the 50K to 70K mileage range. The estimated cost for full repair and component upgrade after a failure is around $18K. He also says that if you have a history of servicing your vehicle at the Mercedes dealer then Mercedes has been known to pick up most of the cost for repair and component upgrade, but only after an actual failure. His recommendation is to do the preemptive work, or get rid of the car.

Ingenieur

'dozens'?
as in 12? 24? 96?
adament? lol
the failure rate is well below 1% at this point
maybe 20-30 failures out of over 50,000 units worldwide
and won't climb much since most cars in the range have 30-80k miles
of course he wants you (to have him) do the preemptive work

I've been in touch with MB USA
They know the rate will be so low as to deal with them on an individual basis. There will never be enough for a class action (a given percentage of all units failing). No incentive to do a mass service action, why repair cars that have no issue and will never fail?

I have called 4 dealers in my area and 1 in each LA and NYC and the total so far is 3 repairs (assuming they are being forthright)

it's easy for him to advise someone to get rid of the car
he's not taking the hit

on a side note, it seems that most happed at LOW mileage, if what is posted is accurate, but without an invoice hard to say

Euler

I had a feeling you would start a pissing match over this. Its just your nature. I'm merely restating what I was told. Take it or leave it. I really don't care.

rentzington

Saying they will all fail seems a bit extreme to me, if MB had evidence of this I'm sure they would issue a recall as the liability would be extreme. If the evidence shows its less than 5% (just an example) then it would cost them too much to repair every one

but this is all making me think of taking my new to me 2010 in for a once over and coolant test

Ingenieur

this is how I look at it
if it happens to me 100% rate
if not 0% rate

I'm not gnashing my teeth and hand wringing over it
no knee jerk reaction like preemptive repairs or dumping the car, it's too nice and too much fun

I did hear (can't vouch for accuracy) that MB has a hold on head bolts when ordered so they can track failures and where they go...I looked a few dealers part sites and on the diagram for the cylinder head the bolts are not listed?

CSIKTY3

Just wanted to update everyone.... Mercedes just called me and Fidelity is covering the whole repair!

Ingenieur

Good news

Have they given you anything in writing?

Mort

Good news, that's a load off.

-Marlin-

Congratulations! That's fantastic news. Very happy for you. Must be a huge weight off your shoulders.

I am one in the minority that says this is a big problem that will eventually rear its ugly head. The proof in the pudding for me is the re-design of the headbolts in late 2010. I would have never had bought my car hadn't I known it was clear of the problem. I searched for one year. Heck, I bought a color I didn't even care for, but learned to live with because I knew the engine had the redesigned head bolts. Sorry, but anyone who has one of the affected engines is in denial on this. I would honestly sell my car if I knew what could be coming down the road. All indicators lead to an astronomically expensive failure for the long term keeper of the pre 2011 C63.

I believe the headbolts will stretch, coolant levels will drop, and catastrophic engine failure will take place prior to engines with serial number 60658.

CSIKTY3

not yet, I will get all that once the repair is done. as of now I have confirmed with my warranty company that they are going to fully pick up the tab

CSIKTY3

I guess we will just have to see, I guess thats why we buy aftermarket warranty as I have no worries now that I know my warranty company isnt a pain to deal with.Plus I dont plan on keeping the car forever

glennhl

I am also in the minority that I also think we are just seeing the beginning of the failures. This is a design deficiency. But the option I would choose is to buy the new head bolts for $360 and replace them one at a time myself. And if I weren't good with wrenches I'd find an Indie shop that would swap them out for another $500. I would not wait for a failure and possibly have a $5000 or worse repair bill.

Ingenieur

US Production numbers up to 2008 http://www.6speedonline.com/forums/m...model-usa.html
Looks like 13,000 units
Basically 2007-2008 models
Many of these cars 6-7 years old so many > 50,000 miles

If all will fail wouldn't we be seeing 1000's of failures by now?
If you add MY's 2009 & 2010 the production numbers may triple
40,000 cars
Even a rate of 20% would be 8,000 cars

glennhl

First off, only Mercedes has any clue what the failure rate really is. Second, the failures are going to meet a bell curve. We most likely are in the very early part of the bell curve. Heck, the mean may be over 200,000 miles. Of course the other confusing thing, is it's probably not related to mileage as much as cycles (full heat to full cool down). Either way, we don't have sufficient data to determine where most of the failures will happen. Mercedes probably has sufficient data, but they aren't sharing.

You are trying to convince yourself that you will not have a failure. Truthfully, you probably don't have to worry, but again, for $360 and a few hours of your time, you can remove the worry completely. Even if you have someone else do the work, you are looking at under $1000 max for peace of mind.

I just wish Mercedes would do the right thing and do a recall. They admitted to making a design error when they redesigned the head bolts in early 2011. What confuses me is I'm convinced Toyota or Honda would do the right thing, I just can't imagine why Mercedes won't.

Ingenieur

I agree it may be a bell curve and can only guess where we are on it. But if ALL cars fail it may not: it may be asymmetric rising until all cars have failed then dropping off (since all have failed, lol). Like a rising e^x pulse. I only couch it in these terms because now I know your background .

But looking at the the head bolt thread and documented failures they seem to be happening in the 50k range? Sometimes lower. So I believe we are well into the curve, whatever shape it is. This leads me to believe not all cars will fail, and the percentage will be low.

As to why, I believe break in has a lot to do with conditioning and strengthing/weakening the bolts. I also believe warm up has a lot to do with it, a 'cold' bolt is more likely to fail.

I know I'm covering ground that you already have in arriving at your conclusion (again knowing your background is well thought out) and we agree it a complex system.
We also know there was a better choice of fastener.

consider 6000 rpm or 100 rev/cycle (to make the math easy)
power strokes 50 / cycle or 25 per head
that is considerable elastic cycling!!
from the AMG site mean eff cyl press is ~ 170 psi
piston area ~ 12.6 in^2 so mean force ~ 2140 lbs (force)
if an ideal system peak will be 4280 lbs and at the bottom of the stroke close to 0
so the stress on the head is ~ 8600 lbs
in a 'rectified' sine waveform of 25 Hz with root mean sq of ~6100 lbs, distributed cyclicly over the bolts

the inner cylinders 4 surrounding bolts probably absorb ~ 80% of the total for that cylinder firing (20% each) with the remaining 20% distributed decreasingly to adjacent bolts in an exponential fashion
so the max a bolt would see is 0.8 x 4300 /4 ~ 860 lbs per bolt plus the force from the preceeding cylinders firing (at 90 deg or the mean eff pressure) total ~ 860 + 0.8 (2140/4) ~ 1290 lbs per bolt
the preceeding cylinder would not be adjacent but I still used 80% rather than 20% to get a worst case
this is well within the spec for the bolt, without the cycling considered
but imo the issue is the end cylinders, the end bolts don't get to 'share the load' so they see more stress, and seem to be the ones that break more often?

Now, after having said all that crap, lol. I believe the failure mean will be close to 50k miles, for those that fail, and that rate will be less than a few percent. there are 13,000 2007-2008 in the US, using an average accepted value of 12,000 miles per year, and reducing it by 1/2 to 6,000 becasue these cars are more 'cherished' lol, that equates to 30-40k average per 2007/2008 MY, most likely low.
we would be on a 'steep' part of the curve and be seeing 100's of failures if all wer to fail. It would be an epidemic, hard to surpress, and class action worthy. It si not, at least yet.

I hope you are correct and the mean is 200k miles
I put 6k on a year so that is 30 years down the road.
I hope I'm wrong since that only gives me 3-5 years

Ingenieur

FYI
Just went through all 32(!) pages of the headbolt thread in order to make a spreadsheet when I have time
11 documented failures (or that I 'trust') not many more that I haven't included
only 1 has an invoice
a few have pictures
lowest mileage 26k
high 57k
clustered about 55k (52, 54. 55, 57, etc)
about 1/3 are e63's or other non C63 models

this seems to tell me that a lot of 2007-2008 have hit the 50k range and failures should be occuring in large numbers, 100's if not 1000's???

this site has 310,000 members
and based on post/thread count ratios about 30,000+ in the M156 related forums
this would appear to be a statistically significant sample
the arguement that only regular posters (the usual suspects, lol) doesn't hold water for me because:
1 many of the failure posts are new or not active members with low post counts
2 the timeframe: that thread has been around for a couple years
3 if you google m156 head bolt that thread comes up high priority

there are ~30,000 affected engines in the US
I estimate 20% minimum of those owners have visited this site, 6000 (out of our 310,000 members)
even if we allow 20 failures 20/6000 ~ 0.3% rate
I believe the owners number is higher, closer to 10,000 (or 3.2% of our members which iirc is about the same ratio MB sells AMG/to total MB sales, although enthusiests are more likely to participate in a forum, moreso than the yuppie with the basic C klasse, so 3.2% of M156 AMG owners is still low imo, but conservative) and the number of failures closer to 10, not 20, so the rate is much lower than the 0.3% (to date)

glennhl

Ingenieur, I like your data collection, excellent job. Just a few things. It's possible that the early failures are at 30,000 to 60,000 miles. What would be really useful is if we could figure out how many cycles (full heat up to full cool down) there were in those miles. If the early failures are at 50,000 miles or so, then the mean could be way out there, only time will tell.

Now one clarification, when you torque up a bolt it goes into tension (I know you know all of this). Let's say you torque it up until the bolt is in 10,000 pounds of tension. The bolt acts just like a spring in tension and the only way to put more load on the bolt is to exceed the 10,000 pounds and then the bolt will start stretching more and the load will go up. So if the normal operating loads are under 10,000 pounds the bolt will just sit there all day long with 10,000 pounds of load. So with head bolts, the design would be such that the forces from the combustion chamber pressures were always well below the bolt's 10,000 pounds of tension. If it were more, then the bolt would stretch more and the head would be forced off it's seat and the gasket would lose it's seal. Side note: That's why it's so important to lubricate the threads on bolts. If you don't lube them then you may not get the full tension load when you torque up the bolt.

So the bolt does not get millions of cycles of load from the combustion chamber pressures. The thing that gives the bolt more load are thermal growths. So when the engine heats up and the head/cylinder expands more than the bolt, the bolt load goes up because it has to stretch more. So the only fatigue loading the head bolts see are from heating and cooling of the engine. So if the failures we have seen so far are at around 50,000 miles and the average trip is around 20 minutes in length, then that would be 150,000 cycles.

Now a Metalurgist (someone a lot smarter than me!) had some failed bolts, they could look at the failure and determine what kind of failure happened...high cycle fatigue or low cycle fatigue. But it could also be more complex and be stress corrosion. Again, I'm betting Mercedes knows all of this. And they corrected the problem in 2011.

EDIT: I googled bolt spring scale analogy to see if I could find the way bolt preload was explained to me many years ago. Here's a pretty good illustration to see why bolt loads don't go up unless you exceed the preload:

https://upload.wikimedia.org/wikiped...ng_analogy.svg

Ingenieur

Let's hope it's way out there, lol
I think the 50k mile range is the midpoint IF they are going to fail
You may be onto something about thermal cycling

let's say the bolt is torqued 10 kip of tension
the combustion force must be absorbed/transmitted somewhere
downwards to the piston/crank/etc
equal and opposite outward on the head
the force to the head can only go/transfer/dissapate to a few places
-flex the head and convert to heat
-or the bolts, whose purpose is to restrain the head from said force
so the bolts take some of it, hence they stretch over time and eventually yield
I agree, the force absorbed due to combustion forces are only a fraction of the preload (if designed properly, sfaety factors, etc.) but it does increase the tension, it has to...think of grabbing the head and pulling, strong enough the bolt stretches and tension increases

If it didn't the bolt would not break (unless you believe corrosion is the culprit and weakens the bolt until its static preload/temnsion breaks it, I do not)

I believe the cycling weakens the bolt over time, at the weak spot (invternal torx shoulder) and eventually kapoots

glennhl

I don't think 50,000 is the mid point because if it were then half the cars would fail by then. 50,000 is just the early beginning, the mid point is way out there. That is good news.

Ingenieur

only if ALL cars will fail
if only a sample of cars will fail and the midpoint is 50k, perhaps we have seen the worse and are on the down side of the curve
eternal optimist but objective realist

time will tell

glennhl

That assumes that there is a sub-population, which is always a possibility.

Either way, I think the failures are a ways out yet. But if I were a gambling man, I'd say the worst is yet to come, but will be years away. I just wish Mercedes would step up to the plate. In the scheme of things, if they were proactive and replaced the bolts before they failed, it would be cheaper and they would gain millions in good faith from their customers.