Timing Chain nightmare - SCARY !

[gladie]

Dear community :-)

I have fully refreshed/renewed my E21 318 carb engine (1978).
Timing is then also brand new : new chain, guides, tensioner, spring etc........
I have started and let warm up the engine for a total of about ten times and then went last WE for a very short drive in my driveway.
Suddenly the car stalled in a nasty metalic noise.
I towed the car up to my house and opened the engine covers !!!!!!!!!!!!!!!!

The timing chain was just about to drop, I potentialy (could) have killed the engine with this event.

In fact, on the camshaft sproket, 2 screws have dropped down in the engine (as far as I can imagine as they're not there anymore) and last 2 screws were just about to fall !! I can only guess this saved the engine from the piston to hit valves.

I have 2 questions at this stage :

1) What is the torque value for the camshaft sproket screws ? They are very slim screws. There is a security (don't know the name in english) to prevent the screws to unscrew by itself, a kind of metalic strip, I cannot tell why the screws got loose !! Do you apply Loctite ? I'm not sure but I think I did Loctite the screws, but not sure...

2) I have a big doubt about the chain tension ! In fact, whn I installed it back on the engine, I already thought it was very tight. Could the chain to be too tight and then it applies to much tension and then unscrew the screws ?? And then, how loose/tight (on the tensioner side of course) should the chain be ?? I must double check, but let's consider the chain lenght is correct, is this normal that the tensioner guide (the one in metalic/rubber) is almost touching, if not touching, the aluminium cover ? I mean the chain is so tight/straight, that it seems the tensioner to be at farest position so it as no effect on the tension ?

I must find the reason of why the sproket almost falled and how to make it not happen again !!

Thanks for your help and here are some pictures (1rst picture shows the sprocket almsot falling, 2nd picture shows the remaining screws, 3rd picture shows the chain and tensioner position) :

[alsimma95]

Hey there, good thing you caught it before something more serious happened.. Will definitely want to get those screws out, dropping the oil pan in these cars is doable without a whole lot of work from what I've heard..

The torque value is supposedly not very large, ~10nm. https://www.bmw2002faq.com/forums/to...t-bolt-torque/ . When I did my chain I used some red loctite, others recommend blue but I didn't want to take any chances... 10nm did not seem like a lot and I didn't have much faith in that locking plate design..

I don't remember exactly 'how tight' the chain was, but things seem ok in the pictures you posted... The tensioner should definitely contact the guide, I believe you can manually pump the guide against the tensioner while oil is in the valley to prime the tensioner before you start it next time. But if you already have a guide that is making the chain 'too tight,' you might want to verify you have the guides placed on properly, and that you ordered the right guides. AFAIK for timing chains, tight is good..

[autox320]

Only way I see that happening is if the gear wasn't flush or seated fully when installed. Also the bolts can get "chewy" if reusing old hardware. I use 10.9 for cam bolts they are a little harder and the threads hold up better. Blue Loctite and no lock plate here but everybody has an opinion. I do have new flat spring washers under the bolt heads but those don't really do much but keep the bolts from galling the gear face. My gear is adjustable as well; just a factory one cut with slots by IE. Never slipped and never lost the bolts between seasons. I'd also check the cam threads and make sure they aren't damaged. Thread in a new bolt part way and jiggle it see if lots of thread play before it seats. Usually the cam steel is extremely hard to damage, the bolts will give before the cam threads but never know.

Clean surfaces, no oil; I use brake clean before installing the gear to cam face. New hardware, blue loctite, fully seated and square on the cam. When pushing on the gear it kind of snap seats over the cam center. Seems simple but just be sure. I mean it shouldn't even bolt up unless square on the cam so thinking out loud. I mean I always forget factory has the dowel engaged as well. My setup has no dowel due to changing timing and the performance cams don't have dowels or marks for that matter.

Far as fishing the bolts out just maybe if lucky can drain the oil and try a long magnet stick through the drain plug hole go fishing. Otherwise try a magnet stick down the front cover and get lucky. Otherwise will have to go in for them. It's my only fear when timing a cam is dropping the bolts down in there

Honestly if when put back together can't find anything obvious I'd just spend the extra time to take off the lower cover and inspect the whole works. I mean this is a new engine it's worth the extra mile and overlook a second time.

Factory chain tension should be pretty stiff once the oil is in the piston. Correct chain and parts I don't see how can change this much unless using an aftermarket spring or washer shims.

[gladie]

Thanks to both of you for answering :-) very appreciated !

So the fact the chain would be (or not) over tight is not an issue, and should not be the reason for what happened ?!

I haven't changed the screws, they are 40 yo, do you think I should use them again ? If so, where from to get knew one ? Would prefer to avoid that as finding parts are becoming harder...

Concerning the piston of chain tensioner, which along with a soft long spring, I just re-installed new part, but never paid any attention of any matter of oil with this part. What should I take care of with the piston and the oil please ?

I'm sure the sprocket was installed properly honestly, the only missing part could be that I did not Loctite the screws... I did torque exactlty at 10NM, so very strange it has so quickly felt appart !!!!!

I'm very scared doing all the process and having the same issue :-( because I cannot find so far a good explaination/reason of that fail (except maybe the miss of Loctite)...

[autox320]

Thanks to both of you for answering :-) very appreciated !

So the fact the chain would be (or not) over tight is not an issue, and should not be the reason for what happened ?!

I haven't changed the screws, they are 40 yo, do you think I should use them again ? If so, where from to get knew one ? Would prefer to avoid that as finding parts are becoming harder...

Concerning the piston of chain tensioner, which along with a soft long spring, I just re-installed new part, but never paid any attention of any matter of oil with this part. What should I take care of with the piston and the oil please ?

I'm sure the sprocket was installed properly honestly, the only missing part could be that I did not Loctite the screws... I did torque exactlty at 10NM, so very strange it has so quickly felt appart !!!!!

I'm very scared doing all the process and having the same issue :-( because I cannot find so far a good explaination/reason of that fail (except maybe the miss of Loctite)...

The chain should be pretty tight. In fact it's just possible to put the cam gear on with the slack given when the tensioner is out. If the tensioner is left in it's near impossible to muscle the gear on so there's not much leeway. The gear sits on the cam nose so like a hubcentric wheel the bolts are just holding it to the face. Also has the dowel engaged so it's like a permeant bolt keeping from sheer.

Bolts I'm not sure in your part of the world. Here I use Fastenal or local hardware stores. Basically it's a M6x1.0 bolt x 16mm long. New 8.8 is fine but I like 10.9 metric grade being closer to grade 8 SAE.

I'd loctite and put a bit extra on tightening the new ones. If using a quarter inch drive ratchet it's hard to overtighten by hand. No need for a torque wrench book spec. Tight is tight.

btw hard to see in your photos; just a check but is the locking plate installed on the front of the cam to head? If not that would be culprit.
cam lock.jpg

[gladie]

Thanks Autox320.

Today I checked inside the cylinder with an endoscope : very bad to see the valves knocked the piston !!!
All my restoration is now DEAD !!!!
It is catastrophic !

My main question is now : why the screws are gone ? How is it possible ? If not a question of chain tension (too tight), is it just because I did not Loctite them ???? I did torque at 10NM and I did add the locking plate under the screws, so WHAT ????

I need answer before to decide if either I bin the car, or I start for weeks and month of again hard work and money spent !!!!!!!!!!!!

[autox320]

Thanks Autox320.

Today I checked inside the cylinder with an endoscope : very bad to see the valves knocked the piston !!!
All my restoration is now DEAD !!!!
It is catastrophic !

My main question is now : why the screws are gone ? How is it possible ? If not a question of chain tension (too tight), is it just because I did not Loctite them ???? I did torque at 10NM and I did add the locking plate under the screws, so WHAT ????

I need answer before to decide if either I bin the car, or I start for weeks and month of again hard work and money spent !!!!!!!!!!!!

Man sorry to hear that. Any piston damage? Usually it's just the valves will bend. It's possible to swap valves, check seats and put back together. Piston tops are pretty thick and usually bend valves before making holes.

I guess my first question is was the motor turned over by hand a few times manually before starting? Always before hitting the starter should make full clean unobstructed rotations by hand without hard stops.

It's possible the cam bolts just weren't very tight. Another thought is the chain alignment. If it's off can work things loose. I've not had this issue before so hard to say without taking off the front covers and verifying everything when assemble. Again why I also ask if the cam lock plate is in place. if not it will walk out the front and put angle on the chain. Just thinking out loud.

[cgifool]

Did you bend the tabs of the lock plate against the bolts after you tightened them? I don't think they do much if you forget that step....

[gladie]

Thanks for the replies.

Yes I did re-install the cam lock plate.
Yes I did bend the tabs against the bolt, maybe not properly ? I'm not sur I did Loctite the bolts, maybe it is "the" mistake ?? I hope it was the reason, but very suprised that with the locking plate bent againt the bolts and the bolts torqued to 10NM, how it could happened so fast, after only with starts and 4 miles driven !! My only concern is : could it be from a chain with too much tension on it ?? My worrie is that the chain is not loose at all and the tensionning guide (the one in metal+rubber) is stuck against the cover on the side, like if the tensionner spring and piston had no effect on pushing the guide... Do you understand my worry there ?

I'm really think, to Loctite the bolts next time, but also to weld the bolt with a tiny welding on one side of the bolt ! I cannot afford to have this issue ever again !! What do you think ?

Thanks again

[gladie]

Hi,

I have some news and I need your knowledge/experience please !

I did a compression test, after my timing chain slightly came off because of the camshaft sproket's bolts to unscrew (please read above to have past history).

I have inspected the engine head after removing it, and not much damage (no disaster finally) but the intake valves have slightly hitted the pistons on cylinder 2 and 3 (there is a small impact mark on the piston where it met the valve). I have re-assembled the head and I did a compression test today, here are the results:

Cylinder 1: 142 PSI (10 bars) : perfect
Cylinder 2 : 127 PSI (9 bars) : limit
Cylinder 3 : 130 PSI (9+ bars) :limit
Cylinder 4 : 142 PSI (10 bars) : perfect

Do you beleive the Cylinders 2 and 3 having lower compression level, and they are the ones having shown a hit mark on the piston when I removed the head, do you think it is a problem ?? Do you think I have bended valves ?? Do you think I have to switch to panic mode ????
When I run the engine, I have some problem : some explosion and the engine does not take freely the revs BUT I already had this before my timing chain issue AND I already have trouble with my carburator and it is an element I have to renew next (swap with Weber carbs instead of Solex). This could be a bad carb, or a bad timing advance, OR a bended valves now, so not sure this bad engine behaviour is due to...

Please let me know your thoughts about those results considering all ?
If I had a bended or twisted (even VERY VERY slightly) valves, what would be the results like ?

Thanks !

[gladie]

In other words, a barely twisted/bended valve would imply what kind of loss in compression test ? Would it be a 10% loss or a dramatic 50% loss ?

[gladie]

I just removed the carb to look inside the intake manifold, and I found some carbon deposit/black soot on the inside walls ! Is this really a bad sign of leaking intake valves or just "normal" ?? (but in a recent past I had a very bad timing ignition, which could have done this maybe ??)...

[alsimma95]

If piston hit valves I would say it's pretty likely they're bent, but I don't really have any experience with bending valves yet so take my advice with a grain of salt. I think a properly done leakdown test should be able to confirm whether you're losing any compression past the valves or not. That being said those numbers don't look awful, I would think with a bent valve you wouldn't have any compression..

[gladie]

Thanks for answering.
Yes I also beleive so, if they were bent there probably wouldn't be any compression or so... but I'm about to do a leakdown test, it is best advice at this stage...
Will keep updated ;-)

[320iAman]

compression values are okay, rule of thumb is ~ 10% difference between highest and lowest values is okay. 142-127 is 15, 15/142 is 10.5%, your probably good on valves, if the valve is bend it would not seat properly to the valve seat and thus loss of compression.. from your test results you should be okay, leak down test will check the valve face tightness to the valve seat. If your really concerned, take the head off and remove the valve keepers, spring etc. then remove the valve , then push the valve in if it hits the valve guide, does not move up and down easy with slight oil its bent, if moves easy its not bent, plus you can check the valve squareness with a straight edge of suitable size for light coming thru. What has happened is your valve struck the piston, more exactly the valve kick angle area hit the piston, unless this was pretty hard smack you should be okay, kick angle is ground in to the bottom of the valve to knock carbon deposits of the valve and other select items.

Randy

[OldGray320i]

I'll ask the smart guys to weigh in, but, I recall from the old days you might pull the head, turn it upside down and pour some gas in the chambers; if it leaks out, valve sealing issue, if it holds, you're ok.

IF that method is valid, you're just out your labor and a head gasket (front cover gasket).

[gladie]

Hi ! Thanks for answering :-)

I have just removed head and tested putting water in the combustion chamber once the head is upside down, and I had no leak. But this is not really realistic as no pressure is applied onto the water, but it didn't leak anyway...

I will test the leakdown tomorrow: I push compressed air in the chamber via the spark plug hole, and on the other side with the carb removed (and valves closed) I check if any air comes out from the intake manifold (putting a piece of paper on the exit holes I should see if the outgoing air is moving the piece of paper)...

[OldGray320i]

I think it has to be gas, not water. A quick google says its something to do with viscosity, which makes sense.

In any case, hope the head is ok!

[320iAman]

That will NOT test if the valve is bent and hitting the valve guide ! It tests if the the valve is seated to the valve seat. Take the spring out and inspect the valve guide to see its not broken or cracked, slide the valve(slightly oiled) in and out of the valve guide if it does not bind or wiggle too much your good to go, you can read about cracked valve guides and the problems it creates.

The water test is okay to use this again tests the valve is seated to the valve seat ! Compressed air and water test( bubbles appearing) will test if the valve is bent so the valve is not seated to the valve seat and if not bent if the valve seat and valve needs to be re-seated with valve grinding compound-usually, valve clearance to valve guide is 0.006 to 0.008 " six to eight thousandths of an inch, valve guides are made of brass which is a soft metal.

If you want to be exact-you'd measure the inside diameter of the valve guide and then deduct the valve stem outside-diameter if the difference is 0.006 to 0.008 " your exactly good to go within specifications.


Randy

[gladie]

Thanks, I do understand.
The point is that the head is already back on the engine. Unfortunatly, I considered that a well seated valve was not a bent valve, which seems not to be true, so my mistake... I will unfortunatly keep it like that, as I won't remove it all to check that... if leakdown test is ok (with compressed air) I will then consider it is OK (but maybe not if valve is very sligtly bent inside the valve guide...)

[gladie]

Hi !
So, leakdown test seems ok, as far as I've been able to do the test. ;-)
Thanks all for your feedbacks, advices and help !

[typhoon320i]

this may be a dumb statement, but I'm looking at the photo and wondering if you had the corners bent up on the plate under the bolts as that is what keeps the bolts from backing out. ...although lock-tight sounds good too.