OM617 rocker arm problem? - Page 2 - PeachParts Mercedes-Benz Forum

Stretch · #16 02-05-2011, 05:49 AM

Quote:

Originally Posted by Yak

Has the head been milled? The hole in pic #2 of post 12 looks uneven/shallow. That may just be the pic since it looks deeper in pic#3

I'd check to see if the tower sits flat without the ring in place. Or place the ring in the tower off the head and measure protrusion. More or less than the detent in the head?

I've just checked - it all sits flat...

...back to the drawing board - I need to do some more investigating.

layback40 · #17 02-05-2011, 07:09 AM

Quote:

Originally Posted by Army

I would kill for a surface plate! I am pretty sure the problem is with the position of the holes - well the locating rings set in the holes in the head - but I will check them all out tomorrow. I will check the bolts too. Put it this way I have now used 4 towers and I have got the same / similar result each time...

Thanks for the input Yak

Army,
I have an old WW2 bomber windscreen, its about 3" thick. perfectly flat. It cost me nothing, was on a junk heap. Just as well I am on the other side of the world!! A guy like you should be able to find one !!

Stretch · #18 02-05-2011, 10:59 AM

Quote:

Originally Posted by vstech

heh. how bad would you like one?
I've got a granite one, and a steel one...

shipping might be a killer though...

Shipping will be the killer - thanks for the offer - besides layback40 reckons WW2 bomber windscreens are the bomb...

...perhaps saying I'd kill for one was a bit misleading...

Yak · #19 02-05-2011, 11:18 AM

Quote:

Originally Posted by Army

I've just checked - it all sits flat...

...back to the drawing board - I need to do some more investigating.

If it's flat without the ring, might that ring be too tall for the current conditions of the head+towers? Does it wobble slightly with the ring in place but not under compression from the bolt?

Because the tower leans to one side, you could be getting an exaggerated 'levering' effect of a tiny out of tolerance combination due to very slight milling or manufacturing tolerances near the fulcrum of the "lever".

If the other tower/ring/head/bolt combinations (#1, 2, & 3) on the head all sit flat when fully assembled and torqued, would the risk in swapping a "good" combo into the "bad location" outweigh the potential of not getting one of those back in correctly?

Maybe build a matrix of conditions and isolate the specific part or location combo that causes the problem?

Stretch · #20 02-05-2011, 11:23 AM

I've got the rocker towers to sit flat on the top of the cylinder head. I adjusted the spindle that runs through the towers to get them to fit:- I filed away some of the cut out on the shaft with a rat's tail file.

The spindle (rocker arm shaft) no longer pushes against the rocker tower bolts - making the distance between the towers at the top greater than at the bottom. So I no longer have the bending spindle effect I tried to demonstrate in my first post.

The rocker surfaces are now in line with the camshaft lobe surfaces.

Unfortunately they are slightly misaligned in the longitudinal direction - along the engine's length. It is most noticeable on cylinder #4:-

It is not so bad for cylinder #5:-

I'm a bit concerned that if I run the engine now I'll eventually get a groove in the cam lobe surfaces due to this misalignment.

Stretch · #21 02-05-2011, 11:31 AM

Quote:

Originally Posted by Yak

If it's flat without the ring, might that ring be too tall for the current conditions of the head+towers? Does it wobble slightly with the ring in place but not under compression from the bolt?

Because the tower leans to one side, you could be getting an exaggerated 'levering' effect of a tiny out of tolerance combination due to very slight milling or manufacturing tolerances near the fulcrum of the "lever".

If the other tower/ring/head/bolt combinations (#1, 2, & 3) on the head all sit flat when fully assembled and torqued, would the risk in swapping a "good" combo into the "bad location" outweigh the potential of not getting one of those back in correctly?

Maybe build a matrix of conditions and isolate the specific part or location combo that causes the problem?

Answer #1:- Because I've source some good second hand towers they don't wobble on the rings when they are attached to the head. They are a good snug fit. You need to tap them off with a rubber hammer.

Answer #2:- The old rocker arm spindle was deeply scored - one of the old rocker arms is still stuck on it. I don't want a repeat of that!

Answer #3:- I decided that it was most important to have rocker arms that are aligned with the cam lobe surfaces. I don't want edges gouging into polished surfaces! I've fixed it so that everything is OK in the vertical plane (see above) but I'd like to shift the rocker arms along the spindle slightly by about 1mm (towards the front of the engine).

Do you think it is worthwhile adding in washers on one side of the rocker arm tower - and grinding off a little on the rocker arm on the other side?

EDIT:-

Here is a schematic of the current problem

Yak · #22 02-06-2011, 01:02 PM

Quote:

Originally Posted by Army

I've got the rocker towers to sit flat on the top of the cylinder head. I adjusted the spindle that runs through the towers to get them to fit:- I filed away some of the cut out on the shaft with a rat's tail file.

The spindle (rocker arm shaft) no longer pushes against the rocker tower bolts - making the distance between the towers at the top greater than at the bottom. So I no longer have the bending spindle effect I tried to demonstrate in my first post.

The rocker surfaces are now in line with the camshaft lobe surfaces.

Unfortunately they are slightly misaligned in the longitudinal direction - along the engine's length. It is most noticeable on cylinder #4:-

It is not so bad for cylinder #5:-

I'm a bit concerned that if I run the engine now I'll eventually get a groove in the cam lobe surfaces due to this misalignment.

I looked through the FSM for the valve assembly and there's no mention of any longitudinal adjustments. See attached scans. There's a tolerance for tower width. Are they worn?

Is the shaft asymmetrical by any chance? That doesn't make much sense with a distance error between the grooves in the shaft, but it's a thought.

Have you checked alignment with only one tower assembled? That may help identify if it's a problem with one tower/rocker/bolt combo or the other.

It appears that everything has to be manufactured to spec in order to fit: the cut-outs on the shaft fit the 'barrels' on the bolts, which control the location of the towers, which control the alignment of the rockers to the cam lobes. And since you're replaced just about everything else, what are your bolts like? The barrels smooth? (see part 67 in drwg).

I'm about out of ideas or suggestions.

Yak · #23 02-06-2011, 01:08 PM

The specs didn't scan well.

Relevant numbers may be:

Radial play of rocker arm on shaft 0.02-0.06 (no units given)
Width of rocker arm bearing bracket 24.07-24.20

Rocker arm bracket to head torque 38 Nm

barry123400 · #24 02-06-2011, 01:59 PM

At first glance I find no particular reason not to shift the rocker over using a washer. The outside diameter of the washer should probably be the same as the outer diameter of the rocker arm journal as well. Depends on the diameter of the machined surface on the tower casting. I can think of no other reasonable solution at present.

I do agree though otherwise a groove could form up over time. How long is anyones guess.

Stretch · #25 02-06-2011, 02:00 PM

Thanks for the specifications Yak - I'll check 'em out tomorrow. The bolts are fine - I checked them against a straight edge.

The towers themselves are a bit of a strange set up. The holes are much larger than the diameter of the bolts. From the 7 towers that I have seen, I'm fairly sure that the holes in the towers are not always 100% perpendicular to the tower base: I think this larger diameter hole is used to take up any play in the set up (due to this manufacturing discrepancy).

The shafts / spindles are new - they are made out of quite a ductile steel. They will bend if you try and force the towers into place. They will mushroom over if you are too enthusiastic with a hammer! I found this out on the old shaft that was deeply scored at the rocker towers and rocker arms. The locating rings are made from the same steel as the shafts / spindles, so how my original locating ring got mangled indicates that a lot of force was used to get the original towers to fit.

Generally speaking if these towers get forced into position, most of the damage will appear on the underside of the towers - one of my old towers has a very oval shape where it is meant to fit over the locating ring.

I'm pretty much convinced that the problem lies with the position of the holes in the head. I'm going to get in touch with the dealer to see if they have a diagram with relative dimensions of these holes with respect to the valves - which I am certain are in the right place!

Stretch · #26 02-07-2011, 08:01 AM

Quote:

Originally Posted by barry123400

At first glance I find no particular reason not to shift the rocker over using a washer. The outside diameter of the washer should probably be the same as the outer diameter of the rocker arm journal as well. Depends on the diameter of the machined surface on the tower casting. I can think of no other reasonable solution at present.

I do agree though otherwise a groove could form up over time. How long is anyones guess.

Thanks for the comment Barry - I've just seen it now - you posted just before me yesterday. The only possible problem I can think of by adding a thrust washer is that the retaining spring will get a bit bent.

The only thing that is stopping from making the modification right now is that I would rather reposition the tower slightly - as I instinctively feel removing metal from the other side of the tower or rocker isn't ideal.

How about gouging out a bit more of a circle on the underside of the tower and filling in the other side with epoxy putty? In this way I can adjust the tower position slightly to one side. I only need one millimeter... seems like a mile when I'm working at this scale though.

layback40 · #27 02-07-2011, 09:03 AM

Army,
As I mentioned to you else where, I have been looking into the tower problem & asking questions of others.
Our suggestion is for you to oval the hole through the tower for the bolt. This will allow you to move it over as required. Probably you could do this with an electric drill or better still, a drill press with the tower in a vice.
Prior to this, if you havent already, check all your spare towers to make sure that there is no variation between them & that you have the closest fit tower.
I would not use epoxy to fill the elongation. Even though it would be contained within the hole, oil will find its way in & degrade it.
You could braze some extra metal to part of the side of the locating ring to insure the tower remains positioned correctly. You could then do some shaping with a small grinder. It will be along the same lines as an offset key.
It is clear that your motor has been "played" with in the past.
Hope that helps !!

barry123400 · #28 02-07-2011, 10:25 AM

Quote:

Originally Posted by Army

Thanks for the comment Barry - I've just seen it now - you posted just before me yesterday. The only possible problem I can think of by adding a thrust washer is that the retaining spring will get a bit bent.

The only thing that is stopping from making the modification right now is that I would rather reposition the tower slightly - as I instinctively feel removing metal from the other side of the tower or rocker isn't ideal.

How about gouging out a bit more of a circle on the underside of the tower and filling in the other side with epoxy putty? In this way I can adjust the tower position slightly to one side. I only need one millimeter... seems like a mile when I'm working at this scale though.

Really though you may have missed something. The thing to bear in mind is the distance between the rockers may be currently less than the spring was designed for. . . Adding the washer basicallly restores the original intended spacing of the rockers for the retention spring.

You can measure the distance on other pairs of rockers on the engine to verify that is the case. Tampering with the head positioning of the tower would work as again in my opinion there is not a serious loading factor that really requires the pin other than as a locating refference.

This sideways tower displacement like the washer insertion just would return the spring to its original intended side to side width. The washer is just easier in my mind.

The main thing is the cause was not related to the flatness of the head. You have or will not have stress that you may have had if that had been the case. My first impression was that your issue may have been more serious than it turned out to be.

Also I thought but have not checked. There is a radius or side to side contour on the rocker face I thought to adjust for some lateral missalignment present as well as parellel tollerances in the manufacturing process. Between the rocker face and the cam lobe.

One way or another you are pretty well home free at this point. I was also guilty of not explaining the washer concept properly in the first post. Other people running across the same type of basic problem will now have some guidance on what it is possible to do to correct it from your overall thread.

Stretch · #29 02-07-2011, 01:07 PM

Thanks everyone for helping out with this problem.

I have just seen some photographs of an OM617 head in a new thread today:-

OM 617 overhead head R&R

When I saw these pictures I first thought - hey that's strange why are the rocker arm groups the wrong way round? In these pictures it shows the group of three rocker arm towers at the rear of the engine...

...so I checked the FSM and my Haynes manual for pictures and there aren't any clear pictures there.

So I now have two questions (and I think I know the answers!):-

Should the group of three rocker arm towers be fitted over cylinders 3,4, and 5?

And the group of two rocker arm towers be fitted over cylinders 1 and 2?

I'll be checking myself for this tomorrow.

Please note - before you all piss your sides laughing - I very carefully took these parts off of the engine in the current order, and put them back the same way. I might have suffered yet again from the Mercedes specialist from whom I bought the car... we'll see tomorrow.

jt20 · #30 02-07-2011, 01:10 PM

Quote:

Originally Posted by Army

Should the group of three rocker arm towers be fitted over cylinders 3,4, and 5?

And the group of two rocker arm towers be fitted over cylinders 1 and 2?

correct