Snapped off rocker arm bolt
 

While closing up after valve stem seal replacement on an 83 240D manual, I managed to over-torque and break off a bolt holding down the front rocker arm. While I'd certainly be interested in advice about how to remove the stud (photos attached), my main question is about damage to the head: how likely is it that I have damaged the threads on the head by over-torquing the bolt?

Some information about how this happened will probably be relevant. The other bolts went in and torqued to spec without a hitch. I tightened them by hand most of the way and finished with a torque wrench, tightening them to 28 ft lbs. The fourth one turned many times after I put the torque wrench on, enough times that I was nervous and on the verge of backing it out and starting over. And then I heard, not the click of the torque wrench, but a much louder crack.

The most likely scenario, it seems to me, is that I tightened to spec or beyond by accident before I put the torque wrench on, and so, since I was already over the limit I set on the wrench, I never got any feedback from it. If this is true, then I have massively over-torqued the bolt, extracting it will be very difficult, and I fear for the threads in the head.

It is also possible my torque wrench malfunctioned, in which case the bolt is probably still massively over-torqued. Or perhaps the bolt was weak for some reason. I believe these are stretch bolts, but I don't really understand the physics of stretch bolts, so I don't know what circumstances could lead to their becoming brittle or weak (besides the usual things like overheating).

The reason for not re-using stretch bolts?

The reason is, once you have stretched the bolt once, it never returns to its original size. This weakens the bolt, and if you try to re-use it, you take a risk of snapping the bolt in two, leaving the threaded part in the block, the top in your hand, and a complicated repair required. It's just not worth the risk to try to re-use old bolts.

As long as the threads in the head are OK don't worry about it. If the heads stripped here is thread repairs for that.

The manual has a specification for the stretch type bolts concerning there re-use. Some are one time use and others you measure and see if you can reuse them.

Remove the broken piece with vice grips, install two new bolts and drive on... you'll be fine.

Good luck!!!

Chances are, threads in the head are fine. Just get the bolt out and start over. Worst case scenario, you buggered the threads, and need to put in a helicoil or thread insert.

Time for some vice grips, or maybe grinding a groove in the tip of the busted bolt to use a flat headed screwdriver.... If all else fails, just weld a nut to the top.

It's good to see optimistic replies!

Assuming pimpernell is right, and the reason the bolt broke was that it was a re-used and therefore weakened stretch bolt, then it likely broke off under less torque than I had originally assumed. That means it is less likely to have done damage, and hopefully it will be easier to get out.

I'm also glad to hear that vise grips might do the trick. I was sceptical about this, and was waiting to hear some advice before I tried anything. But I guess I'll give that a go first and see if it works.

Four new bolts are in the mail (I'm going to replace the ones that went back in as well in accordance with pimpernell's observation), and I'll see how things go on the weekend.

Put the vise grips on TIGHTLY and place rags around to catch any shavings if they slip. Ideally and probably unnecessary in this case, use two pair of vice grips with the jaws directly opposing each other. The grips and the fastener form a "T" when correct, the broken bolt is the upright and the vice grips are the top/horizontal of the "T".

Apply equal torque to each vise grip, this centers the torque over the fastener and reduces the chance of twisting it off to one side.

Good luck!!!

Just re-visited the picture...gonna be difficult to get two pair of vise grips in there. One pair will probably work well.

Theoretically, the bolt shouldn't be very difficult to remove. As there is no head on it anymore to hold down any friction on the threads, wouldn't be surprised if most of the broken bolt removal tricks work for this one. You have access to the first 1/2" of the bolt, plus there's not too much stuff in the way around it. Also, the collar around the bolt comes out if you need. It's just a slip fit collar for centering the rocker assembly

Update
 

Update: The old bolt came out easily. Half a turn with a small vise grip, and then it was loose enough to turn by hand the rest of the way. Not sure what conclusion to draw about that; maybe that the old bolt was stretched to its limit, so it broke very easily.

But now I need some advice on the replacement part. The photo shows what I ordered. There was no photo of the replacement bolt with the description, so I was going on part number alone. I'll cut to the chase: does anyone know if the 17mm bolt shown is a valid replacement for the original stretch bolt, or have I ordered the incorrect part?

How I came to order this part is as follows. I had to consult an MB parts list, which had three numbers for this part (see images). The A112 part is still available, but is cost-prohibitive, especially since I wanted to replace all four. So I ordered the N304 part (which was much cheaper) from Pelican (via Mercedes), and this is what came in the mail.

The new bolt is the same diameter as the old one at its widest parts, so where it screws into the head and where it fits into the alignment groove on the horizontal bar of the rocker assembly. The new bolt is slightly shorter, but that makes sense as the old one was likely stretched to capacity. So it is plausible they could be functionally equivalent.

Why does the new bolt look so much shorter and like it came from Home D?

Well, I assume the difference in length is owing to the old one being a stretched-to-max stretch bolt. Can't speculate on the difference in appearance. The new ones came directly from MB. The new ones may not be intended to stretch, hence the simplified design? But that's pure, and uninformed, speculation on my part (and assumes they are the correct part, which is what I'm trying to find out).

1 Are the threads correct,
2 can a socket be used to tighten the new bolt i.e.,
3 is there enough shoulder room and
how much does the new bolt protrude through the rocker arm support?

If 1, 2 and 3 are yes and the bolt protrudes about 3/8" or a similar amount to an unbroken old one it should be ok.

Good luck!!!

I'll do the protrusion comparison this weekend.

What I'm really looking for is some direct confirmation or disconfirmation that MB used multiple bolt types for this application. I found a photo (attached) in a Pelican tutorial that shows a 17mm hex head like the new bolts have, as opposed to the socket head type that I removed. That tutorial was for a 617 engine, not a 616 (probably an irrelevant difference).

Additionally, I found a note (attached) in the FSM regarding cylinder head bolts, saying that MB had moved from cylindrical shanks to waisted shanks in 79. Perhaps they also did the same with the rocker arm bolts.

Subscribed .

I was taught that a wasited bolt would be stronger than the straight bolt but is the FSM says differently I'd be inclined to use it .

FWIW, I'd not want to use a bolt that much shorter .

1) [See highlight] If the bolt had already been tightened to a torque that was greater than the setting on the wrench, then the wrench would immediately click without any further rotation of the bolt.

2) Insert the bolt with washer thru the shaft stand and observe the number of threads exposed. Five or more, no problem. If four, just make sure that the torque wrench is set correctly.

that hardware bolt is not a stretch bolt so what do you torque it? Also may not be room for that hex head in place of the cap bolt. I'd order a new mb bolt if available.

That's the behavior I'd hope to see, but I've never started a torque wrench in over-torque, so I didn't know what to expect.

I decided to go ahead and look at this. Here is a photo. The perspective makes it hard to see the relative lengths, but there are certainly more than five threads exposed. What is special about that threshold?

This is a new MB bolt. The fact that this came directly from MB is what has led to my confusion. Part of what of I'm trying to figure out is whether there are two varieties of bolt MB used for this, and this part number is for the alternate style. There is evidence suggesting this is the case, which would mean the hex head bolt should be an adequate substitute. But no one has yet emerged with a definitive verdict on the two-styles theory. See one of my earlier posts for how we got here.

In the manual section 05-235 removal and installation of rocker arms with rocker arm bearing brackets it has torque 38 Nm

There is no further comment on torques or the bolts. No comment that the Bolts cannot be reused.

For installation it only has: "Set the complete rocker arm group and screw down"

The problem with bottomed out bolts with a click style torque wrench is the a lot of people will go to the click then click it again for good measure (doing nothing useful in a necked bolt but over torquing a straight bolt.

At five threads engagement 86% of the ultimate thread shear strength has been reached. A torque of 28 lb-ft is approx. 60% of the max. that could be used with that bolt. Hence, well within the torque vs. shear limits.

Who said anything about a bottomed bolt? Not the OP.

Yes! And in many ways, this is part of what confuses me, though maybe I'm overthinking.

Here's how I landed here. Puzzle #1: why did my original bolt break so easily if there shouldn't be any issue regarding re-use of bolts (I wasn't yet up to 38 NM/28 FTLBS when it broke, and was amazed at how easily the stub spun out). Possible solution to 1: the bolt was a stretch bolt that was stretched to capacity, and so shouldn't have been re-used. But this leads to puzzle #2: why does neither the FSM, nor any tutorial I've read about valve stem seal replacement or rocker arm removal advise caution regarding this? Instead they all say just to put the damn bolts back in and torque to spec. Possible solution to 2: there are two varieties of bolt MB used here, a stretch variety and a non-stretch. The FSM and most tutorials cover the non-stretch. This would also explain the multiple part numbers. And according to the FSM note I attached earlier, it looks like this is just what MB did with head bolts generally, only they made sure to note it, and perhaps didn't do so with the rocker assembly ones?

So what does this analysis overlook?

Just general stuff.

I have had click type torque wrenches that would get a glitch and allow you to over torque. I have set the torque wrong. On Truck Engines I have torqued bolts or nuts to the same torque and had head bolts snap or strip out the bolt of threads.
You never know what the person who worked on the Engine before you did.

Torque wrenches with a dial gauge generally stay in calibration better but you need to handle them carefully and in some positions you have lots of trouble reading the gauge. If you drop one I have seen the pointers come off or the clear covers get damaged.
Click type Torque wrenches have to be of a high quality in order for them to have decent accuracy. But they still get out of order more then other types do.

Occasional you get a Blot that did not get heat treated correctly.

I listened in on a conversation. A guy owned various earth moving equipment. It is not unusual for them to occasionally pop a bolt. He said that the Chinese Bolts of the same grade as US Bolts were cheaper but did not last as long. I mention this because like Torque Wrenches there is a quality issue involved.

That does not answer your question but does show the variations that are involved.

So this morning I decided to go out to the 240 and look at the mate to the broken bolt (I had screwed it into its socket to avoid losing it). I think the attached photo is pretty illuminating. I'll leave it here without comment for a bit.

You did in post 16, it appears to me.

I mention it not because you would ever do that but because a lot of less experienced folks do it routinely sort of like double knotting your shoe laces (perhaps) in their mind.;)

Perhaps the previous installer did not understand how the stretch bolts work?

Torque To Yield
 

FWIW, over my career I've seen and had more than a few bolts simply give up before they reached the proper torque ~ there's a finite lifespan and it's not always anyone's fault when one lets go .

This replacement bolt gives me the creeps, I'd rather source the correct one be it new or used .


That snapped one is clearly very stretched .

From my experience as well, have to agree 101% on that statement.

I would appreciate that words that I did not write not be ascribed to me.
If the accusation can be substantiated, then post my words, not just an accusation.
If use of the word "bottomed" by me can be found in post #16, post it.

Wasp Waisted Fasteners
 

It comes from decades of experience as a Journeyman mechanic ~ I don't just change parts, I try to learn new things on every job .

Wasp Waisted fasteners go back a long time, to the early 1940's that know of and being an air cooled mechanic (VW's & Motocycles mostly) I learned long ago that preventing broken or stripped out bolts & studs in high clamping situations coupled with large scale and frequent thermal cycles are the most important places to use them .

I know the DieselHeads here are a vast collective of knowledge and spare parts, I bet someone here has the correct original bolt lying 'round......

If I did I'd ship it to you post haste .

In the Factory Service Manual both on the CD and in the printed manual there is only a torque on the rocker arm bracket bolts.

Concerning the click type Torque wrenches. I saw a vid on one of the racing teams. He had the Engine on the stand and used a Click Type Torque wrench and went through the tuque sequence clicked on each bolt 3 time. When he was done he re-did all the bolts again clicking 3 times on each bolt and when he was finished he did the same thing one more time.

One of the good things about a torque wrench with a dial gauge is that you can use a cheater bar on them because the length of the handle dose not change the toque.
However, the longer handle often puts you too far away to read the dial. You, usually need 2 persons.

Torque Wrenches
 

I too use a click typ, American made torque wrench and click it at least twice .

When I worked in L.A.P.D. Air Support we had a torque wrench testing & adjusting rig, I found it very handy, my old wrench from 1972 or so was almost 1/2 pound off after decades of Commercial Service .

Over approximately 25 years of routinely using Mercedes-Benz FSMs on W123-617s, I have discovered an abundance of instances where said FSMs leave out plenty of relevant information (in my view). Why? I have no idea.

As a matter of personal judgment and choice, I routinely replace relevant bolts on said W123-617s with replacements from M-B. A small price to pay to sleep better at night and stay calmer by day. Your mileage, as the phrase goes, may vary.

1) [See highlight] If the bolt had already been tightened to a torque that was greater than the setting on the wrench, then the wrench would immediately click without any further rotation of the bolt.

That is the sentence I was referring. I see that you did not use the word bottom. I meant no negativity toward you at all in either post and I am truly sorry for any negativity I may have caused by my perhaps loose use of terminology.

My original comment was stimulated by me observing guys at the tire shop going to the click then rotating the bolt another 1/8 of a turn. I am sure they think they are going a doubly good job but in my mind that extra turn after the initial click is all over torqueing.

For the record I have the highest respect for your comments on the forum. They are always filled with excellent information spoken in the context of a truly conscious professional.

My comments are based on the perspective of a formerly advanced hobby mechanic but now more of an interested observer and internet groupie. My mission is to offer general advice here and there, possibly make someone smile....not to make anyone frown, especially you Frank.

Factory Shop Manuals
 

Way back when as I was getting my GM training I noticed the lack of pertinent info and asked why were so many common service procedures not being mentioned in what I thought would be the 'bible' of tech/repair info ? .

'Because you're expected tp have taken previous factory training and know these basics' I was told .

Humph, not a good reply IMO .

I'm retired from spinning wenches now yet I'm still learning and sharing my limited knowledge .

This morning I wrote an MB parts vendor for guidance on what the correct replacement part # is. He sent me exactly the catalog listing I posted in post 10, except in English (attached). And to reiterate an earlier point, the grey bolt in the pictures many are so sceptical of is part number N304014010009, obtained from MB by Pelican.

Moreover, I have a trusted Mercedes-only indy mechanic nearby who has been in business since 1970, so diesel W123s were his bread and butter for the middle years of his career. He liberally dispenses technical advice to customers learning to DIY, but I don't like to ask too frequently, lest I strain the relationship. But I finally broke down and called him this morning as well. He gave me only the N304 part as the correct part, and said that these bolts were not supposed to be torque-to-yield bolts, and that some prior mechanic likely installed those in error. This didn't seem quite right to me, as I feel sure I have seen photos of these bolts in tutorials, and they looked like ones I pulled out. He did say there was a relevant VIN-split, but I didn't want to take too much of his time, so I didn't want to ask him what the part # was for the split.

I'm guessing the part # for the split is the A112 number listed in the catalog. I'm tempted to order it for science, but it goes for between 60 and 80 bucks. If I knew it was the correct part, I'd gladly pay for it, but I have no idea what to think now.

Ironically, I just completed a valve adjustment on my other 83 240, which is my daily driver, last weekend, and I should have looked at the head of the bolts on that engine while the valve cover was off. But that was before the N304 bolts had come in the mail, so this issue wasn't on my mind. I'll probably pull the cover this weekend just to see what's there.

The necked-down capscrews as used in these applications are not torque to yield, rather, they are tightened to a specific pre-load within the elastic range of the material. A torque to yield fastener is tightened past the elastic range and into the plastic (yield) range.

A decided waste of money; the necked-down version has been superseded.

Interesting info. My 2 om617 turbos both have the same necked down capscrews that ya broke.

If this kinda thing really bothers ya, drop me a PM. I have a spare head that’s cracked...

I think I'm convinced now that the bolts I bought are the intended replacement part, so I think I'm just going to put those in and see how it goes. But thanks for the offer, and I'll keep it in mind if something goes wrong.

My mechanic is surely not remembering some things correctly, as so many 616 and 617 engines seem to have the capscrews, it seems. In his defense, his son does most of the work anymore, so he probably hasn't taken the rocker arms off a W123 diesel personally in a number of years.

Thank you, I'm happy to be learning things like this. Also thank you for the information about torque vs. shear limits earlier. This was all new to me.

I agree, this doesn't seem like a good reply. It is indeed reasonable to expect people reading the FSM to know "basics", but I don't think some of the omissions we've been considering here fall under the head of basics.

The purpose of necked-down bolts is so the shaft is more elastic, to act more like a spring to maintain a more constant clamping force on something like a head gasket, as parts change dimensions from thermal expansion. Since the camshaft towers have no gasket, that seems unneeded here, but perhaps they used them due to mixing aluminum and steel parts. Most head bolts don't need a neck-down since they are long enough to be elastic. Torque-to-yield bolts are torqued to a value, then marked and turned an exact angle from there (ex. 1/2 turn). That is because the turning torque can decrease once the bolt starts yielding, so if you kept trying to reach a higher torque value, you would just neck down the bolt and break it. This bolt is thus not a torque-to-yield assembly.

Perhaps M-B got reports of snapped bolts so changed to a stronger straight-shaft bolt. Personally, I would have gone to Ace Hardware and looked for a Grade 8 bolt (not sure they have in metric). 28 ft-lbf is nothing, almost 1 hand on a 3/8" socket wrench, so little worry about snapping a quality bolt. Insure you use 2 hands when torquing, keeping one pushing against the head of the torque wrench so you apply a pure couple load and not also a bending load like you would by just pulling on the handle, though unlikely that caused this bolt to snap.