Britbike forum Forums British Motorcycle specific Brands Triumph bulletin board Disaster! What caused this?

Italian after market pistons have used the double hook circlips for as long as I can remember. JCC pistons, which have dominated the US British aftermarket, have used the other single hook circlip pictured for the twenty odd years we have distributed them. The circlip itself is not the cause of this problem.

It could have been how they were installed or the groove they were installed in, but it wasn't the quality or design of the circlip. The double hook design is very popular with people who make pistons for diesel engines. Even Mahle uses the double hook in their diesel pistons. Yes, if this was an engine that was revved to 9,000 + all the time the weight of the circlip could be called into question, but red line for this motor is no were near 9,000.

It was said that the pin didn't move. Well being a free floating pin it has to be free to move. If it doesn't bad things are going to happen. Clearly in the pictures the pin came to rest more to one side than the other. This is not a failure you often see where the wrist pin is a press fit in the rod. But with the pin free to move about one has to pay closer attention to condition of the rod.

"Both pins are a good fit in the bushes..." but if they were a good fit (.00075") they could not be rocked in any direction. When oiled they would be a very light push fit. With the pin being able to rock up-and-down slightly it makes it even more important that the pin be parallel to the center line of the crankshaft.

Given that the pin should be parallel to the crank +- .001" over 6" it is going to be interesting checking the wrist pin bush.

Being aware of, checking for and correcting all possibilities that were responsible for a failure is what professional mechanics are expected to do. It is in no way looking to make things more complicated. As I suggested, a simple check for the home mechanic is to put the pin in the rod and lower it against a pair of parallel bars laid front to back on the crankcase mouth. While not the best way to do this it will give the home mechanic a clue as to whether he needs to seek professional help. And yes aluminum rods can bend and rod bushings can be reamed at an angle.

The checking of the connecting rod for bends or twists, as I suggested, is essential. It is even more important when a new rod bushing has been installed. And no the pin should not rock in the bushing!

Gudgeon pins are often a tight fit in the pistons, when the piston is cold. Also, I don;t see how it is possible to claim with such confidence that the circlips were not in any way faulty or sub-standard. That photo posted by Triton Thrasher showed a distinctly dodgy looking circlip, proving that they do crop up. Finally, not all JCC pistons come with any 'hook'.

With John's reassurance that the wingy circlip was ok, I might have used it, or maybe not. We're responsible for our own roadgoing scrapheaps and the original type made me feel better.

I had a bout of circlip pushings-out many years ago. I blame slight play in the big ends, but I've never gotten anyone who knows anything about engines to agree with that.

The wear pattern on big end shells can often confirm the circlip problem was actually a bent rod. As the pin moves side to side it rocks the rod. The outside edges of the bearing shell will display abnormal wear. I got a picture of just such a bearing shell that illustrates this in this computer some where. It was from one of those learning curves people talk about. I will try to find it and post it.

Yes, but that is not an issue here. When warmed in use they will have near .002" clearance more than enough to be classified as a "Floating Pin."



For that matter there is no way to prove, or have confidence, that they were faulty. That is why you examine all the other possible causes. Just because you have not experienced such a failure, or your curiosity over the yearsexposed to the reason circlips can fail doesn't mean it isn't true.

May I quote PE Irving, "Occasionally an engine will exhibit a tendancy to hammer the circlips loose, more on one side than the other; in such a case examine the alignment of the small end bush for both squareness and absence of "twist"; the latter is difficult to detect but may be the unsuspected cause of the trouble." Tuning for Speed, PE Irving 1948.

There are other clues, but they require examining the rod bearing shells. Often you will see where there is abnormal wear near the edge of the bearing. This happens when the rod rocks back and forth to the action of the wrist pin. One other

This motor has only done maybe one thousand miles since a rebuild and the big ends, small ends and main bearings were all replaced at that time.
I think John is on the right track about the small ends

davy

I agree that measurements are the key to understanding what happened here.

But the following tends to provide more support for the small end theory.



Having clearance in a vertical rocking direction would support the idea of the small end not being perpendicular to the crank. I would expect it to try and tilt the piston which should wear the bush on the top edge during the up stroke and the bottom edge of the opposite side during the down stroke. Because this is more point contact than full contact across the bushing, I'd expect those ends to wear quickly. That wear should allow the pin to rock in the bushing.

regards,
Rob

Listen guys... I am not trying to be a bully here. Just that I was taught in High School not to take things at face value. Sometimes your first guess is just that...

Enough of the cryptic stuff John, just say what you think.

davy

Tasman, NickL mentioned crank end float, how much do you have ?

Regard Peter

Again, thanks everybody for the input. I'm narrowing this down.

I've got the motor out of the frame and tomorrow I'll break it open and see what horrors lie within.

Peter, I'll measure the crank end float tomorrow. What should it be?

Tasman, I'm more at home with unit engines where the crank is held in position by the timing side ball race on all but the later t140's, so I would say the end float should be minimal ie the side to side clearance of the main bearing. If its more then either the bearings are loose on the crank or loose in the cases although they may not become loose until the motor gets hot. refer to the earlier reply from Excaliber, he's covered the important issues I think.
I welcome comments from forum members with pre unit knowledge on crank location etc.
What happened to all the old parts that were replaced ? They may be worth a look at for evidence.

Regards Peter

Triumph cranks do not have end float as such, but require a certain amount of clearance to prevent any side loading of the bearings. Depending upon year, and bearings used, the crankshaft is located by one of the bearings. Some models it is secured by the timing side crankshaft pinion nut and on others the alternator crankshaft nut.

Yes, with age and abuse outer bearing races wear and the bearing that is supposed locate the crankshaft comes loose and it is possible for the crankshaft to move. But this is not the norm and something that needs to be repaired. While a certain amount of primary chain misalignment will be tolerated by single row and duplex chains bad things happen when triplex chains are run on misaligned sprockets.

So lets explore the theory that you need to shim your crankshaft to the .003" to .017" listed in the Workshop manual. First, other than theorists who read this for face value and post to web sites, there is no other reference to shimming the crankshaft in any manual, factory literature or Service Bulletin I know of. If there is please make me aware of it. Is there anything wrong with shimming the crankshaft, especially when you are looking for hp and want the rods centered in the bores. Certainly not, but for a street engine it is rarely a problem.

I think we can agree that all that is written in the Workshop manuals should not be taken as fact. Take the T140 where the triplex primary chain requires that the sprocket alignment should be less than .010" and the factory supplied reference to, and parts for, setting this alignment. The Workshop manual list as acceptable .003" to .017" "end float." If in fact the crank actually "floated" .017"depending upon initial sprocket alignment you could end up with double the recommended sprocket alignment.

Other brands, like our cousins the Norton and BSA twins, use very different bearings and mechanical set-ups. The crankshaft is free to move from side to side in the crankcase. Unless something is very wrong, Triumph crankshafts are fixed (with clearance to prevent side loading for the bearings). There is no downside of shimming the crankshaft, but as is often quoted here, "there are thousands of Triumph running around running perfectly OK" that never had the crankshaft end clearance measured or shimmed. IMHO it is much to do about nothing.

Tasman: In the range of potential piston failures caused by a circlip yours is relatively mild. A lot more can happen. These can be so bad that the circlip almost seems to vaporize, or is broken up into small pieces with severe damage to the piston and cylinder wall.

Now, you either caught it soon after it happened or the reason for the failure was subtle. If the clip was truly installed properly I don't think the damage would have been so subtle. the clues for the failure might be subtle (the rocking pin for instance). But once the circlip gets a bit of room to work in the groove it isn't long before problems start. Yes, the circlip created the damage you see, but it is what caused it to become loose in the piston's circlip groove in the first place is what you must examine. Given the nature of this failure I suspect you will have to look closely at the rod bearings, and connecting rod. The rock in the pin bushing is the mechanical equivalent of having a bent rod.

It is common to take a quick look at the piston and make a quick assumption. People say it is obvious that it was the type of circlip or how it was installed. But I have learned to look beyond what seems the "obvious." Especially in situations where there are know "other" reasons for this type of failure.

And yes this type of circlip has its critics. Where they are used in high revving engines or where there are exceptionally high pistons speeds any type of wrist pin clip can fail. This is why in racing where the circlip ends are placed is a concern. When the piston changes direction the circlip tends not to stop moving with the piston. It springs closed and can either break or enlarge the groove leading to eventual failure.

Because of the mass of these circlips this type of circlip would not be recommended to be used in a high rpm engine with a long stroke. The T100 is not this type of application and this circlip has been used in British pistons for a very long time. If one looked deeper into failures of this nature in the past and examined the other factors that can cause this type of failure the "common" knowledge that these are bad would be different.

When I have experienced this type of failure I always ask, "what did I do last." It is human nature to blame the other guy but I am not ashamed to say a lot of failures I have experienced were the result my lack of experience, not the part.

The plain wire circlips (with no ears attached) are a better idea. If the ends of the pin have a 45 degree chamfer, any force from the pin will push the circlip harder into its groove instead of pushing it out of the groove.

A twisted conrod (as viewed from above) could still cause problems in an extreme case. A mandrel through the small end, and viewed from above, would not be parallel to the crank axis. I've seen rods bent in the other plane (too many times), where the small end was 0.020" or more left or right of where it should be and the small end and pin out of parallel with the crank axis, but the circlips still stayed in place. I put most of that type of bending down to heavy handed owners hammering the pins out of pistons, and I now always check rod alignment. You could get the same effect by reaming a small end bush out of square, and not parallel to the crank axis.

I don't believe that the timing side main bearing will (or can) hold the crank securely in one position, just because it's clamped tightly to the crank. You may or may not even have a clamping washer fitted, to clamp that bearing to the crank. If you have a roller drive-side main, you should use the clamping washer and shorter pinion gear. The outer race of the timing side bearing bearing is NOT held tightly by the crankcase when the engine gets hot. Outer races can and do move and rotate, etc. How much they can move depends on end float, which is what you would measure if you could slide the crank left and right through the inner bearing races. Sometimes measuring that can take some force and ingenuity. You won't easily measure it if the crank is still clamped to the main bearings.

The crank will have about 0.005" more end float when the engine is hot, so I always go for the minimum of 0.003". Any shims to achieve that normally end up between the drive side bearing inner race and the crank shoulder, to keep the conrods central in the cylinder bores. That's not necessarily always the case, but just more often than not.

Here's a few photos of what I've got to contend with.

When I took the circlips out of the pistons one of them had broken and the broken section was lodged in the gudgeon pin.


Damage to the cylinders is bad enough to warrant a rebore but as I'm already at forty thou over I'll be opting for a resleeve back to standard.


The gauze strainer in the sump did its share of minimising damage and the spin on filter on the return did the rest.


Unfortunately enough metal evaded the filters to ruin the big end bearing shells but thankfully the crank journals don't look too bad.



So now I can order the parts I need and thoroughly clean everything that I'm keeping. I'll get the crank and rods checked for straightness and find reputable repairers to do anything that I'm not 100% confident doing myself.

I've got over the initial anger and depression and beating myself up over this and I'll be using the down time to fix up a lot of niggling little issues that have been bugging me. Onward!

When looking at this you must remember the stock Triumph wire circlips are made from .048" wire. The wire circlips used here are typically made from .060" wire. The grooves in the pistons are cut to suit the wire size. You should never replace a wire circlip with a square edge seeger circlip as the pin will spit them out faster than you can say seeger.

If you were looking at replacing the ones that came with the pistons you would have to find ones made from the same wire size and relaxed diameter. Failing to do this would cause problems of its own. This isn't just get any old circlip and your good-to-go.

As far as the "shimming" the crankshaft, removing the bearing must be done in a workmanship manner. It is easy to damage the bearing, or the threaded end of the crankshaft with a puller, as you take it off to install shims. Of course if the bearing is a loose fit forget I said this. For practical reasons I have always been reticent to do this. While in itself there is no damage done by installing shims, to do it without the proper tools and experience you risks causing more problems than you "fix."

And remember the shims do not always go between the drive side bearing and the crankshaft. To maintain connecting rod alignment one has to note which side was used for reference when the engine was ORIGINALLY built.

Up through 1953 the timing side ball bearing was retained by a circlip and the crankshaft was referenced to the timing side. Then shims would go on the drive side. In 1954 the circlip was removed and crankshaft was referenced to the drive side. The shims would go on the timing side. Later engines using the drive side roller reference moved back to the timing side side.

There are two things to note here: One — if when the engine gets warm enough to free the bearings so it will float the noise from the crankshaft would drive you crazy. With time it would only get worse. You would think that the pistons were trying to swap bores. Two — Triumph had enough experience with this engine to remove the circlip and rely on the fit of the bearing in the crankcase to firmly locate the crankshaft. For nearly 30 years they never looked back at the circlip.
John

PS: and have the rod wrist pin end bushing checked for size and squareness.

John, I find what you say about bearing / housing fit very reassuring. It is worrying to read that the outer races quite normally become loose enough to rotate freely when the cases get hot. I most definitely prefer your account of things, and I hope very much that you're right!

My T100R crank is timing side located and after the first 50 miles I rode it, it started making a horrible hollow bell like noise at low speed and load.
First problem was the chamfered washer between the crank bearing and pinion was fitted the wrong way around allowing a lot of potential end float and second was the bearing was moving in the bore as the cases expanded and giving around 27 thou actual end float.
I orientated the washer correctly and shimmed the end float to 4 thou cold and it's been a gem ever since.
Fwiw even at that excessive end float the circlips showed no inclination to displace themselves. and no other abnormal wear was evident

davy

Main bearing outer races do rotate in their housings, on a hot engine, even with a good interference fit (good is between 0.0015" and 0.003"). The bearing races aren't skidding, but slowly rolling around in a housing bigger than the race. Differential expansion is often enough to allow this, but you also have firing forces stretching the housing (to make the hole even bigger) on each firing stroke. Even imbalance forces on the crank can be as high as the combustion force. No amount or grade of Loctite will stop this happening.

Here's a discussion and evidence of it,that was on the Norton Bulletin Board a few years ago. Main bearing outer race rotation in housing

So, this is why Paul Dunstall pegged the main bearings in his Dunny Racers Details of this are in his booklet "Norton Tuning "
I'm giving this serious thought for my Triumph racing project. Of course, the real answer is how the Goldies, Manx Nortons, and the sublimely beautiful AJS/ Matchless G50''s addressed this!

Pete, I read that when it was first posted and could not relate to his findings. He lost me in the first sentence: "Just to re-confirm what I had found in the past I pulled an old virgin 750 bottom end off the shelf and checked the end play." The pictures show an engine that is anything but virgin. His assumptions with this engine are made without confirming the condition of the bearing housings.

Triumph crankcases do not operate at 220° F. I use 220° F as the temperature I aim for when installing new bearings. At that temperature the cold bearing falls into the crankcase and quickly locks in place as the bearing reaches the case temperature. How about a test at a more practical temperature on cases where bearing fits confirmed. When was the last time you boiled water on a Triumph crankcase?

The movement provided by the maximum allowable end float shown in the Triumph Workshop manual could easily exceed the maximum allowable misalignment of the triplex primary chain. That asks more questions that it answers. Just saying, you haven't made me a believer. Good basis for an article in VB.

All this is really interesting, and much of relevance has been discussed. Now, my 1932 MatchlessBigtwin has the little ends offset so the cylinders are in lnline (1935 these engines were redesiged wich used compound conrod/bigend assembly which obviated offset little ends). Now, on a low revving (26 hp @4400 rpm),as long as certain parameters were met, this could be gotten away with. But, at higher revs, and because of the physics involved, this could cause the gudgeo to place heavy loads on one of the gudgeon circlips. Just the same result as John Healy says is caused by bent conrods, badly reamed little end bushes, etc.
I honestly believe the problems with this Triumph will be found in the gudgeon area, due to bad little endsm orbent conrods. I concur with JH.

Triumph crankcases do not operate at 220° F. I use 220° F as the temperature I aim for when installing new bearings.

I'm not sure I'd fully agree with this as after a 30 minute run to the coast and back this afternoon that included a blast up to 90mph and a fair bit at 75 t0 80mph I measured my external crankcase temperature at 195*F and the return oil filter external temp at 185*F after the bike had sat idling down for a couple of minutes.
I think hard run on a hot day it will easily exceed these numbers (In fact I know it has in the past and I'm pretty sure the bearings loosen up enough to move well before the 220 degree number mentioned under loaded running conditions)

davy

NickL - Keep in mind that flywheel is from a 6T and is in a T100 and is balanced to 75%. The balance job was done by a specialist who races Triumphs in VMX. It may look scarey but it works well. I wouldn't race on a bolt up three piece crank period.

Those holes must cause a lot of oil drag.

davy