Valve failure = piston demolished, 1300fwd

[Stupps]

Hello all,

Infrequent contributor to this forum but thought I'd share some pictures of a depressing piston destruction event (sorry to Sideways regulars about cross-posting).

I bought a 1300fwd a few years ago, then spent two long years slowly rebuilding the engine. Fully back on the road since April, it had given me various niggling problems but after 1800 miles I thought I was just turning the corner and could trust the bloody thing. 

Then on Boxing day, 15 miles after checking the plugs and valve clearances (everything appeared healthy) I was bowling along a country lane when there was a loud clackety-clack, loss of power, and I limped to a halt outside a pub. 

Took the plugs out and could see from no.4 that I had a major issue:

 

Got the head off yesterday, and surveyed the damage.

It turns out that the exhaust valve had broken, and whatever is left of the valve head must be sitting in the sump. Hopefully it didn't do any major damage to the crank on the way down. Can't drop the sump pan on these cars so I'll have to pull the engine out to fully assess the situation. 

Here's what was left of the valve stem

And here's that damaged plug next to no.1 plug. It looks a shade lean, but I wouldn't have thought too bad. 

 

The question is: how to I make sure this doesn't happen again?! The valves were standard items from James Paddock, and I have another batch from the same place to go into my Spitfire head. I'm a bit hesitant now though. Presumably it was a manufacturing flaw, although I wonder if the engine was running a bit hot and if that might have contributed. Anyway hopefully it won't take me another 2 years to get it back into service. 

Cheers, 

Pete

[Guest]

Hello Pete,

Thanks for posting this and sorry to see the damage.

Looking at the valve stem it appears to have sheared and that, as you say, may well be a flawed item. Of course the obvious question is, how many of these valves are flawed and have you got any more of them with no knowledge of knowing that, if that is indeed the case.

Might it be worth taking out the other valves and replacing them with others sourced elsewhere - any unused, I would return.

JP has a good reputation and I think the company will be concerned about this.

Just my initial thoughts.

Regards.

Richard. 

[RogerH]

Hi Pete,

it would be nice to see a good picture of the fracture face square on and in focus. That may show what happened.

Are these valves made from one piece or is it a stem/crown fabrication. If the latter how are they welded together.

 

If it is a one piece job then that is a strange failure - too square.

 

Roger

[Nigel Clark]

Hi Pete,

Very sorry to see the problem with your rebuilt engine. On a standard 1300 the valves aren't highly stressed. Most likely a manufacturing fault as had been said, very frustrating.

With his knowledge and experience, Roger may well be able to shed more light on the reason for failure.

Good luck,

Nigel

[poppyman]

I would be having a chat with Paddocks as said above..... That sure looks a faulty item to me.  Never seen a break like that.

Tony.

[johny]

I have wondered for a while where replacement valves are coming from these days as the main manufacturers can't be interested in producing such small numbers now. Was there a name on these items?

[thescrapman]

I have the remains of 3 big saloon engines that have suffered same failure, all at least 20 years ago, so heads dropping off valves is not new. 

And boy does it make a mess.

I would replace the con-rod just in case.

[Pete Lewis]

without a microscopic view  it looks like a crack has emanated in the stem from a forging defect or inclusion  and works across it till failure   hence the straight line

fracture , not a tear.

needs someone with the right gear and knowledge to give a reason behind this

pete

[RogerH]

A decent photo of the cracked surface (square on and in focus) will greatly help to understand what has happened.

A forging defect could well start the process but when the stem gets thin enough it will tear off with a raised edge (tensile failure etc)

To go straight across suggests some sort of geometry process hence my question about how are they constructed.

 

Roger

[Stupps]

Thanks for the feedback everyone, Colin I'm glad to hear it isn't totally unheard of. I'll certainly change the conrod, it's a bit mashed.

I'll speak to Paddocks today just to pass on the info, can't imagine they'll accept responsibility and it was probably 3 years ago that I actually ordered the valves. I will ask about who makes them though - don't recall seeing any manufacturer's details anywhere. The big question is how I treat the other valves from the same batch, and then the ones I ordered 2 years later which are a different batch (different size and collet type). Am inclined to ditch the other 3 exhaust valves, and possibly refit the old original ones as they weren't in bad shape. 

Spent quite a while trying to get a photo of the broken end in focus, below is the best I managed. Roger apologies it's not square on! Lighting for photography is tricky in the garage and I could do with a nice macro lens... 

Not sure how much of that damage was done after the head had broken off, there was evidently a lot of metal flying around for half a minute.

 

To be honest I don't know, but think these are two-piece valves. Is there any easy way I can check with one of the undamaged ones? Here's a pic next to an unused inlet valve (from a different set but also from James Paddocks) and the break is aligned exactly with the apparent change in materials.

[RogerH]

Hi Pete,

that is a nice pic.  At the 3-o-clock position there is a small lump. This may be the last bit holding on - but not for long.

Normally when a crack face propagates across the stem it will leave little ripples on the surface (Beach marks) until the stem fails.

There are no beach marks suggesting it failed in one go

There is no reason for the crack to go straight across the diameter as it has done if it was growing slowly.

I don't know if two part fabrication is the norm but here is a simple article suggesting it is  https://www.howacarworks.com/engine-valves

It may be possible to test to see if they are two part. The stem may be magnetic (high carbon steel. The crown may not be magnetic (stainless steel) however there are magnetic SS's.

 

I would suggest that the stem failed at the weld interface.

This is not good as it is very difficult/impossible to inspect.  Only quality production methods would avoid this issue.

 

Roger

[poppyman]

A set of Tranco valves on fleabay at the moment. Dont know if the same model Pete? https://www.ebay.co.uk/itm/Triumph-Exhaust-Valves-1147-1296cc-Herald-1200-1250-And-1300-1961-66/202643693655?hash=item2f2e816057:g:5cEAAOSwfHVbM7wJ

Tony.

[Stupps]

Thanks very much for your insight Roger, most appreciated, your last paragraph sums up why I'm worried about all the other valves!

56 minutes ago, RogerH said:

I would suggest that the stem failed at the weld interface.

This is not good as it is very difficult/impossible to inspect.  Only quality production methods would avoid this issue.

Seeing it in the flesh I thought that lump at 3 o'clock might be a ridge next to an indentation left by post-break impact from the flying valve crown; but in any case I take your main point that it failed abruptly at the join.  I'll have a play with a magnet just to see if there is a clear difference between head and stem materials.

Tony thanks for that link - I might well give them a go, they've got the right part number. Do we think those old Tranco valves would be alright with unleaded? I'm not sure if the newer valves are harder, or they're just labelled as 'unleaded' to draw money out of fools like me.

[JohnD]

Stupps,

You already have expert advice, so this site may be only for additional information: https://www.supertechperformance.com/technical

John

[Colin Lindsay]

40 minutes ago, Stupps said:

Tony thanks for that link - I might well give them a go, they've got the right part number. Do we think those old Tranco valves would be alright with unleaded? I'm not sure if the newer valves are harder, or they're just labelled as 'unleaded' to draw money out of fools like me.

The valves themselves aren't usually unleaded, just the valve seats, unless they're now being made with harder materials than originally supplied. If your head has been converted to unleaded those will be fine, but any lead memory from years of use will now be long gone from your head. The last two heads I've worked on got original Stanpart valves - still quite cheap if you can find them online; I've no idea if they're any better than those currently available but I'm wary of modern versions being inferior. It's not a widespread problem that I've heard of in normal road use so possibly you just got the one flawed valve out of a large batch. I've no idea how you would tell if the rest were okay without having each one microscopically examined or otherwise inspected.

[RogerH]

I imagine that the stem is welded to the crown by spinning one on the other. This generates massive heat. The metal melts and the weld take s place.

If the process is good (tried and tested) then  you could be happy there would be 100% success rate.

However who knows what the cheapo guys do.

As for testing I would (again) imagine they do a destructive test (Tensile pull) now and again (not for the cheapo guys) to see where it breaks.

As for Non Destructive Testing (NDT)  you are seriously limited. An external flaw may show up with Dye Penetrant (DPI) or possibly Magnetic Particle (MPI) but don't count on it.

An internal flaw may be detected with UltraSonic Testing (UT) but again don;t count on it.

More likely the weld was poorly completed and was doomed from day one.

 

Roger

[JumpingFrog]

I think there's only really one company supplying valves off the shelf now, which will be County/XRN, I don't think they do any actual manufacturing. AFAIK they supply everyone be it Canley's, Paddocks or Rimmers... Maybe Moss/TriumphTune ones will be different. However, there are other companies like G&S valves who are well regarded and will make you custom valves, no idea on the cost and you'd have to provide the profile and specs you want (maybe they would advise and copy a sample valve?)... Also no idea about minimum quantities...

Probably more information than you ever wanted to know about valve materials:
http://www.gsvalves.co.uk/assets/g-s-technical-infomation.pdf

Edit:
Old topic on Triumph-Exp, good read containing comments from a certain marmite Triumph Tuning guru:
https://www.triumphexp.com/forum/spitfire-and-gt6-forum.8/hardened-valves.949405/

[RogerH]

Excellent link.  Page 28 shows some good failures.

 

Roger

[Stupps]

Some really useful info in those links, thanks gents. That G&S manual is very interesting, a lot of it over my head but I'll save it for future reference. Was particularly interested to read that weak valve springs could contribute to stem breakage like mine, due to valve float. Not something I'd thought of.

I've ordered some NOS Tranco exhaust valves off eBay to use in the rebuild. Not that I necessarily think they'll be better, but I'll have some peace of mind that I've done something to avoid repeating the same failure. Will stick with the modern inlets.

I now feel a lot better informed about valves! Nothing is ever straightforward is it...

Thanks very much for all the info.

Pete

[PeteH]

Hi

It looks as if it failed where there is a change of section. Without the old valve head to check against it is a moot point, but there could have been a microscopic "notch" where the section transitions?. "notches" (rapid change of section) are a notorious point of failure and one reason why radius undercutting at the base is a common device when turning. I suspect that friction welding of the head to the stem was used in the manufacture, welding head to stem was becoming very common in the manufacture of Large Marine Diesel valves, "back when", as forging, especially when in excess of 12" dia, was extremely expensive by comparison. NDT will be unlikely to confirm one way or the other as to the quality of the other valves. I can only speculate but, unless a trick of light, I would hazard a guess the "fracture" started from the 3 O`clock position as the surface appears darker, suggesting a progressive fracture?.

 

It`s a bit of a long shot, but if you can get hold of (say) dye penetrant kit?, testing that area on the "good" valves, might? show any incipient defect, but as I say long shot?, and not necessarily reliable either.

[poppyman]

On 04/01/2020 at 11:56, Stupps said:

 

I've ordered some NOS Tranco exhaust valves off eBay to use in the rebuild. Not that I necessarily think they'll be better, but I'll have some peace of mind that I've done something to avoid repeating the same failure. Will stick with the modern inlets.

I now feel a lot better informed about valves! Nothing is ever straightforward is it...

Thanks very much for all the info.

Pete

Hi Pete, i have used Tranco valves many many times over the years, and found them excellent. I have even bought main dealer valves (Ford) and Tranco were in the box......

Tony.