Valve train failure.

[86rocco]

Over the years, I've seen numerous modes of failure but this one is a first for me, the valve stem broke at the retainer groove. Any comments? Anybody had this happen to them?
BTW, this is out of an Esslinger head for a Ford 2.3L

IMG_0196.jpg

[jfholm]

Ed,
Yes I have seen this happen. The keeper grooves are square and can have a stress riser start in the corner of the keeper groove. It really happens the most with valve train float or the valves are bouncing. You will see that most high performance heads at now coming with rounded grooves for the keepers.

Years ago when we were racing on the sand we had a air cooled VW engine and when we redid the valve job you would almost have to chisel the keepers off of the valves. We thought we needed to lighten up on the spring pressure. Gene Berg in California told us we did not have enough spring pressure and we were getting valve bounce and float. We went to stronger springs and never had an issue again. When we would take the heads apart the keepers would fall right off of the valves.

Another thing is the spring harmonics you have to deal with. That is why the Spin-O-Trons are so popular now. The pressures may be the same on two different springs but due to their wire diameters, angle of the coils, etc. will make them act very differently on the same cam.

[86rocco]

I'll have to check and see how much time's on this head, not that much I think. Anyways, the head was designed (including cam selection) and built by Esslinger expressly for the the type of racing it was being used for, so, I would find it surprising, given the vast experience they have with this type of engine, that there would be a fundamental flaw in the valve train design, maybe they got a batch of substandard valves.

[Darcy R]

It's difficult to see what happened first, without having the parts in a persons own hand for close inspection, but I would be taking a close look at the marks on the edge of the retainer, and corresponding contact marks on the underside of the rocker/follower. This contact would take place after breakage occurred as well, but that looks like a lot of contact marks for the little damage visible to the head of the valve. This type of valve stem breakage can happen from retainer to rocker/follower contact. Is there any signs of contact on others that did not fail?

[86rocco]

There are some nicks on the retainer, lash cap and also the valve spring, but the rest of the valve train parts look perfect even the rocker arm from the offending valve, so, my best guess is that those nicks happened after the stem broke. Needless to say, the rest of the engine will be getting a very thorough inspection but based on the initial tear down, surprisingly little damage was done to the the rest of the engine, it appears pretty much the best one could hope for after a failure like this.

[jfholm]

Ed,
Also check the lash cap depth. If the lash caps bottom on the keepers before the inside bottoms on the end of the valve then it will push the keepers down a little in keeper groove. Even if it is just a few thousandth of an inch it can let the keepers move and when the valve gets back to the top and there is valve lash the keepers get slapped into the top of the keeper groove and eventually this can cause fatigue. Just make sure the lash caps are not bottoming against the keepers. Just a thought. If they are there will probably be shiny marks on the bottom edge of the lash caps.

[Hotz]

this is a fissure? normalization problem when hardened

[86rocco]

That looks VERY familiar! So you're suggesting it's a manufacturing issue related to the heat treating, that makes sense to me as we've found no other issues that might have contributed to the failure. Anyways, the head has since received a new valve guide, all new valves, springs and retainers.

[rwdford]

Definitely a problem with both the square groove design and the material quality or heat treatment of the valve

Radius grooves are way safer than square locks, the radius spreads out the stress over a larger area where as a square corner creates a high stress point where all of the pressure will go, in the UK and EU 4V heads use single radius grooves never square locks, even with large diameter small stem race valves

Something I would look into is if you were getting valve bounce on that valve and the rest of the valves, inlets are heaver and a lot more prone to bounce, checking the spring pressures would be a good idea to see if some of the springs had weakened over time, when valves bounce it places a huge strain on the locks, valves and retainers, several times the pressure at full valve lift

The 2.3 will roller cam and 11/32" stem valves need a lot of spring pressure for high rpm's, in the region of 145 to 155LB seat and closer to 400LB at peak lift if spinning to 8.5K

You were probably just unlucky with that valve but I would certainly check all the above, fitting new valves springs, valves, retainers and locks all round was a very good idea

[86rocco]

The last couple of race seasons, it seems a lot of people I know have been having issues with Esslinger heads, the build quality just doesn't seem to be what it once was, another one of their heads which we looked at recently had hydraulic lifters that would pump up at about 6000rpm, it turned out valve springs that were WAY out of spec. I'm not sure what's going on over at Esslinger, I hear they had a change of ownership, maybe they lost some key personnel as a result whatever the reason, we're now finding it prudent to double check every single new part we get before it goes into a car. It's been a real pain!