Over the years i have read literally hundreds of people asking why Cosworth specified a maximum head depth of 138.68mm and can they risk going thinner?

Lots of people reply that it is solely to do with compression ratio, and whilst that is certainly an issue, it is not the most important one. In my opinion (Although i am not an engineer) it is also largely to do with structural rigidity, heat distribution and valve seat security in the head.

I very recently had cause to dismantle a YB which has suffered a catastrophic engine failure due to the loss of a valve seat whilst running and it happens to be a very thin head so i can use pictures of it to illustrate teh difference between a new head and a heavily skimmed one. (Not necessarily the cause of a seat or any other type of failure of course, but good for visual education)

Hopefully, these two pictures will tell a thousand words about how to tell if the heads very skimmed or not without measuring it. This will save my tired old fingers.

New unskimmed head direct from Cosworth:





New head cropped around inlet valves:
(Please look at the material to the right of the valve seat)






Over Skimmed head cropped around inlet valves.
(Please look at the LACK OF material to the right of the valve seat)





Actual size of the over skimmed head measured face to face at the front.



As you can see, the skimmed head has very little, if any, material to the right of the valve seats, this means when it gets hot the retaining tension around the valve seat could be somewhat uneven and the hammering of the valve against the seat at high speed could cause it to come free, thus destroying the engine! It also of course means that the compression ratio is higher and the valves are extending further into the cylinder than normal, and they catch pistons as standard as it is if you over rev them.

Hopefully, this will help people in the future when they are buying expensive heads that have been over skimmed by unscrupulous traders because you dont know how skimmed they are until you have seen an unskimmed one as per my pictures. Slight skimming is fine, as Cosworth naturally allow for this, but beware how far you go.

Please, if you know of people who have had such damage recently, dont bring them up in this topic as it may affect any ongoing legal cases if they were to think this was their head...

 

Interesting stuff Stu. So is there anything that can be done with an over-skimmed head. Repaired maybe or just fit for the bin?

 

Nice info.

Could the installation of a different material, say Inconel (for it's high temp properties), stop this from happening?

 

you do have to bare in mind that i doubt their are very many cossie heads running on the recommended limit! (mine is as new as its never been apart )

 

Aint much head left to the right of the valve.lets hope people take notice of this,could be majorly exspensive if not,thickness of head is down a lot aswell.wud say it could not be repaired and fit for scrap?

 

Chatting to Speedy Sheady today, he reminded me of something else to do with valve seat failure that may be worth a mention on this topic. When porting these heads, you must not take too much material away from the actual valve seat area as you compromise its seal against the head and this can DEFINATELY cause valve seat failure.

As standard, the seat throat is about 27mm IIRC. Its acceptable to extend to maybe 28/29mm, but after that you really are compromising the valve seat security.

Ive seen one fail recently at around 31.6mm, which means the seat lost 2.3mm of its material all round during the porting process! No wonder it fell out when it got hot!

 

Personally I would just like to say thanks to Stu for explaining that in virtually laymans terms for me, that post makes the world of difference to people like myself who are at the bottom of a very steep and expensive learning curve.
Cheers

 

squish tastic pics

 

So It deffo down to failure of the head and not other reasons?

 

I believe there is something that can be done to over skimmed heads to "reclaim" them, but im not 100% sure on the feasibility.

 

Very helpful thread, cheers.

 

Many thanks, I'm glad it helps.


Wrong topic maybe? Nothing in those pictures has failed...


Yes, a few things, generally i believe you can actually recess the valve seats further into the head and re-machine to suit. There are complications wit doing so but its cheaper than a new head which is of course, currently priceless due to being made out of unobtainium. Karl Norris used to be the man for this job, not sure if he still gets involved though.

 

this was explained many moons ago with, from what i can remember, the same sort of pics

i was always under the impression that once it was skimmed into the leveo of the valve seat it was over the aceptable limit to use but no one has, as far as i have seen or can remember, explained why this was the case

if there is a proper reason rather than just being "becaue someone said so" it should stop all the issues surrounding this

 

Would that include taking a bit off the top of the valve (the tappit end) ?

 

Thank you stu imho these days if its not MSD,NMS,MAD,SCS,Reyland & (a few other apologies if i havent mentioned you) if its not any the well known proven tuners best avoid now as ppl are having to repair their car twice or 3 times before the go to any of the above as there are that many ppl kidding themselfs on that they are experts.

 

i am struggling to understand this a little, valve seat i presume is the area that the valve sits into which is cast into the head? when you say fall out, is it a part of the head that breaks away or is there some kind of insert that sits in there?

 

When the material around the seat gets hot, it expands and as a result allows the valve to fall out.

The more meterial there is, the heat wont be as high as there is more material there.

Correct me if im wrong Stu.

 

good thread, and its something to bare in mind!

 

well if that is the theory, i completely disagree

 

Excess heat does play a part in the valve falling out, im pretty sure of that.

 

What a was trying to say was wiv the head being so thin around the valve,obviously when its gets hot the the metal expands? am i right? thus it wud cause the seat to drop causing head failure and worse damage or am i on the wrong track here lol

 

valve seats are hardened metal (not sure which metal, deffo not alloy) there inserts pressed into the head.

 

The valve seat is an insert Twellsie.

 

remember chaps, we are dealing with two materials, which two different co-effluence of linear expansion

 

ah right, thank you

 

Anything's fixable..

 

i would still suggest that with the head being Ally the reduction in strength by reducing the material that little would have very little if any effect on thermal expansion, as Ally is such a good thermal conductor heat transfer is very uniform irrelivant of shape (to a degree)

 

Good topic

 

The valve seat is a special hardened material that can take a red hot valve hammeriing into it at high speed all day long without failure. It is an interference fit and thus requires thermal stability and structural integrity of the surrounding area to retain it.

 

Good thread with even better pictures!

Alex

 

which i would suggest with the large amount of material behind it like it has, i believe it will still maintain its structual rigidity, is this based on the fact that the valve throat (if thats what you call it) is impinging on the hemispherical area?

 

the valve seats are removed by heating the head to around 200 degrees C at this temperature they normally fall out on an aluminium head with the aid of a very expensive piece of equipment called gravity.
Valve seats normally fall out/ come loose for the following reasons
1 head has been run hot at some time in it's life
2 long term heavy detonation at some point in it's life
3 over bored throat during the modification process of porting
4 poor machining of the head at production
5 insufficient interference fit at production
6 4 and 5 after market valve seat replacement.
The interference for the seats on a yb head is 8 thou.the seats must be frozen and the head heated to install them.
skimming slightly below the recommended limit is not normally a major factor
the exhaust seat is more likely to give problems than the inlet seat, as the inlet is cooled by the incoming charge and is on it's seat during the hot exhaust cycle, unlike the poor exhaust valve

 

 

lol lol lol at tony

 

I'm sure they are made from Beryllium.

 

Easy enough to sink the seats in deeper to salvage a head below the limit IMHO

Obviously can only go so far still though.

 

And when they get hot do they get tighter or looser in this instance.....

 

I agree martin I think they are made from nickel berrylium alloy

 

I`ve seen probably no more than 3 heads this thick or more in the last 8 years....




I`m with Tony on this one,poor modification/repair has far more to do with this problem than overskimming........

Stu, you have just scrapped 70% of Cosworth heads in the world with one post

 

looser chip