Race Engine Forum UK

This head is from a Thema turbo 16V. The clearances were correct and the engine is all original. I had a thought, someone may have fitted NA valves at some stage.

Saw through the mid section of the stem on the chipped one - halfway between the radiused region behind the valve head and tip. It should be hollow and if you immerse in water, a rapid exothermic reaction will ensue if sodium filled...

GC

There are more easier and less destructive ways of testing.

Not that it matters much to the valve in question....... :!:

True, but that has to be the most fun way of testing it. Sodium reactions can be quite impressive!
Reuben

I'll have to give that a try!

My engineer told me the problem was most likley caused by excess carbon build up.

I was referring of course how to test the damaged valve not a new one.

Nothwithstanding that:

'There are more easier and less destructive ways of testing'

Apart from visual - they have a distinctive stem - such as?

GC

Basically test for magnetism and weight:

Lancia 16v Exhaust valves:

Sodium filled:

Has A362 written on it.

Only the tip is magnetic.

Is 3-4g lighter in weight than non at 52-53g (because it has a hollow stem)


Non sodium filled:

Non magnetic (anywhere).

Has a slightly waisted stem.

Heavier than sodium filled at 56-57g.



The sodium filled valve is filled 50/50 with an inert gas (Nitrogen) and Sodium which becomes liquid at engine running temps and slops up and down with the movement of the valve transfering the heat from the head to the stem and into the guide.



Hope this helps,
Martin.

Very good Martin, thanks.

GC

Thanks Martin. Integrale and Thema turbo valves are different though. Thema valves are half the price so there must be some difference there.

They are indeed different Steve, Thema S2 turbo and N/As have non sodium filled which is why they are much cheaper and you have one with a piece burned out of it.

Martin.

Well thats the answer to a few questions!

Thema turbo valve is still twice the price of the NA valves so I wonder what the difference is there.

As rule I don't publish part numbers but in this case I consider it important for clarity:

The part number of the OE 28.4mm dia sodium cooled 16v Lancia Integrale valve is 7650916. The cost today retail from Fiat dealer is £48.18 and they are going up at least £3 a year.

As far as my OE CD Rom part number records can tell me, the normally aspirated and turbo Thema have ex part number 7687460, in other words not the same valve.

GC

I would like to revisit this thread after a friend sent me a pic of his head (from his engine that is) after it suffered a loss of compression on one cylinder at a track day, below is what he found, the car is an integrale and if the aforementioned 'dimple test' is 100% then the exhaust valves must be Sodium filled.

Anyone any ideas on the likely cause of this?



Too much boost and not enough intercooling would be my guess, detonation? Not any obvious signs in the pic, but maybe the pistons would tell a different story.

In my view this is the result of excessive exhaust gas temperature, lots of potential causes for this, including worn valve-seat sealing (see below). I cannot see the other side of the head face which is where detonation shows up (usually) but irrespective I've never known detonation do this. I don't know the engine spec so if it's not highly tuned the age of the valves and their life cycle may be a factor.

Looking at material and its contribution to failure. The head material is 21-4n austenic stainless steel (BS970. Pt 4 349S52). Between 500 - 800 deg C it loses half its tensile strength and 1000 hr test data for that material shows that at 800 deg compared with 650 deg C say, its creep strength is reduced by 75%. Creep is, well, stretching.

I'm going to talk about loads (I mean stresses) on the turbo valve train for a moment.
When you compound that loss of strength at temperature with the bending, compressive and tensile stress that the valve experiences during the cycle (and bearing in mind that the valve come down on a seat randomly coated with carbon and other combustion deposits every time it closes) it's not hard to see why they fracture. The higher the cylinder pressure at ex valve opening the worse the problem. The product of cylinder pressure x valve area can place a very high bending load on the valve head - the cam is trying to open it and the cylinder pressure is fighting it. And yes they can and do bend, how much depends to an extent on the condition of the valve guides (worn guides also aggravate the distortion of the head as it slams down on the seat) and inlet valves definitely bend too if there is high temperature, high back pressure and inlet-opening well before tdc.

Then there's the internal (hoop) stress generated by thermal expansion of the valve head itself. Bear in mind that this thermal expansion (even though the valve head is er, round), cannot be assumed to be equal all the way round the rim. If the valve-seat combo is in poor condition ex gas leakage in one area can lead to local overheating of the valve and expansion of the valve head in an irregular way. I mean it can go out of round. This in itself can be an issue irrespective of ex gas temperature and can lead to cracking.

These things stack up.

GC

Thankyou for your reply Guy - much appreciated.

I'm now pondering over whether any damage has been done to the turbo from the bits.....

Martin.