Mtu 4000 series cylinder head problem.

Fuel quality is a possibility. EGT's that are too high or load is too high, are another possibility. Bad valve seats, or something wrong with the geometry in the valve train on your engine, or bad valve adjustment can be another.

If it’s just the seat only and not the valve itself then I suggest the valve clearance of the exhaust valve is too close , ie not a wide enough gap .
As they warm up and run well with EGT s and load for the generators because the exhaust valve is not quite shut or shutting fast enough gas escapes and over time the heat destroys the valve seat .

Or the timings a fraction out .
It may meet factory specs with the marks available but a batch of duff exhaust cams , cam gears means the E valve isn’t closing as it should at the right time .....staying open a bit .

Or the E valve springs are not correct spec .You have not said if the original valve gear has been transferred across? Assume naked heads and they transferred everything across with the original closing error(s) ?

what sort of hrs are we talking about?

How ever on reflection 304 heads in 3 yrs it’s unlikely to be a valve timing error so many times .
Generators running UNDER loaded most likely .
You need hight temps / reasonable high EGTs to get the best burn and highest concentration of “ protective “ by products .
A low loads the protection is not there .
There’s and old adage with diesels ( any ) “ run em hard “

more here ....

https://www.sciencedirect.com/science/article/abs/pii/S0043164896069451

Pre-ignition/detonation will cause burned exhaust valve seats as well. So if the fuel is too high a cetane number and the engines are experiencing detonation it could cause what you're seeing. But having these failures among 19 different engines points to something that is common between all of them in your situation. Could be underloaded, could be poor fuel/lubricity, or etc. Also, MTU has been known to have a batch of inferior parts now and then that they let slip through the cracks and then tried to get out of warrantying later, but miraculously have an "updated part" when you buy a replacement part. First 150 sets of 16v2000's had piston rings stamped and installed upside down and many failures, however on one boat they tried to blame the crew saying they had too much oil in the engine, until another boat came to the same marina with the same engines and also had a major failure requiring overhauls in the first year and low hours.

How many engine hours have these generators gone before having exhaust valve seat failures?

“ closing errors “ I mean gas is prematurely escaping because the E valves are not tight enough closed within tolerable timing .
How ever now you disclose 4/6 K hrs each that’s a different ball game in terms of getting to the aetiology of this .

Then they ( 304 ) were replaced as routine of which 50 were found with issues .

Coupled with Capt J ,s additional info on “ a batch of inferior parts “ ,my contribution of excessive low load hrs ( not enough protective layers forming ) and now yours hrs contribution I don’t think we are ever going to get a single definitive answer .

eg if due to “ defective parts “ poor lubrications.......say a valve guide wears then the stem will move , kinda vibrates about = poor seal = gas getting round = prematurely wear etc etc + not quite the right fuel burning at low ish loads = inadequate layer or valve seat protection by products etc etc .....Then toss in 4000/ 6000 hrs ........chasing shadows.


So the heads came off in routine maintenance....if I understand you correctly?
If so what’s the problem ?

MAN s exhaust valve lash is 0.6 mm , the inlet is 0.5 .
They tend to wear open ie get bigger hence the tapperty sound .
The exhaust clearance is always larger because they heat up more and thus expand a bit more .
But none of this is the cause btw ....assume the lash is set routinely, with MAN every 400 hrs ?

How often have those MTU s valve had there valve lash checked / adjusted?

Right / Ok gotcha .

Sorry I,am ( and suspect mist on here ) unfamiliar with the MTU service schedule.......thought the heads come off in routine maintenance.
I now see your anxiety.

So let’s eliminate “ open E valve / bad timing “ as wear means they actually prematurely close as the gap widens .Power is lost as the gas mix is not optimised ...buts that’s not the issue here .

I think we are back to low loading .....assuming none or little of Capt J contribution.

Could also be weak valve springs and the valves are floating.

Agree .
4/6 K hrs is about 10 years + normal leisure boating ....arguably near 20 years for some .So a big rebuild if they were in a boat wouldn’t really raise an eyebrow .
Obviously in a industrial application.....and inside 2/3 years I can understand the OP dissatisfaction/ disappointment.

Fiammetta42 may be on the right track with insufficient loading, or other operational issues. Bear in mind that vic13r said in his first post that the engines are used for generators. I'm guessing that with that kind of equipment he may be using them for an island/resort power supply? Those engines should be rated at over 2,000 KW.
Still not clear is if the heads are supplied with installed valve, and, exactly what the failure is. Are the seats cast iron head material or inserts? Are the seats burning or mating poorly enough to affect compression? It would seem to me that combustion temperatures are too low, otherwise the discussion would be about valves instead of seats. Any valve material unable to transfer heat to the seat should be burning away if the engines are loaded.
Interesting that the issue began with the installation of new replacement engines and that the factory has not become involved. The dealer should have brought them in. Some of the replacements should have been warranted, attracting the attention of the manufacturer or at least the distributor.
Are the engines direct replacements for the previous installation, as the same make and model as the predecessors? If not, are there environmental issues like insufficient or cool enough intake air? A real concern may be exhaust flow, as in new engines bolted to an existing exhaust system designed for different engines.. Are the engines equipped with exhaust temperature alarms? Did the previous engines lead a happy life and die of old age, or were there issues with them, as well?
What kind of fuel are they running on? Can there be a veg-oil blend? Is used motor oil added into the fuel?
Interesting to get some answers, and maybe failed parts photos.

The computer should be able to tell how they were loaded throughout their life. I'm not very knowledgable on the 4000 series but I think at 5-6,000 hours MTU calls for an upper end rebuild and a bunch of things, perhaps this is when the valve seats were noticed?

Sure would like to see that generator plant! 19 engines! Generating European power @ 50hz, those things are maybe running at a steady 2,000 RPM? Maybe no electronics needed for constant duty gen power. Looking forward to some more info.