Menu Sign In Contact FAQ
Welcome to our forums

Extra 400 upgrade project - call for advice


No this is the procedure we defined. No reprogramming at all.

LSZH, Switzerland

Thanks Peter

Will take a look at the sources you mentioned. Re Cylinders: I have not yet had the discussion with my engine shop in Canada. From the email exchange we will first replace only on (the last ECI) and observe.

LSGG, LFEY, Switzerland

You won’t believe this, but after 3 short flights we had our third cylinder failure.
The third failed cylinder was an overhauled one that had been delivered as a spare from Canada to replace another failed cylinder.
The failure mode is a crack in the cylinder that lets the coolant leak in.
The only symptom is that the CHT starts climbing.

Positively NO significant EGT event, but the EGT curve of this specific cylinder – while following that of the others- sits a little bit lower (100 – 150 F) than the hottest EGT.

I was PIC, so I cut power and immediately aborted the flight when this happened. Coolant temperature (shown on a steam gauge ) was not in the red, but I can’t say if it was climbing abnormally fast (this occurred while the AC was climbing at high power, so you’d expect coolant T to rise anyway) and we have no log…

My engine overhaul supplier in Alberta had done extensive integrity tests on the spare cylinders before shipping. After analyzing the EDM data log, they suspect something wrong with my spark plugs. They say they have talked with Continental and both believe that a plug is slowly failing, resulting in high plug temperature and causing detonation.

These plugs are fine wire Iridium Champions. All of them had been cleaned and bench tested in Canada as part of the rebuild process and my supplier had considered them serviceable.

I have exchanged emails with Champion. They challenge the common belief that any plugs showing above 5 Kohm when tested with low voltage should be replaced.
Their point is that the plug’s resistance at low voltage and temperature is irrelevant, what’s important is that it makes a nice blue spark, which implies that its resistance is OK under high voltage. This makes sense to me…

Anyone here who could provide advice?
I could share access to my engine logs at Savvy Analysis.
Please keep in mind:
- this is a TSIOL 550 C turbocharged water cooled engine
- completely rebuilt but the cylinders have “only” been overhauled, not replaced
- pistons, pins, exhaust valves, valve guides, tappets, camshaft etc… new, intake valves re-used.
- engine TT is 1200 hours, less than 15 since rebuild


LSGG, LFEY, Switzerland

If you search here on
champion resistor
you will find some old threads on the resistor topic.

I agree with their statement that the low voltage test doesn’t mean anything, with their old resistors. But I thought the changed the design away from that type of resistor anyway…

Just use Tempest plugs. They never had this problem. Champion spent years covering theirs up. I am about to order 12×URHM38S for my new engine. I have used them since 2003 and they show no wear after 1000+hrs but I am changing them as a precaution.

Whether a duff plug explains your problem I don’t know. In the air cooled world almost nobody repairs cylinders; they throw the old ones away.

Shoreham EGKA, United Kingdom

Thanks Peter for your response.
I have found tons of information on the “presumably faulty” Champion spark plugs which got me even more confused… a mixed bag really.
Sadly Tempest themselves told me that they have no plugs for this particular engine. Volumes too low.

Champion confirmed to me that they did change the design, during our exchange they minimised both the motives and the impact of the change.
They have gone silent when I told them that we have extensive engine damage and asked them to help us find out if the plugs are in any way defective.

We have sent a failed cylinder to Dachsel, a renowned German overhaul specialist for inspection. Let’s see what they find (or not).
I have also asked my shop in Germany to let me know the (low voltage) resistance of the plugs that were firing the failed cylinder…

LSGG, LFEY, Switzerland

When I first saw this chart, I thought “wow, these cylinder temperatures are a low” – but then, this is a watercooled engine!

If that had happened in an air-cooled one, what would have been the chances of this destroying the cylinder?

[and on an unrelated question – isn’t a hot spot causing “only” pre-ignition (ignition happening too early but with a normal burn, and hence higher peak cylinder pressure and temperatures), but not detonation (abrupt detonation of most of the mixture in one go, and hence abrupt pressure rise)? I thought that detonation of this magnitude kills pistons and conrods…

Biggin Hill

Cobalt since I was PIC on that flight I can answer your question: pre-ignition would cause a noticeable roughness and a drop in power. The engine was running smooth and the only symptom was this diverging CHT. I do not think it was pre-ignition.

At this point, I suspect the water cooling hides the thermals stress that the inner cylinder barrel is subjected to. In an air cooled engine, CHT’s would shoot up instantly. In the water cooled it seems that either:
- The plug misbehaves from startup and causes extra heat. The coolant absorbs the additional heat so that CHT simply doesn’t move until the cylinder cracks, coolant flows in and washes away the oil film and causes a huge increase in friction
- The plug only starts getting incandescent after sometime and the CHT does move right away.

What puzzles me is that EGT remain unaffected. One would expect to see some impact of the detonation…

Of course this is ONE path.
It could be that the cylinders are simply bad (for whatever reason) and the CHT only shoots up after they crack and coolant starts flowing in.
But given the fact that both replacement cylinders were (allegedly) subjected to extensive NDT, I like to dismiss this theory.

LSGG, LFEY, Switzerland

I have always understood that

  • detonation is the flame propagating a lot faster than it should
  • pre-ignition is a normal combustion but triggered too early (e.g. by an incandescent “something” in the combustion chamber)

and that while the former doesn’t do an engine any good and if allowed for too long (minutes) can destroy the engine by excessive CHT and possibly melting a hole in a piston, the latter will destroy the engine in tens of seconds or less.

In a certified engine running the correct fuel, detonation is supposed to be almost impossible to achieve unless the CHT is allowed to go wild (say 500F) via thermal mismanagement (e.g. a Vx climb for a long time).

Re the spark plug, I would think that if it looks OK and tests OK, why would it go incandescent?

Shoreham EGKA, United Kingdom

Hi Peter

This is what I re-read every time I need a refresher on these topics:

I have no previous experience, because the IO-360 in a DA40 is thermally fairly trivial.
The heavily turbocharged TSIOL 500 (39.5 in take-off MAP and 37.5 in. in the climb) must surely be producing extreme combustion temperatures.
What I have been told is that a failing plug (damage not yet visible) may be overheating. No idea why, but that’s what my Canadian guys are saying.

I thought like you: why would it go incandescent?
Another way of asking this question is: what can possibly cause a (thoroughly tested) cylinder barrel to crack if not excess heat
What can be the cause of excess heat apart from something going really hot and causing detonation?
The investigation of the fuel injection system shows no signs of problematic air/fuel mixture… one would see the EGT shooting up if the cylinder was too lean…
Steep learning curve here…

Last Edited by Flyingfish at 02 Dec 09:13
LSGG, LFEY, Switzerland

Flyingfish wrote:

- The plug misbehaves from startup and causes extra heat.

The problem with a bad plug is that it can develop a hot spot (incandescent if you prefer) and the hot spot can light of the mixture on the power stroke BEFORE the mag fires the plug, thus causing PRE-IGNITION. Cylinder head defects or FOD can cause the same . So the plug does not really produce any “extra heat” per se, but rather the pre-ignition event does – big time !

Flyingfish wrote:

The heavily turbocharged TSIOL 500 (39.5 in take-off MAP and 37.5 in. in the climb) must surely be producing extreme combustion temperatures.

That’s not what I would qualify as “extreme”. Cabin class Cessna Twins (and many others) routinely run @ 2700/41 inchs at TO.

Flyingfish wrote:
what can possibly cause a (thoroughly tested) cylinder barrel to crack if not excess heat

Most likely is indeed heat, but manufacturing defects could be a cause as well.

I seriously doubt that any of the cylinders were actually “testted” but rather thoroughly INSPECTED .

Last Edited by Michael at 10 Dec 08:46
Sign in to add your message

Back to Top