After landing yesterday, the engine nearly stopped – as if the idle RPM was set way too low.
I have had that happen before, years ago, but it was adjusted and was OK afterwards.
But this time is much worse and on a closer look it turns out that the fuel flow near idle is 4 GPH when it should be about 2.5 GPH, so the engine is getting a massive rich cut.
In flight, 2hrs at 10.5 GPH, it was perfect.
If it was fuel starvation, there could be multiple causes but I think this has to be the RSA5AD1 servo.
I didn’t try it at higher powers because I wanted to taxi to the pumps and then go back to work, but I have no reason to think it failed just after touchdown.
It was only due to the fuel totaliser that I could see the difference between 4 and 2.5 – the analog gauge is too bad so far down.
If I lean it drastically, to return the fuel flow to say 2.8 GPH, the engine runs fine, but the mixture setting is far too sensitive to be useful.
Obviously the plane cannot be flown so I have some “interesting work” to find a hangar 
But one can change the servo outdoors in about 2hrs. This one was installed last May, as a precaution ($3000) following a suspected fuel contamination incident. It came directly from the very top US specialist servo builder.
Any views?
One engineer I spoke to, who I know knows his stuff, says there are some diaphragms in these servos and if one of them splits, you get this. I know that the altitude compensated version (which mine isn’t) has a catastrophic failure mode in the baro sensing diaphragm which, when it splits, stops the engine enroute – unless you are very quick in reaching for the mixture lever and lean it aggressively to restore the fuel flow.
I would take a guess that it is a split diaphragm between the unmetered fuel and metered fuel chambers, I think it is unlikely to be a problem in the air metering side of the system at those power settings but I would not rule it out.
I’ve had the diaphragm split in an altitude compensating servo. Whilst the engine did rich cut the power returned from 8000’ downwards. By 3000’ it ran normally without manual leaning. It split at FL180, probably the first time the aircraft had been that high so it was just a bit too much for the diaphragm.
Some of these diaphragms are pretty old so it’s no surprise if they fail occasionally.
OK; thank you all. I too think it must be the servo.
I am getting a new (overhauled) one, specially built for the fuel flow to be at the top end of the tolerance band (for lowest CHTs in climb) from the best US builder. Should take a week or so.
It was the servo.
The original one is going back to the USA for an overhaul and return, so I have a spare one on the shelf.
I will report on what they find inside. The only thing I can see from the outside is that the mixture control is much stiffer than the one on the new one, and it squeaks a bit as if something was rusty. Accordingly, the mixture control lever now moves much more freely.
On the one that’s just gone in, I asked them to adjust the fuel flow at the top end of the tolerance band, and sure enough I now see 24.0 USG/hr during initial climb from sea level (+5C at SFC, QNH 1021). Previously I was getting 23.2 and the original one was doing 22.7. This translates to a significant difference in CHT during climb.
The report on the old servo is: nothing found!
They say there could have been a bit of debris in there, which stopped the valve from fully closing.
However I have had no explanation for the very tight mixture control, the cause of which must obviously be easy to determine once you have the unit dismantled.
Various conclusions are possible in this case…
Didn’t I mention last time somebody had fuel injection problems that carbureted aero engines do have significant advantages? People tend to be scared by carb icing horror stories and forget that it is a much more robust design by principle and carb icing is very installation dependent.
The fuel servo is a rather delicate component and it is a single point of failure. Injectors often get clogged up but usually one at a time and it’s easy to fix yourself.
