(followups set to r.a.owning)

In rec.aviation.owning Robert M. Gary wrote:
The fuel shop fixed my fuel servo by replacing the spring that holds
the fuel regulating diaphragm. The tech said the previous spring was
very stiff and he replaced it with one that was softer. Does that
indicate that the previous spring was broken (i.e. stiffer than it
should have been) or does Lycoming (Airmotive really) have a variety
of springs for adjustment purposes??

I don't know anything about fuel servos. But I have played with springs
before. For compression springs made out of a piece of wire that is
coiled up, the common failure modes seem to be: 1) breaking into pieces,
2) getting less springy over time - in other words, when new, a 10 lb.
force caused 0.25" deflection, but now a 10 lb. force causes 0.5"
deflection, and 3) taking a set - the springiness is about the same but
the free length is shorter than when new. I guess getting _more_
springy over time (10 lb = 0.25" new, 10 lb = 0.1" now) is possible but
I don't think it's very common.

If you look at a (car) engine manual, they often specify the
out-of-squareness, free length, and deflection under load of the valve
springs. Out-of-square is just putting the spring on a known flat
surface and making sure it is really vertical, or vertical to within
some small tolerance. Free length is the overall length with no load.
Deflection under load is either given as "it should take 10 to 15 pounds
to compress the spring 0.25" from its free length" or "a force of 15
pounds should compress the spring 0.2" to 0.3"." You can even get a
little jig with a lever and a ruler so you can apply a certain pressure
to the spring and measure the deflection.

I don't _think_ this is the case in a servo, but if you have a bunch of
springs, sometimes relative sameness is better than absolute conformity
to a spec. For instance, if the spec says the free length should be no
more than 1.25" and no less than 1.0", and you have four valve springs
at 0.990", 0.985", 0.995", and 0.993", then you're out of spec, but the
engine will probably still run sort of OK. But if you have four springs
at 0.990", 0.985", 0.720", and 0.993", then it'll probably run like crap.
If it's rebuild time, you'd replace all the springs in either case. But
if you're trying to figure out why it's running poorly, in the second
case you can probably blame it on the bad spring, while in the first case
you might keep looking. (Don't get me wrong - once you figure out why
it's running poorly you'd probably wind up replacing all the springs in
either case, since they are all out of spec.)

Sometimes you can get different springs to adjust a mechanism - either
of different free length or different spring constant. Springs with
very different free lengths are easy to tell apart by eye. If the
springs are the same length but have different spring constants, they
are sometimes color coded so you can tell them apart in the parts bin.
On older equipment, if there's only one spring, you sometimes see service
manual directions that say in so many words: "If you still can't get it
adjusted, maybe you got a widget from Monday morning and a spring from
Friday afternoon. Try a different spring and see if that helps." In
other words, they are admitting to tolerance buildups in the
manufacturing process.

Going on to the fuel side of things, it seems like a couple of common
problems include extra air (naturally aspirated engines) and crap in the
fuel lines. "Extra" air is an intake leak that allows air to get in
without going past the air measuring device (carburetor, fuel servo,
mass airflow meter, whatever) and leans out the mixture. This might
come and go when a hose flexes or similar. Crap in the fuel lines can
be pieces of the Dorito you dropped in the tank while fueling up, but
it can also be gunk deposited by the fuel when you let the engine sit
for a long time. Pieces of Dorito may eventually get sucked into the
engine, caught by a filter, or just fall out unnoticed when the fuel
line is disconnected. Old fuel gunk tends to be a little more
tenacious, but can also dissolve into the new fuel over time.

In your particular case, it seems to me that the procedures and
specifications ought to be written down somewhere. There should be a
defined rebuild procedure. There might be a spec for the spring and
there should definitely be a spec for the performance of the entire
servo assembly when run under defined conditions on a test bench or
a test engine. Even if you can't legally use that information to fix
your own servo, I would think (ha ha) that you ought to be able to get
a copy of it somewhere.

If you buy a $50 rebuilt alternator for your car, it will probably come
with a little tag that says "Alternator type 123 S/N 456 rebuilt by Acme
Corp. put out 60.1 A at 14.05 V at 3000 RPM when tested on 7/8/2009."
Sometimes there's even a cute little graph of output vs. RPM. Maybe
it's all printed by a random number generator, but if Acme can give you
all that for $50, your fuel servo shop ought to be able to do something
similar, IMHO.

Anyway, I think I agree with what I understand to be your basic premise,
which is that springs don't suddenly get stiffer on Tuesday. Ruling out
everything else can be a lot of work, though.

Disclaimer: This is based on experience with fuel systems, engines,
and general machinery in ground vehicles and in fixed installations.
I don't have an A&P; I don't even have a TG&Y. Your mileage may vary.

Matt Roberds