Maintenance
Posted: Sun Sep 22, 2019 10:43 pm
As a pilot, you will deal with maintenance all your life.
From their perspective, pilots are people that break airplanes. Maintenance
gives pilots working airplanes, and they get back broken ones.
That's not really fair ... I'd rather fly an aircraft that's just about to go
in for maintenance, instead of one that's just come out of maintenance.
An outsider would think that's silly, but an aircraft that's just come out of
maintenance is an unknown quantity. Dangerous, actually. And the more
maintenance it's received, and the longer it's been apart, the more likely it
is to try to kill you.
My religion is that the same guy that took it apart, has to put it together.
When a different guy puts it together, you are in for a world of hurt. Mechanics
will say I'm full of shit, because they can look at some book as to how it
goes back together, when they assemble it from a collection of parts in buckets.
Hold on a fucking second. Ever heard of a documentation error? There are
plenty of changes over time to aircraft manufacture, and trust me when I say
you want to put an airplane back together EXACTLY THE SAME as it came
apart. The same pieces of metal touching each other, in the same arrangement.
An extreme example is a new airplane. The most dangerous of all. It's never
been flown, and they have special pilots - production test pilots - fly them for
the first time.
Similarly, I would recommend that only senior, experienced pilots test fly
aircraft that are just out of heavy maintenance. Hopefully they've got a bit
more systems knowledge and flying ability to deal with stuff that breaks.
It's important to know that there are maintenance-induced failures. That is,
stuff on the airplane was working, before maintenance broke it. There is a
risk, whenever maintenance is performed, that you will take one step forwards
and two steps back.
I hate checklist maintenance - blindly fixing and changing stuff, which may or
may not be broken. Not only is it inefficient, there is a real danger that maintenance
may break something that was working.
I firmly believe that you should only fix what's broken. If it isn't broken, why fuck
with it? Sure, inspect it, but do it in a manner that has minimal disturbance.
For example, in Canada, if an aircraft doesn't fly for a year, TC wants you to start
pulling jugs to inspect the inside of the engine for corrosion. I agree that internal
corrosion is a problem, but capricious jug R&R can cause serious problems. Far
better to use modern technology to inspect the engine internals for corrosion.
People think that maintenance is like candy or sex or cocaine - more is better. No,
you have to do a cost/benefit analysis on every maintenance task, to be sure that
you are doing more good than harm.
This brings us to a very important subject.
Diagnosis.
People are really shitty at figuring out, what the hell the problem is. I'm not just
talking about mechanics - I'm talking about doctors, engineers or anyone else that
deals with complex systems.
There's a guy called Mike Busch who incredibly makes a living doing remote diagnosis.
People tell him what's going on, and he tells them what to tell their local mechanic to
do. This is incredible, when you think that Mike never sees the airplane, which is quite
a disadvantage. Yet, he does a better job of diagnosis than your local mechanic.
I learned a long, long time ago, that if you want something important done right, you
have to learn to do it yourself.
I remember, in the late 1980's, picking up the Maule from annual on Toronto Island.
I noticed that I couldn't pull the control column all the way back - they had installed
a vacuum filter blocking the flight controls. And what I didn't notice until later, was
that they had welded the exhaust without a jig, so it didn't fit any more, so two of
the cylinders only had one stud on. Fuck me.
Some more Transport Canada AMO maintenance .... we had some very expensive
radios installed by a very big name shop. Comm radio never worked worth shit.
Turned out they had plugged the RG-58 coax connector for the VOR antenna into
the comm. Smooth.
Had a G430 installed when it first came out, into the M20J. Strikefinder didn't
work when it came back. They'd damaged the wiring for it.
Especially avionics/instrumentation, I do not know of a single competent shop.
If there ever was one, you couldn't afford it.
For straight and level aircraft, I guess that sort of clown show is ok. It's merely
horrible. But for an aerobatic aircraft, you had better do some higher quality
maintenance, or you're not going to live very long. Needless to say, this is not
mentioned anywhere in the regulations, but it is the difference between living
and dying, so it might be important.
If you're a white shirt with a checklist, well, you have to fly whatever shit they
put on the ramp. My condolences.
From their perspective, pilots are people that break airplanes. Maintenance
gives pilots working airplanes, and they get back broken ones.
That's not really fair ... I'd rather fly an aircraft that's just about to go
in for maintenance, instead of one that's just come out of maintenance.
An outsider would think that's silly, but an aircraft that's just come out of
maintenance is an unknown quantity. Dangerous, actually. And the more
maintenance it's received, and the longer it's been apart, the more likely it
is to try to kill you.
My religion is that the same guy that took it apart, has to put it together.
When a different guy puts it together, you are in for a world of hurt. Mechanics
will say I'm full of shit, because they can look at some book as to how it
goes back together, when they assemble it from a collection of parts in buckets.
Hold on a fucking second. Ever heard of a documentation error? There are
plenty of changes over time to aircraft manufacture, and trust me when I say
you want to put an airplane back together EXACTLY THE SAME as it came
apart. The same pieces of metal touching each other, in the same arrangement.
An extreme example is a new airplane. The most dangerous of all. It's never
been flown, and they have special pilots - production test pilots - fly them for
the first time.
Similarly, I would recommend that only senior, experienced pilots test fly
aircraft that are just out of heavy maintenance. Hopefully they've got a bit
more systems knowledge and flying ability to deal with stuff that breaks.
It's important to know that there are maintenance-induced failures. That is,
stuff on the airplane was working, before maintenance broke it. There is a
risk, whenever maintenance is performed, that you will take one step forwards
and two steps back.
I hate checklist maintenance - blindly fixing and changing stuff, which may or
may not be broken. Not only is it inefficient, there is a real danger that maintenance
may break something that was working.
I firmly believe that you should only fix what's broken. If it isn't broken, why fuck
with it? Sure, inspect it, but do it in a manner that has minimal disturbance.
For example, in Canada, if an aircraft doesn't fly for a year, TC wants you to start
pulling jugs to inspect the inside of the engine for corrosion. I agree that internal
corrosion is a problem, but capricious jug R&R can cause serious problems. Far
better to use modern technology to inspect the engine internals for corrosion.
People think that maintenance is like candy or sex or cocaine - more is better. No,
you have to do a cost/benefit analysis on every maintenance task, to be sure that
you are doing more good than harm.
This brings us to a very important subject.
Diagnosis.
People are really shitty at figuring out, what the hell the problem is. I'm not just
talking about mechanics - I'm talking about doctors, engineers or anyone else that
deals with complex systems.
There's a guy called Mike Busch who incredibly makes a living doing remote diagnosis.
People tell him what's going on, and he tells them what to tell their local mechanic to
do. This is incredible, when you think that Mike never sees the airplane, which is quite
a disadvantage. Yet, he does a better job of diagnosis than your local mechanic.
I learned a long, long time ago, that if you want something important done right, you
have to learn to do it yourself.
I remember, in the late 1980's, picking up the Maule from annual on Toronto Island.
I noticed that I couldn't pull the control column all the way back - they had installed
a vacuum filter blocking the flight controls. And what I didn't notice until later, was
that they had welded the exhaust without a jig, so it didn't fit any more, so two of
the cylinders only had one stud on. Fuck me.
Some more Transport Canada AMO maintenance .... we had some very expensive
radios installed by a very big name shop. Comm radio never worked worth shit.
Turned out they had plugged the RG-58 coax connector for the VOR antenna into
the comm. Smooth.
Had a G430 installed when it first came out, into the M20J. Strikefinder didn't
work when it came back. They'd damaged the wiring for it.
Especially avionics/instrumentation, I do not know of a single competent shop.
If there ever was one, you couldn't afford it.
For straight and level aircraft, I guess that sort of clown show is ok. It's merely
horrible. But for an aerobatic aircraft, you had better do some higher quality
maintenance, or you're not going to live very long. Needless to say, this is not
mentioned anywhere in the regulations, but it is the difference between living
and dying, so it might be important.
If you're a white shirt with a checklist, well, you have to fly whatever shit they
put on the ramp. My condolences.