Three Different Kinds Of Mechanics

[Colonel]

As a pilot / aircraft owner you should be aware of the different kinds of mechanics.

(1) The Parts Changer. This guy isn't bad. He just knows how to remove and replace,
and that's not all bad. But I remember when a 540 had been parked, and the valves stuck,
and the mechanic pulled both magnetos for overhaul (shoulda bought new ones). They
came back, and somehow they didn't fix the sticking exhaust valve.

(2) The Guru. He's been around for decades, and he's seen it all. This guy is the top of
the heap when it comes to mechanics, because our aviation technology is moribund in
the extreme. We are using engines designed 50 and sometime 100 years ago. This is
actually good, because the total lack of change allows someone to accrue immense
knowledge which is still applicable today. He is a pattern matcher and he's seen
your problem before and solved it before. He is cherished because he will not do as much
unnecessary maintenance (and possibly inflict as much damge) as The Parts Changer (1)
above.

(3) The Scientist. You have never met one of these. You have probably never heard of
one of these. I doubt of the existence of even 1 or 2 of these in all of Canada. This is
a mechanic that uses the Scientific Method to solve problems. He poses a hypothesis
then conducts the most rigorous, simplest test in order to collect good data to learn
more about the system. This is not taught to anyone in Canada, and if by some freak
of nature, this Unicorn teaches himself about Science, he immediately learns that he can
be paid 10x or 100x as much in some other pursuit that rewards the application of Science.
Like an exotic sub-atomic particle, his existence in aviation can be measured in nanoseconds.

You actually want someone that is all three of the above. For example, let's say you have
a spark plug go bad. You might change them all, because another one in that batch might
fail tomorrow, and you might switch suppliers (Champion to Unison, for example).

The Guru is really useful. Don't knock pattern matching. It is a sign of frontal lobe development
and is missing in your pets and politicians. If you have a sufficient knowledge base and the
landscape never changes, this can be very effective.

You will probably never meet a Scientist, or even ever hear of one. Mike Busch might be close,
but he gets an awful lot of help of looking at data previously collected by engine monitors, and
he's been doing this for decades, and qualifies as a Guru.

Mike actually has a couple of degrees in mathematics and built a software company from scratch
and sold it, back when the earth was cooling, and Canadians probably think he is a moron. He
does not think I am a nice young man.

[Colonel]

Case Study:

I have mentioned the dude with the Great Lake biplane down the hangar row
from me. His mechanic has changed everything but his rudder and tailwheel
trying to fix his starting problem.

Eric and I were staring at each other as he tried to start it. It sounded like a
shotgun going off - the unburned charge was (probably) firing out the exhaust
(but it could have been the intake, blowing out the filter which then gets ingested).

Given the above, I constructed a hypothesis that his spark was firing at the
wrong time, and the magnetos needed investigation (alternate hypothesis - no
spark and hot carbon spots).

I talked to him afterwards. More passports than James Bond - British, Canadian
and American. I mentioned that he had might have a spark timing problem.

He replied that was unlikely because he had recently replaced his magnetos (and
they were changed again after that conversation) so he thought that was unlikely.

I got sad at that point, because even though Canadians think I'm a moron, there is
something in engineering called the "bathtub curve" which represents failures
over time. Frequently, new parts fail. Pilots die on the first flight after maintenance
all the time, and sometimes it's because the new parts are bad - even with
perfect maintenance practices!!



Ok. I gently suggested to him that he unhook the p-leads from both his magnetos
for a ground test only - very dangerous. Prop is hot all the time. Then get in and
crank it. You have no p-leads or ignition switch in the system any more. Does it
behave the same?

Then get a wire with crimp connectors at both ends, and conduct two tests - one
magneto grounded for each test - and see how it starts on only one mag. See,
both of his mags are impulse coupling, so each test would duplicate a normal
Lycoming, with the ungrounded (hot) impluse coupling mag firing one spark plug
in each cylinder.

The three tests conducted above would tell you if
(1) the p-lead and switches were shorting to ground
(2) which magneto was firing at the wrong time (confirm hypothesis?)

He was uninterested in science, and his mechanic kept changing parts. Including
the entire fuel system and ignition system (again). I shudder to think of the cost
of parts and labor, and the downtime, and the risk posed by the unnecessary maintenance.

See, you don't have to give a shit about science if you don't care about money or
living. You will likely never meet a scientific mechanic. You might see a unicorn
one day, too.

[Colonel]

Anyways. As a former mathematician a template that I try to extrude reality through is vector spaces.

A vector space has many dimensions. We can measure the competence of a tradesperson like a mechanic, doctor or lawyer using perhaps two orthogonal dimensions:

1) competence (entire above discussion) and
2) character

Because people like simple scalar numbers, we can calculate the norm of the vector space
using the Root Mean Square of the two above like so:

SQRT(competence^2 + character^2)

Obviously this is lossy, but people like to compare simple integers, and 3 > 2 for example.

Anyways, feel free to shit on me for sharing the above. I get it that I am different (always
have been) and there is supposed virtue from having a GPA of 2.0 ... lots of company there,
right in the meaty part of the bell curve.

[digits]

The local guru in the area is also pretty close minded. He's seen it all, so obviously he is right and you are wrong. Pretty annoying when he is spending your money.
Even though he hasn't really worked on your airplane type and model, they are all basically the same, and thus an issue on type X 15 years ago is exactly what you are experiencing on type Y now. At some point, being a guru might work against you.

I've had good experiences with young guys willing to learn and look into your issue.

[Colonel]

Gather information from all sources.

A terribly important lesson for young people to learn, is that you can learn a lot
of very valuable stuff from people you don't like. And even idiots, which can be
a terrible blow to the ego. I tell myself the idiot is merely mindlessly repeating
something that some unknown genius told him. Helps me sleep better.

Anyways. One has to choose if one is going to be a Man of Science™ or a Man
of Fashion™. Different objectives. One gathers truths. The other seeks external
approval.

[Squaretail]

(1) The Parts Changer. This guy isn't bad.

With this guy you just need to remember that you need to troubleshoot, then tell him what you want replaced. Probably can't make repairs though.

(2) The Guru. He's been around for decades, and he's seen it all.

Know a few of these guys. Occasionally a problem if you do manage to find something outside of their experience. May try to apply a previous solution that may not be applicable.

(3) The Scientist. He poses a hypothesis
then conducts the most rigorous, simplest test in order to collect good data to learn
more about the system.

Dad is like this. Though with automotive. Frustrating when you're young, when you think he is happy your car is broke down since its a puzzle to figure out. The older I get, the more I can appreciate the method. He also combines that with the ability to rapidly give probability of the problem's nature with a monetary/time cost of conducting he experiment, then put them in order. I think this is the thing that drives me crazy with mechanics I run into, is when trouble shooting, don't exhaust the cheap and quick possibilities before wading into the time consuming and expensive ones.

[Colonel]

when trouble shooting, don't exhaust the cheap and quick possibilities before wading into the time consuming and expensive ones.

True. When shotgunning parts at a problem, start with the cheapest ones.

From a logic standpoint, you must always be re-examining your axioms (assumptions)
that your logic is built upon. Though nobody uses logic any more, so that probably doesn't
matter.

[Colonel]

Probably can't make repairs though.

Don't get me started. No one repairs anything, any more -
they just remove and replace with new. Often this makes
sense because the new part will cost pretty much as the
repaired part, and if you don't know about the bathtub curve ....

You should get on the Savvy Aviation mailing list. Mike Busch
makes a living making strategy (being the brains) for A&P's in
the field, whom undoubtably detest his interference - especially
when he is right.

A&P's today always want to replace with new. Cylinder, engine.

Some of the time they are right, but other times Mike recommends
repairs (valve guide cleaning, valve seat lapping, replacing valve rotators)
that are unknown to the new A&P's, which can sometimes work pretty
well after he coaxes them into learning something new. Actually something
old (and forgotten).

Changing an engine because you need to clean some lead out
of the exhaust valve guides is pretty silly, but is normal these days.

It's interesting to read what Mike has to say about changing cylinders -
he thinks that A&P's are unqualified to do so in the field, and they often
do horrible damage to the engine. Note that this is the diametric opposite
to TC's guidance which is to pull cylinders frequently.

In fairness to A&P's in the field, Lycoming's instructions in the Direct
Drive Overhaul Manual leave out an awful lot of essential information.

Hint: ever re-tensioned the spokes on a bicycle wheel? Doing that should
be essential training for anyone that's going to change a cylinder. Also,
learn about torque vs tension. Everyone thinks the former is a superb
proxy for the latter, and that simply isn't true.

[Squaretail]

when trouble shooting, don't exhaust the cheap and quick possibilities before wading into the time consuming and expensive ones.

True. When shotgunning parts at a problem, start with the cheapest ones.

Quickest ones are also frequently the cheap ones too, when you start factoring in shop rates. For example when it comes to electrical issues, I usually start with the battery. Its cheap (relatively) and quick to replace if there is any question about it. So are its terminals. At the very least a weak battery may have compounding effects on your problem.

The main issue I find is people just don't know how to logically troubleshoot. They do it in a random fashion as opposed to using a process of deduction. More people should read the Sherlock Holmes stories. Or Batman comics.