Murphy's Tractor:
Whatever can go wrong, will go wrong.
20 July 2009
There really was a guy named
Murphy.
The exact story varies but the
most likely version comes from Edwards Air Force Base back when Chuck
Yeager was trying to break the sound barrier. The late
1940's was a busy time at Edwards with a lot of important research
happening there. One team, lead by Dr. John Stapp, who became known as
"the fastest man on Earth," was using high speed rocket sleds to study
g-forces and rapid decelerations.
During the project, Stapp and his team were visited by a guy named
Edward Murphy. Murphy had some fancy new electronic strain gauges that he
wanted to use to collect better data. Unfortunately, after flinging an unsuspecting
chimpanzee down the rail, the strain gauges showed a G-force of zero.
If you've ever seen a rocket sled in action, it's obvious that the
G-forces involved are nowhere near zero. A little diagnosing revealed
that the gauges had all been wired backwards.
As a team of engineers concerned with safety (their research later helped make
car seat belts standard), they always tried to make things as fail safe as
possible. They coined the term Murphy's Law to describe the concept
that whatever can go wrong, will go wrong. To help prevent this,
it's a good idea to design things so they can't be put together wrong.
If the electrical connectors on the strain gauges had been different
shapes, different sizes or even different colors, then the problem
would probably have been prevented. This is called Murphy-proofing and
most engineers are aware of the concept. But apparently Chinese
tractor engineers have never heard of Murphy.
I own a Chinese tractor. It's a horrible tractor. It came with the
farm and I like to think I'm smart enough that I'd never buy such a tractor.
In fact, as soon as I get it running again, I plan on selling it.
The tractor only has a couple hundred hours on it. That's very little
time for a farm tractor. Most of those hours are from chipping pruned
coffee branches. Chipping is high RPM, low torque work, kind of like
freeway miles in a car. Yet my tractor has
still needed a complete overhaul, twice.
This latest overhaul has been extra difficult. To start with, the
tractor didn't even make it back to the barn before the engine seized.
I had to get my neighbor to bring his tractor over and tow my tractor
back to the barn. It didn't take long to realize that it wasn't going
to be an easy fix, requiring the entire engine to be taken apart.
Even with experience, it's not an easy job. Car engines are
basically held in place with nothing but four large bolts. Remove
those bolts, along with a few other pieces, then you can lift the
entire engine out. That's not the case with a tractor. Instead of
removing the engine from a tractor, you remove the tractor from the
engine.
Step one is removing the bucket. On some modern tractors it's a 30
second job but on my cheap Chinese tractor it involves swearing, lots
of wrenches and a heavy duty chain hoist. Step two is removing the
entire front half of the tractor. That includes the wheels, radiator
and half of the frame. Step three is removing the engine from the
back half of the tractor. I've had enough practice now, I can take the
whole tractor apart in about four hours if I hurry. Of course putting
it back together takes a bit longer.
Trying to prevent my tractor from breaking like this again, I took the engine
block completely apart then had the local machine shop modify it a bit.
After a thorough cleaning I put it all together again. I was hopeful that
it would start right up but of course it didn't.
The engine would crank and crank but refused to start. This normally
means no fuel so I checked and rechecked everything. Having fixed the
fuel system on several other occasions, I was very familiar with it and
I knew that everything was working correctly. The only possible
explanation is that the timing was off.
There are things called timing gears inside the engine. These gears
ensure that the pistons, valves and fuel injector all work in unison.
If the gears are off by even a single tooth then the engine won't run.
I was very careful with the gears and I was positive that I had put
them back together correctly but there was no other explanation.
Tearing the entire engine apart again revealed that the gears were
indeed correct. So how could the timing be wrong if the gears were
correct? It didn't make any sense.
Note how the fuel pump gear in my tractor has three symmetrical bolts.
|
The gears in other tractors use an asymmetrical pattern so they can't be installed wrong.
|
Murphy's law, part one: The gear that attaches to the fuel pump is
connected with three bolts. I've seen this gear on other tractors and
the three bolts are usually off-set in an asymmetrical pattern so it is
impossible to put the gear together wrong. The Chinese engineers used a
symmetrical pattern and of course I somehow managed to put the gear on
240 degrees out of sync.
Many, many hours later, after assembling the entire tractor again, it
finally started up. I ran it for about 10 seconds before shutting
it down because there was absolutely no oil pressure. Running an
engine without oil is not a good thing to do. Being an experienced
diesel mechanic now, I know important things like this.
The most common way to lose oil pressure is with a leak. I had
assembled the engine carefully and there were no leaks, not even a
drop. That left the oil pump as the likely problem. I had checked the
pump carefully before but I checked it again anyways. It looked fine
and I could even spin it by hand to see that it was working. Despite
this, everyone was certain that it had to be the oil pump so I took it completely out of
the engine. That meant I had to take the entire engine apart again.
Two days later, when I finally had the oil pump out, I put it in a bucket
with a couple inches of oil in the bottom then connected the pump to my
pneumatic hammer and took it for a little speed run. As expected,
it flung oil everywhere. The oil pump works fine.
Murphy's law, part two: I checked every single nook and cranny
of that engine for any possible leaks or clogs and everything looked
fine. When I finally found the problem I was relieved, embarrassed
and angry all at the same time.
The oil filter housing is two pieces. The main piece holds the oil
filter while the smaller piece is nothing but an extension. I now
realize that it is possible to put this extension piece on backwards.
There is no way to know if the piece is on backwards other than carefully
examining the oil passages through the inside of it. Once installed,
the only way to tell if the piece is backwards is by the fact that there
will be no oil pressure. Too bad there's no markings or anything to
indicate which way is correct. A little bump, slightly different
shape or off-center bolts would have made it impossible to install the
piece backwards.
If you're ever designing a tractor and there's a part that must be
fitted in a particular direction, please, please design the piece so it
can only be installed in one way, otherwise I guaranteed that
eventually some novice such as myself will come along and assemble
things backwards. Not only did these mistakes make me feel stupid,
they also cost me far more time than I can afford.
Repeatedly taking an engine apart isn't really good for it. Bolts,
gaskets, hoses and other parts get worn out in the process. I'm
currently waiting for a few new parts to replace the ones that keep
breaking. Hopefully, once the parts arrive, I'll be able to put
everything together and finally get the tractor running again. You'll
know if I do because the tractor will be for sale.
|
