Last month, I mentioned the last sale of the season. And guess
what? My collection has increased by one! The new one's a
vertical Crossley VOE, a hot-bulb, single-cylinder engine that
looks to be about a 6 HP, but is in fact an 11 HP, achieved by a
pleasingly complicated air-scavenging system. Originally someone
else's restoration project, this Crossley came to me in
do-it-yourself assembly form - one large chunk of cast iron and
several large containers of assorted parts. Thanks to the Internet,
we had no trouble finding a complete manual to download and print
out, which proved invaluable in putting the jigsaw back together. I
know I've covered the subject of flat belts before, but
it's the thing I get more 'please help' e-mails about
than any other topic. I looked up the last time I mentioned belts,
and it was 32 articles back, so I don't think I've overdone
the subject - yet.
And that brings up the one thing we don't seem to be able to
find on the Internet: belt dressing. If anyone knows of a source
for solid sticks of rosin-based dressing, contact GEM so the
information can be passed on to the engine-collecting world.
And now, on to the the discussion that attracted my attention
this month on the Stationary Engine List:
On more than one occasion I've been asked at a show,
'What keeps the belt from slipping off the side?' - or -
'What keeps the belt on?' Can anyone give a simple
The simple answer is that the pulley's crown (center)
section is traveling at a higher surface speed than its outer
edges. The belt gravitates to the portion of the pulley that runs
fastest. That's what keeps a flat belt on the pulley. If you
made a tapered pulley and tried to run a belt on it, the belt would
walk right up the taper and off the high side of the pulley.
I agree with what you're saying, but I don't understand
why it will gravitate toward the larger-diameter, higher surface
speed part of the pulley. It seems logical that the smaller
diameter would be the path of least resistance, and the belt would
drift to that direction.
There's a simple way to understand why a belt will gravitate
toward the larger-diameter side of a crowned pulley.
Make an outline sketch of the crowned pulley. Viewed from above,
lightly sketch a belt that's about half the width of the
pulley. Draw it off to one side so the edge of the belt is at the
edge of the pulley. The pulley at the other end of the belt is
directly in line with the first pulley, so as you sketch the belt,
show it centered on the second pulley.
Belts are flexible. Taking this into consideration, finish
drawing the belt. The crown of the 'half-off' pulley will
put a curve in the belt, when viewed from above. Right?
Now, at the 'straight' portion of the belt, before it
begins its curve, draw a line down through the center. Where does
that line point? It points toward the center of the
As the straight portion of the belt travels toward the
'half-off' pulley, it's headed toward the center. Thus,
you can see that the - belt's curvature is caused by the crown,
which in turn causes the approaching belt to aim toward the center
of the pulley.
If you're still not convinced, draw a reverse-crowned pulley
and do the same exercise. If the belt is slightly upset off the
center of the pulley, it will run off the pulley. The explanation
is clear as mud until one draws the sketch.
The belt is tighter in the crown area. The crown's extra
grip pulls the belt into the center. I read here someone
experimented with running the belt in a trough, meaning the pulley
center was low and the edges were high. The belt walked right off
If the belt slips, it'll go the other way and throw itself
off the pulley. That happens all the time when I cut firewood. 1
get an odd-shaped piece that binds a bit, and I throw the belt.
The belt's behavior on a crowned pulley can be explained by
complex math and big words, but a simple demonstration is much
clearer. It's an experiment you can do right now while sitting
at your desk.
Take off your belt and lay it flat on your desk. Now take a
cylindrical soda can, lay it perpendicular across the belt, and
then fold your belt over the can. Note how the top side of the belt
lines up with the bottom side. That's great, but if the belt is
near the edge of the can, there's nothing to make it want to go
back to the center.
Now get something similar in size to the soda can, but tapered
or rounded. A water glass will work nicely. Start as before, with
the belt flat on the desk and the tapered water glass perpendicular
to the belt. Fold the belt over the glass. Whoa! The top part of
the belt now points in another direction. It points toward the big
end of the glass. This is the same effect that keeps the belt
tracking toward the fat side of a pulley.
Now consider a belt's behavior when slipping. If the pulley
is perfectly flat and the whole system is perfectly aligned, the
belt will slip, but stay in position. If the pulley is crowned, the
belt will drift more quickly. It takes more belt to make it around
the larger circumference, and there's extra belt left over on
the small circumference.
Since the the surface closest to your reading area is
probably covered with sketches on the back of old envelopes, belts
and a selection of household items, we'll move on to a slightly
different belt question.
How tight should a flat belt be around a pulley? I'm going
to run a buzz saw with an old Briggs 14 HP for the first time.
Belts don't have to be tight at all. In fact, a loose belt
will still do what you need. Many times mine actually flop pretty
good, and I wonder how they stay attached.
When I used my Witte to power a buzz saw, 1 had the belt tight,
but still lose enough to slip the belt on and off the saw pulley
when not running.
My belts aren't too long, maybe 16-20 feet total length, or
8-10 feet belted up. You'll need to experiment on your own.
It's important to use belt dressing -you'll often have
trouble keeping a belt on the pulley because it doesn't have
As I recall, a 100-foot belt self-tightens to about the right
tension under its own weight. Shorter belts have to be
I would've thought the Briggs was too fast for a buzz saw
unless you have a speed reducer. You'll need a pretty small
pulley on the engine, and that may be hard to keep a belt on. Many
people use hit-and-miss engines at maybe 300-500 rpm with a
slightly larger pulley that's 12-18 inches.
I got it going! The pulleys were kind of rusty, so it grabbed
real well at first. Then it started to slip, so I used some belt
dressing and had a great time. After a while, the Briggs started to
work hard. When I shut it off I noticed metal shavings under the
gear unit. Closer inspection revealed the shavings were coming from
under the blower housing - this doesn't look good! On the
bright side, I have two spare engines in the shed!
I always knew I had a good reason for purchasing another engine
... and another ... and another. They're 'spares!' And
don't forget to contact GEM if you know about a supply of solid
belt dressing. Everyone seems to prefer it to the modern
And finally, take care of your engines, belts and saw rigs when
you're cutting firewood this winter. Happy holidays,
Engine enthusiast Helen French lives in Leicester, England.
Contact her via e-mail at: Helen@insulate.co.uk You can join the
Stationary Engine List on the Internet at: www.atis.net