One of the big problems with our mechanical marvels is gearing.
Things work just *fine* where you have a fixed need: Water pumps, for
example. Just find a big enough pump, and a motor strong enough to
supply the power needed ... and then a fixed gear (or belt system) to
speed up or slow down the output of the motor to match the need of the
pump, and you're all set.
MOST (not all) of the motors we have these days work *very* well
indeed at a fixed speed and power output; though they can sometimes do
OK for short periods of high-load on startup. Other times though, the
load is too heavy ... and we have to introduce some kind of variable
gearing.
Cars and gasoline motors fit this description. The only motor that
works worth a damn over a wide range of speeds is the series-wound DC
electric motor ... and for years they were used to run Trolley cars
for just that reason. However, parallel or "shunt" wound electric
motors are far more efficient; and it and the internal combustion
engine both do their best at fixed outputs and speeds.
Which leads us to gear-boxes in cars, and "automatic" transmissions.
A low gear-ratio allows lots of power for starting ... but the engine
winds up too fast to give much power ... so we shift gears to get
going faster ... do it again when going medium speed, and once more to
our "running" or "top" gear, where the speed of the motor is nicely
matched to the "cruising speed" of the car.
What's really needed isn't a whole *set* of gears that we shift
between (automatically shift between on "automatic transmissions), but
a smoothly variable gear-ratio that has a practically infinite ratio
of choices to match engine to road speed.
There have been many such invented; but none of them handle truly
heavy loads worth a damn, or they are big, bulky, cumbersome, and very
complicated (like the wobble-plate transmissions used in some lawn
tractors). What's needed is something approaching the simplicity of
two gears meshing ... or at the very least not much more complicated
than a four-speed gearbox, and having very little slippage and wasted
efficiency.
By slippage, I mean if you slowly turn the power gear the output gear
should turn at close to a fixed ratio, continuing to do so as you
increase the speed.
Efficiency is obvious: No large percentage of the input power should
be wasted by turning it into heat because of the interior of the
"gears" doing work not directly involved in turning the output shaft.
Size is also obvious. A GOOD device of this nature shouldn't be any
larger (and preferably smaller) than an old-fashioned four-speed
gearbox that handles the same power.
The ratios handled should also at least match a four-banger.
Handling zero through reverse *would* be nice, but not necessary; as
reverse *can* be handled separately.
There *is* such a transmission. I invented it over 40 years ago, but
never had the money (nor the personal gumption) to develop it. If
anybody wants to develop the process and make lots of money off it,
feel free to steal it or even contact me directly for more details
than here. I declare no interest in rights, and put it in the public
domain; because I simply cannot afford neither time, effort, nor money
to do so myself. Anybody willing has my blessing.
The *basis* for the idea is simple; and several attempts have been
made by many people to make a decent device. It's hydraulic.
Basically hydraulic pumps and motors work nicely to transmit power
while retaining fair efficiencly. The problem is: Hydraulic motors
and pumps at their most efficient are fixed-ratio devices just like
gears are. The moment you start moving away from fixed ratios you
lose the advantages. But there *is* a way around this problem, or I
wouldn't be writing this.
A very good pump (and motor too!) is what's described as a "gear
pump". It's very simple: You take two gears, drive one of them with
your motor, and surround the two gears with a housing. That's it!
The outer parts of the driven and drive gear push fluid towards one
side, while the meshing of the two gears prevents it from flowing back
the other way, as the gear-teeth of the other gear of the set blocks
that. Gear pumps using one gear within the other are found in almost
all automobiles ... used as oil-pumps. Very simple, handle high
pressures, last almost forever.
Take one gear pump, and feed the output to another gear pump, and the
other gear pump becomes a motor with a fixed output ratio depending
purely on size of the gears to the first pump.
Note: That's SIZE of the gears ... the spur-size and volume moved per
revolution, NOT the physical diameter like with two gears meshing. If
you take two gear pumps of identical size, cut one in half so you end
up with four gears on one of them instead of the original two, and use
those two new, smaller gears to make another pump, the new pump/motor
will turn twice for every turn of the bigger pump.
That is the basis for this transmission.
It takes a bit to visualize; but you can *surround* part of a long fat
gear (think of a rolling-pin; but not quite that long and fat) with a
metal shape that fills in the spurs, yet rotates with the gear inside
it. Take two of these sets of gear and filler, one for each side.
The fillers can slide in and out. Along with the rotating filler, a
matching blank that blocks where the other gear *would* have fit also
slides in and out. With a set of these on each side of the gear-pump,
moving in and out in unison, you can change the *effective* size of
the gear-set ... and thus the ratio between it and a fixed pump.
Or (of course) you can use TWO such variable-ratio pumps for more
extreme ratio possibilities. This method works with both internal and
external gear pumps; though the design is slightly simpler with
external gear-pumps while the efficiency of directly driving one
internal pump from another makes the slightly harder design of the
internal pump well worth the extra effort. With no external piping
like external-gears require, an internal design will likely end up
much simpler overall, while taking quite a bit more thought in
designing the particulars. The external design though, is much easier
to do "proof of concept" with, and work out bugs, problems with
sealing, and controls. Controls *are* a problem in design; since the
moving parts that control where the flow goes must resist the same
pressure that pushes on the gears to turn. A balanced design can
accomplish this; but it takes some thought. For proof of concept
though, fixed blocks of metal can temporarily be used, or a screw
mechanism.
If anybody is interested, either contact me for more details (free) or
just take the idea and run with it. *I* certainly ain't going to. I
can't afford to.
--
_____
/ ' / ™
,-/-, __ __. ____ /_
(_/ / (_(_/|_/ / <_/ <_
|
|