InsaneHydraulics - Sergiy Sydorenko © 2009-2012 All Ridghts Reserved
Every now and then I get an email regarding the video
I uploaded more than a year ago - the one that shows me trying to
"overpower by hand" several hydraulic motors - and rightfully folks ask
what the big idea is behind the "masterpiece". Better late than never - I
says - and give the following clarification to the old but not
forgotten IH clip.
It all started with a letter from a Canadian hang
gliding enthusiast describing an interesting experiment he'd conducted
with the hydraulic winch they were using for tow gliding (I had to look
it up, the tow gliding I mean, and it did look like tons of fun). The
man was looking for a way to make the rope tension during tows as stable
as possible, and one of his experiments involved tying the rope to a
tree and then measuring the rope tension with a strain gauge while
recording oil pressure in the motor line. The winch consisted of a
series 2000 Char-lynn orbital motor with the drum directly attached to
the shaft. A seemingly strange phenomenon was recorded:
- rising the pressure to 2000 psi resulted in approximately 200 lbs of rope tension, however
- lowering the pressure in the motor line did not reduce the rope tension! With the pressure dropping to 1100 psi the rope kept
tensioned at 200 lbs!
- It was only when the pressure dropped to 1000 psi that the rope tension would start to drop.
- Also, pulling the line against the motor with the relief set at 1000 psi required 200 lbs of rope tension.
One peculiar phrase from the letter: "I have talked to
numerous hydraulic engineers and they say that this is not possible".
It was clear that this interesting phenomenon (and it was
new to me at the time) needed some additional looking into - which lead
me to conducting a series of experiments with orbital motors, some of
which I filmed and combined in the video mentioned above.
The main purpose of the experiment was to test how various orbital-type hydraulic motors generate static torque,
and confirm if the described "torque lagging behind pressure" (or
torque hysteresis, if you will) is present and is that big. The
test rig consisted of a lever with a weight at the end attached to to
the shaft of the tested motor. The experiment would start with the lever
in vertical position (see schematics)
- the lever angle would indicate the static torque. The idea was to
increase and decrease the pressure slowly, and record how the
angle of the lever would change, as well as experiment (by moving the
lever by hand, which was the easiest but definitely not the safest
way...) how the stalled motor would respond to a changing load.
From an ideal motor one would expect a direct relation
between the shaft torque and the line pressure - increasing the pressure
would result in an increasing angle of the torque lever, an decreasing
it would result in a decreasing angle - and I was expecting to see
something similar, with minor hysteresis maybe, however - I was
greatly surprised to see that the torque hysteresis was HUGE - just like
in the experiment the man described! The relation between the motor
shaft torque and the line pressure was absolutely not linear (I tested
several motors, all produced similar results, the first part of the
video shows one of them).
After giving this phenomenon much though I came up with a
theory - this huge torque hysteresis is the direct result of the orbital
motor mechanics. Under the stall conditions the combination of static
friction and the orbiting rotary group deformation creates significant
friction forces which can be comparable to the torque created by the
pressure differential - which makes them (again - in this particular
situation - stall) poor motors and good brakes. You could say
that they behave kind of like a mechanical gearbox, in which the
friction forces work against you when you drive the load, and for you when you hold the load.
In other words - the mechanical efficiency of an orbital
motor under stalled condition can be very low, like very-very low -
which automatically makes these motors a poor choice for applications
where you require precise control of static torque.
You might say - hey, these motors are made of steel, you
know, you can't deform a chunk of metal, now, can you? Well, yes you
can. And you already know this from the fact that some orbital motors
can stall due to over-torque of the end cap screws, for example. To show
you how the seemingly rigid bodies of these motors can be deformed I
conducted a similar test with an orbital motor mounted in a vice - an
example of an incorrect installation (the second part of the video). As you can see - under the same circumstances the motor behaved much worse - with even bigger mechanical torque hysteresis.
The last part of the video shows a smart (but not cheap)
choice - a bent axis type piston motor - in which the torque creation is
more direct and the mechanical friction of the rotary group is very
small as compared to the rotary group of an orbital motor. As you can
see - the behavior of an axial piston motor under stall is much
smoother! The torque predictively increases with the increase of
pressure and drops when the pressure drops, load changes also result in a
very smooth movement.
Conclusion: orbital motors have very low mechanical
efficiency in stalled condition, which makes them not suitable for
applications where precise torque control under zero speed is
required. When such motors are used in winching applications where
the load can stop the winch and "pull against" the motor - the rope
(chain, line or else...) tension peak during the "zero speed" transition
is something to expect. A good solution for such an application to
smoothen out the tension peaks would be to use a different type of
hydraulic motor (bent axis or a radial piston), or use an
alternative mechanical solution, like a spring-pulley system or else.
I, personally, never came across a hydraulic motor
application which would require creation of static and controlled
torque, but if I ever do - I will definitely consider a bent-axis or a
radial-piston motor over orbital. In any case - it was worth experimenting with it.
P.S. This is my own theory - it may be not entirely
correct or even plain wrong, so any ideas or experience in the field of
"static tension creation" will be much appreciated...
P.P.S While watching the video, please, concentrate on the
pressure gauge reading and the lever position - don't mind the mate
shuttling in front of the camera...