Creating an electronic pull tester and i can't figure out how I should vary force electronically
[imagine a lat bar workout machine. You set the weight stack to 50 kg and you start to pull. The weights will lift indicating that you have pulled with 50 kg F]
like this I want to be able to choose and create that force electronically (and on a much smaller scale).
I can't just use a force gauge, because I must make a mechanism that prevents overloading. So, If i set it to 50kg of force. I cannot pull 55 kg. The mechanism should only resist a max of 50 kg...
in reality I need to range force from 0.5 to 2 kg.
Ideas:
PWM a geared DC motor with a lever attached to the shaft. Pull in the opposite direction that I m trying to turn the motor and viola. BUT since the motor only rotates when I pull against its desired rotation, it tends to just hum and warm up... Plus it doesn't seem to be consistent.
A servo. Servo is set to 90 deg. When i pull it passed its holding torque it will return. Vary the force by pulling further from the center of rotation... voila! but I don't know if this is really bad for a servo...
What distance do you need it to work over? A voice coil/electromagnet setup can't do a long distance but it will be very easy to control. Linear actuators and geared motors will need some feedback control as the friction in the gears will be a big component of the force.
My guess is that you need a geared DC motor that can produce the maximum torque that you require while stalled without overheating the motor. That may mean using an over-large motor and using PWM and or a lower supply voltage to limit the current. What I am thinking is that you would limit the current well below the motor's "raw" stall current in order to protect the motor.
You can then get lower torques by using a lower PWM duty cycle.
@robin - good idea, but large geared motors will need too much force to overcome gearing alone. Plus i noticed that it isn't consistent. This is still my favorite option. So i am goign to try it.
@ Morgan - I only need to move 2-5mm. As soon as I feel or see the release/pull it means I passed the test. I agree about the actuators and motors. I am going to try the electromagnet option. I just need to find a cheap way to vary the voltage to them. Unless you can use pwm? I think you use voltage to adjust mag force if i am correct. Maybe both.
Calibrate using known masses/loads and plot a curve.
Use excels "somewhat" handy "fit a trendline and show the equation" feature to have a nice simple termed equation between your limits for on the fly calibration based calculations.
I can't just use a force gauge, because I must make a mechanism that prevents overloading.
Yes you can, a force gauge MEASURES force, not produce it.
How much space have you got and how fast will you be doing force changes? ie step changes or constant force.
We need to know are you doing a static test, the sample just sits there and you apply force and measure strain, small deflection.
Or is it dynamic, as you apply force its physical length changes considerably.
Magnetic force is a function of current. Since we're looking at DC or very nearly, the current will be a simple linear function of voltage. PWM can be used too.
For 2-5mm, a medium-to-large speaker can do it. That's a very cheap way to acquire a good electromagnet. It won't be able to maintain 2kg force for very long before overheating. As a laboratory device, that would cost hundreds of dollars, but that price buys you continuous operation.
GuyJustHere: @robin - good idea, but large geared motors will need too much force to overcome gearing alone. Plus i noticed that it isn't consistent. This is still my favorite option. So i am goign to try it.
Maybe use a strain gauge to give you feedback.
And choose a gear system that meets the project requirement.
I have attached pics to show how we currently do it.
It is an adjustable weight on a shaft that is lifted. SEE BELOW.
@ Tom - I have plenty of digital force gauges, but we tend to go beyond the desired force. Even after it beeps.
@ Morgan - yep
@ Johny - exactly what we were thinking
@ zoom - that is another method we currently use. But as you pull, the force increases. Meaning you will put excess force on the test piece.
So, anyone know of a nice enclosed ac/dc PS with adjustable voltage?
Also, these will probably be on 12 hours a day. Only disengaging the metallic block during testing. which should happen 1,000 times a day. how do i prevent overheating?
Hi,
How does that system overload >55?
You just put a limit block on how far the weight can be moved away from the fulcrum, OR change the weight so it is heavy enough to allow 54 at max distance.
That is a great KISS system.
Robin2 suggestion is possible.
The system you have is reliable and already working.
Motor and chain on the base plate, some brackets on the chain to go up either side of the weight and push it back and forth.
Limit switches on either end to set min and max travel.
I think the motor idea mentioned is a good one. As Robin2 mentioned, you'd want to make the motor and h-bridge large enough not to overheat while the motor is stalled. You might be able to get more consistent results if you could monitor the current going to the motor. I'd think the current should be proportional the the PWM duty cycle but I'm not sure. I think monitoring the current may be a way of overcoming the inconsistency problem. Perhaps the h-bridge is less efficient as it heats up? If you monitored the current to the motor (not to the h-bridge) you might be able to catch any discrepancies caused by temperature changes in the h-bridge (or other changes which would cause the current not to be consistently proportional to the PWM duty cycle).
Adding a way to monitor the temperature of the motor might also help to make the system more consistent. As with monitoring the current, perhaps monitoring the temperature of the motor you'd be able to figure out some sort of algorithm which would correct for the inconsistencies you've noticed.
I've thought of doing something similar to test motors and h-bridges. I think it would be interesting to measure the torque a motor was able to produce at various PWM settings and using various h-bridges.
I just thought of another option. Rather than powering the motor, you could use some sort of heavy duty variable resistor to act as a brake on the motor. When the terminals of a DC motor are shorted together, the motor is very hard to turn. As you increase the resistance of the connection between the terminals the motor becomes easier to turn. I'm not sure if I understand what you're trying to do well enough to know if this would be useful or not.
@tom. Yes, the current system is great, but it is very hard on the eyes. and takes up alot of space. Also, it takes a lot of machining time to build. I am really trying to make an electronic system.
@ Duane - I love the idea of a pot and varying resistance, but I have tried. It allows you to turn the motor a bit before engaging any sort of resistance (also, slow turns see very little resistance) . Maybe i didn't have the right motor/ resistor. Now, i love the idea of measuring motor current. Maybe after some analysis i could make a formula to easily compensate for inconsistencies. Hmmm. Wouldn't a hobby servo work best for this? It would try to stay at the position being sent. maybe i would overheat that too...
The dynamic braking won't work for what you are doing. Once the motor stops moving, you lose all braking force. That is why it is called dynamic. It will not produce a static holding force.
