2
$\begingroup$

I have a Zumo Shield. It is small and tracked, and contains two micro metal gear motors soldered to it.

In theory, these motors are identical and do seem to rotate at the same speed. However, the left side seems to have considerably less torque than the right. The left side will stall on small obstacles and the robot ends up veering off course.

Another test of turning force, if I set both motors to spin at the same speed, while the robot is off the floor, I can put my finger on the track and cause the left to stop with a lot less pressure than the right.

I know it's never going to be perfect and I need some sort of feedback loop will be required to keep it straight however there is a considerable difference between the two.

What is the likely cause of this difference? Soldering? Motors? Gearing? Motor drivers? How can I go about diagnosing and therefore fixing this problem?

$\endgroup$
  • $\begingroup$ make sure that the motors are the same model ... remove the wheels ... one could be binding $\endgroup$ – jsotola Nov 26 '19 at 2:57
  • $\begingroup$ yes, already done that $\endgroup$ – Dave Hillier Nov 26 '19 at 9:29
  • $\begingroup$ Normally the feedback you get from motors is from encoders, do you have on your platform? $\endgroup$ – nionios Nov 29 '19 at 9:45
  • $\begingroup$ yes, but even still the motor is considerably less torque. $\endgroup$ – Dave Hillier Nov 30 '19 at 23:21
1
$\begingroup$

This is not a complete answer, but one generic technique for diagnosing differences between devices which should be identical is divide and conquer. Swap items until you see where the problem follows.

For instance, if you swap the motors and the problem stays with the motor, you probably have a faulty motor. If a different motor with the same amplifier exhibits the same problem, it's probably the motor driver. If swapping them and neither of them works properly, or both work, it may be (or have been) the soldering.

Once you have identified the subsystem which is at fault, you can then continue to investigate and find a more specific subsytem or component.

For instance, if it appears to be the motor drivers, you can try swapping the amplifier chips. If the problem follows the chip, replace the faulty one, if both chips work as well as each other, try measuring voltages across the chip, current flowing through the chip, the resistances seen between the pins of the chip and the motor and power supply connectors etc.

Similarly with the motor module, you can try swapping wheels (as you've done) gearboxes, encoders (if they have them) and so on.

The key is to build a hypothesis about what might be wrong, conceive of a test which can prove (test) that hypothesis and then use the data from that test to either identify the component which needs to be fixed, replaced or investigated next.

Rinse and repeat until the root cause is found.

$\endgroup$
  • 1
    $\begingroup$ I havent got to the root cause but its something to do with supply voltage i.e. when the batteries get low its far worse on the let. I'm leaning towards a design flaw with the zumo but whatever easier just to keep the batteries charged and correct with feedback $\endgroup$ – Dave Hillier Dec 6 '19 at 18:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.