Arduino SMC3 Dual monster moto boards for 100watt 12 volt motors HELP PLEASE

Gday guys and girls
Im a motorsport simulator builder im melbourne i use the arduino r3 uno and smc3 code to run a 2 DOF sim platform… i use a single monster moto board for this and run 70watt 12volt motors which work great for what i need… ive been given a new project from simworx here in Melbourne… long story short i now need to run 100watt 12volt motors which means running 1 single mm per motor so a total of 2 in the control box hooked up to 1 arduino r3 uno… ive got wiring pics but as far ive had issues … i can get one motor work no worries but when i go to the 2nd motor there no power until i ground the wire i use for interdependently grounding the motors… which normally i dont use motor 1 has the wire cut and placed in the harness because it has 12volts to it… but motor 2 wont work unless that wire is grounded which in turn draws alot of current and the motor shakes while operating and when i stop the motion input…

this is what ive followed
cheers

im using a single 30amp power supply to run both boards could this be the issues
should i run 2 psu 1 per board ?

Hi Evan,

Perhaps remove the motors and monitor the motor control outputs with a DMM / DSO. Double check you don’t have wiring issues, as it happens.

Monitor the voltage and current of your power supply under load. Ensure it’s not browning out. Back EMF can/will damage sensitive digital devices, so you will want a well-filtered supply for 5V. It’s always best to have an isolated supply when digital electronics are in the mix - ensure a common ground between the two supplies. The ever-popular Pololu D24V6F5 5V Step Down which is often used in quadcopters is rated for massive 42V, but that’s still not enough for the EMF of some high powered motors.

However you approach this, bear in mind 100W of power is serious business. From power supply through to seemingly harmless electronic motor breaking. One wrong move and you can fry a lot of things, almost instantly. Be careful! When I am prototyping with high power devices, I like to be 100% sure that things are working as intended by observation with a DMM / DSO / simulation. That way, if something is wrong, I can get in front of it (9 times out of 10) before it all goes south.

So in short, no quick fix for you. Just some trouble shooting tips. I hope it all pans out!

1 Like

thanks mate but no help i just thought some one have done what im doing im adding a second psu now fingers crossed

Is the problem consistent when you swap the logic-level wires to the motor drivers? ie. now driver A is driven as if it were driver B and vice-versa. This would trivially rule out whether there is actually a weird code issue or perhaps the reference returned by a pot is not stable.

Weird motor-shakes are hard work. Normally, I’d say that the micro is resetting if there are shakes, but it sounds like it’s only a problem with one motor. To avoid ground loop problems, make sure you’re using a star-grounding pattern when working with high current.

If you haven’t already, I’d add to the list of good practices EMF filtering across the motor terminals. This may not be the source of your problem, but it’s definitely a nearly-free improvement to the reliability of any DC-motor project.
This can be in the form of a small capacitor across the motor terminals, or a ‘triangle’ of capacitors using the motor case as a virtual ground. There’s a sketch below showing how to do this with a DC motor. The dots are the motor terminals.

2 Likes

thanks mate your info is helpful… the main issue is there are 3 brush housins in my wiper motor 2 have wires 1 has a terminal going to the body… i cut the wire and solder a wire on and run it to the outside harness… i normally dont use this wire… but this project with dual boards 1 motor works fine but the second does nothing till i earth the wire i cut then the shaking starts… the motor thats working has the same wire so i grounded that and that motor starting shaking to… so i no grounding that wire is no good… but with motor 2 if i dont ground it i have no motion… the motor cliped out after afew mintes under load as it got hot…there is also 12v to that wire… ive tried all possible motor wire connections and this is the only way that the motor dont just spin out of control… ive build many motion sims useing a single board to run 2 motors no issues… ive got a clone board in there with the core one and its the core board thats not working but the clone is…ive swaped motor wires and it just take the issue to the other motor

It sounds like such a tricky issue to have. As @Graham mentioned in his previous post, from here, the only way to reliably troubleshoot the issue would be to get a scope out a visualise every signal, the control lines, the power, everything. You’ll be able to see any noise, ripples, bad signals etc…

Unfortunately, using products outside of their intended use can raise some tricky issues to work through.

If all else fails, consider using a two-channel motor driver rated to handle your current requirements in its standard configuration. This will simplify your wiring, and reduce the likelihood of ground-loop issues.
I’ve used to MD03A from Pololu in a large robot before. Its a monster for the price. The only drawbacks are:

  • you will need to make a wiring harness to carry logic signals from your microcontroller to the board (it’s not a shield). I suppose this isn’t really an issue because you’re already doing the same for two shields.
  • There are only two mounting holes, so it’s happiest mounting on standoffs, using the header connection on one side as a support.

If you wish, you can close-the-loop using current feedback from the driver - this can be helpful to diagnose if your motors are stalling or failing.

If you decide to go for this solution, pay particular attention to the constant-duty current rating (infinite run time). If your motors are about 100W @ 12V = 8.5A then you’re about on the limit for the VNH3SP30 so it might be better to opt for the VNH2SP30 model, linked. Adhering a couple of small heat-sinks to the h-bridge chips never hurts too.