last day (16 days later) » 

1:53 PM
1
Q: Motor Driver keeps dying/frying driving a worm gear motor

TheCodeNoviceI am looking for some consultation about controlling a24VDC worm gear moto. I have blown 4 h bridge motor drivers thus far and while I have been trying to keep my project price tag low I may need to step up in quality but I want to make sure something else may be going wrong. I am a ME by trade b...

 
What is the frequency of the pwm? Generally I’d put a varistor (MOV) across the motor terminals as close to the motor as possible. Something like a 20mm 30V device.
 
@TheCodeNovice, Just now I skimmed the Pololu motor driver you are considering, to get a rough idea of your requirements: (1) Pololu G2 High-Power Motor Driver 24v13 (6.5V to 40V, cont 13 A without heat sink, no over-temperature, over current shut-off) pololu.com/product/2992. My first impression is that it is not newbie proof, because over temperature and over current can fry the driver! :)
 
@Kartman my pwm frequency is 10kHz. Do you have a link of MOV so I know what I am looking for?
@tlfong01, Would it over heat with out running the motor? I do not think that worm gear motor I am using will run close close to the rated 13A that is why I thought I could slide by without it. Do you know of a nice noob level h bridge brand I could use?
 
Google MOV 20mm 30V
 
@TheCodeNovice, well, I did fry many gear motors (do't ask me how many, but I do have spare gear heads, just in case! :)). One of the many reasons are (1) physically overloading, (2) gear stuck or crashed, (3) start current which might be many times of operating current (so Pololu suggests ways to avoid bad things happening, by using big cap near PSU, or slow slow starters).
For motor newbies, I usually recommend the following Q&A: How to use motor drivers with H-bridge and PWM input, to control direction and speed of DC motors? - EESE, Asked 2020jul16, Viewed 1k times electronics.stackexchange.com/questions/510755/…. I usually newbies to start with plain, old, very power inefficient, stupid, L298N bipolar H-bridge, 4A driver, which has built in Schotty flyback diodes on board/module to protect back EMF.
You can start messing around with the very popular yellowish, cheapy TT130 gear toy motor with operating current around 350mA, and stall current less than L298N's max 4A (So it is very difficult for newbies to fry the driver!). After you have gained some experience and confidence, you can bravely go up using mid level MOSFET H-bridges, and perhaps catch up me, the motor driver ninja, with my BTN7971B, max 43A, with current control, current sensing etc etc. You can search "BTS7970", "BTN7971B" for my related answers. Ah, I missed my locking down morning tea. Good luck and see you later.
 
1:53 PM
@TheCodeNovice This is one of those times when I really tend to prefer BJTs. Not because BJTs are better. They aren't. They are just different. But because it is harder to design MOSFETs for bullet-proof performance in motor bridges like this and if you are buying cheap products chances are you are also buying lower training and experience levels in designing them, too. I'm just a hobbyist and not a well-trained engineer, though. So I'm biased somewhat towards "easy to get right." Take that into account.
@TheCodeNovice Aside from all the ways the MOSFET gate can acquire far too much charge with inductive motor loads -- reversable moreso -- MOSFETs also hate sudden junction temperature increases because of over-voltages across their internal body diodes. (Avalanche!) When the MOSFET is switched OFF the inductor's reverse EMF passing through that MOSFET body diode heats it way past its limited junction temperature and boom. An external high power diode across the DS terminals can help. I guess I'm saying that design experience matters -- especially with MOSFETs and inductors.
 
@tlfong01 if you want to convert this great info into an answer I will accept it.
 
Gil
Do you have dead time when switching?
 
@Gil, yes, that is my worry when I studied the OP's two drivers he frird. His drivers use 4 discrete transistors to form a bridge (I am not sure if the module has hardware "break before make" mechanism so that one transistor switch off before the other switches on, otherwise there would be a "shoot through" current and fry something. The modern H-bridges have built in hardware shoot through through prevention. The two drivers the OP is using are either discontinued (for a reason) or seemingly old design. So I am recommending the newbie proof, L298N BJT NPN H-bridge, with built in flybacks.
 
@tlfong01 If you are stuck using a motor for the application what would be a suitable H Bridge? The L298N appears to only handle 12VDC motors.
 
What kind of power source do you use for all this? Maybe you have undervoltage under some conditions and driver can't drive gates properly and transistors can't open completely, so they blow up?
 
1:53 PM
@Kamil I am using this amazon.com/gp/product/B01IOK5FM0. I have this powering the worm gear motor, stepper, stepper controller, beaglebone black and a load cell digitizer. Outside of the motor nothing really draws much in terms of amperage
 
Gil
MOVs will adsorb transients but they are not extremely fast, I prefer semiconductor devices designed for this purpose. I would also suggest he check his circuit and be sure all of the H-Bridges have appropriate pull up or down resistors on the micro as they will be unknows during startup and large transient can occur especially if both drivers on one side are activated.
 
@TheCodeNovice, Your Amazon EagleWell SMPS looks good. I will be including it in my reference list, and we will study it if necessary. I have various PSUs in my lab, mostly 5/12V, up to 48V, from 1A to 21A. I will be using 12V only for this motor driver project, and perhaps you 24V on your side, so we can compare and contrast.
@Gil, yes, I agree that MOVs are not fast enough. I usually use 1N5822, sometimes in parallel in big current cases as the OP. Back EMF Schottky Flyback Diode Selection Notes - Rpi.org.forum raspberrypi.org/forums/…. Cheers.
@TheCodeNovice, Well to avoid motor getting stalled and drawing huge current, you can use H-bridge drivers with over temperature and over current protection: (1) I use ployfuse and slow blow fuse as the first line of defense. (2) I also use current sense circuits (yes, eg, for L298N and BTN7971B) to detect over current, actually current monitoring for feedback speed/stall control.
Ref 7, 8, 18 of my answer describe how to monitor motor current, for over current alarm, and/or PID speed control etc. Appendix D of Ref 8 explains why wrong wrong wiring/control might burn the driver. BTW, do you use any BBB code to control the motor?
Your Pololu motor driver (Ref 15) also has current sense circuit to do stall alarm or speed control.
Just now I searched my junk box and found a 24V 13rpm worm gear motor, lying there for perhaps 5 years, collecting dust. I am now thinking of playing with it, using my BTN7971B motor driver: i.imgur.com/MDDbhBa.jpeg. Stay tuned. Cheers.
Now I am doing a preliminary test. I used a multimeter to measure the worm gear motor's resistance and found 28Ω. So the operating current should be I = V/R = 12V/28Ω ~= 400mA. Then I use a 220VAC to 12VDC 1A PSU to drive the motor. I found the motor moving slowly, as expected: youtu.be/8RdyY_lDuRQ. Next step is to use BTN7971B to drive the motor.
BTN7971B driving the worm gear motor is smooth: youtu.be/SHsiKPyaaW4. So I have shown the OP how to do it. Perhaps the OP can get a L298N and a BTN7971B and test his 24V wrom gear motor. I will lwait for him to catch up and see if he has any other questions. In the mean time I will carry on the following: (1) use PWM to adjust speed of motor, (2) Use Rpi Thonny python and ADC to measure the current of the motor. This following up work is not asked by the OP, so is out of topic, and not reported here.
 

  last day (16 days later) »