Hello everyone. First time here. Long post, but I try to give as much info as possible.

I am controlling a DC Brushed motor with a PWM Controller Board, which is connected to an ESP32. I have burned my second Controller. I know why the first one burned, but for the second one I just have guesses. I am very new to electronics, and I'd appreciate your input.

• The brushed DC Motor is rated 1.2 HP, 900W, 12V. There is no amperage rating, but I understand 900/12 = 75A
• The controller was rated 100A continuous.
• The controller is powered by an external 12V Switching Power Supply connected to wall power (120V)
• Just before the controller burned, I noticed the power supply shut down, and restarted, which suggests the overload protection in the PS kicked in.

I read that when a DC motor slows down, it can act as a generator, and supply high current back to the system. I read several recommendations to wire a same-volt battery parallel to the power supply, but I am not sure if this is a good idea, or how to do it.

From one controller documentation I read:

If a power supply that cannot sink current is being used (bench top and AC to DC switching power supply), the input voltage will rise when the driver is regenerating (motor is slowing down). Thus, it is important to connect a battery with same voltage in parallel with the power supply to absorb the current generated by the motor. Else, the input voltage might rise to a level where SmartDrive160 will be destroyed permanently.

Also from another controller supplier, they show how to connect a battery in parallel with an ATX power supply here

From the above posts I understand this might be the cause of the burn, as I did not have anything to absorb the regenerated current (if there was any)
For my questions, which I have many, but I'll start with a few:

1. Is this even a good idea to wire a battery in parallel to the power source?
2. If so, what kind of battery should I use?
3. Will the power source also charge the battery?
4. When the controller/motor draws high amps, will that cause any problems with the battery?

More background info:
The DC motor will run for one minute, twice a day.
The system needs to be connected to power 24/7 . It will be used only twice a day, but what time of the day it will be needed is not set.
The system is in a remote location, with no-one to monitor it. For any maintenance, I need to drive 6-7 hours to get to it.

When I start the motor, my code on the ESP32 gradually increases the PWM duty cycle from 0 to 98% over 500 milliseconds, and when I stop it, I gradually reduce the PWM duty cycle from 98% to 0% over 333 ms.

The PWM switching frequency is 15KHz, which is recommended for this controller.

The DC motor has a worm drive gear box, and it is pushing a 2000 lbs load which is on wheels. To give an idea of the force needed: If I disconnect the drive system, I can hardly push the load myself while standing on a ladder (it is high above ground), so I'd assume I am pushing at most with 50lbs if I'd put a scale.

When the controller burned, the heat sink wasn't even warm. It was a part not under the heat sink that burned.

The first controller burned because I set it to 100% PWM duty cycle, and the documentation mentioned not to exceed 98%. I'm dissapointed it did not have any protection against it, but anyway, I should have read the doc more carefully.

Thanks much,
-Ozer.

 

I read that when a DC motor slows down, it can act as a generator, and supply high current back to the system.
-Ozer.

Just about all DC motors generate, This opposes the applied voltage.
What it sounds like is you need suitable current protection devices of some kind, what is in place now for protection?

 

I would go for a much higher powered controller. The motor stall current will be a lot higher than the running current and even tho you are ramping the PWM up, the higher current needs to be taken into account. And for such a load, just ramping up for 500mS may still be too fast.
Try running it on a car battery for your testing, with an added trickle charger connected when you finally install it. As it runs so infrequently, you will not need a bigger power supply.

 

What it sounds like is you need suitable current protection devices of some kind, what is in place now for protection?

There is no protection. My setup is, 12V DC Power Supply -> Controller -> DC Motor, and an ESP32 connected to the Controller with its own 5V power. In the attached pic the ESP32 pin header is disconnected, but that is pretty much the entire setup.

 

500Ms is not a very long ramp up time, can you take a leaf out of the TM units that take a few seconds to ramp up, Some use a method of using a SCR to slowly switch the HV DC from 0 to max over the required period.

 

I would go for a much higher powered controller. The motor stall current will be a lot higher than the running current and even tho you are ramping the PWM up, the higher current needs to be taken into account. And for such a load, just ramping up for 500mS may still be too fast.
Try running it on a car battery for your testing, with an added trickle charger connected when you finally install it. As it runs so infrequently, you will not need a bigger power supply.

dendad, are you saying not to use the Power Supply at all? <Looking up what a trickle charger is/>. Ok, makes sense <after reading about Trickle charges/>. So just pick up a 12V Car battery, and a trickle charger from Amazon.
Also, increase ramp up time, and slow down time. And find a higher rated controller, which might be the challenging part.

 

500Ms is not a very long ramp up time, can you take a leaf out of the TM units that take a few seconds to ramp up, Some use a method of using a SCR to slowly switch the HV DC from 0 to max over the required period.

I can easily increase the ramp up time. The slow-down needs some consideration, as I need to stop the system at a specific location. I have limit switches that I will use to slow down and stop. I will experiment with those, but will keep in mind to start/stop it over a few seconds.

 

Quite a problem you have.. Now I have questions: what were the conditions when the controller burned up? Also, what are the conditions for running when the controller does not burn up?
By conditions I mean how much load , approximate speed, increasing or decreasing speed? Does the controller require an added heat sink?? Are all of the motor and power connection wires sized for the required current?
AND, as for a DC motor running as a generator, that only happens while the motor drive power is less than would be required to spin it at whatever RPM it is turning, that is, when the motor is being driven rather than driving.
That would be like coasting down a hill in a motor powered vehicle

 

AND, as for a DC motor running as a generator, that only happens while the motor drive power is less than would be required to spin it at whatever RPM it is turning, that is, when the motor is being driven rather than driving.
That would be like coasting down a hill in a motor powered vehicle

Not in my experience, where do you think the BEMF comes from to limit the current on a driven motor,
From one of my very OLD academic Note books.

" As an armature turns, the conductors cross the magnetic field and a voltage is generated in them.
This voltage is the same whether the armature is turned as a generator, or as a motor. When it's turned as a motor, this voltage is called the 'back EMF "

 

Quite a problem you have.. Now I have questions: what were the conditions when the controller burned up? Also, what are the conditions for running when the controller does not burn up?

To clarify what I'm doing, and to show the operation of the motor check out this video
In case you are wondering, it is (will be) an observatory.

This was the first time I ran the system under load. And a few minutes later the controller burned. I didn't even run it the entire length, because I was checking for any obstructions, etc. But of course the frequent start/stop action might have contributed to the issue.

 

Not in my experience, where do you think the BEMF comes from to limit the current on a driven motor,
From one of my very OLD academic Note books.

" As an armature turns, the conductors cross the magnetic field and a voltage is generated in them.
This voltage is the same whether the armature is turned as a generator, or as a motor. When it's turned as a motor, this voltage is called the 'back EMF "

OK, certainly the counter EMF is in opposition to the drive voltage, and that limits the running current. I was referencing the terminal voltage, which becomes apparent when the motor is being driven by an outside torque or inertia. The counter EMF while the motor is running limits the current, it is when the motor is not delivering torque that it ca be damaging. Certainly I should have been clearer nin my comment.

 

Certainly that is an unusual application. I have an idea but no simple answer.
If the drive package has an input that will allow speed control, then slowing it to a stop instead of just switching the drive off could be a solution. If you are using a mains power switch to stop and start, that could be the problem, because it is certainly not controlling when power is suddenly removed. I did damage a motor drive doing exactly that, with an "Emergency Stop" circuit that simply cut power. Slowing to a stop could be the simple solution.

 

I think try again with a car battery (not a gel cell).
A car battery can absorb the back EMF from the motor ok. Just do not disconnect the battery when the motor is running (unless there is a major failure).

 

Ramp up and ramp down definitely can blow up a power supply. If you need it to stop at a specific point than maybe a full on controller maybe with braking resistors. Its probably outside of the budget but a 180vdc motor and a 2hp controller from automation direct https://www.automationdirect.com/ad...gh_voltage_(>_50v)_dc_drives/gsd8-240-10n4x-u. The largest 12vdc motor controller at automation direct I saw was rated at 20a.

 

Ramp up and ramp down definitely can blow up a power supply. If you need it to stop at a specific point than maybe a full on controller maybe with braking resistors. Its probably outside of the budget but a 180vdc motor and a 2hp controller from automation direct https://www.automationdirect.com/adc/shopping/catalog/drives_-a-_soft_starters/dc_drives/high_voltage_(>_50v)_dc_drives/gsd8-240-10n4x-u. The largest 12vdc motor controller at automation direct I saw was rated at 20a.

I would like to see an explanation of how ramping speed up and down in a reasonable rate can cause a problem. A reasonable accell and decell are the best way to go, under normal conditions.

 

reasonable wont. an unreasonable acceleration or deceleration based on motor ratings and load will. Trying to get a 3 ton load up to 3000 rpm with a 1/2 hp motor in a half second will. If its not a smart speed control and it cogs the motor, that whole phase difference is dumped into the controller as back emf.

 

The ramping that I am suggesting is much slower, so that the decell is during a foot or two of travel. But as the drive was apparently sized to go with the motor, and as I believe it is a DC/universal motor, it should have plenty of torque. Also, a correctly sized drive will have built in overload protection. One certain thing is that there may be an additional capacitor to absorb the spikes of over-voltage from the PWM operation. I recall destroying a modular stepper driver that way. So the part that burned up may have been the zener diode.
Does the TS have the instructions that came with the drive?? If no instructions came with it, take that supplier off the list!!

One other easy thing will be to use a smaller sprocket on the gear-motor shaft. That would be a simple way to reduce the load quite a bit.

 

There is simply no way that a Printed-Circuit-Board can deal with ~100-Amps-Continuous-Current.
Even 100-Amps "Peak" Current is still a huge maybe.
It's going to smoke at some point.
~10 or ~12 gauge Wire is just not gonna get it.
The Wire, by it's self, will generate enough Heat to cause Component-Failure on the Board.

If mechanical "Shock-Loads" can be minimized, ( to prevent mechanical breakage ),
a simple automotive Starter-Solenoid, and a Lead-Acid Starter-Battery may do the job just fine.
.
.
.

 

LQ is right!! Based on personal experience integrating a PCB designed and built by others into a complete product package, even if the PCB conductors do not evaporate, the voltage drops will be excessive. And the product i was producing only had to operate for less than a minute. It was the power package for lights and film cameras inside a vehicle for crash testing. The PCB survived because of #12 wires soldered on top of the high current traces.
This is a prime example of the flaw of hiring outside circuit designers who lack adequate real world experience. The project was dumped on me shortly after I was hired to replace one who had left before the project was to be assembled.. Quite an interesting start at a "startup" company position.

 

If 75A is the running current then the start-up current will be several times that. Unless the power supply is rated for the start-up current it will go into self-protect shut down. Unless that happens quickly enough the controller will have to handle the start-up current. Seems it can't.