Power Mosfet Motor driver voltage regulator

[gyrojeremy]

Hi all, I'm looking for clarification on this treadmill driver circuit and also what I hoping to achieve with the components.

1. Treadmill PWM driver board is a King-I BC-1070(6) though may not be relevant as I think I may have damaged something while trying to get the motor to run without the control panel.
2. DC Motor is 220Vdc 6amp 4000rpm 2hp.
3. Arduino Nano or Uno to provide PWM.

So I plan to stick with what I can learn from Mr Google. It took me a while to come to this conclusion but I think this treadmill motor driver is using the irfp450's for the voltage regulator as well as speed control for the motor. (Wow that sounds really dumb when I read it back) Let me try again.

Input voltage to King-I BC-1070(6) is 238Vac. It runs through a bridge rectifier and pumps out 324Vdc. The irfp450's have 324Vdc on the Drain. The max voltage of the dc motor is 220Vdc.

So what I'm asking is, Do you think the driver is using the irfp450's but only opening the gate enough to allow a max voltage of 220Vdc to go to the motor?

[gyrojeremy]

My other question is why would they need 2 IRFP450 when they are rated for 500v? The motor has a locked rotor amp draw of about 9amps.
IRFP450 Date Sheet

End goal is to build a closed loop PID PWM motor controller that will keep a constant set motor speed during loading and unloading of motor. It's for a CNC Lathe project, spindle motor to start with.

[WattsThat]

So what I'm asking is, Do you think the driver is using the irfp450's but only opening the gate enough to allow a max voltage of 220Vdc to go to the motor?

Nope. The irfp’s would release their magic smoke if you did that. PWM at a 50% duty cycle effectively outputs 50% of the input voltage. There are gate drivers on that board that either must be used or duplicated in some way in order to drive those mosfets from an Arduino. You’ll do way better to simply reuse the motor control board you have. It is speed controller and a good one at that. It maintains a constant speed even with the intermittent loading of a human body jumping up and down on it. That’s a way bigger load change than you’ll subject it to in a mill.

How did you try testing the motor that you think may have damaged it?

FYI, the maximum torque you’ll get out of that motor is about 2.7 lb-ft or 3.6 Nm.

[JohnRob]

I don't know any specifics on your board but can hazard a guess:

Check if the FET's are connected in parallel, If not I am guessing one is a switching device and the other is taking place of the "catch" diode. Normally referred to as a synchronous rectifier.

Regarding voltage. The 220 volts is likely a "running" voltage and you are able to run above that for short periods of time (i.e. when PWM duty cycle is low). Also I suspect the 384 volts will drop dramatically when it starts to get loaded by the motor.

[gyrojeremy]

By gate drivers do you mean something like a transistor with 10v drain connected to the gate of the fets?
If so then yes that's what it looks like.

Before I started pulling components off, I was able to test a few things and see the motor running for short periods. I'd removed the pulley with the magnet so it only ran for maybe 10sec at a time before it through up errors and stopped. After that I unplugged the control panel and the other driver for the angle motor so I was left with AC power input and the Main motor driver. This left me with the 7 pin plug on right bottom of the board. I figured I could put a pot across some of those pin and control the motor speed that way. There is a 12v pin and I tried jumping that to each of the others to see if the motor would start but nothing worked. When I reconnected the other driver and control panel I couldn't get anything to work.

The 2 irfp450's are in parallel but I'm not sure what the thing on the end does. It's connected to the the fets but not the same way.

During the testing phase I didn't really know what I had nor was it any good to me if I could get it running. It wasn't till I started researching how to build motor drivers that I realized that board had everything I need on it. Now I fear the damage is done.

If I could have tested those 7 pins while it was working I may have had better luck but I didn't think of that till it was to late

[MarkT]

gyrojeremy:
By gate drivers do you mean something like a transistor with 10v drain connected to the gate of the fets?
If so then yes that's what it looks like.

No, you'll destroy everything that way!
A gate driver is a chip dedicated to properly switching the highly capacitive gate of a MOSFET or IGBT
rapidly and not letting back-feeding from the gate/drain capacitance from destroying the gate oxide.

Gate currents are typically 0.1 to 1A or so, in order to charge up and discharge the gate fast enough
that the device doesn't burn up through ohmic heating.

Gate driver chips for high/low MOSFET pairs often have dead-time enforcement and undervoltage
protection (without either of these you'll soon have a blown up MOSFET or two). They also
use the PWM to bootstrap the high-side MOSFET gate 12V floating supply.

Before I started pulling components off, I was able to test a few things and see the motor running for short periods. I'd removed the pulley with the magnet so it only ran for maybe 10sec at a time before it through up errors and stopped. After that I unplugged the control panel and the other driver for the angle motor so I was left with AC power input and the Main motor driver. This left me with the 7 pin plug on right bottom of the board. I figured I could put a pot across some of those pin and control the motor speed that way. There is a 12v pin and I tried jumping that to each of the others to see if the motor would start but nothing worked. When I reconnected the other driver and control panel I couldn't get anything to work.

You played around until you blew something up. The way to not blow things up is to understand
the circuit first. There are lethal voltages involved - I hope you took all the right precautions.

The 2 irfp450's are in parallel but I'm not sure what the thing on the end does. It's connected to the the fets but not the same way.

Free wheel diode across the motor perhaps.

During the testing phase I didn't really know what I had nor was it any good to me if I could get it running. It wasn't till I started researching how to build motor drivers that I realized that board had everything I need on it. Now I fear the damage is done.

Yup - time to learn about motor controllers and figure out how the board worked, and perhaps replace
all the burnt out semiconductors?

If I could have tested those 7 pins while it was working I may have had better luck but I didn't think of that till it was to late

Google "reverse engineering" ?

[gyrojeremy]

Lol. I know just enough to not hurt myself but not enough to stop me from doing some damage. When I discharged that big cap I got quite a fright at the light show resulting from it.

It would take me months of research and scratching my head to try figure out how that driver works and lets face it, circuitry doesn't really float my boat (I wish it did) so it would be frustrating to say the least.

Might be time to put that one in a box somewhere till I figure out what a semiconductor is... lol. Or find another treadmill, take what I learnt from this round and apply it to the next one. Eventually I'll get it right.

Thanks for you help. I appreciate your time and knowledge.

[MarkT]

gyrojeremy:
Lol. I know just enough to not hurt myself but not enough to stop me from doing some damage. When I discharged that big cap I got quite a fright at the light show resulting from it.

Next time try a 1W 100 ohm resistor on the end of a grounded probe (or even a tungsten light bulb), less
unexpected spot-welding happens...

It would take me months of research and scratching my head to try figure out how that driver works and lets face it, circuitry doesn't really float my boat (I wish it did) so it would be frustrating to say the least.

Might be time to put that one in a box somewhere till I figure out what a semiconductor is... lol. Or find another treadmill, take what I learnt from this round and apply it to the next one. Eventually I'll get it right.

Thanks for you help. I appreciate your time and knowledge.