Help repairing DC variable speed motor controller driver

[TheBigM]

I am trying to repair a board from a ball pitching machine using a single DC motor.

Here is the driver schematic that I was able to take off the board (no power or MIC control is shown and the driver itself is incomplete).

All voltages are derived from 110VAC rectification and DC/DC converters - no transformers are used; all DC voltages were and are present and operational.

The only failed components were the two driver FETs and the main power fuse (not shown).

Upon replacing them the board would power on, the LED display would show me desired output speed, turning the knot controlling would increase the number on the display, i.e. it looks like it's working.

However the output voltage would vary wildly. I do not have a motor on site at the moment, so I was wondering if the circuit may need inductive load to work properly. I do have active load, but I do not have a motor at this time.

Attaching active load makes no difference.

I naturally replaced the FAN driver itself as well as the current sensor - no change in output.

There is no damage in the gate circuits that I can tell and am used to repairing power supplies, so I can usually tell.

Any ideas on what I might be missing?

Few things that bug me:

1. Why use different value gate resistors for the MOSFETs? I've never seen that and I am certain it is OEM, the board has not been worked on before.
   
   
2. Why use the switching relay at all if all NC, NO and COM contacts are shorted on the PCB? Again, it's OEM. If it was used as a cheap coil only then what the heck is the role of the coil in that circuit as I can't figure it out.
   
   
3. What is the role of the coupled inductor? Is it sensing or is it used to control the output? One of the two back to back diodes there might be a zener.

Note the schematic diagram is incomplete. All power and MIC sections are omitted (since power is present and the MIC appears operational) and there's more after the J? pin marked as input on my schematic.

Thank you!

 

[Tony Stewart]

Check your schema connections to VB and measure it with a DMM. It must the output switching pulse to charge Vb above 12V.

I wonder if the coupled L uses Q102 to create the Vb boost before the motor starts.

 

[TheBigM]

Thank you for jumping in. I was beginning to lose hope someone would.

Your participation alone already helped by forcing me to discover a few bugs in the schematic. I don't use KiCad often so I didn't catch a junction it had placed where it didn't belong (though the fault was mine - line was running right through a component's exposed end) and I also noted I've flipped D101.

The way it is now makes a whole lot more sense for the coupled 1:1 inductor to be meant to power Vb.

Shouldn't it need load for that though?

As far as testing goes, Vb swings widely and the Fluke can't get a reliable read, which also tells me there's no point in trying to scope it - if it was periodic then the fluke would be able to generate some average.

But I have my isolation transformer now and can check driving signals...if I know what to expect.

Thanks again!

 

[Nigel Goodwin]

Connect it to the motor, SMPSU's often don't work properly without a load - they 'hunt' - which sounds like what you're experiencing.

Using a scope would provide more information than a simple meter, and you'd see what was happening, rather than guessing.

 

[TheBigM]

I thank for your input as well. Do not presently have the motor unfortunately (it is with the client, who is remote), but

I've tried resistive (correction: "active") load, which is what I do in the rare cases when a SMPS behaves oddly for no apparent reason.

Did not make any difference, but looking at the schematic and thinking over what Mr. Stewart said it does look like load should be needed so that the coupled inductor can be used as Vb voltage source.

 

[Nigel Goodwin]

I thank for your input as well. Do not presently have the motor unfortunately (it is with the client, who is remote), but

I've tried resistive load, which is what I do in the rare cases when a SMPS behaves oddly for no apparent reason.

Did not make any difference, but looking at the schematic and thinking over what Mr. Stewart said it does look like load should be needed so that the coupled inductor can be used as Vb voltage source.

What sort of resistive load? - it might require an inductive load, but if not it would require a fairly substantial resistive load.

 

[TheBigM]

I meant "active" load, of course, not "resistive"...all load is resistive....my apologies.

Some - though not many- plasma TV power boards require load on the Vs output, which is usually around 210V DC and I use 150 Ohm power rheostat instead of classic bulbs.

It may need less than 150 Ohm in this case since the output is limited to ~90V DC, but frankly if the load was the issue I would have expected even 150 Ohm to work fine.

It may indeed expect inductive load...if it does it'll be my first. I did ask on other forums and the prevailing opinion was it shouldn't need load in the first place to regulate the output voltage.