Maker Pro
Maker Pro

Need help diagnosing H-Bridge

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The H parts have a higher guaranteed minimum current gain than the L parts.

I would suspect that either will work, but to be sure, get the H part.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Ok, while I'm waiting to get my transistors, I have a question for you Steve. I am running this steering wheel with a bench power supply since I have modified it with different drive motors and have raised my voltage from 24; sometimes I run it at 30V, sometimes up to 40V. Amperage maxes out about 14A-15A. The part of the board that gets hottest is the drive MOSFETs, and I was wondering what your thoughts are about paralleling them to help insure I don't get too close to current limits. Would this help keep things cooler also? I have looked for a possible replacement MOSFET that would have lower on resistance, but haven't found anything that would be a direct fit. Any suggestions?
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Anybody have any advice on this subject for me? I tried PM'ing steve and for some reason I seem to have turned invisible to him!
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
EEEK!

I would strongly advise you to run the device ONLY from the voltage it is designed for.

If you really want to run a higher voltage/current device, then take the logic signals from the existing board and run them to an external H Bridge with its own driver.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Well, I looked up the components and they are safe (rated) up to 40V, or I wouldn't have done it.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
It strikes me that you are controlling a much larger load than these units were designed for.

A device rated at 15A and 40V might not like 15A and 40V. You need to look at the safe operating area.

The other thing is that if you just change the voltage to the motor (essentially across an H bridge) you need to ensure that the "upper" mosfets get sufficient gate drive to both turn on and turn off.

If you increase the supply voltage to the entire circuit, you need to ensure that the circuit will operate correctly at the higher voltage.

Unless you know it's going to work, it's best to proceed very carefully.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Right, I have slowly increased things and monitored heat levels (with temp probes) and checked voltage, etc. to some other components and it had been working great for several months until I overvolted it by accident.

Some of your stated concerns are also the reason I asked for some different recommendations on better/higher capacity MOSFETs. If you think it would be better to add a whole different h-bridge I will have to consider that. Can you outline for me why it would be better to go that route?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Shall we get the thing running again before we decide to modify it?

Probably the best move.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Sure, that is probably the best route. Parts are on the way. (Now I feel like a pest, sorry).
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
No, that's OK.

But best to start with something working (if we can) before we modify it.

Tell me about the original load used? 24V but what current -- and it was a motor of some sort I presume?
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Original motors were two Mabuchis with about 3A stall current. The OEM power supply is a 24V, 5A switcher. The ones I'm using as replacements stall at about 4.2A.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Well, I looked up the components and they are safe (rated) up to 40V, or I wouldn't have done it.
The components may be rated for 40V, but the circuit design won't necessarily work at 40V. Many component selection decisions, including resistor values for example, are made based on the supply voltage. Without a schematic, I can't say for sure that the circuit would need changes to work at a higher supply voltage, but I definitely would not assume that it wouldn't.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Well, I understand where you're coming from, but I have already used it at 40V, often for several hours per day and nothing died until I accidentally turned it on at 52V. I slowly increased things while watching temps and everything held up well. The ancillary circuits besides the motor bridge are all regulated, so I'm pretty sure the only stressed member there would be the reg's. I am a little leery about longer term on the h-bridge, hence my questions.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Crud. Got my transistors and went to R&R the fried one. Removed it and found that a leg was missing...expected that was possible due to the short. However, the pad and the trace both also vaporized back about 1/16" away from where the pad should have been. As this is a multilayer board, is there any way to tell if there are any connections going down into the board? I'm sure I could flatten a piece of copper wire and epoxy it to the board to replace the trace, but I'm not sure how to find out I there were any connections going into the board and if so, where they go?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
If it is a multi-layer board, connections between layers are done with vias. These are tiny plated holes. Pretty easy to spot. They look like a hole for a component leg, only a lot smaller.

Vaporized traces are never a good thing.

You can use a short length of wire to connect the end of the trace to the pin. I would use wire-wrap wire as it is thin and easy to solder.
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
Ok great, there are no holes (vias) on the trace path in the area that it's burned off. Hopefully I can get the track repaired tomorrow and get to measuring and comparing those base, collector and emitter voltages. I got a really good deal on a new multimeter off of a local guy on Craigslist, an 87V with TL81a lead set, so the new leads should help prevent this happening ever again.

The area in which the trace vaporized is extremely close to both a ceramic resistor and the leg of another transistor, right between them with only about enough space as the trace is wide, so it would be difficult to route a wire unless I left it hanging in mid-air. The trace comes into the leg from the opposite side of the transistor, so it has to go under to get to the leg. That's why I think the flattened wire re-trace would probably be best. Do you agree?
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
If it's really close to other components, use a short piece if insulated wire (wire wrap wire is insulated).

The last thing you want is another short!
 

eKretz

Apr 8, 2013
251
Joined
Apr 8, 2013
Messages
251
It's ok to have it up in the air? I was under the impression that that was a no-no.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
That's fine, but I'd prefer not to if possible. If you have to (or if it has to snake around) it may be best to tack it to the board in a couple of places. Hot glue is OK for this as it is just a melted plstic.
 
Top