Diagnosing Treadmill shorts

Thread Starter

natacon

Joined Feb 19, 2018
7
I have a treadmill that keeps tripping the power circuit as soon as it is turned on. It was working fine up until a few weeks ago and it is regularly maintained, silicone spray under belt etc. When I took it apart, I found a long screw had fallen into the space beside the controller board and I suspect this has shorted something. There was a scorch mark near the bridge rectifier and the main cap looked slightly swollen, otherwise no visual damage to the board. I replaced both of these components but the problem persists. As soon as I turn it on at the wall, it blows the breaker. I've isolated the motor and tested it with a drill battery and the motor seems to run fine. I've checked for shorts against the frame and everything looks fine - no wires frayed or out of place.

Any suggestions as to what to try next.
 

tranzz4md

Joined Apr 10, 2015
310
Approximately when was it built? I'm guessing from what you've described that there may be an SMPS type controller, and it's toast too.

Info is everything though. How about a couple good pic's of that board, the machine, some spec's, etc.
 

MaxHeadRoom

Joined Jul 18, 2013
20,652
There is commonly two types of T.M. drive, SCR bridge and PWM/Mosfet.
If it has a normal four diode bridge then it is most likely PMW, if you have replaced or checked the bridge, check the power Mosfets etc.
Is it these you have checked already?
Max.
 

Thread Starter

natacon

Joined Feb 19, 2018
7
Thanks for the responses and sorry for the delay in getting back to this.
Here's a few pics of the board in situ and removed:
IMG_20180220_072537.jpg
IMG_20180220_072856.jpg

the scorch mark on the base and the bridge rectifier that I replaced.
IMG_20180220_073103_1_edited.jpg
IMG_20180220_073111.jpg
IMG_20180220_073138.jpg
This is the bridge rectifier I replaced. I was told at a local electronics store that it was 15A 600V and ordered the replacement based on that.
I also replaced the big capacitor.
IMG_20180220_073240.jpg
 

Attachments

Thread Starter

natacon

Joined Feb 19, 2018
7
Max, sorry to be so clueless, but are the mosfets the rectangular black plastic components on pins that are bolted to the aluminium bracket? In line with the bridge rectifier?

My brief online search based on the identifiers printed on them has one as a G4PC40W 705P and the other one I can't seem to find. The markings on it are NG43AF with PURG3060 underneath it. Any ideas? Ideally, I could pick these up on the way home so that if they test bad, I can swap them in.
 

MaxHeadRoom

Joined Jul 18, 2013
20,652
Yes, the power semi's are attached to the heat sink.
The IRG4PC40W is a IR mosfet, the other could be a rectifier/diode, does it have two pins or two of the pins connected together?
See what they are designated on the board.
Max.
 

Thread Starter

natacon

Joined Feb 19, 2018
7
It has 2 pins. I've removed them both to get a better look.
I tried testing the mosfet (as per a youtube video I found) and it seemed to test ok. Not sure if I was doing it correctly though. I had my multimeter on diode mode and couldn't get continuity between any of the mosfet pins which is good right?
IMG_20180220_231247.jpg
 

ebp

Joined Feb 8, 2018
2,332
I know some people are awfully fussy about their exercise togs, but I had no idea there was a distinction between treadmill shorts and general-purpose running shorts.

OK. Put down the stick. I'm leavin' I'm leavin!

Well, not just yet.
Max has you well on your way, so just a couple of comments, which may be telling you things you already know:

Use some fresh thermal interface goop (the white stuff). Clean the old compound off carefully first.
It looks like the diode and FET have a thermally-conductive electrically-insulating sheet between them and the heatsink. Usually if you use the insulator you don't use thermal compound, too, so I haz some bafflement.

Snug up the mounting hardware before you solder those parts in. It can be hard to get them to seat flat on the heatsink if you solder first.

It looks like you pulled out a bit of the plating that goes through the hole for the right-hand lead of the FET. It's easy to do. It is unlikely to be any sort of problem, but when you solder the part back in, check to see if the solder flowed well to the top side of the circuit board. If it didn't, it might be good to solder from the top side, too.
 

IamJatinah

Joined Oct 22, 2014
122
Good work.... lets back up a bit to the swollen capacitor. This is a critical part, when this degrades the motor drive needs to deliver more current to maintain motor power, speeds and regulations.

The blowout burn marks are similar to what one would find near a high-current cracked/broken connection point, or at times when exiting hot-gasses spew from ripped semiconductors....

The FET is a 40A part, the kickback diode is also beefy, and I might imagine the failing cap, which swells due to internal heating during charge/discharge operations while under use, plays a role in killing FET's, especially in single-fet driven motor controllers. The 15A bridge may also fail, as most bridges I see for 115vAC controllers are no less than 25A surge capable, up to 50A bridges for the fast peak surges seen by controllers of this type.
Under use, the treadmill can take 10A-13A (AC wall current)with surges exceeding this. The bridge rectifies that AC into DC at a high potential, and motor currents can get quite high. 15A amp may work at 220vAC.
Also, if we rip that FET hard enough, we should consider the gate-drive circuits as well, as just yesterday I found a FET shorted to the Gate, which back-peddled back into the predrive circuits that set up the switching drive for the FET. Lastly, nothing attached to that metal frame should be conductive, meaning all devices, bridge, FET, Diode will need to be isolated from that metal mounting frame, or immediate breaker pops could be expected. Most all motor controllers use "Off-Line" rectified drive circuits that have HOT grounds and are not isolated from the main Line AC. This is hazardous to humans, and these devices attached to any frame need to be isolated. Good Luck...
 

Thread Starter

natacon

Joined Feb 19, 2018
7
Good work.... lets back up a bit to the swollen capacitor. This is a critical part, when this degrades the motor drive needs to deliver more current to maintain motor power, speeds and regulations.

The blowout burn marks are similar to what one would find near a high-current cracked/broken connection point, or at times when exiting hot-gasses spew from ripped semiconductors....

The FET is a 40A part, the kickback diode is also beefy, and I might imagine the failing cap, which swells due to internal heating during charge/discharge operations while under use, plays a role in killing FET's, especially in single-fet driven motor controllers. The 15A bridge may also fail, as most bridges I see for 115vAC controllers are no less than 25A surge capable, up to 50A bridges for the fast peak surges seen by controllers of this type.
Under use, the treadmill can take 10A-13A (AC wall current)with surges exceeding this. The bridge rectifies that AC into DC at a high potential, and motor currents can get quite high. 15A amp may work at 220vAC.
Also, if we rip that FET hard enough, we should consider the gate-drive circuits as well, as just yesterday I found a FET shorted to the Gate, which back-peddled back into the predrive circuits that set up the switching drive for the FET. Lastly, nothing attached to that metal frame should be conductive, meaning all devices, bridge, FET, Diode will need to be isolated from that metal mounting frame, or immediate breaker pops could be expected. Most all motor controllers use "Off-Line" rectified drive circuits that have HOT grounds and are not isolated from the main Line AC. This is hazardous to humans, and these devices attached to any frame need to be isolated. Good Luck...
Wow! Thanks for the info. I think I understand although I'm not sure how to test the gate-drive circuits properly.

I replaced the blown cap, the bridge rectifier and the FET. I also checked the leads throughout for fraying and/or potential shorts against the frame. All looked good, but it blew again when I powered up. If I disconnect the motor, it doesn't blow the breaker. The motor runs fine when tested from a drill battery.

I'm a little out of my depth I think. If someone has the patience to guide me through this I'd be very grateful.
 

IamJatinah

Joined Oct 22, 2014
122
Hi Again, Well, there are a few things we should verify to go any further. I don't know if you have checked / replaced the kickback diode with any shorted FET as they usually die together. Then, have we verified the motor leads have no current path to Frame ground? Then, if these things are good, that predriver circuit that feeds the FET may be in trouble. If we don't set up a good switching signal to the FET, we will rip it.

I see some numbers on the PCB I will research.....also this is running on a 220vAC line. I suspect predriver circuits have taken a bit of a hit.

Lets not rip any more FET's until we find further issue's, did the bridge survive the last FET failure?
 
Last edited:
Top