I also checked the motor and there is no short to ground. The resistance over the motor is 1 Ohm. Any ideas would be highly appreciated.

Thank you

 

I have yet to get hold of one with the 4 flash error and diagnose the actual reason for the error, the Youtube videos explaining the LED codes are too vague to be of any good.
Max.

 

Thank you for the response Max. My experience has been exactly the same. I thought my googling skills were failing me. It has to be a motor controller issue.

 

I would like to give some feedback on the 4 flash error on the MC2100ELS-18W controller. I found a short on the K50DDTP (IKW50N60DTP) IBGT between the collector and emitter. I replaced the IBGT and the treadmill is working 100%. It appears as if the 4 flashes indicate an over current protection state.

What I observed was happening, before I replaced the IBGT, is that when the start button is pushed the motor starts to immediately turn at full speed. The sudden inrush of current is high enough to trip a 16 Amp circuit breaker. That is with other appliances also drawing current from the same circuit, so I am not sure exactly how much current was drawn. If the treadmill is the only appliance on the circuit it would not exceed 16 Amps but would immediately go into the 4 blink state. Pressing the stop button would cancel the state.
It appears as if the IGBT failed due to overheating. The back of the IGBT shows signs of melting. It is mounted to a very substantial heat-sink with a silicone insulator between the IGBT and the heatsink. It is held in place with a spring clamp in stead of a screw. The main problem is that the mating surface of the heat-sink does not make contact with the full surface of the IGBT. A part of the IGBT sits below the heat-sink.
Mounting the IGBT higher should address this issue but it also appears as if the spring clamp does not hold the whole IGBT flush against the heat-sink. This appears to be the case when one looks at the damage to the IGBT. A screw would undoubtedly be better.

The damage on the IGBT is clearly visible on this photo. The smooth line on the bottom quadrant of the IGBT is the only part that does not show any heat damage. The part below the line sits below the edge of the heat-sink. It seems as if that smooth line was the only part making proper contact with the heat-sink. It is not that obvious on the photo but metal flowing upwards from that line is clearly evident.

 

Thanks for the update!
Max.

 

Where I can find the resistor number 32 or alternative? Is an (MRE .01 ohm 1watt) Part#
MR3FB10L0. Is normal to have continuity ?

 

DigiKey 15FR010E

 

The Q2 Mosfet was shorted I need to replace that resistance? thanks

 

Max, should the status LED be ON if no connectors other than AC power are plugged in? I wonder if I can debug the board that's not connected to the treadmill.
Thank you.

 

The LED should be lit with power applied to the board and nothing else plugged in, I test them stand alone with a 20Hz PWM generator I have made up with some custom boards to provide PWM and stop start into the HD2 socket.

 

Thank you Max, you've made my life easier - no need to run to treadmill every time I need to test.
Is heatsink also optional if there is no motor connected?

 

If you are not drawing current through the IGBT or a little current the heatsink should not be necessary.

 

Thank you!

 

Agreed, I run them off the heat sink when testing and with a motor on the bench with no flywheel attached, i.e. no load.
The built in acceleration is quite slow also.

 

Looks like the MOSFET was the problem. I also lubricated the deck under the belt. After it started running, it would slow down when you stepped on the belt. It still does a little at very low speeds of 1 mph and under.

I had the exact same problem with my treadmill running full speed for about a second and then shutting off. I followed this post, found the MOSFET, and replaced it. Treadmill now works great. Thank you for the post.

 

Does it start in high speed, or start to ramp up as it should?
I have 3 versions here, one of them a REV B. There is no temp sensor, just the CB1A/B for O/L switch.
And a MOV1 next to it, that is across the supply, so don't short that out.
Max.

Sorry to revive this thread, but my issue is as you first state (runaway start, dimming lights, and stop - display and election still work). Somehow that is not addressed anywhere else I could find... you obviously have seen this before. Thoughts?

 

Sounds like a shorted Mosfet /BJT.

 

Sounds like a shorted Mosfet /BJT.

Yeah that's what I thought too. Thanks that should help narrow my troubleshooting.

 

Hi all, first of all thank you for this forum I've learned a lot. I hope it's okay to keep an old thread alive. I'd like to get some feedback on how to proceed.

I have an MC-2100 Rev B in a Pro-Form XP 542E treadmill with a completely shorted mosfet... Q2 IRFP250N. I replaced it with DigiKey part IRFP250NPBF-ND which temporarily fixed it. Everything seemed to work fine then a day later it failed when starting it with no weight on it (when I pressed Start (or picked a speed) the motor started at full speed for less than 1 second then stopped). Yup, Q2 shorted again.

Any suggestions for what's causing it to fail / what I should do next? I'm assuming parts on the MC board have drifted too far out of spec - e.g. I'm eager to blame capacitors, but I'd like to get feedback on this before placing another DigiKey order.

I bought 2 mosfets so I replaced it one more time and took current measurements with the treadmill plugged into a Kill A Watt meter. It appears to me the treadmill is in good mechanical condition based on amp draw numbers I found on a fitness website. While I was testing it the motor sounded good and the walking belt was operating smoothly.

Test performed at 4 mph with Incline set to 0 (level)

Unloaded (nobody walking on it):
2.45 amps @ 121.0 volts

Loaded (me walking on it):
5.25 amps @ 121.0 volts
Bonus: I put a Klein clamp meter on one of the DC motor wires and got at most 10 amps but I'm not sure this is accurate given that it's pulsed DC.

After 15 minutes or so of it running at 4 mph, Q2 shorted again. I was on it fast-walking for probably the last 5 of those 15 minutes when it suddenly raced to full speed then stopped. Counting the original that's 3 mosfets destroyed.

To judge the walking belt friction I tried pushing myself away from the console (with the machine off). It's difficult and I didn't glide much, but it's much easier than every treadmill at the gym. I can barely push off with those. Deck lubrication wouldn't hurt (I'd have to order some), but I doubt that's the problem since on one occasion it failed during start-up with no weight on the belt. It seems fairly easy to move the belt by rotating the flywheel by hand. But I'm not a treadmill expert.

I also wonder if it could be an issue with the heatsink. I re-used the pink (silicone?) sheet that was on the back of the original mosfet and placed it on the new mosfet in the same way and tightened it to the heatsink fairly snug. I didn't use any other products. Do the pink sheets wear out over time and not conduct heat as well as they age? Or could it be breaking down and conducting electricity? It looks okay, visually at least. I think I heard a faint crackle sound coming from the MC board while testing. I also think I smelled burning electronics. I touched the face of the mosfet (the printed text) right after it failed and it was warm but not hot.

I reviewed the reverse-engineered schematic referenced on this forum.

I completely removed the MC board from the treadmill and placed it on my desk (no connections to anything) and made the following measurements in-circuit with a Fluke DMM:
(I just assume the diode bridge rectifier and everything before it is okay)
(Also Q1 - I don't know how to test but assume it's okay judging by its role in the circuit and the fact that the treadmill functions normally, albeit for some minutes, when I replace Q2)
R9 measures 3.472k
R14 measures 25.23k
D13 diode mode measures 0.418V in one direction, 0L in the other
R32 measures 1.0 ohm
R29 measures 99.1 ohms
R30 measures 99.5 ohms
C2 capacitor mode measures ~ 7000uF

Unsoldered and measured this out-of-circuit:
C5 (1000uF 200V) measures 890uF on Fluke capacitor setting

I'm wondering if C5 (or C2) are bad despite measuring close to the rating. I notice ebay listings for rebuilt MC boards state "the capacitors upgraded to low impedance/high reliability capacitors" which further makes me question the caps. I should note my board is either an original or a replacement from a Sears technician (a family member called for service many many years ago).

With the motor disconnected from the MC board:
Motor's negative wire to treadmill frame (ground screw) - infinity
Motor's positive wire to treadmill frame (ground screw) - infinity
Measuring across the motor's positive and negative wires - 1.0 ohm
Measuring across the motor's blue wires (I assume this is a N.C. thermal overload switch) - 0.2 ohms

Thanks for reading!
--Jay

 

How old is the TM? Have you checked the brushes and any carbon dust in the housing on the com end?
If you could monitor the current when in use under different circumstances it may show something!