Hi community, I’m glad I found you on the internet.

So what I am facing: After pressing the start button, the treadmill begins to accelerate to the default 2 km/h, but then it slows back down. It speeds up again, then slows down significantly. It repeats this in a regular cycle of about 3 seconds.

If I set, for example, 10 km/h, it speeds up to roughly that speed, then slows down again, speeds up again, and slows down again in a regular cycle continuosly. Shunt resistor looks ok, but i cannot measure its exact value because it is normally small.

Nobody is standing on the treadmill, and there is no mechanical issue.
The reed sensor signal reaches the motor controller board, and from there it goes to the console board.
There is no break in the cable from the HD2 connector, and the 5V and 9V DC supply voltages are present and stable.

I also tested the motor separately by powering it with a 12V 5A DC supply, and it ran smoothly and evenly. Incline works well.

The LED on the motor controller is blinking rapidly — I believe this indicates that it is receiving a PWM signal (speed signal set point) from the console board. The board microcontroller is a smd CY8C24223A. The IGP is G4PC40K.

What could be the problem? Thanks.

 

If I set, for example, 10 km/h, it speeds up to roughly that speed, then slows down again, speeds up again, and slows down again in a regular cycle continuosly. Shunt resistor looks ok, but i cannot measure its exact value because it is normally small.

Did you check the belt speed sensor?

 

Did you check the belt speed sensor?

I have already tested the reed switch circuit, and it works well — it detects the magnet correctly. The signal from the sensor also reaches the control panel, coming from the motor controller board through the microcontroller and the optocoupler. Although the control panel (user interface) doesn’t actually use this signal — the distance counter works even without it. The control panel only sends commands (aka pwm signal); it doesn’t regulate anything.

I also checked the PWM signal coming from the control panel to see if there was any pulsation in it. There wasn’t. Changed only accordingly the "set speed". The LED on the motor controller board always flashes quickly and evenly.

I tried using a dummy load by connecting a 1400W iron instead of the motor. I measured the treadmill’s current draw and the voltage going to the iron. This way it operated smoothly and evenly. When I changed the speed on the control panel, the voltage and current changed gradually and steadily. There was no pulsation at all.

Then I reconnected the motor. The treadmill started operating with pulsation again. I tried holding the belt to see how it reacts to load. The LED began flashing slowly, indicating it was approaching peak current. Interestingly, the pulsation stopped — the belt speed felt smooth. But then the house circuit breaker (16A) tripped. The IGBT blew. I hope that’s the only thing that failed. Now I’m trying to reproduce the original fault condition . I have a “similar-rated” IGBT at home (IGW40N65F5), but I’m not sure if it’s worth trying.

Some video:
https://drive.google.com/file/d/1IamLLNQVNQWV2icw2OgUkyqRyyIU5_5o/v...haring
https://drive.google.com/file/d/1KcXACHlEHExCsVW-wHbjlUQ69GQ2emsB/v...haring
https://drive.google.com/file/d/1bs6hxuA0qjcJAu04mu3OGlBNTv7RK54f/v...haring
https://drive.google.com/file/d/1NYvc1X81eQXfcAiSrlJUDwvRHYOR860N/v...haring

 

Oops, meant to reply elsewhere .... Deleted...

 

Did this get resolved? I am experiencing the same issue. The screen no longer works on my treadmill, so I've wired in a Raspberry Pi into HD2 of the mc2100. I'm generating a 20Hz square wave from the PWM of the Pi. When i turn on the signal the motor starts running smoothly and LED is flashing normally, but after 5-10secs it slows down and stops, then a few seconds later it starts up and repeats the cycle.

Not yet. In fact, during my experiments I ended up blowing the IGBT as well ...See my comment above for details.

 

I replaced the IGBT that was accidentally damaged, and the original fault pattern has returned. I had some time to experiment and observed the following behavior:

The treadmill accelerates in short, kick-like bursts of about 1 second, repeating roughly every 3 seconds (well above the set speed), then slows down to around the target speed or slightly below. This cycle repeats continuously. At very low speeds, e.g., 1 km/h, I can manually move the treadmill-belt after it has accelerated. If the speed is not allowed to drop below 1 km/h, the impulse-like acceleration does not occur.

From the microcontroller side, a 15 kHz PWM signal can be observed going to the IGBT driver IC. During acceleration, the duty cycle increases and then gradually decreases almost to zero. This indicates that the microcontroller is controlling this continuous acceleration cycle. The faulty pin of driver IC is alway on high (no fault).

I measured the 0.01 Ω current sense resistor (used by the microcontroller) and it is within spec. I also recorded the voltage across it with an oscilloscope. The signal is very noisy, but it clearly rises during acceleration.

I am not sure how much this noise actually affects the system. As a test, I plan to place a 100 nF capacitor across the motor brushes.

I have already disassembled the motor. Both the commutator and brushes appear to be in good condition. Adjacent coil resistances on the commutator are around 0.9 Ω, and continuity to the frame is intact.

Here are two videos showing the current sense resistor voltage and the 15 kHz PWM signal:
https://drive.google.com/file/d/1XixyK4PtUtpaxO33OqYIgEuRQAg9irX2/view?usp=sharing
https://drive.google.com/file/d/1IPYMAfDQW4UWLNrDRlN7PRlSP8xEMCVc/view?usp=sharing

If anyone has insights or suggestions, I’d appreciate your input.

 

You need to bench test the board using a 20Hz PWM input via one of the cheap variable units available.