Maker Pro
Maker Pro

How does this circuit work?

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
Hi!

I'm trying to repair an old treadmill that is not working. It's an Aeorobics Inc Pacemaster 870X. After a period of not using the treadmill (summer time), it stopped working. Currently, when the start time and start speed are inputted and then the start button is pressed, there is a short delay and then a message shows on the display saying there's a dead sensor and the tread does not move (I'm assuming the problem is not the sensor since there is no movement of the tread to sense). I have troubleshooted it extensively and have replaced a few parts ($50 worth!) to no avail. Rather than go into all the details of that, can anyone help me understand how the circuit works that powers the motor? I've drawn the relative part of the circuit and have attached it.

When the user inputs the speed and distance and presses start, the relay closes and provides power to the Crydom L512F SCR diode. When power is supplied to the gates, a DC voltage is applied to the motor. The gate signal seemingly is a 120VAC signal that goes through an optical isolator. The isolator is turned on when a VN 2222 transitor is turned on with a signal from the microprocessor.

What I don't understand is:

1. What voltages at the Crydom gates would be expected? DC or AC. And is that appropriate? I don't think I've traced the PC board incorrectly. The specs for the Crydom are here: http://www.crydom.com/en/Products/Catalog/l_3.pdf

2. How does the speed of the motor vary? Does the microprocessor turn the transistor on and off at different frequencies such that the effective voltage to the motor therefore can vary?

I have found that if I ground pin # 2 of the opto isolator, the motor and tread works, but quickly shuts off and a different message is displayed that would indicate that the speed of the motor is not correct (not surprising). So I'm afraid that the signal from the microprocessor is not correct, but I don't have an O scope to look at it. But could the microprocessor be bad since many of the other functions work, eg the display, the incline angle, the diagnostic messages, etc.

Any suggestions to further troublshoot? I've replaced the Crydom module and the opto isolator.

Thanks in advance for any help!


EDIT: Minor change to schematic
 

Attachments

  • Treadmill rev3_schem.jpg
    Treadmill rev3_schem.jpg
    58.2 KB · Views: 2,448
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I wouldn't assume that the microcontroller is trying to start the motor and then discovers the sensor problem. I suspect it is refusing to start the motor because it has already detected that there is a problem with a sensor. This is consistent with the fact that the motor will start and run if you bypass the microcontroller's control signal.

If you want to be sure, connect an LED with a series resistor (e.g. 470 ohms) from the microcontroller's output to ground (i.e. gate to source of the VN2222) and try to start it. If the microcontroller is trying to start the motor and then giving up, the LED should blink.

You've already confirmed that the optoisolator and the Crydom unit (and the motor) are working because when you bypass the VN2222 it starts up. Presumably then the micro detects movement of the tread and turns off the relay as a safety measure.

It's possible the microcontroller is faulty or damaged, or there's a break between its output and the VN2222, but there's nothing to indicate this. My guess is there's another sensor that the microcontroller checks before it starts the motor, and that sensor is faulty.
 

Jamie7

Sep 28, 2012
29
Joined
Sep 28, 2012
Messages
29
DougB,

The micro senses the AC Voltage and turns the opto on and off at different points in the cycle of the AC which will vary how much voltage is present to the motor through the Crydom unit. This is how the speed of the motor is determined. For example when the AC Voltage is at 80% there is around 96 Ac volts applied to the motor. If the speed is decreased then for example the AC Voltage is at 50% then only 60 Vac or there abouts is present at the motor. (example only ACV not linear)

Since the micro must turn on the opto at some point in thne AC to get a feedback from the motor moving through an encoder or similar, I would suspect that since you get the opto to turn on, that the VN2222 N-Channel Enhancement-Mode Vertical DMOS FET is likely bad. If you don't have the means to check the signal to the transistors base, then just replace it.

The reason the motor comes on for a second when you do the test of shorting the opto to make it come on, and then the relay disengages is due to the fact that the
micro knows when it is calling for the motor to run. When you bypass the FET
the micro sees the motor start to run by the encoder feed back and shuts the relay off due to the fact it itself was not calling for it to run by activating the VN2222 FET. So this is howI figure it is probablythe transistor and not the micro.
In a rare instance the output pin of the micro couldbe bad but probably not in this case.

So, let me know what you find out...

Best regards and good luck, Jamie7 SORRY ABOUT THE MIS-STATEMENT ON THE FET.
 
Last edited:

Jamie7

Sep 28, 2012
29
Joined
Sep 28, 2012
Messages
29
DougB,

Does this machine have an emergency pull string stop sw circuit?
This would attach to your person and then plug in and activate a switch
or sensor to let the micro know the safety switch is activated and be deactivated
should you fall or fail to keep up the pace and it pulls out deactivating the machine.
Perhaps this "sensor" or switch is bad.

Just a thought... Good luck, Jamie7
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
KrisBlueNZ:

connect an LED with a series resistor (e.g. 470 ohms) from the microcontroller's output to ground
Good suggestion! I'll do it and let you know what happened. It may be a few days before I can, however, so sit tight!

Jamie7:

Does this machine have an emergency pull string stop sw circuit?

Yes it does and it seems to work OK. I attach the pull string, the display lights up as usual, then I input the speed and distance, then hit start. The treadmill then makes a beep as usual, signifying that it is starting. I then hear the relay click, then a few seconds later it displays the error message.

The reason the motor comes on for a second when you do the test of shorting the opto to make it come on, and then the relay disengages is due to the fact that the
micro knows when it is calling for the motor to run. When you bypass the FET
the micro sees the motor start to run by the encoder feed back and shuts the relay off due to the fact it itself was not calling for it to run

But I don't ground it until I figure it should be calling for the motor, so I think it should be looking for the motor to be moving at that time (right after the relay clicks as above). But then, I think, it shuts done for a different reason, ie the speed isn't correct??? All that is conjecture though. Also, if I move the belt by hand at the appropriate time, it will not alarm as long as the belt is moving.

I'll be posting again when I have more info based on your ideas. Thanks to you both. By the way KrisBlueNZ - Gotta love New Zealand! I've been there two times and would love to go again. We even looked at how we might immigrate, but you guys didn't want us if we didn't have skills you need!

Thanks again,

Doug
 

Jamie7

Sep 28, 2012
29
Joined
Sep 28, 2012
Messages
29
DougB,

Thanks for the update and when you try what KrisBlueNZ has suggested, that
should tell you whether the VN2222 is a possible candidate or not.
If there is a signal there and the opto is working then it is just a matter of trace, wiring or the VN2222.
I hope this will be the case. What you did supports my thoughts that the VN2222 is bad if there is a signal at the gate.
I look forward to the results.

Well, have a nice evening...

Best regards, Jamie7
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
More Info

Well I finally got the parts and got time to check them out.

I replaced the VN2222 FET - no change in operation.

So, I placed a LED on the gate to the VN2222 and found no signal. I traced it back to the microprocessor and found that the output of the micro (Pin 16 = P26 output of timer 1) is ALWAYS high. Then it goes into an inverter, then to a resistor and diode and then to the VN2222. So, apparently, what's happening is that the micro is supposed to turn off pin16, which in turn makes the output of the inverter high, which then turns on the VN2222. When I apply 5VDC to the gate of the VN2222, the motor runs, but runs fast. So, quickly the treadmill turns off, alarms, and reports "race sensor" which probably means the motor is going too fast - not surprising since the signal is constant.

Does this seem to definitively say that the micro is bad. It seems hard to believe since everything else seems to work coming out of the micro. Is it possible that the program is corrupted and it is not the micro? It looks like I can get a new micro for $10, so I guess that's my next step?

Thanks for any input.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Did you read my previous post fully? I suspect the micro is deliberately not enabling the motor because it has already detected a sensor problem, because the sensor is faulty, or because it is detecting a genuine problem. This is consistent with the results of your tests so far.
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
Did you read my previous post fully? I suspect the micro is deliberately not enabling the motor because it has already detected a sensor problem, because the sensor is faulty, or because it is detecting a genuine problem. This is consistent with the results of your tests so far.

Sorry, I guess I didn't address that. If I manually move the tread after the relay has energized and the micro expects the tread to move, it will not alarm as long as I am moving the tread. Once I stop moving it, the alarm sounds. So it seems to me the micro is seeing the sensor when the tread is moving, but if the tread is not moving for some other reason, it thinks the sensor is dead because it is not getting a signal. Do you buy that?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
If I manually move the tread after the relay has energized and the micro expects the tread to move, [...]
I don't understand. You said the micro never enables the MOSFET, so I assume it would NEVER expect the tread to move, right?
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
I don't understand. You said the micro never enables the MOSFET, so I assume it would NEVER expect the tread to move, right?

Yeah, I guess I assume it thinks it told the tread to move, but that signal didn't get out causing it to move because the micro is damaged.....
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
My gut feeling is that it's not likely to be a single faulty output pin. I think it's more likely that the micro is detecting a problem before it even tries to enable the motor, or there is some other problem that's preventing it from generating a PWM signal on that pin. This is just based on probabilities, and it's possible that that one pin is faulty, sure.

Since that pin is probably a PWM output, to control the motor speed, it might be worth checking whether there are any clock frequencies provided into the microcontroller that would affect the PWM output. For example, there could be a mains synchronisation signal from the power supply that the micro uses to control the firing angle of the SCRs, and that signal might be missing. Also there could possibly be another clock at some higher frequency that the micro needs for its PWM generation.

Do you have the part number for the microcontroller?
Do you have a full schematic? Have a look for a mains-frequency signal coming from the power supply section that the microcontroller would use to synchronise the PWM signal going to the motor controller.
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
KrisBlueNZ:

Thanks for those thoughts. I just realized that replacing the micro means desoldering all 64 or so pins - that won't be fun... or possible? So I'll take a look at your suggestions. I'm not ready to give up on this yet! I don't have a schematic, but I've been tracing the PCB and will fill in the areas you suggest.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
If you replace the uP you will have an unprogrammed part. And that it guaranteed not to work.
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
Since that pin is probably a PWM output, to control the motor speed, it might be worth checking whether there are any clock frequencies provided into the microcontroller that would affect the PWM output. For example, there could be a mains synchronisation signal from the power supply that the micro uses to control the firing angle of the SCRs, and that signal might be missing. Also there could possibly be another clock at some higher frequency that the micro needs for its PWM generation.

I don't have the equipment to check this out (both physically and mentally!).

Do you have the part number for the microcontroller?

The micro is a Hitachi HD6303YP. The signal from the micro going to the inverter and then to the VN2222 comes off of pin 16 which, I believe, is P27 (TCLK) within the micro (I was mistaken earlier when I said it was P26 of TIMER1 because the datasheet is very hard to read). Does this add anything that would help me? Otherwise I'm at a dead end. Although there are many used treadmills available, I don't like giving up. I may be ready to though!!

An additional bit of info is that when I ground pin 16 (which is usually high) of the micro, the treadmill starts and stops the same way it does when I put 5V on the gate of the VN2222. So there doesn't seem to be any problem between the micro and the VN2222.

Thanks again for any help.

Doug
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Ah, the 6303. That was the first micro I worked with when I started my previous job back in 1988!

Could you post in-focus photos of both sides the board with good illumination on it? If possible illuminate it at a slight angle and photograph it straight so the photo has no reflection of the light source.

I understand your testing so far. It looks like there's no problem after the micro's output; when you force it low, the motor runs. We need to know why the micro isn't driving that pin low.

My guess is that there is a signal that the micro needs. Probably it will be a mains-frequency signal that comes from the power supply. This signal will allow the micro to work out when to turn that pin on and off, to make the motor run at the desired speed.

Don't give up! It could just be something small.
 

CocaCola

Apr 7, 2012
3,635
Joined
Apr 7, 2012
Messages
3,635
Don't give up! It could just be something small.

I know this was touched upon earlier but I would be looking for any and all possible fail safe sensors and making sure they are working as well as the signal from them is actually getting to the micro...
 

DougB

Oct 13, 2012
33
Joined
Oct 13, 2012
Messages
33
I know this was touched upon earlier but I would be looking for any and all possible fail safe sensors and making sure they are working as well as the signal from them is actually getting to the micro...


Thanks CocaCola. That was KrisBlueNZ's impression too, but the fact that I can manually move the tread at the correct time (when the relay kicks in) and delay the report of a bad sensor for as long as I'm moving the tread, seems to say that the sensors are working. Frankly, I only know of one sensor that is looking at the speed of the motor. I don't know of any others.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Doug, yes I think you're right about that. My first thought was that the micro was detecting a problem and refusing to start the motor, but since you can move the tread and the micro thinks everything's alright, that seems to imply that there's no other sensor problem.
 
Top