Proform Treadmill runs with Tach disconnected!

Thread Starter

sdtrent

Joined Dec 4, 2011
7
Hello,

I have trolled and read everything I can find (big thank you to Rom Rollins for the hand drawn MC-60 schematic!) but I cannot figure out what is going on with a Proform J8 treadmill. The belt stutters when you walking. I started down the usual path but it turns out the motor runs (drive belt disconnected) in some kind of open loop but still appears to control the speed somewhat. I feel like I must be missing something obvious.

Proform J8, PB-121 Power Supply Board, MC-60 Motor Control Board, Icon 2.5 HP 130 VDC motor (red, black, and two blue wires)

Observations:
1) with the drive belt off (so no signal from the reed sensor that the walking belt pulley is turning) the motor will still run without going to max speed.

2) the speed changes with the speed changes via the control panel, up and down. It stops driving the SCR's when reducing the speed.

3) I can stall the motor by pressing on the flywheel with my shoe. The current limit comes on at a steady 14amps.

4) the system appears to ignore the signal from the reed sensor.

5) voltage on motor seems to track with speed setting, around 12V at 1mph, 24V at 2mph, 36V at 3mph.

Ok, running in open-loop with no speed feedback would explain the jerky behavior when walking. But this raises two questions.

1) where in the heck is it getting the speed signal to know when to turn the SCR's on and off? It is not just firing at a fixed angle verses speed setting because it will coast with no SCR firing until the speed drops when you lower the speed.

2) why does the system not drop into an error condition after X amount of time with no signal from the tach. (I am not going to change any firmware on the interface panel so this is more of a philosophical question)

I cleaned the brushes and the contact area just for good measure with no impact. Any suggestions on closing the control loop short of replacing the interface panel?

Thank you,
-Shane
 

SgtWookie

Joined Jul 17, 2007
22,230
I've actually been putting together an LTSpice simulation of the MC-60 (in my spare time)...

I've had to make some guesses here and there, as Trolley1/Tom Rollins did not include values or reference designators for a number of components.
Link to thread where Tom's schematic resides in reply #2:
http://forum.allaboutcircuits.com/showthread.php?t=18895

However, it's starting to work... have a look. Perhaps you can explain some of the inputs that I don't know anything more about than what I copied from Tom Rollins' schematic.

For example, the 6/8 MPH TR - what does he mean by that? There must be some kind of sensor that I don't know about.

RPSI doesn't make any sense either. Maybe it's RPS 1 instead. There is an RPS 2 in the lower right corner.

Anyway, it's somewhat involved, and I've had to find or make PSPICE models for a number of the components - I still don't have some of them, but at least it's doing something.

Perhaps you can help fill in some of the blanks/questionmarks?

Don't expect this to be a "quickie" project. Getting this LTSpice model constructed should help a lot of people; well over 1,200 people have downloaded Tom's schematic, and it would be helpful for people to see what waveforms at various areas are supposed to look like.
 

Attachments

Thread Starter

sdtrent

Joined Dec 4, 2011
7
SgtWookie,

Noble effort on the LTSpice model. I believe the 6/8 MPH is a trim resistor to set the board to a maximum speed of either 6 or 8 mph. The designator looks more like 6/8 MPH IR on mine board (photo attached).

It appears the MC-60 is used in many different products and can be customized by trimming resistors to modify the behavior for a specific product/model.

I believe I follow a good portion of the schematic but I am still working on my understanding of the fuctions of U1-2 and U1-3. I am also not sure what U2-1 is doing and what signal (W) is feeding R1 and R18.

I hope this helps some.

-Shane
 

Attachments

jimkeith

Joined Oct 26, 2011
540
This is what I believe is happening here--most drive controls have a feature called "armature backup" that enables the control to switch to armature feedback in the event of tach failure. There are two ways to implement this--one is to use a simple diode OR circuit that takes the higher analog signal--the other is essentially full-time armature feedback with 10 to 20% tach trim.

When the tach signal fails, the motor generally increases speed by about 15 to 20% or so--and can run lumpy under some conditions.

I looked at the schem, and did determine that it does have armature feedback (A+ tied to R29), but the circuit is so convoluted that it defies logical analysis--cannot seem to find where the tach ties in.
 

SgtWookie

Joined Jul 17, 2007
22,230
Noble effort on the LTSpice model.
I'd like to get it finished and move on, but I keep running into snags here and there. Just today I got the simulated MOC3052 more or less working, and SCR's Q2/Q3 firing. I still haven't found decent models for the D4020L diodes, so I just threw in something that had the current rating.

I believe the 6/8 MPH is a trim resistor to set the board to a maximum speed of either 6 or 8 mph. The designator looks more like 6/8 MPH IR on mine board (photo attached).
Interesting. Wonder why yours says IR?

It appears the MC-60 is used in many different products and can be customized by trimming resistors to modify the behavior for a specific product/model.
That's what I have been thinking, too. Would be nice to know at least one of the models' values...

I believe I follow a good portion of the schematic but I am still working on my understanding of the fuctions of U1-2 and U1-3.
U1-2 is U1b in my simulation. A- is the low side of the motor & 5.5mH inductor; it's current sensing.
U1-3 is U1c in my simulation. At one point earlier this evening, I was getting a triangle ramp from around U1c's output; but at the moment things are on the fritz; I think it's time I re-booted my computer. I've had a LOT of windows open, and probably have some memory leaks.

I am also not sure what U2-1 is doing and what signal (W) is feeding R1 and R18.
Signal W comes from VR1; it's the speed control pot. It's a value between 0v and 12v.

I'm not quite sure what U2-1 is doing at the moment.

I seem to be having some troubles with the LM324 models that I am using. I was in the process of switching from ST Microelectronics' models to National Semiconductors', but the problems just got worse.
 

Thread Starter

sdtrent

Joined Dec 4, 2011
7
jimkeith,

I had just figured out the armature feedback myself. I went out and gave the flywheel a spin with everything off and the voltage seemed about right. So the thing is controlling to a specific voltage across the brushes and triggering the SCR's when that voltage get's too low. This explains the coasting down with no SCR firing when the speed set-point is reduced.

Now to figure out how to tie the speed feedback from the tach back into the circuit. I may just try to piggy back a micro-controller onto the thing that will read the speed set-point and tack and jack up the duty-cycle when the tach lags the set-point. I am sure there will be some lag and gain issues but I do not follow the circuit well enough to had any idea how work the tach back in into the system in the analog domain. But I do feel like I follow the circuit a little more each time I look at it.

-Shane
 

Thread Starter

sdtrent

Joined Dec 4, 2011
7
SgtWookie,

My system does not have a speed pot and the SPD CNTL signal on the LED flickers. I guess this is a pwm signal from the control panel that get's smoothed out by the 22uF on the input of U1-4 and turned into a DC voltage proportional to the set-point. Clearly it is time to break out the scope and do some probing.

Thanks guys,
-Shane
 

SgtWookie

Joined Jul 17, 2007
22,230
Say, I just noticed that your resistor RPS 1 is missing! Bottom left corner of the board; there WAS a resistor there at some point, but it looks like it was broken off.

It's 5.4k on Tom's schematic.
 

Thread Starter

sdtrent

Joined Dec 4, 2011
7
SgtWookie,

I found the document below that provides more information on the clipped resistors.

<<Configuration

The controller is configured at the factory to operate at different speeds and with different consoles. Four resistors are used to configure the controller. They are labeled RPS1, RPS2, RPS3, and 6-8 MPH. The function of each resistor is described below:

RPS1- Clipping this resistor causes the controller to operate in the same manner as the older 101757 and 108580 controllers. This allows the controller to operate with older opto-isolators on some power boards.

RPS2- Clipping this resistor lowers the Current Limit (CL) setting of the controller by approximately 2 amps.

RPS3- Clipped only for use with older treadmill consoles programmed specifically to work with the MC-30. Clipping this resistor allows the speed to be increased without first having to press another button on the console. This resistor must not be clipped under any other circumstances.

6-8 MPH- Clipping this resistor changes the IR Compensation from a 15 KV motor constant to a 17.6 KV motor constant.
This allows the use of motors of different specifications.>>

http://www.google.com/url?sa=t&rct=...OsCKcyJdA&sig2=AWf9ZzAGlhtHqkqLSY9v5A&cad=rja

-Shane
 

Thread Starter

sdtrent

Joined Dec 4, 2011
7
SgtWookie,

Can you tell me if the max speed is when the speed control pot is set to 0V?

And I would love to see what the simulation says is happening at the collector of Q7. This might be enought reason for me to try my hand at LTSpice again.

Thanks,
Shane
 

SgtWookie

Joined Jul 17, 2007
22,230
It appears that more current flows through the motor when the speed control pot output is low, but greater than zero. If the wiper of the speed control pot is shorted to ground, the current through the motor is only a couple of mA.

There seem to be a number of shortcomings with the simulation; I haven't worked them all out yet. There are a number of components that I don't have values for - or even the correct referenced designators; I've had to simply make random guesses for a number of parts, which I don't like to do.

I've only been running the simulation for 1 second at a time; it takes several minutes to complete, even using a couple of CPU's at 2.5GHz.
 
Top