Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Treadmill motor / voltage, controller board issue

Status
Not open for further replies.

animateme

New Member
Hi,
I am having some troubles with my Nordic Track EXP1000 treadmill and hope that someone here can help me determine the probable issues. The problem started when the treadmill would stop running in the middle of a workout session, about after 25 minutes of operating. Only the motor would stop running but the console would stay on. I noticed that the motor and the transformer were extremely hot and there was even a light burning smell. After about 5 minutes of cooling off, the treadmill could be again restarted. I visually checked the motor while the treadmill was running and noticed sparks coming from the inside. I also checked the voltage, which is supposed to be at 95V max, and the reading I got on the voltmeter fluctuated rapidly between 10 to 100V at maximum speed.

I assumed that the brushes were probably worn and were causing these sparks and maybe also the voltage fluctuation. I proceeded to replace the brushes and took the motor to a service shop where they resurfaced the commutator with a lathe.

After installing the serviced motor, the sparks are still there and I now find that when someone is walking/running on the treadmill the CUR LIM light comes on constantly and the motor begins to speed-up and slow-down irregularly, making a noise as if it was being forced. Also, when there is load on the motor from a person on the walking belt the sparks are much greater. I checked the voltage again and it is the same, fluctuating rapidly between 10 to 100V, getting different readings in between.

I checked the AC voltage going into the control board and it is at 110V, constant. The issue is with the voltage feeding the motor. For example, readings are: 40, 10, 80, 100, 20, 10 V, all these within 3 to 4 seconds.

I am positive that the issue is not the walking belt as this was replaced a few months ago and the treadmill had been running fine after that.

Could the issue be on the motor? Could it be the motor controller board? Is it normal to have these fluctuating voltage readings? Any suggestion on what else I could test or try?

Thank you very much in advance for any help provided.
 
I'm not sure about the fluctuating voltage readings (it may be completely normal, probably due to flyback from the coils in the motor - the commutator sparking lightly is also likely normal); was it a digital multimeter? Try testing with an analog meter as well. You might also want to check the voltage with the motor out of the loop, to make sure that is constant (hmm - you may need to provide a load, depending on the controller - if the controller won't run without a load, a bank of lightbulbs could serve as a resistive load). When they serviced your motor, did they check the bearings (worn, alignment, etc)? What about the bearings for the belt rollers?
 
With fair certainty, the voltage reading fluctuations are the PWM waveform used as speed control. I'd put an o'scope on it and vary the speed and load to see what's up.

This sure sounds like thermal shutdown. Is there a fan on the heatsink assembly? Does it spin freely? Can you unplug it and feed it from a separate power supply? I've seen thermally controlled fans on treadmills. Good Hunting... <<<)))
 
Thank you for your replies. Yes it was a digital multimeter and unfortunately I don't have an analog one or an oscilloscope. I tried measuring the voltage with the motor out of the loop and I got similar fluctuating readings. The bearings all seem to be fine.

The shutdown is in fact thermal. There is a thermal switch inside the motor that is connected to the control board. The motor itself has a fan but it never overheated in the past.

Question - if the motor was ran at the service shop at 110V AC, when in fact it is a 95V DC motor, could they have damaged it? I am not sure if this is what actually happened but I would like to consider this as something that might have occurred.

Thanks.
 
That it runs at all doesn't make me suspect motor damage firstly. But it depends on how it runs. Not that it Couldn't be the issue, but the motor shop should know how to check the windings and test it.

That it's thermal: is the motor thermal switch causing the shutdown? If so, how does it run with low/no load? How does the motor sound? Any kind of stuttering or stacato pulsing indicates winding troubles, most noticeable at low RPM and at load.

Another consideration: is there a rotation/speed sensor on the motor? They're usually an optocoupler with a disc or tab on the shaft. Make sure its' clean, at very low RPM you can see the pulse on a meter. Erratic sensor behavior could explain all of this. G.H... <<<)))
 
Last edited:
Thank you again for your reply. I asked about the possible damage to the motor by running it at 110V AC instead of the 95V DC because I got the sense that they don't have a way to get the 95V DC, from a comment they made. They actually told me that it would not hurt to test the motor at the 110V AC, but I am not sure if this is right or not. So, could they have damaged it if they did run it at the wrong voltage and current type?

I am almost positive that it is the thermal switch causing the shutdown as I am able to start it again after it cools of for a few minutes. With no load, the motor sounds fine, it is when it gets the load of the weight of someone on the treadmill walking belt that it starts to act and sound irregular. Can you please explain "winding troubles"?

There is a rotation/speed sensor on the motor. It's similar to those for bicycles, like a magnet that senses every rotation. This seems to be clean. Still, I am not sure if the electronics behind this sensor could be working properly or not. Any way to check this?

Again, the issue is that the motor sparks while running, and the sparks become quite heavy when the motor has a load. Also, the speed fluctuates constantly with the load and after a few minutes, it overheats and the thermal switch shuts it down.

Thank you.
 
You wrote:

I assumed that the brushes were probably worn and were causing these sparks and maybe also the voltage fluctuation. I proceeded to replace the brushes and took the motor to a service shop where they resurfaced the commutator with a lathe.

Hmm - extra sparking could be caused by a resurface commutator and new brushes (they were brand new, right?), as the brushes would take some time to wear in. Afterward, light sparking would be considered normal on a brushed motor, slightly heavier sparking with load.

After installing the serviced motor, the sparks are still there and I now find that when someone is walking/running on the treadmill the CUR LIM light comes on constantly and the motor begins to speed-up and slow-down irregularly, making a noise as if it was being forced.

Can you describe the noise? I am still wondering about the bearings; when the motor is running without being connected to the belt, how does it sound? Do you hear any clicking or grinding noises? "Winding troubles" could be many things, but one could be a loose or shifting winding on the rotor, which could cause issues. Do you have any way to remove the motor, mount it on a bench or stand of some sort, and apply a makeshift "pony brake" to the shaft to simulate a load?

Can you remove the motor and grab the shaft (or pulley/flywheel, as the case may be) and try to shift it from side to side (mounted on a bench or other sturdy and fixed surface is best) laterally, and fore/aft axially, checking for play in the bearings. What about when you spin it by hand (or run it from a high-current 12 VDC source, like a car battery, IF it is a DC motor - it will run, just slower than normal) - do you hear any abnormal noises.

There isn't a lot a play between the motor's rotor and its magnets (is it a permanent magnet motor, or universal? I own a treadmill motor I picked up at a Goodwill store - it has permanent magnets - but I suppose a universal motor could be used - heck, even an AC motor is possible) in the case; if it has a bad bearing (or a loose winding) that throws it off balance under load, the rotor can collide with the magnets or field coils.

When you took it to the motor repair shop, did you describe the symptoms, or did you just tell them the commutator was sparking? I'm just really surprised if they didn't check the bearings, give it a load test, check the windings, and check the rotor/field for scrapes, etc that might indicate an issue...
 
Thanks again for your response.
The brushes were new but I ran the motor for about 3 hours so I assume that these would have settled by now, right? I took them out and they now have the commutator's curved shape.

This CUR LIM light is supposed to mean that the motor is drawing the amp limit from the control board. The noise is like if the motor was struggling to operate, electrical noise maybe? It is not a grinding noise or any type of friction noise nor a clicking noise. I can simulate the load, as you are asking, by just removing the belt that drives the treadmill's walking belt and operating the motor at the same spot where it is mounted. What would I be looking for with this test?

As for the bearings, if I just spin the shaft with the motor unmounted it spins freely. I would think that these are fine. I tried shifting the shaft side to side and there isn't any play. I don't have any car battery handy to try running the motor but even if I did, I would assume that it would run fine as it does with the treadmill's control board. The issue is mainly when there is load on the treadmill, that's where the sparks, the fluctuating speed and the struggle noise occur.

I did describe the symptoms to the repair shop individual but for some reason I get the sense that they are not the most competent people around, but they are the only source that I know around my area for this. Should I take it in again ask them to look for these specific issues? How would they test the winding?

Thanks again.
 
Thanks again for your response.
The brushes were new but I ran the motor for about 3 hours so I assume that these would have settled by now, right? I took them out and they now have the commutator's curved shape.

This CUR LIM light is supposed to mean that the motor is drawing the amp limit from the control board. The noise is like if the motor was struggling to operate, electrical noise maybe? It is not a grinding noise or any type of friction noise nor a clicking noise. I can simulate the load, as you are asking, by just removing the belt that drives the treadmill's walking belt and operating the motor at the same spot where it is mounted. What would I be looking for with this test?

Well - you'd be seeing if you can see or feel what the motor is doing when it makes the noise under load (which I assume - I might be wrong - would be better than with the belt connected and/or a person on the belt, since there'd be less noise from those sources).

As for the bearings, if I just spin the shaft with the motor unmounted it spins freely. I would think that these are fine. I tried shifting the shaft side to side and there isn't any play. I don't have any car battery handy to try running the motor but even if I did, I would assume that it would run fine as it does with the treadmill's control board. The issue is mainly when there is load on the treadmill, that's where the sparks, the fluctuating speed and the struggle noise occur.

Given what you tried, I'm wondering if maybe (for some reason) there is a short between adjacent coils off the commutator (this could be another "winding issue"); I'm not sure if this would cause the issues you are seeing though, but it might.

I did describe the symptoms to the repair shop individual but for some reason I get the sense that they are not the most competent people around, but they are the only source that I know around my area for this. Should I take it in again ask them to look for these specific issues? How would they test the winding?

Well - visual inspection, followed by a continuity check between the communtator's contacts. Something else I am wondering is if - maybe (long shot here) the brushes aren't the right size for the motor; if, for instance, the brushes installed are wider than what is required for that motor, the brushes might be contacting more windings than necessary, increasing the current draw of the device (and perhaps causing the "struggle" noise you are hearing as the extra coils on the rotor are energized at the wrong part of the rotation cycle...?

I'm not sure what you can do regarding the repair shop you are using; you sound like your kinda stuck with what you have. If you had a spare set of brushes, and knew how to change them (and how to disassemble/reassemble the motor - its not normally difficult, but each motor is different - I've got this motor that is really difficult to put back together because the bolts that hold the bearing plates to the stator casing are made of steel, and they pass right between the poles of a large pair of permanent magnets; keeping the bolt from sticking to one magnet or the other while passing the bolt thru to the other end can be an exercise in frustration and patience, let me tell you) - you could install the extra set of brushes, run the motor in, then you could remove the rotor, and match the new brushes up to the commutator, to see how many pole contacts each cover (they should be the same). Then, you could use an emery board to file bevels on the corners of each brush to remove contact with two of the contacts, taking out two poles, then reassemble the motor, and see if it runs better (because fewer poles would be involved, taking less current, and maybe fixing the struggle issue).

That's all a real long shot, though - based on a guess that the motor wore out its brushes down to the nub previously, and the wrong sized set was put in afterward...
 
I'd also look for any cross conducting in the commutator. After lathe trueing, you Must clean up the spaces between commutator contacts. A piece of a jewelers' saw blade works well. Smoothing/polishing after that is always good: 600 grit paper then 1000.

Set the magnetic sensor gap as close, but definitely clearing everything. You can see the sensor respond at (very) low speed on a meter. It depends on test equ available.
Good Hunting, this one isn't obvious.
 
I would almost bet on one or more open windings on the rotor. Testing is pretty simple -- with the motor out of circuit and the brushes removed, simply use an ohmmeter to check for continuity from any one commutator segment to the segment 180° around the commutator. Each measurement should show approximately the same resistance as any other; an infinite reading indicates an open winding. Make sure that there is no continuity between any given segment and the segments directly adjacent thereto.

While testing for continuity, be sure to examine the commutator for proper undercutting of the mica segment separators as was already mentioned. Proper undercut will have a depth that is equal to its width, i. e., the space between commutator segments. Ideally, the trailing edge of each segment should overhang the undercut slightly, which will help to reduce arcing (sparking). Bear in mind that depending upon how much material was removed from the commutator surface, your brush spring tension will be affected...
 
Treadmill trouble

Some info that might help. Did you take the treadmill through calibration? Did it pass? The manufacturer knows how to tell instruct you the through calibration process? Test your resistance for the motor with it unplugged. The motor should have connections to do this. The manufacturer will have the specs for that motor and its resistance.

Speculation: you need a new motor, your speed sensor needs to be replaced, or your motor control board is not communicating with the motor and its a matter of replacing some cables. There also appears to be a power issue. Its unlikely you have a dedicated power source going to your treadmill. Treadmills generally don't like to share power. They are power hogs. This could be causing the fluctuations in the voltage to the motor. It appears that you are towards needing a new motor.
 
Something else I am wondering is if - maybe (long shot here) the brushes aren't the right size for the motor; if, for instance, the brushes installed are wider than what is required for that motor, the brushes might be contacting more windings than necessary, increasing the current draw of the device (and perhaps causing the "struggle" noise you are hearing as the extra coils on the rotor are energized at the wrong part of the rotation cycle...?

Thanks, actually, the brushes are the right size as they have to fit into a special sized compartment in order to install them. The brushes I took out were not completely worn out but I just decided to change them assuming that maybe they were no longer making good contact and that was causing the sparks.
 
I'd also look for any cross conducting in the commutator. After lathe trueing, you Must clean up the spaces between commutator contacts. A piece of a jewelers' saw blade works well. Smoothing/polishing after that is always good: 600 grit paper then 1000.

Thanks. The repair shop did clean the spaces between commutator contacts. They did tell me that this needed to be done after resurfacing with the lathe.
 
I would almost bet on one or more open windings on the rotor. Testing is pretty simple -- with the motor out of circuit and the brushes removed, simply use an ohmmeter to check for continuity from any one commutator segment to the segment 180° around the commutator. Each measurement should show approximately the same resistance as any other; an infinite reading indicates an open winding. Make sure that there is no continuity between any given segment and the segments directly adjacent thereto.

Thanks, this seems to be simple to test. I will give this a try.

Ideally, the trailing edge of each segment should overhang the undercut slightly, which will help to reduce arcing (sparking).

Can you please explain this a little further? I am not sure I get the right idea.
Thank you.
 
Some info that might help. Did you take the treadmill through calibration? Did it pass?

Actually, I just found out about this calibration procedure but have not tried it yet. I will give it a try.

Speculation: you need a new motor, your speed sensor needs to be replaced, or your motor control board is not communicating with the motor and its a matter of replacing some cables. There also appears to be a power issue. Its unlikely you have a dedicated power source going to your treadmill. Treadmills generally don't like to share power. They are power hogs. This could be causing the fluctuations in the voltage to the motor. It appears that you are towards needing a new motor.

The cables seem to be fine and the power is not being shared. This treadmill has been on the same power outlet for years now and it used to work fine up until a few weeks ago. Therefore, I rule out the power issue. That would leave the motor and the motor control board, which is what I am trying to determine which one of the two is the issue, or maybe even both. I would hate to replace one of these to later find out that it was actually the other component that needed replacing! At what point would the motor be considered unrepairable? Thanks.
 
In a typical motor of the type that you have, each winding or "loop" of wire on the rotor is terminated at each end by its connection to a copper bar, which is formed to comprise a "segment" of the commutator. Normally, the opposite ends of any specific winding are physically located 180° apart on the commutator. Each segment is separated and insulated from those on either side by a thin strip of insulating material, usually a piece of mica.

The mica is harder than the copper, so as the commutator wears, the mica (if not undercut) will end up being higher than the copper. This will cause the brushes to lift off the copper surface each time they pass over the mica. Of course, this creates a gap between the brush and the copper, resulting in heavy arcing which in turn causes pitting of the copper surface. The result of that is accelerated brush wear. Also, in an extreme case, mica that is significantly higher than the copper can cause breakage of the brushes.

If the mica is undercut too deeply, copper and brush dust will accumulate in the cut, which can eventually create a short between adjacent segments. A cut of the proper depth will generally allow the collected dust to be thrown out by centrifugal force as the rotor turns.

If the mica undercut is not deep enough, the copper segments will likely wear to the depth of the mica (and probably beyond) before the normal brush life has elapsed. Of course, this will then necessitate additional commutator service.

If the undercut is too narrow, a thin line of mica will be left on one or both sides of the cut. As already discussed, this has the same effect as not having undercut the mica at all, as the mica will end up higher than the copper with commutator wear. Another possibility here is breakage of the mica, with the possibility of the broken-off piece becoming lodged between the brush and the commutator. Typically, cut width is specified to be 0.003" wider than the thickness of the mica separator. Properly centered, this will result in a small amount (0.0015") of copper being removed from each side as the cut is made.

If the cut is too wide, heavy arcing will usually occur. Depending upon specific commutator and brush configurations, a wide cut can also cause brush "cocking" and breakage.

To address the specific point about which you asked -- that is, the overhanging trailing edge -- some motor manufacturers specify a "keystone" shaped undercut, wherein the physical gap at the commutator surface is slightly narrower than the width at the bottom of the cut. This serves two purposes. Firstly, it helps to ensure that there is no mica remaining that will cause brush lift as already described. Secondly, it helps to prevent brush cocking and breakage by providing a smoother transition from one segment to the next as the rotor turns.

The trailing edge is the edge of the commutator that the brush is leaving as it moves onto the next segment. In a bi-directional motor, either edge of each segment will be the trailing edge, depending upon which direction the motor is turning. In this case, a "true" keystone cut is recommended. In the case of a unidirectional motor such as a automotive starter motor, a cutter is used that makes a "modified" keystone cut, with only the trailing edge being angled, and the leading edge is left square.

Cuts of this type are somewhat difficult to make with a saw blade as has been suggested. Several tool makers produce special tools for the cutting of the mica in this manner. If using a saw blade, don't worry about the overhang that a keystone cut would provide -- just be sure to remove all mica on both sides of the cut to the full depth of the cut that you are making. As I mentioned in my previous post, typical cut depth is equivalent to the mica width. Also, it is better to use the saw blade "backwards", i.e., with the teeth pointing towards you so that the blade cuts as you pull it towards you. Do not put force on the blade -- the weight of the blade is usually sufficient to cut the mica. While a jeweler's saw will work, satisfactory results for most larger motors can be obtained with a standard fine-tooth (24 to 32 teeth per inch) hacksaw blade. snap the rounded end off the blade to allow it to cut properly at the machined shoulder of the commutator.

I hope that this helps to clarify the subject; post back if you need further explanation...
 
use an ohmmeter to check for continuity from any one commutator segment to the segment 180° around the commutator. Each measurement should show approximately the same resistance as any other; an infinite reading indicates an open winding. Make sure that there is no continuity between any given segment and the segments directly adjacent thereto.

While testing for continuity, be sure to examine the commutator for proper undercutting of the mica segment separators as was already mentioned. Proper undercut will have a depth that is equal to its width,...

I checked for continuities and I do get just about the same resistance readings between each set of commutator segments separated at 180°, with no infinite readings at all. However, I do get continuity between each and all adjacent segments. This is no good, right? Does this mean that the segments are shorted? Is there an easy way to fix this?

I checked for the mica segment separators, as you suggested, but I don't seem to find any... it looks like there are only gaps between the segments but I don't see any material filling the gaps (see pictures). Is this normal or are the mica separators missing? Also, I notice that the depth of these gaps seem to be deeper than the width (again, see pictures), I am assuming that this should be normal for this motor as this is how it has always been. If any, the depth has been decreased when the commutator was rectified with the lathe.

Once, again, any comments will be greatly appreciated.

**broken link removed**

**broken link removed**

**broken link removed**

**broken link removed**
 
Sure looks like Bad arcing. I think I see some surface copper melt. I'd clean it up with a little 600 then 1000 grit paper. Blow the dust out afterward.

I wonder if this isn't insufficient brush pressure. Badly worn brushes put the springs at their lowest pressure. Light pressure can cause extreme heat, distempering the springs. I've seen springs turned into soft wire. I can't describe how to check this other than just replace them. <<<)))
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top