Very Noisy Motor

[ArthurH]

Hi, I hope someone can help. I have previously asked for assistance and been very pleased with the help and advice the members have provided. This time I recognise it's a bit cheeky of me to ask the question of this forum, as it isn't exactly workshop related - but it is a motor and / or bearing issue, driving shafts via pulleys and belts, so not too far from the mark .

My wife has a treadmill; she used to use it for 30-45mins at a time, two or three times a week. But then it started getting a lot noisier, and eventually, it was too much for her to bear. I suspected the bearings of the 180v 1hp motor, but I found another whole motor being sold second hand not far from us, so I bought and fitted that. Although not silent, it was much quieter. It lasted perhaps 6 months, then it too made the same loud noise. To make this easier to follow, I'll call the original motor 'A' and the replacement 'B'.

I bought new bearings from eBay for motor A - they were not too hard to fit. It seemed quite quiet for a few minutes, but then back to full volume.

I did some reading up online and persuaded myself that the bearings must be low quality, perhaps cheap Chinese bearings. The advice was to buy good quality - probably American made bearings. I did that, fitted them to motor B, swapped the motors over - made no difference - still very loud.

I'm now thinking it can't be the motor, it must be the treadmill belt rollers. It's not easy to run the motor disconnected from the belt - there is a sensor on the belt roller that if not generating a signal puts the treadmill into an error state and stops the motor.

I bought and fitted four new belt roller bearings (and lubricated the underside of the belt), but as I had suspected, it made no difference at all.

Spinning motor A by hand I thought I good detect some roughness, so I again researched bearing quality on-line. I looked at SFK, but the one that was generally recognised as best was the Japanese 'Toyo' bearings. I bought a pair of them (genuine hologramed items), fitted them to motor A and swapped the motors back. No difference at all.

I would like to run the motor on its own - but this is where I'm out of my depth. The motor is marked as 180v (I'm surprised the plate doesn't say AC or DC?) and I don't see how I can provide that directly to it.

Can anyone help?

Many thanks, Arthur

 

[MARK.B.]

Not a pro in any way but given that you have already had this plugged in and running, i think it will be AC and as it has not been damaged on your 240 supply before it should be ok to test on 240v.
As i said above,i'am not a lecky and someone in the know should be along soon to give pro advice.

 

[ArthurH]

Thanks Mark - I'm also hoping that it will be OK, but I'll just wait a little longer before doing so

As I re-read my post I realise I hadn't really described the noise itself. It's not the sort of noise that I would have naturally associated with worn bearings - it is a whine, but no rumble. As you would guess, the treadmill has a built-in slow start. Most interesting is that the noise is quite bad from the very moment the motor starts rotating, even when only doing just one or two RPM. It does increase in both volume and pitch as the revs go up, but not by as much as you might expect. In terms of volume, even with no one on the machine, you would need to shout to make yourself heard over the noise.

 

[Myfordman]

Not a pro in any way but given that you have already had this plugged in and running, i think it will be AC and as it has not been damaged on your 240 supply before it should be ok to test on 240v.
As i said above,i'am not a lecky and someone in the know should be along soon to give pro advice.

Absolutely not. it needs to be run on max 180 v and that is not readily available without special equipment.
The whine is likely to be the PWM speed controller and nothing to do with the motor.

 

[Terry - Somerset]

Google the motor model number - I think it may be DC so connecting directly to 240V AC probably not a good idea.

I think the treadmill will have a control board which does all the clever bits - providing DC, variable speed, safety cut outs etc.

I'm no expert - just playing with google.

 

[Fitzroy]

Take the belt off, get someone to turn the bed manually by walking on it to trick the sensor and then run the motor? If that’s possible you’ll be able to isolate the sound.

 

[Ttrees]

My wife has a treadmill; she used to use it for 30-45mins at a time, two or three times a week. But then it started getting a lot noisier, and eventually, it was too much for her to bear.

Can anyone help?
Many thanks, Arthur

Simples, buy her the ear muffs and tell her get back to work and get dem darn things sanded already!

 

[MARK.B.]

happy to be corrected

 

[ArthurH]

Thank you all for your suggestions - much appreciated

Terry, yes the treadmill does have a controller board - it is shown in the first photo just to the left of the motor.

Fitzroy, I'll give this a go a bit later today

Ttrees, sooner you than me

Myfordman, I would like to pursue your suggestion re the PWM (Not sure what that is...) In case it were to help, I have included more photos of the boards that are connected to the motor. Can you help me confirm that is (or isn't) where the noise is coming from?

Thank you all again

 

[TheUnicorn]

I wonder if it is a viable option to extend the leads between the control board and the motor, meaning you could move the motor away, preferably somewhere where you can muffle the sound. That way you could see if the sound is the motor or the board? The only issues I can see with this would be if the extension wires created a resistance that confused matters.

 

[guineafowl21]

Possibly repeating some of the advice, but you have three areas of origin for the sound:

Electrics
Motor
Treadmill belt/rollers

Pop the motor drivebelt off and drive the treadmill belt as fast as you can, either by walking on it or by hand. If no noise, it’s the motor or electrics.

By swapping the motor and getting the same noise, you have a suggestion it’s not the source of noise, unless the motors have a defect or ‘pattern’ failure. Does the person who sold you the spare motor also have the electrics?

The treadmill speed sensor is likely in one of the rollers - can you take the treadmill belt off easily, and just spin a roller to imitate drive? In which case, you can try each motor on its own.

Try the screwdriver stethoscope method - tape up the shaft to avoid shorting anything, then hold the handle to your ear as you probe each component. Don’t probe the green board where the mains comes in, with the two coils and three yellow capacitors - that’s just interference filtering. Focus on the motor, rollers and those two transformers.

It looks like a DC motor. If it’s PWM noise, the pitch will usually stay the same regardless of speed. Describe the noise exactly - how does it change with time, and with increasing speed? We need to know about pitch and intensity.

 

[ArthurH]

Again, thank you for your suggestions.

TheUnicorn - I was able to do a variant of what you had suggested, I couldn't move the motor away, but by unwinding some of the motor power leads from the ferrite ring and providing a new earth connection, I was able move the circuit board away - see photo:

This enabled me to prove there was no sound coming from the circuit board.

I also discover that if I cut the power to the treadmill when it was running, so that the belt and rollers were spinning the motor, the noise stopped. The belt and rollers made very little noise. This implies it is the motor under electrical load that is making the noise.

Guineafowl21 - I have tried the screwdriver 'stethoscope' but it didn't really help. The noise coming from the motor is definitely from the commutator end (interestingly the opposite end to where the belt is exerting its force onto the bearing), but it is the commutator end that has the main ventilation holes for the motor - so more sound is likely from them anyway.

I did record the sound of the motor on the attached sound file. I have put it in a zip file as mp3 is not an allowed format (?). This is from starting, up to a belt speed of 6mph - 75% of full speed. Ignore the gentle repeating noise from the treadmill belt itself, that is a little belt adjustment I need to make.

Thanks again

 

[MorrisWoodman12]

In my limited experience of lathe motor variable speed drives (I've fixed two. It's the remit of a retired electronics engineer) the motor is wired to the 'DC' side of a bridge rectified in series with a power FET(field effect transistor). The FET is turned on and off rapidly. The ON pulse width is varied to adjust the amount of power to the motor and hence the speed. This is a PWM or pulse width modulation drive. So the motor will be DC. As the PWM frequency will remain the same any audible whine from it should remain the same. It will probably be well above 10KHz so you'll be lucky to hear it anyway.

 

[ArthurH]

Guineafowl21 - did the attached mp3 file give you what you requested?

 

[guineafowl21]

Guineafowl21 - did the attached mp3 file give you what you requested?

Yes - I couldn’t play it initially, but this time got it to go.

There appears to be a reasonably constant pitch noise throughout, which gets swamped by a louder, rising pitch whirring which I assume is the bearings and general mechanical noise.

Without getting a scope on there, I can’t tell if it’s PWM frequency or not, but I guess it doesn’t matter as you’ve established the noise is coming from the motor, and have eliminated the bearings. Something is vibrating in response to the electrical drive.

It’s worth a check that the motor mounts, end-covers and magnets are securely fastened. While you’re at it, carefully examine the rotor for loose windings. I assume you’ve already checked the commutator and brushes for wear and sticking.

 

[ArthurH]

Thank you.

I thought it might be appropriate to dismantle one of the two motors and post some pictures. Nothing looks wrong to my (untrained) eyes. What I find most baffling, is that the noise, to all intents and purposes, is identical from both motors. Both motors initially worked fine (i.e., quietly) for years and or months, then eventually both started making the same loud noise.

The motor is bolted directly to the treadmill chassis (four large HT Allen bolts) there is no rubber bushing. The brushes on both motors look the same, healthy with plenty of remaining life, and slide out to reach the commutator with ease. The end covers have fool-proofing lugs on them - and given the number of times they have been on and off, it is unlikely I would misalign them every time.

 

[guineafowl21]

Both motors initially worked fine (i.e., quietly) for years and or months, then eventually both started making the same loud noise.

Yes, which is why I initially wondered if it was the motor at all, unless there’s a defect or common failure in both.

It could be that some components on the driver board have drifted in value or failed, altering the PWM waveform and causing some sort of resonance. Do you know of a tame EE locally who could scope the output, and maybe test the board?

If you like, I could post a link to this thread on the electrical engineering forum I’m on as well.

EDIT: The input wiring appears to be the wrong way around. Line goes straight to the filter board, while neutral is led through a thermal switch, the power switch, then some sort of reactor. Not the end of the world, I suppose, but a suggestion of poor quality or someone fiddling, perhaps.

 

[ArthurH]

Thanks yet again for your very quick response - I really appreciate it.

Since I've retired, I no longer have access to the sort of EEs that you describe - so I would be grateful if you could re-post this link.

Regarding the input wiring issue you describe, are there any benefits in me making changes?

 

[guineafowl21]

Thanks yet again for your very quick response - I really appreciate it.

Since I've retired, I no longer have access to the sort of EEs that you describe - so I would be grateful if you could re-post this link.

Regarding the input wiring issue you describe, are there any benefits in me making changes?

Thread posted - I’ll link it in if there are any replies.

Reversed L/N is a safety, rather than a running issue. Switching and fusing should be on the L side. I’d leave it for now until we make some progress with the fault.

 

[guineafowl21]

Some replies have come in: Noisy treadmill motor

Briefly:
- Belt and rollers offering too much resistance to motor, putting it under excessive load. You may consider that part of the drivetrain ruled out, but it does make sense and explains why two motors did the same thing.

- Stretch the brush springs to improve contact. Easy and worth a go.

- TomG has done a spectral analysis of the noise. I think the most relevant is the 10 sec one.

Adjust the belt properly (as you mentioned it needed doing above), try the brushes and get back to us?