DC MOTOR NOISES (Thrustmaster T500RS sim racing steering wheel)

[perkabrod]

Hey, have a listen to this DC motor, what is the whining/clicking noises heard? Something faulty, dc motor, h-bridge, position sensor?
In the video I just connect Thrustmaster steering wheel to computer without any game running, that's why the noise can only be heard after I have offset the steering wheel from it's 12 o'clock position and the control circuit is bringing the wheel back to center.

But while playing a game, the noise is heard (and lightly felt) no matter the direction of angular movement. It's a slight notchy feeling, like a stepper motor. I have not heard this type of noise from watching other videos on youtube some I'm curious to find what's causing it. No warranty left, wheels are around 500 EUR new 

Video demonstration
https://youtu.be/6a0iiH1Ne-U

[Armadillo]

I think is bearing noise, either from the motor or the drive bearings.
I would take out the motor to an external system, let the motor drive dummy load and trace it in an oscilloscope for any sign of abnormal current ripple in the DC power supply line to eliminate the motor.

[perkabrod]

Noise is only heard when power is connected. I will try to remove the motor (but still connected electronically) and see if the same noise is apparent under no loading from belt drive. Will post report once I have done that.

[Cyberdragon]

Sounds like a faulty and arcing commutator. Either your brushes or the commutator contacts themselves (or both) have pitted (or corroded) and started arcing.

[Armadillo]

Noise is only heard when power is connected. I will try to remove the motor (but still connected electronically) and see if the same noise is apparent under no loading from belt drive. Will post report once I have done that.

Hi, the bearing needs to be loaded for the test.
The belt pulled the bearing sideway [side loading]. If you tighten the belt too much, the bearing life will be impaired. That's while correct belt tightening is important. The balls can be wear or the ball spring race can be warped.
If unloaded, not a true test but if it sounded, good for you.
If it don't, you still can't conclude or exclude.

Edit: maybe try grease/oil your motor [example Real Four tournament grease]

[perkabrod]

...I would take out the motor to an external system, let the motor drive dummy load and trace it in an oscilloscope for any sign of abnormal current ripple in the DC power supply line to eliminate the motor...

What should I hook up my oscilloscope probes to, the + - on the motor or something else?
What am I looking for in a good working DC motor versus one with problems (due to worn brushes/commutator etc)?

[Armadillo]

Firstly is your motor EC/BLDC motor? Brushless or Brushed?

[perkabrod]

Firstly is your motor EC/BLDC motor? Brushless or Brushed?

Brushed DC, 24V, datasheet, model 236 should be close but label says 289 which cannot be found on their website
https://www.reichelt.com/sz/en/index.html?ACTION=7&LA=3&OPEN=0&INDEX=0&FILENAME=C160%252F1.13.044.236_DS.pdf

[Armadillo]

If there is a commutator problems, your motor will jerk, the torque would be jerky and usually sparking at the commutator.
In your case, your motor rotate smoothly, So don't waste your time on this idea.

I noticed that there is a encoder at the back of the motor. The click click click sound could come from the encoder. Why don't you remove encoder and try it again

[perkabrod]

I noticed that there is a encoder at the back of the motor. The click click click sound could come from the encoder. Why don't you remove encoder and try it again

Thank you very much for your input. Yes, encoder not part of the Bühler motor, I will remove it and investigate, hopefully lot cheaper to replace if it should be the encoder

[perkabrod]

Angular position hall sensor removed, what's next? Feels like it's time to probe something with the oscilloscope?
For this video I did some post fx on the audio channel prior to uploading, a level compressor and noise reduction. Hope you "like" how clear the noise is in this video;)

https://youtu.be/IKybn8OyKF8

[Cyberdragon]

If there is a commutator problems, your motor will jerk, the torque would be jerky and usually sparking at the commutator.
In your case, your motor rotate smoothly, So don't waste your time on this idea.

It's not moving smoothly anymore. It's making noise when it's not moving at all, which means it's having trouble starting. This could definately be your commutator.

However, it could also be an electronic issue. If your motor is getting electrical noise, there may not be enough power in the spurrios impulses to make it move, but enough to make the coils vibrate like a speaker. Probe the motor input with the scope.

[Armadillo]

Hi;

When the pinion rotate, it sounded like the motor is rubbing onto something. Mechanical noise.
I would suggest you take out the motor from the system so that the noise can be isolated and traced.
Also the speed controller seems to be trying to rotate a stuck rotor.
You need to power the motor not from the speed controller, but external power supply, best is variable voltage power supply. Adjust voltage from 0v to 24 volts to just the motor, both unloaded and loaded externally situation.
Hope to see the video soon. 

[perkabrod]

...

Probe the motor input with the scope.

Will do.

[perkabrod]

I would suggest you take out the motor from the system so that the noise can be isolated and traced.

Yes, this is on the way and should hopefully be in my mailbox tomorrow. Should make it easier to remove motor, no need to disassemble the whole drivetrain...hopefully

[Armadillo]

It's not moving smoothly anymore. It's making noise when it's not moving at all, which means it's having trouble starting. This could definately be your commutator.

Yes, it did sounded like a pwm induced commutator chattering noise onto a stucked rotor when the encoder is loosely/initially coupled. Example feedback from encoder not in circuit and transient.
So hopefully the feedback, pwm drive can be excluded from the test by using external DC power supply.

[perkabrod]

The rear shaft with the magnet was very accessible after removing the angular sensor, so I tried spinning the shaft. VERY little resistance, I was a bit concern of mechanical resistance before when I move the steering wheel, but due to the gear stages, there was very little resistance moving the actual shaft on the motor. So any high resistance would then be because something happening inside the motor itself. Or weird noise which the scope will show. Have to sleep now, girlfriend already sleeping and she doesn't like dc motors. Will update tomorrow, thanks for helping!

I have also sent video to Thrustmaster tech support and the people that made the angular sensor, though I don't think they will answer

[perkabrod]

By the way, the torque can be set on computer control panel. One of the first things I tried was comparing noise at maxed out and minimum torque setting. For small movements of the wheel, fraction of a degree, absolutely nondifference in the chattering noise. Not sure that info help much but at least it's out there now.

[Armadillo]

Hi, noise reduction article you could try.

[perkabrod]

Nice document! The single cap method shown in figure 1 is used, now I know why there was a tiny conponent (painted black/gooped) clamped to the motor case with cable ties.

There's also hystheris on the angular position sensor that might be off, can't recall what the datasgeet said about how to set it though...

[Armadillo]

Just assume that black/gooped component is faulty [which is quite common] so you add the external capacitor to it. Best is also to try the 4 capacitors version.

https://www.pololu.com/product/1166

[perkabrod]

Will do. I should have equivalent caps at home.

[perkabrod]

First post here
https://www.xsimulator.net/community/threads/thrustmaster-t500rs-motor-upgrade-65w-to-85w-success-01-23-16.7797/page-3

There's a PCB soldered to the motors in these steering wheel products. I can access the + - on my motor without removing the motor nor pcb, do you think it is safe to hookup 0-24 VDC from my bench power supply directly to + -, leaving everything still connected to the rest of the circuitry, h-bridge etc?

Would be way easier to try that than waiting for the angled screw driver to remove motor and to desolder the pcb

Not sure what that little thrustmaster pcb soldered to the motor does, there's a separate pcb as well with the h-bridge, diodes etc.

[Armadillo]

The first post is from you?

Better take out any wirings to the driver circuit/H bridge because the bridge could be faulty.
At this time, anything is the suspect because I don't exactly know it to be the truth..... LOL.
You want positive conclusive test to isolate so best don't mix.

 

[perkabrod]

Thanks, yes good point.
Will wait for the angled screwdriver, desolder the circuit board and test directly on the motor terminals, with a diode in parallell.

No I haven't wrote anything in that thread about swapping in a larger motor, but I have read it

[perkabrod]

Okay. Obviously I don't have enough arms to pull out a proper oscilloscope test, so bare in mind that this link is captured with the ground probe floating. The red + probe is connected to + terminal on DC motor.
https://youtu.be/gVAbqVpBr9I


There's continuity (or 4-5 ohms) between dc motor + and - terminals (measured while the PCB is connected)

Here's a photo taken on the scope while having both probes hooked up, ground on motor housing and red probe on + terminal on motor.


When measuring with the multimeter I got -12 VDC at the motor housing, with the - terminal on motor as reference. I also measured the heat shield on power transistor closest to the large capacitor (on a pcb located near the left padle shifter, this gave me 12 VDC but I cannot remember if it was + or -. The other power transistor heat shield measured 0 V, still the same reference (DC motor - terminal).

I did try a few different capacitors (0,1 - 1 uF) in parallel with motor but no difference what so ever in noise.

Duty cycle 50% = motor standing still
=> I don't get it... I thought the duty cycle should be 0% when no movement, and then 0-100% (0 to +24 V square wave with duty cycle aka PWM) to move one direction, let's say clockwise CW, and 0-100% duty cycle negative (0 to -24 V PWM) to move CCW.

Replying to myself: Locked Anti-Phase PWM control
In locked anti-phase PWM control, the amount the voltage signal “swings” from a positive voltage to a negative voltage is used to control the motor.  In this case, forward and reverse motion depend on the duty cycle of positive voltage versus negative voltage, with the 50% duty-cycle equaling a stationary motor.  For example, if the voltage across the motor is positive for longer than it is negative, the motor will move forwards.  The same is true for the opposite: if the voltage is negative for longer than it is positive, the motor will move backwards.  Therefore, if the positive duty cycle is higher than 50%, the motor will move forwards; if it is lower than 50%, the motor will move backwards.

source:https://www.google.se/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0ahUKEwiB5_7a59_YAhXTbZoKHTkYCcUQFgg2MAE&url=https%3A%2F%2Fwww.systemseng.cornell.edu%2Fse%2Fintel%2Fteam%2Fupload%2FMotor-Controller-Guide.docx&usg=AOvVaw33H71VprwAx6MwX1PdmByg

[Armadillo]

I can tell you the motor sounded angrily powerful when its turning.   
Sure don't look or sound like a commutator problem to me.
I noticed only when the pinion slowly rotate and you placed the microphone so close that I can hear the rubbing click click click sound.
When it stood still I cannot hear it, example there is no chattering.
If you powered it on and at 50% pwm, pinion not moving and it chatters, then its commutator chattering otherwise IMHO, sounded like loaded motor bearings.

The scope signal looks healthy as far as I can visibly see.
The 50% PWM at full peak voltages, I think is for a purpose to hold/grip the steering wheel in place. Otherwise it's a waste of energy for non gamer like me.

Hey, considering the loudness of the drive vs the clicks [10,000 : 1], I say it's a "acceptable". 

Edit: The NETT average voltage during holding should be "ZERO", it would be this remnant voltage that could cause the unnecessary chattering and commutator wearing. For normal turns, the optical feedback signal will make sure the PWM turns the motor.
So Check that the Positive voltage and the Negative Voltage excursion/drive are equal resulting in 50% Nett Zero current and voltage.

[perkabrod]

Yes, this is no toy It's powerful enough to make you slip on the steering wheel while playing. Yes, no noise when in a "idle position", noise is only heard when you rotate the wheel just a fraction of a degree. The annoying thing though is that when playing, that noise will be heard 90% of the time because of the delicate movement while in a long corner for example.

Maybe the pcb on the motor terminals is for the noise reduction. Hard to see but I see a few tiny ceramic disk capacitors. I also figured that the gooped looking cap is probably a temp sensor, it is strapped on to the motor housing with cable ties.

Maybe have to live with the noise Tomorrow the angled screwdriver will be delivered, if I'm in the mood I might carry on and remove motor and investigate...

I wrote that the cap didn't reduce the noise but I realized I did it wrong, connecting one leg of the cap to motor terminal + and the other leg to motor housing, and the housing is not connected to the negative terminal. So there's some hope still that a cap is faulty, will look at the pcb on the motor terminals, one motor is removed.

Do you know if the dc motor noise reduction document you linked, also work with locked anti-phase motor control, like in this case?

[Armadillo]

Regarding the capacitors, Yes it will work with the anti-phase motor control. To the cap, it's just delta V. The purpose is to reduce the slew rate of the pulse / average the drive. Considering the plus and the minus voltage, the capacitor voltage selection should be 100v or more. 0.1uf type is more to reduce the radio frequency noise. 1uf and up would be more for averaging.

But you should resolve the following which I think is more critical.;

Edit: The NETT average voltage during holding should be "ZERO", it would be this remnant voltage that could cause the unnecessary chattering and commutator wearing. For normal turns, the optical feedback signal will make sure the PWM turns the motor.
So Check that the Positive voltage and the Negative Voltage excursion/drive are equal resulting in 50% Nett Zero current and voltage.

Why the Vmax 18.32v is greater lesser than Vmin -27.71v? The H-Drive merely invert the power so should be same. You should check.
The Vavr is 2.01v, it should be zero during idle.

[perkabrod]

...You should check.
The Vavr is 2.01v, it should be zero during idle.

That cheap oscilloscope based the values graphically, I need to zero/tar before measuring in order to get readings I can trust. I measured today with DMM and there's 6 mV between + and - terminal on motor. I also probed the 4 power transistors and couldn't find any noise or weird voltages (measured this with motor case as reference and got reading of 12.12 VDC, 24 VDC etc.)

I also tried a 1 uF between MOTOR terminals but no difference in noise. Right now I'm doing a recalibration of the center point, found a guide on how to do that. http://ts.thrustmaster.com/download/accessories/manuals/t500rs/t500rs_autocalibration_centering_process.pdf

I will make another video then which has some interesting stuff that I didn't think of too much before, but could potentially mean something for any of you with experience

[Armadillo]

That's OK.
Looking forward to your interesting video. 

[perkabrod]

https://youtu.be/-4KfLH-VoCI

In games there is no return to centre, because the steering wheel will just follow the input from the car physics, so the electrical noise will be heard when turning both CW and CCW. I re-assembled the base plate so that I could do some sim racing tonight, hope to find an end to this soon, or accept the noise

[perkabrod]

Such a lovely screwdriver <3 It arrived today and now the motor is out, will desolder the pcb and test motor with bench supply... Update to follow.



The reason I purchased this screwdriver was because I can come away with not having to disassemble the gear stage with pulleys etc. Lots of screws and potential problems (pulley tensioner etc).

[perkabrod]

Motor tests
https://youtu.be/yHzX8w0N_uA

Short video on below 1V showing a little stuttering
https://youtu.be/U5mJKIpSu3s

[Armadillo]

I would say it's bearing noise. I can hear the noise even when the pinion is not loaded.
Change the bearings using proper tools.   
If you can open it up, then you can tell what bearings to buy or you can try check with manufacturer.
I would recommend boca ABEC 7 type bearings.

[Armadillo]

Or consider to change the motor.
Look for 775 type motor but you need to check the mechanics, pinion etc...
DC motor should not cost a hand or a leg.

https://www.ebay.com/itm/1PC-DC-24V-21000RPM-High-Speed-Large-torque-DC-775-Motor-Electric-Power-Tool/331883102354?hash=item4d45c64092:g:Y8gAAOSwJhFZgaKU

[perkabrod]

Or consider to change the motor.
Look for 775 type motor but you need to check the mechanics, pinion etc...
DC motor should not cost a hand or a leg.

https://www.ebay.com/itm/1PC-DC-24V-21000RPM-High-Speed-Large-torque-DC-775-Motor-Electric-Power-Tool/331883102354?hash=item4d45c64092:g:Y8gAAOSwJhFZgaKU

The periodic noise was very easily distinguished irl. I need to reuse the pinion and rear magnet so ig I start removing tjose, I will be able to open the dc motor and look for bearings. If it is not serviceable, at least I know what motor to buy.

Well, the Bühler motor cost a leg or two  but I have the datasheet, also, in the thread I linked about someone who replaced to more powerful motor, there was a suggestion for a cross reference chinese motor...

[Ozz]

Can I get an update. Im having the exact same sound issue. How did you fixed it? OR did you able to fix it for good

Cheers