3 Phase Class D amp for DIY BLDC motor Drive

Hello everyone,

I've got the AA motor and the shaft coupling parts I told about a few posts ago.

They are a good fit. Tight, although not too tight and without any cogging or misalignments. A perfect fit, IMO, and this way I can use the original VPI pulley and belt (I will buy the 600 RPM HRX pulley, the one on the picture is the 300 RPM two speed pulley).

The only "ïssue" is that the motor shaft is a little too long, so the final assembly got a little too high, but I don't think this would be a real problem (and I'll leave it like that unless it is easy to move the shaft, but I'm not sure about that).

I intend to use the original VPI SAMA housing, maybe with a new top plate for a better fit.
 

Attachments

  • IMG_20180512_063315350.jpg
    IMG_20180512_063315350.jpg
    565.2 KB · Views: 423
  • IMG_20180512_063400881.jpg
    IMG_20180512_063400881.jpg
    511.2 KB · Views: 417
Last edited:
Hi John,

I had the BLWR up & running for quite a while before finally getting the BLWS, Lots of testing different pulleys and reassembly. The noise was mostly from the bottom bearing and I replaced both with these last Friday.Amazon.com: 5x16x5mm Hybrid Ceramic Brushless Motor Ball Bearings (2): Industrial & Scientific The motor is much quieter now and we'll see if that lasts.

I had made the pod on an eccentric for adjustable belt tension but will make a new one centered since I'm now running stock size pulley.



I've started work on the pod for the BLWS, a bit smaller than what Pyramid designed as I have to work within the existing dimensions of the Oracle. Will report back on findings.



Cheers,

Gregory



Is the above referenced link for bearings that fit BLWS motors? Also assume there are two bearings per motor.
 
Member
Joined 2014
Paid Member
I have a couple questions.

Is one of these two motors inherently better than the other? I have both and the BLWR seems to exhibit a bit of cogging when turned without power applied - the BLWS does not exhibit this characteristic.

Is there a lower practical limit to the frequency at which these motors will operate? I'm using a commercial pulley on a BLWS to drive a VPI Scout. I'm operating about 18.8 H in order to drive the correct rotational speed. Is this OK?

Thank you.
 
The two motors are nearly identical specs, but the BLWS does seem to be smoother.

I've run the motors as low as 1Hz for start up and they work beautifully. You must change the voltage with frequency though to maintain the proper power consumption. 600RPM at 20Hz is normal for this motor (it has 4 poles); 18.8Hz would not need any voltage compensation, it just means your pulley is slightly larger or the belt is slightly thicker than expected.
 
The two motors are nearly identical specs, but the BLWS does seem to be smoother.

I've run the motors as low as 1Hz for start up and they work beautifully. You must change the voltage with frequency though to maintain the proper power consumption. 600RPM at 20Hz is normal for this motor (it has 4 poles); 18.8Hz would not need any voltage compensation, it just means your pulley is slightly larger or the belt is slightly thicker than expected.

Hi Pyramid- I've seen you say this before, can you expand on how to work out the voltage compensation needed, if operating at a different frequency? I've tried working through the various formulas I've found, but multi phase electrical stuff goes way over my head...
 
I haven't been able to find a usable formula for 3 phase motors. The closest I remember is a post by Mark Kelly:

Effective load=coil resistance x forward voltage / (forward voltage - back EMF).

I doubt this is accurate for 3 phase.

I've tried modeling something in an Excel spreadsheet, but the results weren't even close to the measurements I was getting and the complexity of the spreadsheet was starting to get unmanageable. I'm sure there are a number of DSP apps that address this, but the math is not trivial.

Probably the most direct approach is the empirical method: Measure the DC current between the power supply and amp driving the motor; multiply by the efficiency of amplifier (~50% for class AB, 90-95% for class D) and the result should give a close approximation of the power consumption of the motor. Adjust the output voltage to keep the power consumption constant at both speeds (or what ever speed you want to run at).

Because the back EMF generated by the motor opposes the drive signal, the impedance of the motor effectively increases with speed so the amplitude of the drive signal must increase accordingly. As you push the speed higher and the drive level increases, you must take care not to stall the motor or the current will suddenly spike as the back EMF disappears completely.
 
Thanks for the explanation Pyramid. For some reason (Dunning–Kruger? Domain dependent thinking?) I always assume that things in the electrical world are neatly reducible to theory, with nice analytical solutions. Guess at that point it would no longer be an engineering discipline, and the fun would be over :)

Maybe I'll try and add a live power display to my unit, as a self-education project...
 
Thanks Pyramid for making this project available.
I built the SG4, the MA-3D and I'm using a BLWS motor. I'm planning to use them in a Thorens TD 125, but with the current pulley dia (0.738") I need 279RPM/376RPM for 33/45. The frequencies are 9.29Hz/12.54Hz.

What should be the nominal and reduced motor voltages for these RPM values?
 
Thanks Pyramid for making this project available.
I built the SG4, the MA-3D and I'm using a BLWS motor. I'm planning to use them in a Thorens TD 125, but with the current pulley dia (0.738") I need 279RPM/376RPM for 33/45. The frequencies are 9.29Hz/12.54Hz.

What should be the nominal and reduced motor voltages for these RPM values?

Nominal
9Hz: 7VPP (2.5VMRS)
12Hz: 7.6Hz (2.68VRMS)

Reduced
9Hz: 6VPP (2.14VRMS)
12Hz: 6.58VPP (2.33VRMS)
 
Hi Pyramid, I have a question regarding the BLWS231S-24V-2000 motor.
The second motor I got, got nosy also, as my first one. This happened after the frequency of the controller got high up, without me taking note.
Is there anything I can do to fix that or I should buy a third motor?
If the motors are so sensitive, maybe we should have a lower limit on the max frequency from the controller?
 
vgeorge-

Not sure I understand. Why was the frequency of the SG4 high? It should not damage the motor regardless of which frequency you use. If the motor runs noisy at high frequency, does the noise go away when you return to normal freq?

What are the actual frequencies you are using for 33/45 RPM and at what freq did the motor become noisy?