Electric Motor Secrets

[lighty]

Ah yes, self-aligning is the proper term. It will probably help a lot judging from my own bitter experience.

[Jetijs]

I mean such ones:


They are self aligning. I think that this could solve my problems

[lighty]

So far the only solution I see is to replace the bearings with ones that can handle a little axis shift.

Something like self-adjusting ones?

[Jetijs]

The problems keep on coming.
Today I wanted to measure the efficiency. I started the motor and let it run for a while till the bearings warm up. I have not run this motor for longer periods of time before. After 5-6 minutes I noticed, that the speed is slowly dropping and the current consumption is rising slowly. The motor became slower and slower till it reached about 1000RPM and I stopped the test. I noticed that the bearing holder on the commutator side has become hot, I could barely hold my finger on it. At the same time the other bearing holder was only moderately warm. This indicates that the heat might come from the friction in the bearings. When I started the motor some minutes later, it started with high RPMs but then started to slow down again. This is weird. Also I now have a bolt on each bearing holder so that I can tighten it so that the outer part of the bearing can not rotate inside of the bearing holder:



If I tighten it too much, the speed drops, if I tighten it not enough, the outer part of the bearing starts to rotate inside the bearing holder and the motor becomes very loud and amp draw increased.

So far the only solution I see is to replace the bearings with ones that can handle a little axis shift.
This is frustrating

[Peter Lindemann]

Correct

Peter,
I watched your video about the dynamo meter again and a question arised. Why do we need the wheel? Why can't we just load the shaft with that leather strip? I mean if we know the shaft diameter, we can calculate the circumference. If my shaft is 8mm in diameter then the circumference would be 2 x phi x R, or 2X3.14X4=25.12mm. Right? and we can easily convert this to foots.
Of course I will build the wheel, but it is not as easy as I thought and it will take more time.


Jetijs

Jetijs,

You are correct. The wheel is not necessary. You just need a "known circumference" to do the calculations. The major benefits of the wheel are to save one step in the math, and to lower the amount of side force on the shaft. Try what you have, and see if you can develop consistent results over a series of test. If you run three tests, and the measurements are all within 2% of each other, then you can be pretty confident of the lowest number.

Also, if you rig up a hinged frame to hold the scales, then one person can hold both scales quite still, and you can get the rpm reading and other inputs reliably.

Good luck.

Peter

[Jetijs]

Peter,
I watched your video about the dynamo meter again and a question arised. Why do we need the wheel? Why can't we just load the shaft with that leather strip? I mean if we know the shaft diameter, we can calculate the circumference. If my shaft is 8mm in diameter then the circumference would be 2 x phi x R, or 2X3.14X4=25.12mm. Right? and we can easily convert this to foots.
Of course I will build the wheel, but it is not as easy as I thought and it will take more time.


Jetijs

[lighty]

If not, you can still run the stator coils in PARALLEL (each coil with its own FET), to reduce the inductance of the coils and reduce the rise-time even further.

@Peter

I used UCC3732x drivers extensively for the last few years and they can each easily drive two IRFP450 (or even more but with some modifications). I also sent adequate number of MOSFETs to Jetijs so he has enough MOSFETs to try driving each coil separately. He may have to change gate resistor value (if needed) but in this particular circuit I believe it will be enough to add another MOSFET in parallel with the existing one without changing values of the components.



@Jetijs

Good work indeed. I love the elegant way of solving your disbalance problem. I thought you would go for a new rotor plates but this is faster and cheaper (and more elegant). Congratulations!

Also, if you need help setting up the parallel configuration contact me the usual way.

[Jetijs]

Peter,
Here is the circuit I am using now:

http://www.emuprim.lv/bildez/images/viskautkas_labs/visadi_strunti_1/shema10%20(1).GIF

The blue lines represent the recovery part, I think I drew everything right.
I will do the mechanical energy measurements. I already have two small electronic scales that I can use. I will make the brake wheel out of acrylic. I just need to figure out how to attach the wheel to the shaft.
Thanks,
Jetijs

[Peter Lindemann]

Further Refinements

Elias,
my logic circuit positive voltage regulator only allows me to go as high as 35V max. But at this point I don't think that I will need to increase the speed more. Now I need to pulse theoptortigger LEDs and find the correct frequency and duty cycle at which I can get the best recovery. Then we will measure the efficiency of this motor. And yes, the motor has a considerable torque, I mean I can't really stop it by hand. At the same time I also have a small induction motor that is rated 120W and I can easily stop that motor with hand, my motor currently consumes about 50W at 24V . I just like how the motor gets at its full speed almost immediately
BTW, thank you, Grace, for the blessings

Thanks,
Jetijs

Jetijs,

I looked at your latest video. Awesome!!! The fact that the motor jumps to top speed very quickly is also an indicator of its torque production.

Before you change anything else, you may want to build a dynamometer, like I show at the beginning of my DVD, to start measuring the mechanical energy produced by the motor. You are at the point where looking only at the electrical efficiency of the motor is not enough.

The energy economy of the motor is as follows. It takes a certain amount of electricity to produce the magnetic fields in the stator pieces. Once running, the machine produces TWO BY-PRODUCTS from these magnetic fields. The first product, is the motion of the iron rotor coming into alignment with the stator pieces, and the second product is the electrical energy given back when the magnetic fields collapse. Individually, each of these energy products is "under-unity" in the classical sense, but their COMBINED VALUE can become greater than the original amount of energy expended, due to the geometry and timing of the systems.

Also, if you would please publish your current wiring schematic, I would like to see that at this point.

Before you start flashing the LEDs, you may want to look at the FET timing on your scope as the motor runs. At those speeds, you may be within the optimal window of operation already. If not, you can still run the stator coils in PARALLEL (each coil with its own FET), to reduce the inductance of the coils and reduce the rise-time even further. So, there are still a number of options for "tuning the motor up" at this point.

I sure wish I was in Latvia, right about now.

Great work!

Peter

PS. Thank you, Grace! Your heart-felt support has been a blessing, right from the start!

[Jetijs]

Elias,
my logic circuit positive voltage regulator only allows me to go as high as 35V max. But at this point I don't think that I will need to increase the speed more. Now I need to pulse the optortigger LEDs and find the correct frequency and duty cycle at which I can get the best recovery. Then we will measure the efficiency of this motor. And yes, the motor has a considerable torque, I mean I can't really stop it by hand. At the same time I also have a small induction motor that is rated 120W and I can easily stop that motor with hand, my motor currently consumes about 50W at 24V . I just like how the motor gets at its full speed almost immediately
BTW, thank you, Grace, for the blessings

Thanks,
Jetijs

[elias]

Enjoying!

Hi Jetijs,

I enjoy seeing your motor in operation. Your motor should have a great amount of torque to it. Have you tried running it at around 40v? It should Rock! I am looking forward to see the perfection of your motor. Thanks!

Elias

[Grace]

For Peter!

Just wanted to wish You and your amazing work here a Happy Belated Anniversary! Your work is Love and Supported! You are Blessed!!

[Peter Lindemann]

Same Behavior

Thank you Peter
Your words are encouraging

Here is a video of how my motor works at about 12v. The RPMs are about 2400.

YouTube - Lindemann attraction motor V2.0 video 5

After this test I made the needed changes in the drive circuit so that I can run the motor on higher voltages. When the changes were made, I first connected the motor so that only one phase was working. The power supply was set to about 12v when I switched it on. The motor started to spin fast. Then I increased the voltage slowly till it reached 20V, the RPM's were about 3000 and that is on one phase only Then I connected also the other phase and turned the power supply ON. Now the motor spinned even faster at about 4300 RPM. Here is a video of this test:

YouTube - Lindemann attraction motor V2.0 video 6

After that I increased the voltage up to 24V and the RPMs now reached 5300
I also noticed that at 12v the amp draw was at about 2.2 or so, but as I increased the voltage, the RPMs increased but the current draw remained mostly the same, I think that this is because as the speed increases, the impulse duration becomes smaller and this makes a better recovery. The output in all these tests was connected to the capacitor on the input section. I also noticed that the shaft became warm after these short tests, I don't know so far if this is because the bearings heating up and transferring some of the heat to the shaft or the silicon steel laminations heat up because of the eddy currents.

Next I will try to pulse the optotrigger LEDs and see how it works
Thanks,
Jetijs

Jetijs,

Yes, I have seen the same behavior in my unit here. The voltage rises, the speed rises, but the input current stays the same.

Great work!!

Peter

[Jetijs]

Thank you Peter
Your words are encouraging

Here is a video of how my motor works at about 12v. The RPMs are about 2400.

YouTube - Lindemann attraction motor V2.0 video 5

After this test I made the needed changes in the drive circuit so that I can run the motor on higher voltages. When the changes were made, I first connected the motor so that only one phase was working. The power supply was set to about 12v when I switched it on. The motor started to spin fast. Then I increased the voltage slowly till it reached 20V, the RPM's were about 3000 and that is on one phase only Then I connected also the other phase and turned the power supply ON. Now the motor spinned even faster at about 4300 RPM. Here is a video of this test:

YouTube - Lindemann attraction motor V2.0 video 6

After that I increased the voltage up to 24V and the RPMs now reached 5300
I also noticed that at 12v the amp draw was at about 2.2 or so, but as I increased the voltage, the RPMs increased but the current draw remained mostly the same, I think that this is because as the speed increases, the impulse duration becomes smaller and this makes a better recovery. The output in all these tests was connected to the capacitor on the input section. I also noticed that the shaft became warm after these short tests, I don't know so far if this is because the bearings heating up and transferring some of the heat to the shaft or the silicon steel laminations heat up because of the eddy currents.

Next I will try to pulse the optotrigger LEDs and see how it works
Thanks,
Jetijs

[Peter Lindemann]

Awesome!!!!

Jetijs,

Your patience and persistence in solving each and every problem that has arisen are an inspiration to everyone on this forum. Your studies of how wide the power impulse can be to create the largest energy return for your design are a model of good engineering practice, and I thank Lighty for guiding you through that phase of the work. Your continued examination of the rotor imbalance situation has lead to the solution and a highly balanced rotor with a tight air-gap. Again, only persistence and careful observation, coupled with practical deduction and hard work have lead to the solution.

You are an inspiration to us all!

Your next set of tests should be quite interesting!

Peter