Please look over my plan

[H500]

But again with many people at the higher performance end of the CNC scale preferring to drive their 8wire motors with bipolar half-coil, there is no performance advantage, that is exactly the performance of unipolar.

I don't understand why anyone would want to do that. Even if the drive was not capable of supplying more current, there are reliability gains in running the motor at a lower temperature.

Unipolar offers benefits in reliability with only 4 larger semiconductors used to switch the motor current rather than 8 smaller semis and the complexity of high side drivers. Also bipolar requires motor current to pass through 2 semis, which causes double the heating of the driver (which can be more than double as high side hbridge driving is known for inefficiency).

While this is true, the power loss in the driver is very low compared to the motor losses, if a discrete power stage was used. I have a motor and driver on my desk right now. With 3.8 amps, the transistors are barely warm a small heat sink. The motor is uncomfortably hot.

But I understand your point. Simplicity does have it's appeal.

[kreutz]

I have not come across anything, practical or theoretical that supports the belief that unipolars can reach higher speeds than bipolars. If you look at typical datasheets, bipolar half coil, parallel and unipolar have the same inductance. In theory, they should all reach the same max speed.

At other speeds, the lower coil resistance of bipolar parallel motors let you pump more current through without heating the motor up. So it will give you up to 40% more torque if the rest of your system is good for it.......

I have to disagree on the bipolar parallel having "intrinsically" more torque than unipolar drives or either of them being able to reach more speed. It all depends on unipolar drive technology.

DIY Unipolar micro-stepper Mardus-Kreutz drives were developed years ago on this forum (free and open source). Even when they were developed as an educational design, they perform at the same level of bipolar parallel drives including high speed and same torque due to RMS vs peak current compensation on all micro-step modes.

They were tested driving high inductance motors (test motors were 8 wires 6.8 mH / coil). Motor heating is not a problem even at less than 1 rpm on any of the step modes available (Full, half, 1/4, 1/8, 1/16, 1/5 , 1/10). M-K drive specs are 80 Volt 8 Amps but tests have being done driving 1 A motors too. No noise come from the motors at any speed, not even at standby. Standby current reduction is standard.

I don't sell them (never did). This is not a commercial.

Best regards,

kreutz

[H500]

I have to disagree on the bipolar parallel having "intrinsically" more torque than unipolar drives or either of them being able to reach more speed. It all depends on unipolar drive technology.

Yes, but if you applied the same current leveling techniques on a bipolar drive, you would be able to develop a higher peak torque and rotational power out of a given motor than the unipolar one.

Or are you using other techniques?

[kreutz]

Yes, but if you applied the same current leveling techniques on a bipolar drive, you would be able to develop a higher peak torque and rotational power out of a given motor than the unipolar one.

Or are you using other techniques?

I disagree again. If you do the same on a bipolar motor you will saturate the magnetic circuit, demagnetize the rotor and/or the motor's useful life will be appreciably shortened.

E.g.: Let's say You have an eight wire hybrid stepper motor you could wire either bipolar parallel or unipolar. When I increase 40 % the peak current on the motor wired unipolar the peak magnetic field (on the poles) equals the one specified by the manufacturer for the bipolar parallel configuration. They specify 40% less current on the unipolar configuration because of the heating effect in the full step mode which is equivalent to a DC current flowing all the time (at low speeds and in standby). The r.m.s. value of a DC current is equal to its peak value.

On the M-K drives setting Full Step mode automatically reduces the set current to the specified by the manufacturer for the unipolar configuration in order to protect the motor. There are more measures like standby current reduction that help reducing motor temperature. The automatic current reduction (in Full Step mode) doesn't really affect the motor performance because the user builds (DIY) the M-K drive in order to use it in micro-step mode.

Two things will kill your hybrid stepper motor (excepting perhaps completely disassembling it without closing the magnetic circuit in advance): Overheat and over-current. Both affect the magnets' field intensity, together they invite disaster.

I also used (in the M-K design) reference current waveform morphing (vs speed) in order to reduce stepper resonance.

Best regards,

kreutz

[H500]

I disagree again. If you do the same on a bipolar motor you will saturate the magnetic circuit, demagnetize the rotor and/or the motor's useful life will be appreciably shortened.

I never seen formal specs on the demagnetization current, but the following comments from vendors indicate that it is substantially higher than the overheating current.
High Torque Stepper Motor
Parker Hannifin, Compumotor Division, North America - A Complete Family of Motion Control Products

Heat appears to be the limiting factor.

I also used (in the M-K design) reference current waveform morphing (vs speed) in order to reduce stepper resonance.

I was under the impression that its purpose was to maximizes the RMS current to the motors. I experimented with it on my DSP drive, but it resulted in a reduction of the top speed under load. I'm not sure why. I believe it interfered with the anti-windup / phase lead algorithm.

[kreutz]

I never seen formal specs on the demagnetization current, but the following comments from vendors indicate that it is substantially higher than the overheating current.
High Torque Stepper Motor
Parker Hannifin, Compumotor Division, North America - A Complete Family of Motion Control Products

Heat appears to be the limiting factor.

I was under the impression that its purpose was to maximizes the RMS current to the motors. I experimented with it on my DSP drive, but it resulted in a reduction of the top speed under load. I'm not sure why. I believe it interfered with the anti-windup / phase lead algorithm.

Hello;

At the speed range the morphing takes place there is already torque loss due to the L/R constant of the motor and back-EMF influence (motor's phase current is not sinusoidal anymore). The actual torque improvement comes from the slight phase lead the new reference waveform imposes. It also helps on controlling resonance problems.

Tell me more about your DSP drive, it sounds interesting. Is it based on Microchip's DSPIC application note?

M-K drives were developed just to prove a point, the use of the Attiny2313 (with its limited resources: 2Kb flash, 128 bytes RAM) was a challenge but is also a limitation for future improvement. There are many things that could be greatly improved using a more powerful micro-controller. Anti-resonance was never a design goal although the drive's performance is really good in that area. I am still reluctant to include anti-resonance on an open source drive,it will give Gecko's competence an edge "for free" . I consider that it is not good for our economy and is a lack of professional ethics considering the help Mariss offers in this forum.

Best regards,

kreutz

[H500]

Tell me more about your DSP drive, it sounds interesting. Is it based on Microchip's DSPIC application note?

Yes. It is based on Microchip's reference design.
AN1307 - Stepper Motor Control with dsPIC DSCs - Application Notes - Details
It controls the current using a PI loop rather than a chopper. I really can't tell whether it performs better than the CPLD driver. So far, the primary advantage is that the dsp has plenty of resources available for experimenting with other techniques. I like to play with field oriented control, but 40 MIPS is likely too slow.

I am still reluctant to include anti-resonance on an open source drive,it will give Gecko's competence an edge "for free" . I consider that it is not good for our economy and is a lack of professional ethics considering the help Mariss offers in this forum.

I agree, but I think Mariss's enduring advantage will be his ability to innovate.

[kreutz]

Yes. It is based on Microchip's reference design.
AN1307 - Stepper Motor Control with dsPIC DSCs - Application Notes - Details
It controls the current using a PI loop rather than a chopper. I really can't tell whether it performs better than the CPLD driver. So far, the primary advantage is that the dsp has plenty of resources available for experimenting with other techniques. I like to play with field oriented control, but 40 MIPS is likely too slow.


I agree, but I think Mariss's enduring advantage will be his ability to innovate.

Hello;

Did you use Microchip's development kit? I don't think 40 MIPs will be the limitation, you can offload the A/D conversion and filtering (need to be really fast) and Park's and inverse Park's transforming either to tables in RAM, FPGA or analog custom engine.

Best regards,

kreutz

[H500]

Did you use Microchip's development kit? I don't think 40 MIPs will be the limitation, you can offload the A/D conversion and filtering (need to be really fast) and Park's and inverse Park's transforming either to tables in RAM, FPGA or analog custom engine.

I did my own simplified board based on their schematic. I replaced the power section with a ir2103 / irf540 bridge. Their coded needed to be modified to work as a step/dir drive.

I think the transforms and PID loops will need about 150 mips to achieve 3000 rpm. Mariss has provided sufficient hints for me to guess how he will do the math with an ingenious analogue circuit. However, I will refrain from speculating publicly to avoid potentially helping his competitors. I have no plans to market a product, so I will simply do it the brute force way with a faster processor. Learning is my primary motive.

[kreutz]

I did my own simplified board based on their schematic. I replaced the power section with a ir2103 / irf540 bridge. Their coded needed to be modified to work as a step/dir drive.

I think the transforms and PID loops will need about 150 mips to achieve 3000 rpm. Mariss has provided sufficient hints for me to guess how he will do the math with an ingenious analogue circuit. However, I will refrain from speculating publicly to avoid potentially helping his competitors. I have no plans to market a product, so I will simply do it the brute force way with a faster processor. Learning is my primary motive.

Thanks a lot. I understand your position.

Best regards,

kreutz

[Mariss Freimanis]

"Yes, but if you applied the same current leveling techniques on a bipolar drive, you would be able to develop a higher peak torque and rotational power out of a given motor than the unipolar one."

Not true.

Torque is directly proportional to current. More accurately it's proportional to Ampere-turns. Ampere-turns means current times the turns of wire it passes through.

In a unipolar motor, it passes through a single strand of wire. In a parallel-connected motor it passes through two strands of wire (bi-filar winding). The current still passes through the same number of winding turns, be it a single strand or multi-strand wire; you get the same torque either way.

Mariss

[James Newton]

Torque is directly proportional to current. More accurately it's proportional to Ampere-turns. Ampere-turns means current times the turns of wire it passes through.

In a unipolar motor, it passes through a single strand of wire. In a parallel-connected motor it passes through two strands of wire (bi-filar winding). The current still passes through the same number of winding turns, be it a single strand or multi-strand wire; you get the same torque either way.

Mariss

Um... just to make sure I'm understanding this correctly, all of what you said above applies because you are applying exactly the same current to both the unipolar and the bipolar setup, right? The bipolar setup, because it uses two coils at once, rather than one, has an increased capacity, and so can be driven at 1.414 times the current of the same motor in unipolar configuration, resulting in 1.414 times the holding torque. ... no?

And what you said about it passing through the same number windings really confuses me. I don't see how that is possible when there are two coils in use in the bipolar setup and only 1 in the unipolar.

I must not be understanding what you are saying here, Mariss.