5 phase stepper motor control

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
Hi all, I have a VEXTA PH596H-A-A9 type 5 phase (10 wire) stepper motor (0.72°/step, 2.7A, 0.43ohm, 1mH).
I would like to build the drive for it with these circuits, because I don't know how to program processors.
The ICs I have are L6203, (L6506), L297. The IC in brackets is also there, but maybe the motor direction
change would be easier to solve with L297. I would like to change the stepping speed of the motor with
a potentiometer controlled oscillator. The maximum speed of the motor is 2rpm.
The motor voltage can be 12V or 24V. Any help would be helpful to me in solving the motor control,
because so far I have found little information on the internet about the control of 5 phase stepper motors.
Can the above circuits alone be used to control a 5 phase motor?
Or should I try discrete circuits? Thank you.
 

Papabravo

Joined Feb 24, 2006
21,159
Hi all, I have a VEXTA PH596H-A-A9 type 5 phase (10 wire) stepper motor (0.72°/step, 2.7A, 0.43ohm, 1mH).
I would like to build the drive for it with these circuits, because I don't know how to program processors.
The ICs I have are L6203, (L6506), L297. The IC in brackets is also there, but maybe the motor direction
change would be easier to solve with L297. I would like to change the stepping speed of the motor with
a potentiometer controlled oscillator. The maximum speed of the motor is 2rpm.
The motor voltage can be 12V or 24V. Any help would be helpful to me in solving the motor control,
because so far I have found little information on the internet about the control of 5 phase stepper motors.
Can the above circuits alone be used to control a 5 phase motor?
Or should I try discrete circuits? Thank you.
Probably not. Most IC chips are designed for bipolar drive or unipolar drive with 4 poles. The 5-pole configuration will have more configurations in the stepping sequence and I don't think there is a way to wire it so those chips can be used. It is possible to come up with a FSM in an FPGA, but that is likely to be more work than programming a processor. Send me a link to the motor datasheet and I'll have a look. The last time I saw one of these wee beasties (a 5-pole stepper) was 1979 and the chips you mentioned did not exist.
 

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
The Question is,
why do You need such fine resolution ?
Since your RPM is extremely low,
a much cheaper and easier solution would be a Gear-Reduction, or, Cogged-Belt-Reduction drive.
.
.
.
You are right.
The engine is there, the parts are there, the rest is curiosity. Thank you for your reply.
 

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
Probably not. Most IC chips are designed for bipolar drive or unipolar drive with 4 poles. The 5-pole configuration will have more configurations in the stepping sequence and I don't think there is a way to wire it so those chips can be used. It is possible to come up with a FSM in an FPGA, but that is likely to be more work than programming a processor. Send me a link to the motor datasheet and I'll have a look. The last time I saw one of these wee beasties (a 5-pole stepper) was 1979 and the chips you mentioned did not exist.
 

Marley

Joined Apr 4, 2016
502
You have the circuit showing 5 H-bridges. You also have the waveform graph.
Each H-bridge needs two inputs, one for the top transistor and one for the bottom.
Looking at the graph, each phase can be:
  • High - top transistor on, bottom transistor off.
  • Middle - both transistors off.
  • Low - top transistor off, bottom transistor on.
  • Obviously, at no time will the top and bottom transistors be on together.
So you will need 10 logic signals from your driving system (2 for each H-bridge). From the graph, you can work out the sequence of these signals. Each H-bridge can (should) be a special IC designed for the purpose. These often have circuits inside that prevent top and bottom switching on simultaneously.
I would personally use a simple micro-controller to create these signals but you say you don't know how to program these (you could learn). One IC instead of lots. Possible to easily change if you make a mistake in your design.
But:
Another way to generate these signals is to use "discrete logic" - CMOS logic ICs.
Work out from the graph, how many states there are in a complete sequence.
Use a counter to clock through all these states. The number of states sets the size of the counter. Example: a 4-bit counter will have a maximum of 16 states.
Work out a "truth table" that converts counter outputs (binary) to the sequence of signals required.
Deduce the logic equations of each of the 10 outputs.
Create in actual logic.
Build and test!
Nothing is easy - but it can be done with systematic effort.
Hope that helps.
 

Papabravo

Joined Feb 24, 2006
21,159
You have the circuit showing 5 H-bridges. You also have the waveform graph.
Each H-bridge needs two inputs, one for the top transistor and one for the bottom.
Looking at the graph, each phase can be:
  • High - top transistor on, bottom transistor off.
  • Middle - both transistors off.
  • Low - top transistor off, bottom transistor on.
  • Obviously, at no time will the top and bottom transistors be on together.
So you will need 10 logic signals from your driving system (2 for each H-bridge). From the graph, you can work out the sequence of these signals. Each H-bridge can (should) be a special IC designed for the purpose. These often have circuits inside that prevent top and bottom switching on simultaneously.
I would personally use a simple micro-controller to create these signals but you say you don't know how to program these (you could learn). One IC instead of lots. Possible to easily change if you make a mistake in your design.
But:
Another way to generate these signals is to use "discrete logic" - CMOS logic ICs.
Work out from the graph, how many states there are in a complete sequence.
Use a counter to clock through all these states. The number of states sets the size of the counter. Example: a 4-bit counter will have a maximum of 16 states.
Work out a "truth table" that converts counter outputs (binary) to the sequence of signals required.
Deduce the logic equations of each of the 10 outputs.
Create in actual logic.
Build and test!
Nothing is easy - but it can be done with systematic effort.
Hope that helps.
If you want to go in both directions and avoid illegal states the counter should be one that can count up or down and have a synchronous preload function. All of the illegal states should decode as each half bridge in the middle or all transistors off. The circuit you provided is just a template, it is missing some details.
 
If you move to a bipolar low cost stepper such as a NEMA 17 used in many 3D printers or even a NEMA 23, you can use G-Code and a low cost MCU and Stepper drivers. A source of stepper motors are old dot matrix and laser printers.

For repetitive motion or tasks, you can use G-Code. The electronics are very low cost. Used ATX computer power supplies can supply the DC required. Search RAMPS 1.4 or RAMPS 1.6 on Amazon, Banggood, or Aliexpress. When you find RAMPS, you will often find the ATMEGA based MCU boards as part of a bundle. Arduino boards support GRBL which effectively is a G-Code interpreter.

G-Code can be created manually or with CAM packages. Also the output of joysticks can be used as analog outputs with code in the Arduino interpreting the direction as instructions to move the stepper.

If it is just experimenting, using several available free programs, you can click a button on a screen and cause the steppers to move.

That is what I did to make a 4 axis CNC foam cutter. A reference site for doing such is

www.rckeith.co.uk

A type of driver board is called Botstep. It takes power and step and direction to move stepper motors. Low cost, typically mounted in such boards as RAMPS referred to above.

All of this is NOT applicable to 5 phase steppers. They are effectively orphans in motion control. They are for very specific narrow uses, if still made at all. Find a few old dot matrix or early Canon or HP laser printers and get some 4 wire bipolar steppers out of them. Then you can experiment and learn. Then you can create useful devices for motion control

One company making stepper chips is Trinamic Motion Control. www.trinamic.com
 

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
You have the circuit showing 5 H-bridges. You also have the waveform graph.
Each H-bridge needs two inputs, one for the top transistor and one for the bottom.
Looking at the graph, each phase can be:
  • High - top transistor on, bottom transistor off.
  • Middle - both transistors off.
  • Low - top transistor off, bottom transistor on.
  • Obviously, at no time will the top and bottom transistors be on together.
So you will need 10 logic signals from your driving system (2 for each H-bridge). From the graph, you can work out the sequence of these signals. Each H-bridge can (should) be a special IC designed for the purpose. These often have circuits inside that prevent top and bottom switching on simultaneously.
I would personally use a simple micro-controller to create these signals but you say you don't know how to program these (you could learn). One IC instead of lots. Possible to easily change if you make a mistake in your design.
But:
Another way to generate these signals is to use "discrete logic" - CMOS logic ICs.
Work out from the graph, how many states there are in a complete sequence.
Use a counter to clock through all these states. The number of states sets the size of the counter. Example: a 4-bit counter will have a maximum of 16 states.
Work out a "truth table" that converts counter outputs (binary) to the sequence of signals required.
Deduce the logic equations of each of the 10 outputs.
Create in actual logic.
Build and test!
Nothing is easy - but it can be done with systematic effort.
Hope that helps.
Thank you for your help and your reply. I will stick to CMOS logic ICs.
 

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
If you move to a bipolar low cost stepper such as a NEMA 17 used in many 3D printers or even a NEMA 23, you can use G-Code and a low cost MCU and Stepper drivers. A source of stepper motors are old dot matrix and laser printers.

For repetitive motion or tasks, you can use G-Code. The electronics are very low cost. Used ATX computer power supplies can supply the DC required. Search RAMPS 1.4 or RAMPS 1.6 on Amazon, Banggood, or Aliexpress. When you find RAMPS, you will often find the ATMEGA based MCU boards as part of a bundle. Arduino boards support GRBL which effectively is a G-Code interpreter.

G-Code can be created manually or with CAM packages. Also the output of joysticks can be used as analog outputs with code in the Arduino interpreting the direction as instructions to move the stepper.

If it is just experimenting, using several available free programs, you can click a button on a screen and cause the steppers to move.

That is what I did to make a 4 axis CNC foam cutter. A reference site for doing such is

www.rckeith.co.uk

A type of driver board is called Botstep. It takes power and step and direction to move stepper motors. Low cost, typically mounted in such boards as RAMPS referred to above.

All of this is NOT applicable to 5 phase steppers. They are effectively orphans in motion control. They are for very specific narrow uses, if still made at all. Find a few old dot matrix or early Canon or HP laser printers and get some 4 wire bipolar steppers out of them. Then you can experiment and learn. Then you can create useful devices for motion control

One company making stepper chips is Trinamic Motion Control. www.trinamic.com
Thank you for your help and your reply. I do not want to build a 4 axis CNC machine. I have a 5 phase stepper motor, - given to me by a friend - I have control and drive ICs for 4 wire stepper motors. From their replies, it was clear to me that those ICs cannot be used for the 5 phase stepper motors. I will have to build the drive for this stepper motor with CMOS digital ICs. I hope to succeed.
 

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
If you want to go in both directions and avoid illegal states the counter should be one that can count up or down and have a synchronous preload function. All of the illegal states should decode as each half bridge in the middle or all transistors off. The circuit you provided is just a template, it is missing some details.
"Life is too short to sweat the small stuff."
You're right, but for me it's just a challenge.
I have never seen a stepper motor like this before, so in my spare time I am interested in the control of it and the motor drive.
Now I can see from the helpful answers that, - given my knowledge - I can only solve the motor drive with CMOS digital ICs.
I hope I can get the engine to turn. Thanks for all the help.
 

Papabravo

Joined Feb 24, 2006
21,159
"Life is too short to sweat the small stuff."
You're right, but for me it's just a challenge.
I have never seen a stepper motor like this before, so in my spare time I am interested in the control of it and the motor drive.
Now I can see from the helpful answers that, - given my knowledge - I can only solve the motor drive with CMOS digital ICs.
I hope I can get the engine to turn. Thanks for all the help.
In 1979 when I first encountered one we didn't have chips that would drive it. It had some advantages in step size and microstepping and we ultimately abandoned the effort due the the half-bridge drivers for each phase. They just took up too much room and dissipated too much heat in a small package.
 

Thread Starter

Tóth Csaba

Joined Dec 2, 2021
8
In 1979 when I first encountered one we didn't have chips that would drive it. It had some advantages in step size and microstepping and we ultimately abandoned the effort due the the half-bridge drivers for each phase. They just took up too much room and dissipated too much heat in a small package.
It was a very simple way to turn the engine. I wanted to determine the start and end of the windings. I used every second output of a CD4017 to control 5 power transistors which connected the current to the coils via a current limiter. The unused outputs of the CD4017 did not supply any current to the motor. This solution made the motor rotate smoothly. Next time I would like to use CD4510 to solve the desired waveform and direction change. I still need to figure out the offset of each phase by one clock. I hope to get that right as well.
 
still need to figure out the offset of each phase by one clock. I hope to get that right as well.
Hi Tóth,
5 Phase stepper motors are superior to normal 2 Phase stepper motors. They have more Torque at higher RPM and they have much more constant RPM/Torque graph. They also have better mechanical accuracy and run smoother in full or half stepping modes.

This is why 5 Phase stepping motors are relatively expensive, they are great motors

Around 2008 I designed a simple 5 Phase driver for 10 wire motors. I used The NJM3770A to drive each coil ( so 5x) and a good old 8 bit MEGA8515-P to sequence the coils. In the old days the sequence was done with either the PMM8714PT or the TB6528P both obsolete. You can find the correct coil sequence in their datasheets and also in VRDM motor datasheet of Berger Lahr.

I have plans opensource this project but have not yet had the time. In the meantime I will gladly send you any info you want if you send me an email.

Greezt,

Markus
 
Top