Driving 5 phase pentagon stepper motor with L298 and ATMEGA16

Hi every one
I've tried to replicate L297-driven-L298 scheme for my 5 phase oriental motor pentagon stepper motor, but I've find
L297 only use for 4 wire motors, hence I've decide to use an atmega16 for controlling the stepper motor, I've tied 4 L298
input together as well as 4 output to control each winding with single L298 (2+2 amper for each winding), I've tied currents
sense A & B for the another input to my microcontroller. My plan is following the oriental motor full step pattern table
to energize the winding, in each step I read analogue input of currents sense related to its L298 phase and chopped whether is
exceed from MAX Currents in the loop, then go to next step such as following code

Code:

---


if(FullHalfStep==FullStep)
        {
                switch(step_num)
                {
                        case 0:phase(phase_B,Hz);
                                  phase(phase_E,Hz);
                                  phase(phase_A,High);                                 
                                  phase(phase_C,Low);
                                  phase(phase_D,Low);                                 
                                 
                                  for(i=0;i<Chopping_on_time_us;i++)
                                  {
                                                current=read_phase_current(phase_C);
                                                current*=2;
                                                //print_number(current);
                                                if(current>=Max_Current_mA)
                                                {
                                                        phase(phase_A,Hz);
                                                        //uart_puts("|\n");
                                                        //print_number(current);
                                                }
                                                else if(current<=Min_Current_mA) 
                                                {
                                                        //print_number(current);
                                                        phase(phase_A,High);
                                                        //uart_puts("                H\n");
                                                }
                                  }                                 
                                                             
                                  phase(phase_C,Hz);
                                  phase(phase_D,Hz);
                                  //_delay_us(Chopping_off_time_us);                                 
                                  while(read_phase_current(phase_A)>Chopping_off_Current);
                                  phase(phase_A,Hz);
                                 
                        break;
                       
                        case 1:phase(phase_C,Hz);
                                  phase(phase_E,Hz);
                                  phase(phase_A,High);
                                  phase(phase_B,High);                                 
                                  phase(phase_D,Low);

---

according to reference 5 phase pentagon stepper motor can run as fast as 3000 RPM with excellent power but I cant
reach any performance as claimed, still with theoretical 4 A L298 channel, its get hot and RPM not very hight, I've change the following parameters
but not good success

Chopping_on_time_us
Chopping_off_Current
Max_Current_mA
Min_Current_mA

I've uploaded codes and schematic of my driver if helpful
any comment and help is really appreciated

Performance will be very poor unless your power supply voltage is high enough (around 35-40v). The chopping frequency should be around 20khz., and your code must be fast enough, or the current will very quickly exceed the set value.

Do you have a schematic instead of the pcb diagram?

many thanks H500
I didn't draw any schematic just used single L298 for each phase , connect tied input & tied currents sense to Atmega, I've decide to simulate the L297 performance with Atmega chip.
you mean in my code,current exceed the set value long before I check it and led to overheat and vibration? is there any way to calculate the optimum chopping frequency for specific voltage?

My best regards

When you turn on the coil transistor, the current rises quickly. Your execution loop must be fast enough to check each coil and shut off the current as soon as the set value is reached. Otherwise, the motor will overheat and step roughly. Looking at your code, I think it is not fast enough. At 20khz, you only have 50 uS to check all coils and service the step signal.
Do you have an oscilloscope? If so, measure the voltage across the current sense resistor.

Hi H500
Thanks for reply , some guy says try hardware chopping implementation, it is so fast and realistic, hence I've decide to use opamp such as LM324 to comparable current sense voltage with user set trimpot , but I don't sure , do I need tolerances and close loop feed back for current chopping or not?

My best regards

you will be better getting something better than L298 they are old and redundant only good for playing with they cant handle the current to give good performances

New Pentagon (Bipolar) Driver Chip Set

New Pentagon (Bipolar) Driver Technology

Driver boards for 5-wire steppers

http://inter-eng.upm.ro/2012/files/p...rs/paper62.pdf

that all up was 30 seconds to find there is 4,710 results if you search (5 phase oriental motor pentagon stepper motor)

Hi Daniellyall

Thanks for reply and concern , but always posting a thread is the last step when you have not ready to use solution, those chip are awesome but I wasn't lucky to find it rather than in internet , also I read those articles, how ever there is so rare and completed solution which is used the H bridge or Mosfet array , but I think software and hardware implementation is pretty issue of itself , I would be thank you if you direct me to the person or link which might be helpful

If you are open to other solutions, the tb6600 would be much easier and better.

ok where you are makes it hard to get good stuff the tb6600 is a better chip to use do a search through the arduino sight for that and here.

you need something that can handle a high current more that the L298 can handle if you can get some thing from china you should be ok buying from there just do a search (5 phase oriental motor pentagon stepper motor driver) that should give you all you need to know.

Yes there is better IC rather than H bridge & Fet IC , but this is fun project to pull out somethings works beside somethings make, and as I stated the main issue is to hardware or software implementation of current chopping just like a L297 , but still I cant find applicable try to make 5 phase driver with fets and new IC, and any clue is really appreciated

I forgot that it was 5 phase. The 6600 won't work.

Are you certain it is 5 phase, rather than 2 phase with 5 wires?

H500

Thanks for reply, yes I've already run 2 or 4 phase but 5 phase is something else , as you well know and regards to many article it will run as 4000 rpm with high power, as I stated before there is many mosfet array integrated chip but I've plan to hardware and software implementation by myself. I think hardware current chopping and software pulsing is the close solution ,also I think replace the H bridge Ic with mosfet array , but noting for sure yet

Best regards

Running a 5 phase stepper is difficult and there is very little information available on it. I would strongly recommend dumping that motor and going for something more standard.

Hi

it may of been better to model your stepper driver on a chip set like the Sanken SI-7510 sequencer and fet arrays SLA5073 + SLA5074 or SLA5065 + SLA5068 ( in stock from digikey )

you could build a working stepper driver using the SI-7510 to test your motor and provide data to compare with a second board using your ATMEGA16 emulating the SI-7510
that way you have a way of separating some of possible hardware and software problems you may have in your design

john

I’ve decide to switch from H bright to half bright with array of MOSFET, hence below is the schematic of what I suppose to be work with following assumption
To initiate the lower array just use the gate TC4420, there won’t be problem to turn on by micro
To initiate the higher array use isolated auxiliary voltage with optocoupler
Use opamp for chopping the current whereas voltage of shunt higher than potentiometer voltage (output = 0)
To initiate the high array output of opamp and micro command must be in high state
Pulse and direction control have implemented in software code
Any comet for following schematic is really appreciated

Attachment 279624

Can you post a larger image? The text is not readable.

please find attached PDF

The circuit won't work. The top part of the h-bridge must be driven with a high side driver. I posted a design for a cpld based stepper several years ago. I've attached the schematic. You can copy the power section and modify it for your drive.

Quote:

---

Originally Posted by H500

The circuit won't work. The top part of the h-bridge must be driven with a high side driver. I posted a design for a cpld based stepper several years ago. I've attached the schematic. You can copy the power section and modify it for your drive.

---

Hi

I've made new design with new IC FAN7388, regards to its data sheet , it is capable to drive array of mosfet very smoothly, in low voltage and amper before the chopping switching act ,every things is ok and torque is very low but when I increase the voltage (power) , mosfet temperature increase rapidly and get so hot, and motor vibrating , I think maybe this noise and malfunction due yo high chopping frequency of IC, I attached the schematic of my board

any help is really appreciated

It won't work. There are errors.

-add a pull up on u4 pin1.
-swap vpot and vres
-add a 20 uS time delay between u4 pin 1 and U3. Otherwise it will chop too fast.