Power Supply 40v 2a

[xmen33]

Hi

I went to design a 40V 2A power supply to run 3 stepper motor 3A.
i used L298 to deal with motor, the max volt can drive is 50V with 2A " i get this from datasheet L298 is this right? ".

so i went to design this power supply with regulator .

 

[davelectronic]

2 amp power supply

Hi there. Your supply we plan on or have built is 40 volts at a current of 2 amps, and the motor you plan on running is 3 amps, the current of your supply is 1 amp short, this will stress the supply and it will probably fail ina short time, motors under load often draw a fare amount of current, if it where me, i would build a supply for 5 amps or a little more so you have plenty of power in reserve to meet the motors requirements and ratings. Dave. PS whats your project ?.

 

[xmen33]

my project is small cnc.
what about L298 , it cant handle more than 2 amp.
i thing that 3A is the max ampere that the motor can handle " sanyo denki 103H6707 "

example: i design a 40v 5a power supply and connect it with 3 L298 and 3 stepper , if one of this stepper run only so the current will be 4 - 5A and will burn the L298 and the stepper .
is this right?

 

[Resqueline]

That motor seems to be a three-phase 3A stepper but you have not provided the full part number. There are 4 more digits providing the winding data.
The L298 is a "dumb" dual 2A full-bridge 46V driver. Page 7 in the ST datasheet shows how to connect the duals in parallel to make it a single 4A.
So to run 1 such motor to the full it seems you'll need 1.5 such chips, plus intelligent driver electronics. To me it does not seem ideally suited for the job.

A power supply usually needs to be rated at a higher amperage than the load, but the voltages involved needs to be considered too.
A 40V 5A supply will kill a 3A 24V (= 8 ohms) motor. A 3A 12V (= 4 ohms) motor will also be killed, but will kill the supply too in the process (by drawing 10A).
A 6A 48V (= 8 ohms) motor would draw 5A from a 40V supply and would thus be suitable, although not being fully utilised.
A stepper motor always draws the same current (even if stopped) so the supply needs to be rated at the sum of all the motors.
An unpowered stepper motor does not have a "locked" rotor and can thus be moved by an external force being applied. Not good for a CNC.

Running a low-voltage (= low resistance) motor on a high-voltage supply requires a switch-mode (PWM) constant-current driver.
It will convert the high-voltage low-current supply to a lower voltage - higher (controlled) current drive to the motor.
At the same time the high voltage available allows the motor to be run at a higher speed than usual.

 

[xmen33]

the motor is " 103h6707-0741 " and i have 3 motor will run.

1) This motor is 3A so i think that i need 9 A supply , but what volt power supply need with 9 A to avoid burn this 3 motors ? .

2) what the best way to run this motors " BIPOLAR or UNIPOLAR "?

3) i design a 3 way that i found in many source , i went the best of them

#1) A l298 with parallel chanel , here the pic of it

#2) A driver with mosfet and diode , here the pic of it

#3) A driver with mosfet , diode and zen diode , here the pic of it

4) final : i don't understant the last part of your post " sorry "

 

[Resqueline]

It turned out hard to find data on that motor but one person at least mentioned it having 6 wires with 1.2 ohms from the centre tap(s) and 2.4 ohms end-to-end.
A Sanyo-Denko catalog mentions similar motors having a 3A rating in unipolar mode, and 2.2A in bipolar mode. They had 0.77 & 0.9 ohms resistance.
How big and heavy is your motor, what's the wiring to it like, and what resistances do you measure?
If the 1.2 ohm unipolar measurement is correct then the motor runs on 3A * 1.2ohms = 3.6V and draws ( 3.6V * 3A ) = 10.8W
Allowing for driver losses you'll need a 40W supply for 3 motors, and using chopped drivers you could run off a 40V 1A supply.
Bipolar mode: 2.2A * 2.4ohms = 5.3V drawing ( 5.3V * 2.2A ) = 11.6W which is slightly more than unipolar mode.
I guess the inductance is also higher in bipolar mode, and this may be undesirable for high-speed operation (but I don't know).
You'll need to connect the L298 (& sense resistor) to a L297 or a L6506 to make it a chopped 40V driver. Read up on the datasheet(s).
Using MOSFETs only you'll have use a 3.6V (or whatever) 9A supply. This may also limit the speed at which the motor can be run.
There should be no need for diodes across the MOSFETs (D5-D8).

 

[xmen33]

my motor 1.2ohm from the centre tap and 2.4 ohm end-to-end.

1) u sad that i need 9A power supply , so what the min and max of volt in power supply ?

2) i post 3 way to control the motor , so what the best of them because i need speed and torch,
please after show me the best way , can you check the element of the circuit because i just used general diode without select any type of them .

sorry if u found any mistake in my language .

 

[Resqueline]

1) It depends. Read replies and datasheets, then Google the terms you don't understand...
2) None.. Read my reply again.

No prob's with the language.

 

[xmen33]

i google about the max of the motor, i found someone say that the min is 6.9v and the max 40v
is this right?

 

[Resqueline]

That voltage range sounds correct for a chopped (PWM/ switchmode constant current) driver.
The L298 needs an L297 to make it a chopped driver, and with the correct sense resistors it can run that motor off 40V.
The sense resistors in your pic # 1 doesn't have any useful function at all, and you'd have to run it on 6-7V.
For a discrete, dumb MOSFET driver (pic # 2 & 3) you have to run on 3.6V like I said, but this won't give you the best performance.

 

[xmen33]

Final i finish the design , I have some Q .

First the pic of design ,


Second the Q:

1) I know that Vref = current * R sens, but which current mean ? , Because i found topic at internet calculate it like " motor 3A and put 0.5 ohm sense Resistance so Vref = 3*0.5=1.5 in 4 channel situation
or it = 2* 0.5 =1 because the max in each channel is 2amp "
and i make ic 2 channel "get the way from l298 datasheet " so can it calculate Vref = 4*0.5 =2 ?

2) What the useful that i will get if i connect 100nf cap at each power " vcc and vs " ?because datasheet say that but didn't write why .

3) i will make 3 driver , if i connect sync with each other to provided the RC to the other l297 ics is this better or just make RC to each l297 and forget about sync pin ?

4) What the best diode name for this circuit " 2 channel method ... "

5) Are the sense resistance have a special name or it's just 0.5ohm .

6) please can u check the element in my design to find any mistake .

7) if i not Bother you, can u please Answer this Q by order .

Sorry if i Bother you

 

[Resqueline]

1) The current to be limited is the motor current; max 3A. The sense resistor voltage is not allowed to be higher than 2V, but at 3A this is 6W of wasted power.
And 0.5 ohms * 3A = 1.5V & 4.5W which is also very much. 0.22 ohms * 3A = 0.66V & 2W which starts to get acceptable. 0.15 ohms * 3A = 0.45V & 1.35W = ok.
I would choose to make each sense resistor out of two 0.22 ohm 0.5W resistors in parallel, and having a 0.33V reference voltage.
I would also make provisions for feeding the sense voltages into the L297 via 10k resistors, and having 220pF decoupling cap's at pins 13 & 14.
This is to prevent (any possible) noise from activating the current limit too early. The circuit layout is also critical to minimise ground noise.

2) That is called decoupling capacitors and is a universal, fundamental remedy to prevent short/fast voltage spikes from damaging IC's.

3) To quote the datasheet: "Ground noise problems in multiple configurations can be avoided by synchronising the chopper oscillators."

4) To quote the datasheet: "when the inputs of the IC are chopped ; Schottky diodes would be preferred."

5) To quote the datasheet: "not of a wire wound type". In other words: use ordinary 0.11 ohm 1W metal film resistors.

6) It looks ok as far as I can see, but I have not checked every little aspect nor built & tested it so I can't guarantee there are no problems.

7) Done.. Can I go to bed now?

No bother, it's ok, you seem to have done a lot of good homework.

 

[xmen33]

hi
I just finish test the design and it work but the speed not fast .
I used diode call 1n5822 , used pic16f877a to send the pules " clock " ,the delay is
Delay_ms(20) and i used the pc power supply 12v " i will use 2 power supply to get 24 v" .
I have 4.000 crystal xt so what the problem ? .
if u used 16.0 MHz HS crystal , this will increase the speed or i should change this diode ?

 

[xmen33]

any suggestion ?

 

[Resqueline]

Those 3A 40V Shottky diodes should be fine and of no speed limiting factor.
If the low power supply voltage was the limiting factor you'd experience a loss of torque at high speeds.
I rather believe the limit you're experiencing now lies in the software (& I can't help there). What's the delay thing?
If the PIC can handle 16MHz then you should (depending on the program) get 4 times the speed you have now.

 

[xmen33]

I just buy 18MHz and i will try it after making the PCB .
I really don't know what to say to you, really thanks you for helping me alot
I will post the layout of design after finish it , to help any one .