Current flow controller for DC Motors

[mikebar]

Hi to all.
I'm in the process to control a DC brushed motor 24V 5A continuous (so, at start it will draw much more current).
The motor specifications says that that motor never have to be feeded with more than 5A.
I'm planning to drive it with a single logic level mosfet (60V 45A) on the low side, with hardware PWM
My question is:
there are some suggestions (even better a drawing) for a circuit that control the current and allow more than 5A let's say for 2 seconds, and then if this condition is still there for more than 2 seconds, the power is automatically adjusted not to feed more than 5A?

Even a short-circuit control circuit could be ok, even if I prefer to regulate the current not to exceed 5A.
Any solution will be ok. With PICs or without PICs. Honestly, better without PICs for tasks like that.
Anyone have some suggestions or sources to refer from?

Thanks in advance.

[mackrackit]

The motor will draw the amps needed. So if the driver circuit will handle the starting amps you are OK.

When the motor is running, less amps are required, the amps will drop. The greater the load the greater the amps. An amp meter can be used for feed back to tell how much of a load is on the motor.

I do not think the extra circuitry is needed.

[mikebar]

The motor will draw the amps needed. So if the driver circuit will handle the starting amps you are OK.

When the motor is running, less amps are required, the amps will drop. The greater the load the greater the amps. An amp meter can be used for feed back to tell how much of a load is on the motor.

I do not think the extra circuitry is needed.

Thank you very much mackrackit for the answer.
Actually, what I'm looking for it is some "automatic sensor" that does not allow more than 5A at 24V for the motor. This kind of circuitry could also reveal if the motor's shat it is locked.
I wish to avoid the use of fuses, because one the fuse is blown it need somebody that replace it...
Any idea?

[mackrackit]

What about an automobile/truck circuit breaker. They reset on their own and normally will handle a bit more amps than rated for a short time.

But if the shaft is locked then you do not want it to keep trying? So a latching relay tied in with the circuit breaker might do the trick.

[mikebar]

What about an automobile/truck circuit breaker. They reset on their own and normally will handle a bit more amps than rated for a short time.

But if the shaft is locked then you do not want it to keep trying? So a latching relay tied in with the circuit breaker might do the trick.

Uhmmm..
It could work.
I was looking for some more sophisticated solution (like transistors that regulate the current flow right to the thereshold)...

[skimask]

Uhmmm..
It could work.
I was looking for some more sophisticated solution (like transistors that regulate the current flow right to the thereshold)...

It's called 'foldback' in a regulator circuit.
http://en.wikipedia.org/wiki/Foldback_(power_supply_design)
Totally useless link, but it might get you started...

EDIT:
http://www.national.com/appinfo/power/files/f4.pdf
Better link...

[Acetronics2]

Thank you very much mackrackit for the answer.
Actually, what I'm looking for it is some "automatic sensor" that does not allow more than 5A at 24V for the motor. This kind of circuitry could also reveal if the motor's shat it is locked.
I wish to avoid the use of fuses, because one the fuse is blown it need somebody that replace it...
Any idea?

did you look at Raychem RUE Polyswitches ...

much better than your PIC ... as Dave raised !!!

lol

I think Microchip has a similar App Note about servomotor ... just take the relevant part ...

Alain

[mikebar]

Thank you guys.
It is a starting point.
I'll post some other questions and when I reach the final result, I'll post the schematics.

Thank you, for now.

[sayzer]

If your PIC has comparator, then you can use it to detect current as you wish.
The example below shows the use of an external comparator, but if your PIC has an internal one then you are good to go.


<img src="http://www.picbasic.co.uk/forum/attachment.php?attachmentid=2892&d=1223533980" >
________________

[mikebar]

If your PIC has comparator, then you can use it to detect current as you wish.
The example below shows the use of an external comparator, but if your PIC has an internal one then you are good to go.


<img src="http://www.picbasic.co.uk/forum/attachment.php?attachmentid=2892&d=1223533980" >
________________

Thank you Sayzer, for the hint.
To be honest, I've not clearly understood if GH1 and GH2 are 2 different pins or whatelse. Any suggestions for the filter capacitanze (R and C values) that goes to the comparator? I think to drive the motor with max 5KHz PWM wave. Voltage 24Volts, maximum allowed current to the motor 5A.
Some suggestions also for a good logic level MOSFET for this kind of motor and the name for a fast recovery diodes?

Thank you a lot.

[sayzer]

mike,

This is for reference.

You can play with the values and come up with the right ones much faster then using formulas.

---------OR------------
You can use analog reading to speed up and down the motor for the time you need.
You need a resistor for detecting motor current anyway.

If your analog reading is high, then slow down the motor, and if low then speed up the motor.
You just do it for 2 seconds as you need.

[mikebar]

mike,

This is for reference.

You can play with the values and come up with the right ones much faster then using formulas.

---------OR------------
You can use analog reading to speed up and down the motor for the time you need.
You need a resistor for detecting motor current anyway.

If your analog reading is high, then slow down the motor, and if low then speed up the motor.
You just do it for 2 seconds as you need.

Thank you for the answer.
Analog reading is what I will do if it is not possible (or reasonable/reliable) to do the same without disturb a PIC chip for this task.
To drive the motor, I have in mind to use the logic type mosfet RFP30N06LE for this kind of application. Do you think it is a good choice?
I thought that a fast recovery diode for this kind of application (if memory is still good) should be a BYV28-200. Do you think is a good choice too or you have some better idea?

Thanks in advance.

[sayzer]

Will the motor run in one direction or you need both directions? CW/CCW

If one direction, then I would use IRFZ44. PIC PWM pin directly goes to the gate with 15K to GND. Easy, reliable, affordable.

If you need both directions then I wish you luck :)


For the diode, what is your plan to use it? Where, why, idea etc.

-----------------------

[mikebar]

Will the motor run in one direction or you need both directions? CW/CCW

If one direction, then I would use IRFZ44. PIC PWM pin directly goes to the gate with 15K to GND. Easy, reliable, affordable.

If you need both directions then I wish you luck :)


For the diode, what is your plan to use it? Where, why, idea etc.

-----------------------

Thank for the suggestions.
Yes, I need just one direction: CW
I think to use the diode in order to protect the MosFET from the back EMF while switching on and off the motor because the PWM (like anode to the side where the motor is connected to the GND through the FET and catode to +24Vcc)
I saw from the datasheet of the mosfet you have suggested that the IRFZ44 has an internal avalanche diode. So my one outside is useless.
But I also saw in the datasheet that there is no mention about IRFZ44 to be a "logic level" mosfet. Since I never used it before, it is really enough to drive a PIC pin directly to the Gate and put in the middle of this path a 15K resistor to the GND?

Thank you in advance.

[sougata]

IRFZ44 has an internal avalanche diode. So my one outside is useless.


The MOSFET body diode is not a soft recovery type. While you can live without an external diode putting one may save trouble. Switch Off ringing + Brush Noise + Reverse Recovery Time may kill your MOSFET.

You can use SG3525 PWM controller that solves your whole purpose.

[mikebar]

The MOSFET body diode is not a soft recovery type. While you can live without an external diode putting one may save trouble. Switch Off ringing + Brush Noise + Reverse Recovery Time may kill your MOSFET.

You can use SG3525 PWM controller that solves your whole purpose.

Thank you Sougata.
That's why I've asked if somedy knows a good diode to protect a mosfet that is running a Brushed motor rated at 24V 5A, supposing an intermittent use and regulated with PWM wave.
I searched for the SG3525. I've found the datasheet but also the providers said that it is out of stock.
Have you used it to control DC motors? If yes, could you share some schematic here?

Thank you.

[sougata]

Hi,

I had a crashed HDD last year. (Now I use RAID cause I am too lazy too backup '// Actually my projects never gets finished :D)

I found the PCB layout but seemed to have lost the schematic. Since SG3525 has complementary outputs with dead time. You need to couple (OR) both the outputs and drive the MOS.

The current sense input works on a pulse by pulse limiting followed by a soft start for persistent overload.

[mikebar]

Hi,

I had a crashed HDD last year. (Now I use RAID cause I am too lazy too backup '// Actually my projects never gets finished :D)

I found the PCB layout but seemed to have lost the schematic. Since SG3525 has complementary outputs with dead time. You need to couple (OR) both the outputs and drive the MOS.

The current sense input works on a pulse by pulse limiting followed by a soft start for persistent overload.

Thank you Sougata.
I'm sorry to strees you. Could I ask even a Paint-made schematic of what do you mean?
Maybe you mean take the signal from output A and B and connect to the input of an OR gate?
Then from the output of the OR gate, drive a logic mosfet. Correct?

I've found this info:
http://home.att.net/~wzmicro/3525.html

Is that a good application, in your opinion?

[sougata]

Hi,

My design was almost the same. I used two MOSFETs (common drain / common source) though. Reason:


Better Current Handling
Quicker Turn-OFFs -- Because when you use diode OR-ing you loose the low going edge that turns off the MOSFETs fast


Also I used an active Op-Amp amplifier for the current signal.

The link you posted should work absolutely fine.

[mikebar]

Hi,

My design was almost the same. I used two MOSFETs (common drain / common source) though. Reason:


Better Current Handling
Quicker Turn-OFFs -- Because when you use diode OR-ing you loose the low going edge that turns off the MOSFETs fast


Also I used an active Op-Amp amplifier for the current signal.

The link you posted should work absolutely fine.

Ok, I'll try it first.
Just a last question:
If I want to use an OR gate instead of the two diodes like you said, supposing that I'm feeding the chip with 24Volts, what is the output voltage level from ports A and B of the SG3525? I suppose close to 24volt as "1". So, in this case, wich kind of OR gate I should use?

[sougata]

Hi,

The SG3525 has an internal 5 volts reference but that is only for the comparator, ramp generator. It has got an internal under voltage lockout function whenever the supply drops below 8 volts (i hope i recalled correct)

The output drivers are designed to directly drive MOSFET gates so if you use a 12 volts regulator to supply the chip. You are good to go. The fast turn-off issue becomes more important when you are switching at a much higher speed. So go ahead with the circuit.

[mikebar]

Hi,

The SG3525 has an internal 5 volts reference but that is only for the comparator, ramp generator. It has got an internal under voltage lockout function whenever the supply drops below 8 volts (i hope i recalled correct)

The output drivers are designed to directly drive MOSFET gates so if you use a 12 volts regulator to supply the chip. You are good to go. The fast turn-off issue becomes more important when you are switching at a much higher speed. So go ahead with the circuit.

So, could be ok to power the chip with 12V and the motor with 24V through the MosFET?
Please note that I have to run the motor with 24V dc.

Thanks.

[sougata]

Hi,

Try out the circuit and then if you need some intelligence / monitoring hook up a PIC micro to the same. In fact with an ECCP module on any of the PICs things are easy in software too.

[mikebar]

Hi,

Try out the circuit and then if you need some intelligence / monitoring hook up a PIC micro to the same. In fact with an ECCP module on any of the PICs things are easy in software too.

Ok, I'll try it as soon I can get one SG3525 part.

[mikebar]

Hi guys,
this is just to thank you all.
I've got the SG3525 part. My motor was running using Logic-level rated mosfet. The IRF740 did not produced any result. Motor not running but it make noise like a whistle. Pretty normal, but no turning of the shaft.
Once substituted with a P30N06 logic level mosfet, then voila. Motor running and SG3525 able to regulate the speed up and down.

Thanks again.