Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Control DC Motor Circuit using PWM

Status
Not open for further replies.
G

gauravkothari23

Advanced Member level 2
Advanced Member level 2
Joined
Mar 21, 2015
Messages
642
Helped
5
Reputation
10
Reaction score
4
Trophy points
1,298
Activity points
6,934
Hi All.
I have designed a circuit to control a DC Motor using PWM and 8051 controller. (Circuit attached).
Motor Specifications
Voltage: 240V DC
Watts: 560 Watts
RPM: 2900 RPM

In the circuit i am using a mosfet to drive the motor ON and OFF and controlling the speed of the motor using PWM at the constant speed of Approx 2900 RPM. Where C7, C9 and R31 was not connected Initially.
Will Also explain the working of the motor. A Round flange is connected to the motor Shaft with approx dia of 200mm and 10mm thickness. rotates constantly at the speed of approx 2900 RPM.
It has a manual braking mechanism which is not electronically controlled, where after every 8 to 10 seconds a brake is applied to the motor for approx 50 to 70 mSec. due to which the RPM drops to around 2500, which is acceptable. and again it takes 2 to 3 seconds for the motor to reach 2900 RPM and again after next 8 to 10 seconds again a brake is applied and the cycle continues to around 14 to 15 hours depending upon the requirement.
Now the Brake applied are not so smooth, but it has a sudden braking system which is applied for 50 to 70 mSec. and at the same time the circuit is also feeding the motor to operate at 2900 RPM Continuously, which is not being stopped at the time when brakes are applied.

Initially my problem was, as soon as i turn on the mosfet to start the motor, 5 out of 10 times the mosfet use to get shorted and was not able to control anymore. so later i added 33pF capacitor (C7) to the Gate and GND of the mosfet to avoid inrush current through the mosfet, which solved my problem.
But recently i started facing a new problem that after every 4 to 5 hours of continues operation of motor, the mosfet DRAIN-SOURCE pin or GATE-DRAIN-SOURCE pin gets shorted. have also tested the temperature which was around 70 to 75 degree celsius,

so i added a snubber C9 and R31 across the mosfet. but still the problem with the mosfet continues, it gets shorted within 4 to 5 hours of operation. but the temperature of the mosfet and snubber was around 100 to 110 degree celsius, within 10 to 15 minutes of operation.

can anybody please let me know, what the problem is, why the mosfet gets shorted again and again.
 

Attachments

  • 220DC Drive NEW.png
    220DC Drive NEW.png
    37.4 KB · Views: 181

Sort by date Sort by votes
gauravkothari23 said:
Hi All.
I have designed a circuit to control a DC Motor using PWM and 8051 controller. (Circuit attached).
Motor Specifications
Voltage: 240V DC
Watts: 560 Watts
RPM: 2900 RPM

In the circuit i am using a mosfet to drive the motor ON and OFF and controlling the speed of the motor using PWM at the constant speed of Approx 2900 RPM. Where C7, C9 and R31 was not connected Initially.
Will Also explain the working of the motor. A Round flange is connected to the motor Shaft with approx dia of 200mm and 10mm thickness. rotates constantly at the speed of approx 2900 RPM.
It has a manual braking mechanism which is not electronically controlled, where after every 8 to 10 seconds a brake is applied to the motor for approx 50 to 70 mSec. due to which the RPM drops to around 2500, which is acceptable. and again it takes 2 to 3 seconds for the motor to reach 2900 RPM and again after next 8 to 10 seconds again a brake is applied and the cycle continues to around 14 to 15 hours depending upon the requirement.
Now the Brake applied are not so smooth, but it has a sudden braking system which is applied for 50 to 70 mSec. and at the same time the circuit is also feeding the motor to operate at 2900 RPM Continuously, which is not being stopped at the time when brakes are applied.

Initially my problem was, as soon as i turn on the mosfet to start the motor, 5 out of 10 times the mosfet use to get shorted and was not able to control anymore. so later i added 33pF capacitor (C7) to the Gate and GND of the mosfet to avoid inrush current through the mosfet, which solved my problem.
But recently i started facing a new problem that after every 4 to 5 hours of continues operation of motor, the mosfet DRAIN-SOURCE pin or GATE-DRAIN-SOURCE pin gets shorted. have also tested the temperature which was around 70 to 75 degree celsius,

so i added a snubber C9 and R31 across the mosfet. but still the problem with the mosfet continues, it gets shorted within 4 to 5 hours of operation. but the temperature of the mosfet and snubber was around 100 to 110 degree celsius, within 10 to 15 minutes of operation.

can anybody please let me know, what the problem is, why the mosfet gets shorted again and again.
Click to expand...

Your gate drive risetime with 10K and its input C slows down switching speed
leaving it in linear region where its burning lots of power. Your delay to handle
brake should occur in code, so rise/fall time of mosfet gate is fast.

Also if you are measuring package T with 62 C/W thermal R is quite high, so internal
gate - drain region probably frying. You have this on a heat sink ?
probably melting internally.

What is PWM freq ?



Also you need to calc Igate peak as its charge requirements might exceed max ratings
of BC847. Maybe best to do a sim on the MOSFET drive and load to get a handle on
gate drive power MOSFET switch and conduction losses.




Regards, Dana.
 
Last edited:

Upvote 0 Downvote
danadakk said:
Your gate drive risetime with 10K and its input C slows down switching speed
leaving it in linear region where its burning lots of power. Your delay to handle
brake should occur in code, so rise/fall time of mosfet gate is fast.

Also if you are measuring package T with 62 C/W thermal R is quite high, so internal
gate - drain region probably frying. You have this on a heat sink ?
probably melting internally.

What is PWM freq ?

Regards, Dana.
Click to expand...
The frequency what I am using to drive the motor is 1 khz.
And yes the mosfets are on heat sink, but I have noticed that without snubber the mosfet does not gets hot, an average temperature of 60 to 65 degree Celsius.
But when snubber is connected, both the snubber and mosfer gets extremely hot say around 110 degree.
--- Updated ---

is the snubber what is creating the problem, because with snubber the mosfet and the snubber is getting quite hot .
 
Last edited:

Upvote 0 Downvote
KlausST said:
Hi,

Did you do power calculations?

Do you have a scope to measure Mosfet voltage and current?

Klaus
Click to expand...
have not done power calculations for snubber. i got the snubber values from one of the article in google.

yes, i had measured the mosfet voltage which was around 190 volt and current to drive the motor is around 4.2 Amps.
 

Upvote 0 Downvote
The incredible slow MOSFET turn-on ramp is likely to cause self-oscillation. If you want to advance to usual switching speed, the slow free-wheeling diode must be replaced by a fast one.

Slow dead of MOSFETs is may be caused by continuous avalanche breakdown.
--- Updated ---

gauravkothari23 said:
yes, i had measured the mosfet voltage which was around 190 volt and current to drive the motor is around 4.2 Amps.
Click to expand...
Measured with a multimeter? Mains peak voltage expected above 300 V.
 
Last edited:

Upvote 0 Downvote
FvM said:
The incredible slow MOSFET turn-on ramp is likely to cause self-oscillation. If you want to advance to usual switching speed, the slow free-wheeling diode must be replaced by a fast one.

Slow dead of MOSFETs is may be caused by continuous avalanche breakdown.
--- Updated ---
Click to expand...
i have used 1N5408 as the datasheet says

TYPICAL APPLICATIONS
For use in general purpose rectification of power supplies,
inverters, converters and freewheeling diodes application.

can i use FR107 Diode
if cannot, can you suggest some part no.
--- Updated ---

FvM said:
Measured with a multimeter? Mains peak voltage expected above 300 V.
Click to expand...
The peak voltage what i am getting is 350 to 380V
 
Last edited:

Upvote 0 Downvote
Hi
gauravkothari23 said:
yes, i had measured the mosfet voltage which was around 190 volt and current to drive the motor is around 4.2 Amps.
Click to expand...
My question was not clear.
If you have a scope, can we see a picture, especially turn ON and turn OFF timing, and voltage peaks in both directions?

Klaus
 

Upvote 0 Downvote
KlausST said:
Hi

My question was not clear.
If you have a scope, can we see a picture, especially turn ON and turn OFF timing, and voltage peaks in both directions?

Klaus
Click to expand...
no i dont have a scope.
i tested the voltage with 10M 2 watts resistor on load and keeping the mosfet continues ON. which gave me around 350 to 380V
 

Upvote 0 Downvote
Hi,

Voltage peaks will be generated by switching of the FET and stray inductance.
Even short pulses in the high nanoseconds will destroy the isolation barrier within the Mosfet. Slowly. Maybe it takes months.

Thus a "continous ON" measurement gives not much information.
Also a DVM is too slow to detect peaks.

I can not say whther this is the case. But a scope can tell. So maybe lend a scope.

Other possible solutions for the failure are
* non suitable wiring
* maybe even unsuitbale capacitors are too slow to keep peak voltages low
* bad wiring, bad PCB layout
* ESD during MOSFET handling and soldering

Klaus
 

Upvote 0 Downvote
I do not have a spice model for motor, or your MOSFET, but just R load for 2A
shows a lot of power produced in snubber.

1634039588836.png


Design methods -



What sim does not show is L effects from motor causing possible oscillation in MOSFET
which would cause a lot of Pdiss.

Without a scope you are blind to actual circuit behavior.


Regards, Dana.
--- Updated ---

Just 100 uH L causes some pretty big transients -

1634042737806.png



Regards, Dana.
--- Updated ---

No snubber, just a crappy diode to handle L transient, HV no
longer present. I recommend using at least a fast diode however.

1634044929923.png



Regards, Dana.
 
Last edited:

Upvote 0 Downvote
I would recommend, given the low frequency, that
you use an optoisolator between the control section
and the power switching section. Also that you keep
the motor current loop (incl the supply, switches
and return) entirely independent of the control
section's supplies. If you want armature voltage
feedback for speed, and maybe armature current
for I*R nulling, send those back using matched-
pair optos.

Or look at some of those fancier non-opto isolators
perhaps.

Last thing you want is armature spikes pestering your
control loop (especially if it's a digital loop which
could act really weird once your EMI reaches
bit-flip-inducing nastiness).
 

Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top