Author Topic: MOSFETs and flyback diodes blowing, 555 PWM circuit  (Read 13725 times)

0 Members and 1 Guest are viewing this topic.

Offline darko31Topic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: cs
MOSFETs and flyback diodes blowing, 555 PWM circuit
« on: February 23, 2013, 12:29:38 pm »
Hi guys,

I'm working on 555 based PWM controller for a 36V 600W DC brushed motor, and I'm having trouble with flyback diodes and MOSFETs.
The 555 is configured just like in Dave's video and its working without problems with little fans for computer cases. But when I hook it up to 600W motor, it works until I turn it off. After turning it on again the flyback diode blows out first, and then MOSFET, if I'm lucky enough sometimes MOSFET survives.

The MOSFET I'm using is IRF3710 and I used various diodes, the biggest one I had is 1N5408.

What I'm doing wrong?

I forgot to mention, since I don't have batteries at the moment, I'm using a 24V transformer and bridge rectifier without filter caps, if that helps somehow.
« Last Edit: February 23, 2013, 12:32:06 pm by darko31 »
 

Offline jahonen

  • Super Contributor
  • ***
  • Posts: 1055
  • Country: fi
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #1 on: February 23, 2013, 12:36:20 pm »
Try adding that filtering capacitance (try something like >1000µF for starters) to the supply rail, it is very likely that problems are due to that there is no place where energy can flow without giving rise to excessive voltage.

Regards,
Janne
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2135
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #2 on: February 23, 2013, 03:07:54 pm »
post a schematic!!!!!! :scared:
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 7551
  • Country: 00
  • +++ ATH1
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #3 on: February 23, 2013, 03:19:41 pm »
I'm using a 24V transformer and bridge rectifier without filter caps, if that helps somehow.

Isn't that voltage dangerous for the mosfet ?  ???

Fyi the absoolute max for Vgs is at +/-20 Volt.

Offline Analogtech

  • Contributor
  • Posts: 14
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #4 on: February 23, 2013, 03:52:12 pm »
jahonen is right.increase the filter capacitance on the supply rails and also your mosfet might survive better if you aren't putting 24v on the gate and if all else fails get the largest mosfet and heatsink you can and put more flyback diodes in parallel with the first one.
 

Offline rbola35618

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: us
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #5 on: February 23, 2013, 04:09:52 pm »
Without a schematic it is hard to give advice. Please post schematic. That being said, here is my 2 cents worth.

1.If you are using a flyback converter and if your flyback transformer does not good coupling you will have leakage inductance that will put a spike in the drain of your mosfet. You may have to put a snubber across the primary of the transformer. Below is a video I posted on snubbers that might help if this is the problem. You can also put a snubber across the diode.
Video Part 4


2. Make sure that you are using fast recovery diode. If you use a standard low frequency diode it will hot and short and then take out the mosfet.

3. Make sure that the transformer is not being saturated during turn on. I would do a start screen capture and then see the primary current to make sure that the start of "Toff" of the mosfet, the current surre start from 0 amps. If you see that it has some current then you know that it is saturating the flyback transformer.

Hope this helps.

Robert
 

Offline CarlG

  • Regular Contributor
  • *
  • Posts: 155
  • Country: se
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #6 on: February 23, 2013, 05:35:20 pm »
First thing: schematic please, like the others say:)

Second: do you mean freewheeling diodes instead of flyback ditto?

Third:  600W -> 30V -> 10A? 1N5408? Even with no load on the motor you could easily get more than the 3A rating of the 1N5408. The Fairchild 1N5408 DS doesn't provide single pulse power diagram, so a bit hard to determine the rating even if you know the pulse length. Do you satisfy the SOA curve of whatever diode you're using? What energy do you have to dissipate when you turn off? You have to consider both the total energy and the instantaneous power (for example the motor current * the diode Vf)

Fourth:
Can you verify that you mean what you say, i.e. are you sure that the diode blows when you start again, and not when you turn off?

Depending on the topology, you must, as said before, provide capacitance but also something that can dissipate the motor energy when you stop the driver, maybe at powerful zener from supply to GND.


 

Offline darko31Topic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: cs
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #7 on: February 23, 2013, 07:02:54 pm »
Ok, sorry for not posting schematic in the first place, my bad. It is in the attachment. Thanks for answering.

And yeah, on MOSFETs gate is 12V.

The motor itself draws at 24V with no load about 2.1A.

I cannot verify exactly when the diode is blowing, because simply I didn't had enough of them to test it thoroughly.

Ok guys, thanks for advice, I will try it out.
« Last Edit: February 23, 2013, 07:07:34 pm by darko31 »
 

Offline Analogtech

  • Contributor
  • Posts: 14
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #8 on: February 23, 2013, 07:13:14 pm »
It looks like your diode and mosfet aren't rated high enough a 600 watt motor will destroy them easily
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16386
  • Country: za
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #9 on: February 23, 2013, 07:23:12 pm »
Lose R3 and make R2 1k or less, the mosfet will then not be run as a very poor linear amplifier. You definitely need a beefier diode as D3, something like a 30A unit, and the same for the bridge rectifier supplying it.
 

Offline rbola35618

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: us
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #10 on: February 23, 2013, 07:25:25 pm »
The mosfet is rated to 57A.  I assume this would be good enough only if you have a heatsink. If you don't the thermal impedance is 62C/W(junction to air) Meaning that if you consume 1 watt of power thru the mosfet, the mosfet juction would rise to 62C.

Do you have a scope?  If so scope the drain to make sure you are not exceeding Vds maximum voltage.

For the diode, try putting another in parralell
 

Offline darko31Topic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: cs
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #11 on: February 23, 2013, 07:50:57 pm »
Yes, a heatsink is attached to MOSFET but it doesn't really get hot at all.

I'll try different power supply, properly filtered, the motor itself is supposed to be powered by batteries, but since I don't have any of them at the moment, I'm using that transformer. If it turns out that transformer is the problem, then I'll give one big face palm to myself.

And for the parallel diodes, I'll try it out but I always thought that putting diodes in parallel doesn't add more current capability.
 

Offline hlavac

  • Frequent Contributor
  • **
  • Posts: 536
  • Country: cz
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #12 on: February 23, 2013, 08:06:29 pm »
How about your power rails sequencing?
If the 24V would be still on and motor would be running and you would turn off the 12V and it would fade down slowly due to filtering caps, you would reach a point at about 6-7V where the MOSFET would be only partly on, and great deal of power would be dissipated on it, possibly burning it... but that would probably not burn the diode.
Good enough is the enemy of the best.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 10413
  • Country: nz
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #13 on: February 23, 2013, 08:30:21 pm »
I would add a transistor to drive the mosfet
I'd also add a TVS diode to protect the mosfet gate from overvoltage.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline darko31Topic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: cs
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #14 on: February 23, 2013, 08:35:04 pm »
I turn them off at the same time, both the 12V and 24V rail, and considering that the 12V is regular filtered voltage, its the 24V that is first out of power, then 12V.

Ok, I'll also try adding transistor and tvs diode if all else fails, thanks guys.
 

Offline CarlG

  • Regular Contributor
  • *
  • Posts: 155
  • Country: se
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #15 on: February 23, 2013, 08:46:50 pm »
So, without load, D3 have to dissipate all the inductive + mechanical energy of the motor that isn't dissipated in the motor windings, when you turn the motor off. And what is that energy? How high does the peak power get? Can D3 handle it? Is D3 heat sink properly designed?

If D3 blows, no wonder D1 blows.

As a simple solution, though not good, I would rather suggest that you put diodes in series (than in parallel) to lower the power per device. You get the same current but higher Vf -> more (total) power dissipation -> faster energy decay -> faster stop.

But frankly, it looks to me that you're trying to do this too simple.
« Last Edit: February 23, 2013, 08:49:42 pm by CarlG »
 

Offline darko31Topic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: cs
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #16 on: February 23, 2013, 09:23:59 pm »
Frankly I have no idea what amount of energy is going through diode when the circuit is turned off, but don't forget that is non repetitive current going through diode. According to datasheet the 1N5408 is designed to sustain non-repetitive peak forward surge
current of 200A for 8.3ms single half sine wave. That sounds plenty but its for a very short time.

But I think you're right, when there is no load, the shaft keeps spinning for a brief time, it has to be rather high current that passes trough diode for lets say 5 seconds. In practice the load would be constantly attached to the shaft of the motor and smaller diode wouldn't blow.

So the answer has to be bigger diode.
 

Offline qno

  • Frequent Contributor
  • **
  • Posts: 422
  • Country: nl
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #17 on: February 24, 2013, 09:22:07 am »
Hi,

Its the Energy.
When you stop driving the motor and it is running it becomes a generator.
If you put in 2 amps @ 24 volt the moment the mosfet switches of the you have a 48 watt generator.
Why spend money I don't have on things I don't need to impress people I don't like?
 

Offline hlavac

  • Frequent Contributor
  • **
  • Posts: 536
  • Country: cz
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #18 on: February 24, 2013, 03:50:44 pm »
Another idea, I don't see a 10 ohm resistor between MOSFET gate and DIS pin/6V divider.
When DIS slams to ground you may be getting heavy high frequency ringing on the gate.
Good enough is the enemy of the best.
 

Offline darko31Topic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: cs
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #19 on: March 02, 2013, 03:53:08 pm »
Just to confirm, yup it was the diode, I've used a 25A diode for the flyback diode (or flywheel) and everything is working great.

Again, thanks everyone.
 

Offline kallileo

  • Contributor
  • Posts: 19
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #20 on: December 25, 2013, 11:43:29 am »
Which diode have you used?
 

Offline megajocke

  • Regular Contributor
  • *
  • Posts: 191
  • Country: 00
Re: MOSFETs and flyback diodes blowing, 555 PWM circuit
« Reply #21 on: December 25, 2013, 11:13:07 pm »
So, without load, D3 have to dissipate all the inductive + mechanical energy of the motor that isn't dissipated in the motor windings, when you turn the motor off.

That is not what is going to happen. Why would it? Assuming a permanent magnetized motor:

The transistor stops conducting, the current commutates to the diode and current drops to zero while magnetic energy is converted to mechanical energy at a rate determined by motor inductance and its induced EMF (and the diode voltage drop adds in there too). At the end, the current has dropped to zero, the diode stops conducting and that's the end of it.

Energy could in the general case be returned to the source through the reverse diode of the transistor at a voltage which is below the EMF of the motor, but that can't happen here because of the diodes in the source bridge rectifier. If the motor is of the series wound type, no significant mechanical energy can be returned to the electrical side without reversing the field polarity regardless of that even.

What is usually required is for the freewheeling diode to be able to take the full expected motor current continuously, unless rated torque is not needed at low speed. So it needs to be able to handle at least 17 A. The diode should also be a fast type (which the 1N540x is not) so that reverse-recovery losses will be under control.

Care may also needed to make the power source a low impedance. With the current configuration, energy stored in the quite high parasitic inductance of the source (power line, transformer leakage inductance, wiring, etc.) will cause high voltage spikes each time the transistor switches off, bringing the transistor (or some other component) into avalanche dissipating that energy in it.

Capacitive smoothing on the output of the bridge rectifier is one way of solving that problem.

But your switching frequency doesn't seem to be much higher than 100 Hz. Why such a low frequency? Without knowing your exact motor parameters it's impossible to say for sure, but the current waveform is likely to be highly unfavorable, deep into discontinuous current mode, generating high losses.
« Last Edit: December 25, 2013, 11:18:33 pm by megajocke »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf