snubber diode

Thread Starter

rjjenkins

Joined Apr 16, 2011
214
EDITED - SCHEMATIC WAS WRONG

Hi
I have the circuit below - it uses a buck booster module to charge up a capacitor bank and then discharge it through a coil.
It works - except that when I discharge it, it destroys the booster module, and also the diodes. I have tried putting three snubber diodes in parallel, but the same thing happens. If I don't include the diode(s), and disconnect the booster before discharging, the SCR is fine.
I'm wondering whether it would be better to include a diode that isolates the booster from the discharge circuit - as in the second schematic. I don't want to destroy any more modules!

1701266795375.png
1701266697922.png
 

Attachments

Last edited:

Danko

Joined Nov 22, 2017
1,823
Last edited:

MrAl

Joined Jun 17, 2014
11,271
D1 but more importantly the buck booster module which supplies the 300V
Hi again,

D1 looks to be a diode that is too small for the job. The inductor kickback would be quite substantial and thus there could be a lot of current flow and/or back EMF to blow that poor little diode. I am not sure a diode is a good idea there anyway that eats up some of the energy, but then I am not sure what you are using this for either.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
214
Well the 1N4007 is rated for 30A peak (8mS) so I thought three in parallel might be enough since my capacitor discharges in about 1mS. The current should be about 100A max through the coil, though I have no idea what the back EMF and current might be
The idea is to produce a brief electromagnetic kick to lift a small coin (say) a few cm.
The SCR seems to be able to take care of itself so from that point of view D1 isn't necessary
What about my idea of diode to block the current flowing back into the booster module?
 

Ramussons

Joined May 3, 2013
1,395
Hi
I have the circuit below - it uses a buck booster module to charge up a capacitor bank and then discharge it through a coil.
It works - except that when I discharge it, it destroys the booster module, and also the diodes. I have tried putting three snubber diodes in parallel, but the same thing happens. If I don't include the diode(s), and disconnect the booster before discharging, the SCR is fine.
I'm wondering whether it would be better to include a diode that isolates the booster from the discharge circuit - as in the second schematic. I don't want to destroy any more modules!
View attachment 308650View attachment 308651
How does the capacitor charge in the first place?
D1 should be across L1, not connected to the SCR gate.
 

Ramussons

Joined May 3, 2013
1,395
Ok, The capacitor charges somehow.
What is the discharge path? With the polarities shown, the SCR cannot discharge the capacitor. Maybe another mistake in the schematic?
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
214
Well I think I have explained how the capacitor charges, several times. It's from a buck booster module.
It discharges through the coil when the SCR is triggered.
 

BobTPH

Joined Jun 5, 2013
8,665
Your circuit shows no charge path for the capacitor except through the thyristor. How is that supposed to work?

If you want help, post the exact circuit.
 

Dodgydave

Joined Jun 22, 2012
11,243
EDITED - SCHEMATIC WAS WRONG

Hi
I have the circuit below - it uses a buck booster module to charge up a capacitor bank and then discharge it through a coil.
It works - except that when I discharge it, it destroys the booster module, and also the diodes. I have tried putting three snubber diodes in parallel, but the same thing happens. If I don't include the diode(s), and disconnect the booster before discharging, the SCR is fine.
I'm wondering whether it would be better to include a diode that isolates the booster from the discharge circuit - as in the second schematic. I don't want to destroy any more modules!

View attachment 308661
View attachment 308660
Your circuit won't work correctly, how will the capacitor charge up from the negative side, there is no path from the capacitor to ground until the thyristor is fired, at that point the coil is put straight across the supply which is why it's blowing up diodes. Use the circuit posted in #post 8.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
214
Yes thanks all - the revised circuit I drew was still wrong (i.e. not what I actually had). Careless of me. Mine was pretty much the same as #post 8 but with the diode across the coil. Will try that version.
 

MrAl

Joined Jun 17, 2014
11,271
Well the 1N4007 is rated for 30A peak (8mS) so I thought three in parallel might be enough since my capacitor discharges in about 1mS. The current should be about 100A max through the coil, though I have no idea what the back EMF and current might be
The idea is to produce a brief electromagnetic kick to lift a small coin (say) a few cm.
The SCR seems to be able to take care of itself so from that point of view D1 isn't necessary
What about my idea of diode to block the current flowing back into the booster module?
In order to be sure what is going on we would probably have to have a schematic of the boost module which you said was blowing out.
One point is that if the diode(s) blow out, then there could be a huge negative voltage spike, and a series diode for the module would not help. What would help is a reverse diode right at the output of the boost module in that case, with enough size to take the bang of that voltage and whatever current. That's almost what you are doing though.

When an inductor kicks back, the maximum current will be the maximum current that was flowing right before the kickback. If there was 100 amps through the coil, then the max would be 100 amps, although it could be a bit less. However, there is no limit on the voltage theoretically, so in real life it could go very very high, 1000 volts perhaps, maybe even 2000 volts, until some insulation breaks down and allows some current to flow. The voltage across the coil at that point would be the negative of the voltage just before the kickback. That means if you put 300v across the coil before the kickback, the back EMF will be minus something, and it could be very high like that, -1000 volts or even higher perhaps, until some current was allowed to flow, and then the voltage would be reduced depending on that insulation resistance at that point in time. The problem is, if there is 100 amps before the kickback then it will be looking to keep 100 amps flowing and that means the voltage goes up to a very high negative value.

So guessing, it is probably the kickback voltage that is killing the boost module once the diode(s) blow out.
Maybe the only solution is to go to a higher power diode, like much higher, that can take 50 amps or 100 amps or so.
There is also a lot of energy associated with the kickback, so the diode has to dissipate a lot of energy in a short time.
 

Thread Starter

rjjenkins

Joined Apr 16, 2011
214
OK third time lucky, maybe. Below is the circuit as I actually have it (rather than as how I misremembered it). If I follow Danko's circuit, it should look like the second version. I don't have a 50A diode to hand but the 1N4007 (which I have lots of) is rated at 30A surge (for 8ms). Does the team think it will work? Or should I also keep the diode(s) across the coil?

(This may or may not be relevant, but before I used the booster module, I used the circuit board of a disposable camera. This worked but was slow to charge the capacitor. However, it did not get destroyed when the coil discharged, but the flyback diodes did (so I got rid of them). I don't have the circuit diagram of either the booster module or the camera so it's hard to know what to make of this)


1701337542141.png
1701337792256.png
 

MrAl

Joined Jun 17, 2014
11,271
OK third time lucky, maybe. Below is the circuit as I actually have it (rather than as how I misremembered it). If I follow Danko's circuit, it should look like the second version. I don't have a 50A diode to hand but the 1N4007 (which I have lots of) is rated at 30A surge (for 8ms). Does the team think it will work? Or should I also keep the diode(s) across the coil?

(This may or may not be relevant, but before I used the booster module, I used the circuit board of a disposable camera. This worked but was slow to charge the capacitor. However, it did not get destroyed when the coil discharged, but the flyback diodes did (so I got rid of them). I don't have the circuit diagram of either the booster module or the camera so it's hard to know what to make of this)


View attachment 308728
View attachment 308729
That first circuit looks good, but why can't you just disconnect the booster circuit with a separate switch before you press SW1 ?

Now that I see the circuit drawn like that it appears that the large current surge could be part of the problem, where the booster cannot limit the current fast enough. Disconnecting it would help that too though.

There is a chance you might be able to use a break-before-make DPST switch that will break the booster connection and then make the SW1 connection, thus doing both at once. The switch would have to be rated for over 300vdc however.

You should really have a gate to ground resistor too I think, maybe 10k or something.

Could you do a more direct mechanical setup or does it have to be just like this?

Just to note, you could replace the entire circuit with a thumb (ha ha, just kidding) :)
 

Attachments

Last edited:

Ramussons

Joined May 3, 2013
1,395
When you discharge the capacitor by triggering the SCR, you are also shorting that 330 Volt booster module.
I see no purpose served by that D4 diode in the second schematic. It is also likely to burn off due to over current during the capacitor discharge. The SCR too is likely to fail since it has no protection against voltage reversal from the Inductor.
 
Top