Author Topic: Thyristors problem in a 1800v capacitor discharge system  (Read 441 times)

0 Members and 1 Guest are viewing this topic.

Offline gordi

  • Contributor
  • Posts: 6
  • Country: iq
Thyristors problem in a 1800v capacitor discharge system
« on: October 20, 2019, 02:32:37 pm »
I have two thyristors R1275nc21x ->http://www.westcode.com/r1275nc14x-21x.pdf, they are Westcode thyristor and have some issues in my application.

I want to make a capacitor draining system with 1800V charged and it can discharge to the coil (8uH, 10mΩ)resonantly (with 166us duty cycle and up to 30Hz). But our thyristors(S2) burned, maybe to causes: high dv/dt or high voltage or not suitable snubber.

Of course, I used another thyristor Y89kkm -> http://www.tech-sem.com/english/upfile/201381651945246.pdf for Techsem that higher voltage than “R1275nc21x”, but it has burned too.

It is necessary to mention that Scr2 always burn.

Please guide me select right SCR and snubber or right circuit.




This is one cycle of draining 166us in 1 to 30Hz.


before connect to G K of SCR.
Excuse me, the time step is us, ms is wrong.

Of course, I trigged thyristors gates separately with pulse transformer,2 or 3 pulse every cycle (1 to 30Hz), 5 volts and 10ms Duty cycle and 20us period.
« Last Edit: October 21, 2019, 12:14:11 pm by gordi »
 

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #1 on: October 20, 2019, 03:21:54 pm »
Given that the gate is driven appropiately ( the gate is tricky in thyristors specially with this kind of currents), Id say you're exceeding the peak current, I mean those thyristors cannot withstand that peak current at 30Hz repetition.
Measuring the voltage across with a x100 probe and a battery scope is very easy and should be checked first.

Some questions:

1) how are you measuring the current?
2) does it break the thyristor at 1Hz too or only when you increase the frequency?
3) are the thyristors clamped appropriately?

I have a similar setup at 1800V but with higher capacitance capacitors (around 1000uF), and made one shot discharges over a cable with a thyristor. IIRC I got easily 10kA peak current discharges.
« Last Edit: October 20, 2019, 03:51:40 pm by MasterTech »
 
The following users thanked this post: gordi

Online MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1241
  • Country: us
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #2 on: October 20, 2019, 10:38:52 pm »
Either your gate drive is insufficient for the circuit conditions or you need to replace anti-parallel SCR2 with a fast rectifier. On paper the SCR you've chosen should handle the peak current and rate of voltage/current rise just fine (not including the repetition rate - that's almost entirely a thermal issue).

 
The following users thanked this post: gordi

Offline gordi

  • Contributor
  • Posts: 6
  • Country: iq
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #3 on: October 22, 2019, 10:03:45 am »
I think the gate driver isn't suitable and I should change it.
1-I don't measure current but I see voltage variation across the time and induction voltage from the coil to another single turn coil.
2-it has broken in 1Hz.
3-Yes I clamped SCRs with a big heat sink.



scrDriver_section1




scrDriver_section2
« Last Edit: October 22, 2019, 10:12:47 am by gordi »
 

Offline gordi

  • Contributor
  • Posts: 6
  • Country: iq
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #4 on: October 22, 2019, 10:09:06 am »
Yes I think the gate driver is't suitable for this rate of current and voltage with a big dv/dt and di/dt

Maybe this gate circuit helps me, but I'm not sure.
 t
 

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #5 on: October 22, 2019, 10:14:45 am »

3-Yes I clamped SCRs with a big heat sink.



Sorry but I don't see where the clampling force is coming from, usually something like this is used
https://theelectrostore.com/30115480-Iconopower-New/
 

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #6 on: October 22, 2019, 10:21:21 am »
1-I don't measure current but I see voltage variation across the time and induction voltage from the coil to another single turn coil.


without an accurate current measurement  you're left to whatever it happens, I had to make a coaxial shunt for this. Attached is a measurement I made, 12500A peak discharge (1.06mOhm shunt) in 300us, you can even see the hard reverse recovery of the thyristor, green trace was to get the pulse energy
« Last Edit: October 22, 2019, 10:23:16 am by MasterTech »
 

Offline gordi

  • Contributor
  • Posts: 6
  • Country: iq
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #7 on: October 22, 2019, 10:24:01 am »
I saw the site.
the clamping force is enough, I think.
 

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #8 on: October 22, 2019, 10:29:54 am »

the clamping force is enough, I think.

Certainly you cannot apply 7000kg with that setup
 
The following users thanked this post: gordi

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #9 on: October 22, 2019, 10:43:17 am »
I would suggest you use thyristor modules to avoid the use of clamps at first, there are ones rated for 2400V

« Last Edit: October 22, 2019, 10:45:10 am by MasterTech »
 
The following users thanked this post: gordi

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 456
  • Country: us
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #10 on: October 22, 2019, 12:08:12 pm »

the clamping force is enough, I think.

Certainly you cannot apply 7000kg with that setup
I agree. And those heat sinks are not strong enough.  They will bow.  there are commercial clamps that can indicate the load.   
 
The following users thanked this post: gordi

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 456
  • Country: us
 
The following users thanked this post: gordi

Offline gordi

  • Contributor
  • Posts: 6
  • Country: iq
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #12 on: October 22, 2019, 01:11:26 pm »
I read some article and now I see what you said. the clamping I used isn't enough force for my SCRs.
I wrote an email to Westcode company, they said a suitable clamping force should be 20KN.
Of course, another problem I have to pay attention to it is a suitable gate driver for SCRs. But I think  :phew: .
 

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #13 on: October 22, 2019, 02:28:38 pm »
Of course, another problem I have to pay attention to it is a suitable gate driver for SCRs. But I think  :phew: .

Don't get overwhelmed. Do the following:

-Get modular SCRs, to avoid clamps, westcode has them in 2400V rating.
-Get a x100V probe with a battery scope or a floating scope so you can probe the voltage at the SCR
-Change SCR2 for a diode as suggested before, this diode will need to be big and with a small series resistor or it will break with the reverse peak current
-The driving of the gate should be fine, start at 1Hz and low voltage rail, then keep increasing the voltage till 1800V.
 
The following users thanked this post: gordi

Offline gordi

  • Contributor
  • Posts: 6
  • Country: iq
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #14 on: October 23, 2019, 09:02:24 am »
Thanks a lot,
I can't use Diod instead of scr2, because the application has half pulse and full pulse then the app sometimes uses SCRs separately .
What's your suggestion for current measuring (IL)?
I used shunt resistors with 1.5m Ohm but in oscilloscope, I saw some peak voltage then I removed it.
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 456
  • Country: us
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #15 on: October 23, 2019, 10:40:44 am »
Thanks a lot,
I can't use Diod instead of scr2, because the application has half pulse and full pulse then the app sometimes uses SCRs separately .
What's your suggestion for current measuring (IL)?
I used shunt resistors with 1.5m Ohm but in oscilloscope, I saw some peak voltage then I removed it.

Use a pulsed current transformer.
https://www.pearsonelectronics.com/?gclid=CjwKCAjw9L_tBRBXEiwAOWVVCffqeSdA8B1Mf_0Cn9Gv2Wi22_EQL85GGQtfRf8j5t_SrYxIAhjYMxoCkEkQAvD_BwE

 
The following users thanked this post: gordi

Online MasterTech

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: 00
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #16 on: October 23, 2019, 10:54:10 am »

I used shunt resistors with 1.5m Ohm but in oscilloscope, I saw some peak voltage then I removed it.

Can't use shunt resistors because the magnetic field from the current pulse creates high distortion in the measuring resistor as induced voltage. Either a HF transformer or a coaxial shunt is needed.
 
The following users thanked this post: gordi

Online MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1241
  • Country: us
Re: Thyristors problem in a 1800v capacitor discharge system
« Reply #17 on: October 23, 2019, 06:22:01 pm »
I can't use Diod instead of scr2, because the application has half pulse and full pulse then the app sometimes uses SCRs separately.

If you need to sometimes deliver a half-wave sinusoidal pulse then SCR1 must be capable of blocking more than 2 * Vsupply, and upon rereading the datasheet I see it is actually marginal for full-wave operation (the minimum blocking voltage capability in either direction is 1400V). Needless to say, trying to block 1800V with an SCR that is rated at somewhere between 1400V and 2100V depending on circuit conditions is playing with fire... literally.

The easiest way to measure currents like this is with a Pearson transformer. It is not impossible to use a shunt, but the inductance needs to be vanishingly small (likely a coaxial type, as suggested above) and you need to add a compensation capacitor to the differential amplifier circuit with the same time constant as that of the shunt (which requires very tedious tweaking because the L/R of the shunt involves two exceptionally tiny values). See the attached pic of how to do this in LTSpice.

 
The following users thanked this post: gordi


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf