Author Topic: R.A.E Thyristors  (Read 1451 times)

0 Members and 1 Guest are viewing this topic.

Offline 13Topic starter

  • Contributor
  • Posts: 26
  • Country: us
R.A.E Thyristors
« on: December 27, 2019, 12:10:03 am »
Hello,
I recently purchased 6 RAE YC450PM thyristors. I plan on using them to discharge a very large capacitor bank into a load. Has anyone ever heard of RAE thyristors? The data sheet states they can withstand pulses up to 28,500 amps. Although the RAE scrs are roughly 1/2 the size of a similar spec'd AEG scr so I have a hard time believing the data sheet. I am going to test them and I'll post the results. I have attached the data sheet to the post. As you can see by the picture I have a few scrs  ;D.
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: R.A.E Thyristors
« Reply #1 on: December 27, 2019, 12:26:41 am »
Phase-control thyristors won't be happy in a capacitor discharge application unless there is significant inductance present to limit the dI/dt.

 

Offline 13Topic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: R.A.E Thyristors
« Reply #2 on: December 27, 2019, 12:55:26 am »
Yep, I have simulated the circuit to make sure my di/dt was much lower than the data sheets. I plan on using a 10uH inductor to limit di/dt to roughly 40amps/uSec.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: R.A.E Thyristors
« Reply #3 on: December 27, 2019, 01:44:42 am »
There are improved thyristor structures which support higher peak currents so it does not surprise me that they could be half the size.  But make sure to apply sufficiency gate drive because low gate drive compromises turn on capability.
 
The following users thanked this post: 13

Online floobydust

  • Super Contributor
  • ***
  • Posts: 7679
  • Country: ca
Re: R.A.E Thyristors
« Reply #4 on: December 27, 2019, 03:06:01 am »
Would pulse gate-drive would be bad compared to using DC, in this application? You can hit the gate harder without exceeding gate power dissipation, but between gate pulses the SCR is on its own.
 

Offline 13Topic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: R.A.E Thyristors
« Reply #5 on: December 27, 2019, 04:30:11 am »
To trigger the SCRs I am using this helpful guide by Semikron. I am currently using a capacitor discharged into a pulse transformer which in turn is connected to the scr gate. The pdf also has great information on choosing a RC snubber for protection.
 

Offline 13Topic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: R.A.E Thyristors
« Reply #6 on: December 27, 2019, 04:38:26 am »
Here's a picture of the trigger waveform. The larger spike is gate current where each division is 500mA. The smaller spike is voltage across the gate.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: R.A.E Thyristors
« Reply #7 on: December 27, 2019, 05:43:58 am »
At 40A/us that thing is going to take an eternity to reach peak current!  Are you sure it can even handle that, thermally*?

*Probably.  I mean, SCRs are rated for a shitton of abuse, they have fusing ratings after all.  I'm still not sure what you'd do with milliseconds of surge though.  Unless that's exactly what you're doing, a 10/1000us surge generator for example, ah, but even then the dI/dt isn't anywhere near that.


To trigger the SCRs I am using this helpful guide by Semikron. I am currently using a capacitor discharged into a pulse transformer which in turn is connected to the scr gate. The pdf also has great information on choosing a RC snubber for protection.

If you need more sustained drive, consider using AC excitation.  Pretty standard approach for phase control and inverter application.  The SCR won't much care about mild reverse bias once it's conducting, and the main penalty is a ~Fclk jitter as far as when it finally turns on (which if it was reverse biased previously, isn't going to matter, you'll hardly notice the anode voltage rise in that time).

Shouldn't be very important to pulsed applications though.  Just FYI.

Tim
« Last Edit: December 27, 2019, 05:46:54 am by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: R.A.E Thyristors
« Reply #8 on: December 27, 2019, 04:24:22 pm »
Would pulse gate-drive would be bad compared to using DC, in this application? You can hit the gate harder without exceeding gate power dissipation, but between gate pulses the SCR is on its own.

DC is fine but the problem is insufficient gate drive which slows SCR turn on time increasing power dissipation.  This can result in something similar to secondary breakdown damaging the SCR.

To extend the turn on capability, a pulse forming network can be used in place of simple capacitive discharge to both control the peak gate current and extend the trigger pulse width to turn the SCR on harder without exceeding the peak gate current specification.
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 10174
  • Country: gb
Re: R.A.E Thyristors
« Reply #9 on: December 27, 2019, 05:11:14 pm »
Don't forget the mechanical clamping requirements for those packages. Although for infrequent capacitor discharges, you will probably get away without heatsinking, they require a controlled clamping pressure to ensure proper internal contact between the contacts and the wafer. Perfectly good parts can even measure as open circuit (both ways) when not clamped.

You ought to be able to get the required clamping force from the datasheet of a similarly size part.
Best Regards, Chris
 

Offline 13Topic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: R.A.E Thyristors
« Reply #10 on: January 04, 2020, 06:15:52 am »
Hello,

Update:
My friend and I tested the SCRs with a 15 kJ capacitor bank. They worked well exceeding their specs (whoops). I have attached a picture of the resulting current waveform. To limit the di/dt a 4.2 uH inductor was used in addition to a 20 mOhm load. To trigger the SCR a pulse transformer with a 5 volt, 1.5 amp pulse was used. The current peaked around 35,000 amps.
« Last Edit: January 04, 2020, 06:17:57 am by 13 »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf