SCR troubles. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

SCR troubles.

1 2 next

Move Thread

LAN_403

Arcstarter

Sun May 31 2009, 04:57AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

I recently bought two SCRs, as some might know, from MScooling. I planned to use them in a small 624 joule capacitor bank. I hooked three of the 90 amp 1600 volt SCRs in parallel, with copper strapping. I hooked all 6 gates (i used the two series SCRs in each brick) in parallel.

I hooked a 9 volt battery up to the gate, grounded to the anode. It only pulled .4ma. I am thinking it should be more along the lines of 50-100ma each. Well, i hooked a 22 volt transformer up to the SCR, so that when the gate has a charge it will short the transformer, while i measure the current (small transformer). When i shorted the transformer through the SCR with a 9 volt battery it only pulled around .6 amps. I hooked a 44vdc supply up, and when the gate had a charge it still only pulled 1.6 or so amps. I noticed the 9 volt battery was getting hot and making a popping sound, so apparently the gates are pulling too much from it.

What could be wrong? Maybe they need more current to trigger fully? This much resistance with a 624 joule 800v cap bank would surely kill the SCR's.

Electroholic

Sun May 31 2009, 05:20AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

if you draw up your schematic, you can probably figure it out yourself.
Hint, you are shorting the battery.

Arcstarter

Sun May 31 2009, 05:32AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Electroholic wrote ...

if you draw up your schematic, you can probably figure it out yourself.
Hint, you are shorting the battery.

Well, no. I have the positive of the battery on the gate, and the negative to the anode.

Myke

Sun May 31 2009, 05:41AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

The gate source goes pos to gate and neg to cathode. Why would you ground the anode of the SCR? The anode is supposed to go to the positive end of the load (if there is one).

big5824

Sun May 31 2009, 10:15AM

Registered Member #1687 Joined: Tue Sept 09 2008, 08:47PM
Location: UK, Darlington
Posts: 240

What Myke said :) You need both the loads ground and the triggers ground on the cathode. Be careful your not putting too much voltage into the gate, the puck SCRs im using only take up to 3v and 6v, so check your datasheet if you havent already.

Electroholic

Sun May 31 2009, 03:23PM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

You are using SCRs in series, and you paralleled the gates, do you not see a problem in that?

And why are you using SCRs in series? 1600V is not enough?

As far as triggering goes, you need a series resistor to limit the current.
Battery voltage - Max Gate voltage = Max gate current * R
That would give you the minimum resistance to use.

you really should read more about SCRs in series, or maybe just SCRs in general, search the forum.

MinorityCarrier

Sun May 31 2009, 04:52PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

Your description of what you are doing is confusing and contains conceptual errors. If the battery is 'popping' it has heated up internally and is melting the beeswax used to pot it. You may have destroyed the gates, but maybe not.

The SCR is a current device, discard the idea of "charging the gate".

I think you are confusing the anode with the cathode. SCR turn-on is achieved by flowing sufficient current, from gate-to-cathode. Think of the gate-to-cathode section as a diode (more correctly, the base and emitter of a NPN transistor). Turn-on will not be achieved until the gate to cathode PN junction is forward biased, so you must have a voltage only greater than .6 volts gate-to-cathode. And then there must be a sufficient current source available anode-to-cathode to allow the SCR to go into regenerative self-bias.

For high current non-sensitive gates, there is a minimum holding current for regenerative turn on to occur.

If you are paralleleing SCR's, don't. Protect the gate with a low value resistor (47 ohms or so), and don't bias the gate with more than three volts.

Arcstarter

Sun May 31 2009, 06:58PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Wow, sorry about that :'(. I guess i was too tired to make much sense. I stated all of that backwards. I have the negative of the battery on the cathode. Also, what i meant by series SCRs is inside the brick there are two SCRs which are connected in series. I am using both in series, because one would be 800 volts and that is right at the edge of the voltage i will be using.

I am going to try some other stuff, and see what happens. I will use a different battery, mess with voltages, and other stuff.

I have parallel SCRs before with no problem. These SCRs would likely die if one was used. If it does not share current, at least one will not be getting *all* of it. I used two paralleled SCRs for a tiny 144 joule bank, and with one SCR it would die every 2-3 shots. With these paralleled, it does not die.

MinorityCarrier

Sun May 31 2009, 09:19PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

Paralleleing SCR's effectively is difficult to do. Matched turn-on and forward conductance characteristics of the paralleled devices is required or some sort of equalization network. The magnetic flux of high current switching can also interfere with shared current balance in a parallel circuit, and one SCR ends up handling most of the current, sometimes to the point of thermal runaway. Sometimes only one SCR turns on at all.

For pulse applications, you can test current sharing by putting 100A current shunts on each SCR and compare their respective current peaks.

Electroholic

Mon Jun 01 2009, 12:31AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

Just hear me out here, I know you can hook up an SCR properly, you are just missing this one little series/parallel thing. Before you do anything, can you just try using one SCR from each modules, put 3 of them in parallel, parallel the gates and see if it would switch. My guess is that it would work perfectly.
BTW, SCR modules like that are meant for bridges, so EACH of the two SCRs are rated for 1600V.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob