Protective diode? / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

Protective diode?

Move Thread

LAN_403

andy73

Thu Sept 13 2007, 10:12AM

Registered Member #926 Joined: Thu Jul 26 2007, 04:09PM
Location:
Posts: 9

Hi all, i have my first CG working but i have run into a problem with where i need to place the protective diode, oh and what kind of diode, i have some 400v 3A ultrafast recovery and some 400v 10A ultrafast ones too..

anyway so far i can fire a 5 gram projectile about 8 meters with the set-up in the pic, but when i try to up the wire grade the SCR dies, ok back current from the coil, so i tried a diode and i have so far managed to kill 5 SCR's so my questions are....

where should i put the protective diode? and which way round? if someone could quickly draw the answers on the attached pic it would save the UK's suppy of fresh SCR's and help me out a lot.

once i have the answers i will add the pic to HvWiki so that others dont become serial SCR killers too!! lol

cheers all, oh and are the results im getting any good, details below

fire (muzzel) height: 88 cm, distance of P: 7.5 - 8 meters, projectile weight: 5 grams, cap bank: 4 x 330v - 80uf caps in para,
charge circuit: 2 x cam flashers in para, SCR: 400v @ 8A

cheers

Barry

Thu Sept 13 2007, 01:07PM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

Good job, sure looks like fun!

The protective diode is added in parallel to the coil. See this inductive kickback article.

I suggest using a couple diodes in series, all in the same orientation, alongside the coil instead of just one diode. Using several diodes (up to ten) will help the coil current turn off faster.

Cheers, Barry
There's no place like 127.0.0.1

Steve Conner

Thu Sept 13 2007, 01:18PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I spotted another mistake in your schematic: the SCR anode and cathode are mixed up, and you have no resistor on the SCR gate to limit the gate current. Either of these would explain your SCR deaths, if the mistakes are also in your real circuit.

andy73

Thu Sept 13 2007, 01:52PM

Registered Member #926 Joined: Thu Jul 26 2007, 04:09PM
Location:
Posts: 9

hi all,

cheers for the info, Barry (great site by the way) do you mean like this (see pic) if yes then here's a strange one, that way round is fine with 18 swg+ wire but on 22swg - the SCR gose....bang!!!! but on 22swg, no diode it's fine lots of firing and fun:-)

Steve, this is where it gets very strange.... the C and A are that way round according to the datasheet, and i am using a single mini N size 12volt bat to trigger the gate, as you said with no resistor but the coil i am getting good results from can fire time and time again, so far over 300 shots no SCR murdering, but as soon as i up grade the wire to 18SWG it gose bang!! maybe i will stay with 22SWG as it works fine.

i think the conclusion is dont change it!! go with 22swg wire and stop killing SCR's

OZZY

Thu Sept 13 2007, 06:29PM

Registered Member #511 Joined: Sat Feb 10 2007, 11:36AM
Location: Somerset UK
Posts: 55

I think that we need more information to solve this problem.

1.Gate drive.
You really should have a resistor in series with the gate. However if the bettery is very small then its internal resistance could be limiting the current to a safe level. It would be a good idea to measure the current in the trigger circuit and compare this with whats recommended it the data sheet for the SCR.

2.Current.
The 18SWG coil will have much lower resistance and probably lower inductance than the 22SWG coil, this means that the current will be much higher and the pulse much shorter. It would be a good idea to estimate the peak current by using a circuit simulator like Pspice or barrys RLC simulation. All the tools you need are on barrys website.

3.Wire and SCRs.
Changing from 22SWG to 18SWG is a massive jump, the 18SWG coil could be drawing over 1000A. 22SWG sounds about right for the size of your cap bank. An 8A SCR sounds a bit small, what is the pulse rating for this device? I would be tempted to use a 25A SCR because they are relatively cheep and easy to get in the UK.

4.Diodes.
People refer to protection diodes but what are you trying to protect and from what? Consider the circuit without a diode, if the circuit is overdamped then the current will decay exponentially, there will be no oscillation and no diode is needed. If the circuit is underdamped then it will oscillate, however the SCR will only pass current in one direction so the circuit will perform one half cycle. It will stop when the current reaches zero and the capacitor is negatively charged, this does no harm to the SCR at all but could be very damaging to the capacitors. A diode across the coil will protect the capacitors from being negatively charged. The diode must be able to withstand the full coil current (another reason to work out the peak current). If the diode should fail at a high current (they often explode) then the coil could generate a massive voltage spike that will destroy the SCR.

andy73

Thu Sept 13 2007, 08:04PM

Registered Member #926 Joined: Thu Jul 26 2007, 04:09PM
Location:
Posts: 9

cheers Ozzy, i cant find the data sheet for the bat (12v) but im pretty sure its way below the SCR threshold, but as surrgested i will up grade to a 25A SCR.

i am playing with good old Barry's RLC sim, so ill get back with the peak current specs, but i hink its around the 250 - 300A mark??? but ill get back to you all on that one.

andy73

Sat Sept 15 2007, 04:10PM

Registered Member #926 Joined: Thu Jul 26 2007, 04:09PM
Location:
Posts: 9

hi all, ok iv got some more data, my peak current is around the 207 A mark and the new SCR's rated at 300A peak should be fine.

the gate drive circuit is tiny with no real current, so the SCR's are fine from that side, although i will add a resistor as a back-up safty measure.

after playing around with the coil design i am now getting my 5 gram projectile to hit the wall at the end of my hall which is exactly 9.25 meters away, it is hitting about 15cm front the ground, so all is well, i will get some pics up soon.

cheers all for the help

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