drsstc half bridge / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

drsstc half bridge

Move Thread

LAN_403

Ben Solon

Sat Jan 14 2012, 11:17PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

for my drsstc, i bought a huge 450v 6800uf bus cap only to realize that steve wards drsstc .5 uses two large caps in series with the primary section in between them. stupid me. can i put my bus cap across the entire thing(dc supply to ground) and use two smaller bus caps 500-1000uf for the other section?

if you look at s wards bridge you will see two 1000u caps in series and on 10u cap parallel to that. what i propose is switching the value to a smaller cap for the two in series, and a 6800 to replace the 10u.

if i where to do that, what would the lowest value i can use be? 10u? 100u?

Daniel Kramnik

Sun Jan 15 2012, 05:32AM

Registered Member #3885 Joined: Sun May 15 2011, 12:47AM
Location: Newton, Massachusetts, United States
Posts: 94

You can't switch the 10uF film capacitor for a 6800uF electrolytic capacitor - that 10uF capacitor is there to absorb voltage spikes, not to act as a bus capacitor.

Goodchild

Sun Jan 15 2012, 06:19AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

You know you don't have to use two caps, you really only need to do that if you plan on integrating a doubler with the bridge, which you can do external of the bridge anyway....

You should read up a bit on bridges, you will notice with a DRSSTC that the primary has a series resonant capacitor while this cap is good for resonating with the primary it also blocks that DC component of the output of the bridge making the use of more film caps pointless. You can simply have one side of the tank connected to the bridge output and the other connected to the bridge negative rail.

So keep your 10uF snubbers and slap on your 6800uF 450V cap and make some sparks!

I would however consider adding a doubler or moving to a full bridge or both. This way you can have a higher voltage potential across the tank building up that current in less cycles.

Regards,
Eric

Ben Solon

Sun Jan 15 2012, 03:36PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

sorry if i sound stupid, but i have been looking at many bridges, and i dont see how that works. this that is exactly what i was talking about. i dont see how this type woks. when the top igbt switches, current flow from the supply rail to gnd. when the other igbt is on, current flows gnd to gnd? all that hapens is that the existing charge in the cap equalizes.

i found a bridge after i posted that sorta answers my question:here. aside from the 10uf snubber across the entire bridge, it doesnt help at all to have the other two? because i just pulled two 680uf 200v lytics from an atx.

so there is no reason at all to have the other two other than for a voltage doubler? i dont need a doubler because i have an autotransformer i plan to use. i just thought they where the bus/decoupling caps for a halfbridge.

what is your opinion, just connect tank to ground and be done with it?

Goodchild

Sun Jan 15 2012, 03:58PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Yep just connect the bridge as I was talking about and you will be fine. Ive used that bridge configuration before and I know it works.

Think of it this was lets say you have a 100V buss, if you use two cap design and you measured across the middle of the caps and bridge output you will see a signal that swings from -50V to 50V.

If you used the design I was talking about you will se a signal that swings from 0V to 100V. Ether way you still get a 100V potential across the LC when you switch the bridge.

I wasn't implying that you sounded stupid, I was just implying that you should do a little more research before asking questions. Sorry if it came across that way.

Turkey9

Sun Jan 15 2012, 05:25PM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

But with a single cap, won't only one switch ever have current flowing through it? Doing all the work? The point of a half bridge was originally to mitigate a magnetic offset in cores in SMPS so I guess performance wise nothing would be hurt in the TC, but I would just remove the switch that wasn't doing anything.

Also, are you grounding the negative rail of your bridge supply? I wouldn't do that if I were you. I like to keep it isolated if at all possible, and if you don't use an isolation transformer, you will NEED to have it isolated from mains ground.

In my gut, I feel that you shouldn't just hook the cap across the rails. However, I can't think of any solid reasons for this at the moment.

Ben Solon

Mon Jan 16 2012, 01:13AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

ok, so tank to rail it is. if anyone can second Turkey9's gut feeling, please do so. if not, guess we are going to have quite teh bang!

zzz_julian_zzz

Mon Jan 16 2012, 05:19AM

Registered Member #3964 Joined: Thu Jun 23 2011, 03:23AM
Location: Valenzuela City
Posts: 332

ok, first of all, goodchild's design is correct., output of the bridge (that is: IGBT top Emitter & IGBT bottom Collector) goes to the one side of the primary capacitor series in the primary winding or coil then the other side of the primary winding is connected on the negative rail of your DC supply.

but if you don't want this topology and you want to use the components you already have, another solution can be done to your bridge, that is to replace C4 and C7 with your 680uf/200v each then use a 10uf/400v on the C3; in this way, you are not doubling the rail voltage, connect the output of the bridge (IGBTs) to the primary coil then connect the other side of the coil in between of the C4 and C7,. Now to use your filter capacitor( that is the 6800uf 450v) connect the terminal to its corresponding polarity with the bridge supply and the bridge diode you have and That's it..

just make sure that your driver is working properly, phasing is correct, voltage in the bridge also correct, feedbacking circuit working, It'll be fine.

Goodchild

Mon Jan 16 2012, 06:57AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Turkey9 wrote ...

But with a single cap, won't only one switch ever have current flowing through it? Doing all the work? The point of a half bridge was originally to mitigate a magnetic offset in cores in SMPS so I guess performance wise nothing would be hurt in the TC, but I would just remove the switch that wasn't doing anything.

Also, are you grounding the negative rail of your bridge supply? I wouldn't do that if I were you. I like to keep it isolated if at all possible, and if you don't use an isolation transformer, you will NEED to have it isolated from mains ground.

In my gut, I feel that you shouldn't just hook the cap across the rails. However, I can't think of any solid reasons for this at the moment.

So if what you are saying is true then the extra switch in a synchronous buck is actually very pointless. When the top switch turns on you get 100V across the LC when the low side switch turns on the bus brings the current back down in the opposite direction. Technically you could get away with just a freewheeling diode but that's not as controllable or efficient.

Of course you isolate the negative rail from mains ground!!!! That would be subside for your bridge rectifier and transistor (if they had a chance to switch).

you could also do what zzz_julian_zzz is talking about as well, just more parts that is need is all. :P

Anyway time for sleep!

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