CT phase shift?? / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

CT phase shift??

Move Thread

LAN_403

flyboy7777

Sat Dec 19 2015, 12:41PM

Registered Member #58118 Joined: Sat Dec 12 2015, 12:34AM
Location: Sanger, Texas
Posts: 14

I'm just curious about this... I have a DRSSTC that I'm in the middle of building, and I noticed something odd about when I tuned the phase lead. It seems that when I finally get the phase lead inductor tuned to the point where I get minimum ringing, the phasing between my primary current and bridge output doesn't really look like it's right at 'zero crossing'. I'm curious about why this would be. I was thinking maybe the difference in burden resistance? I'm using a goodchild 2.5 driver board, and the standard 51ohm internal feedback burden, versus my CT for "current sniffing" on my scope that I have at 10 ohms. Using a full bridge of CM300. Any ideas? I want to make sure I have this thing tweaked just right before I go too crazy on power levels. Right now its only a 170 volt bus, but planning on voltage doubler for 340 volts.

Thanks!

Jonathan

Wolfram

Sat Dec 19 2015, 03:33PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

You get a minimum of ringing on the bridge outputs when zero voltage switching is happening. To get zero voltage switching, you need to switch the bridge before current zero crossing, so that there is some current to commutate the bridge output voltage to the opposite rail during the dead-time. Ideally you want to switch early enough that you get good ZVS, but not much earlier than this. What you're observing makes sense. It is still not established if this is the optimal time to switch in a DRSSTC, but it's the best candidate as far as anybody knows.

flyboy7777

Sat Dec 19 2015, 03:49PM

Registered Member #58118 Joined: Sat Dec 12 2015, 12:34AM
Location: Sanger, Texas
Posts: 14

Interesting. Thank for the reply.

Hydron

Sat Dec 19 2015, 05:25PM

Registered Member #30656 Joined: Tue Jul 30 2013, 02:40AM
Location: UK
Posts: 208

You can get phase shifts in CTs, I've done some frequency sweeps to look at it, an example below:

loneoceans

Sat Dec 19 2015, 09:01PM

Registered Member #4098 Joined: Fri Sept 16 2011, 09:26PM
Location:
Posts: 236

flyboy7777 wrote ...

I'm just curious about this... I have a DRSSTC that I'm in the middle of building, and I noticed something odd about when I tuned the phase lead. It seems that when I finally get the phase lead inductor tuned to the point where I get minimum ringing, the phasing between my primary current and bridge output doesn't really look like it's right at 'zero crossing'. I'm curious about why this would be. I was thinking maybe the difference in burden resistance? I'm using a goodchild 2.5 driver board, and the standard 51ohm internal feedback burden, versus my CT for "current sniffing" on my scope that I have at 10 ohms. Using a full bridge of CM300. Any ideas? I want to make sure I have this thing tweaked just right before I go too crazy on power levels. Right now its only a 170 volt bus, but planning on voltage doubler for 340 volts.

Thanks!

Jonathan

How did you construct your CT? It is possible that your signal is phase shifted because of the loop formed by the scope probe and and lead especially since it's a high impedance at the scope side. Proper current probes have a BNC right out of the probe.

Goodchild

Tue Dec 22 2015, 04:54AM

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

Wolfram wrote ...

You get a minimum of ringing on the bridge outputs when zero voltage switching is happening. To get zero voltage switching, you need to switch the bridge before current zero crossing, so that there is some current to commutate the bridge output voltage to the opposite rail during the dead-time. Ideally you want to switch early enough that you get good ZVS, but not much earlier than this. What you're observing makes sense. It is still not established if this is the optimal time to switch in a DRSSTC, but it's the best candidate as far as anybody knows.

This type of drive doesn't allow for ZVS due to the lack of dead time between high and low side switches that would normally allow commutation to happen. Rather the reason phase lead is implemented is this type of driver is to "pre-switch" the IGBTs to allow the IGBT to fully turn on or off before the zero current crossing so that the transition happens right at the zero crossing point in an attempt to achieve optimal conditions for ZCS. This phase lead also weeds out all other propagation delays in the drive and feedback that may otherwise cause the IGBT to turn on late.

If you are seeing a slight amount of lead this is normal and exactly what you should see if the lead is set correctly. The LR Burdon network creates a voltage that slightly leads the current in the tank. This lead in turn translates to early transitions in switching of your IGBTs. The turn-off or on should slightly lead the crossing event.

If you are seeing a large amount of lead you may want to look into what loneoceans mentioned as a possible source of phase error.

flyboy7777

Mon Dec 28 2015, 01:22PM

Registered Member #58118 Joined: Sat Dec 12 2015, 12:34AM
Location: Sanger, Texas
Posts: 14

Thanks for the replies. In response to the question about my CT... It's the standard cascaded 1:33:1:33 configuration, with a 10 ohm burden right at the transformer. There's a 3 foot piece of shielded 2 conductor twisted pair Milspec Tefzel wire attached. Terminated to banana plugs, going through a banana-BNC adapter on the o'scope.

I've actually run this coil now, and It's doing 6'+ sparks. I'm pretty pleased with it!

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