'Terry' Filter Cutoff Adjustment / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

'Terry' Filter Cutoff Adjustment

1 2 next

Move Thread

LAN_403

Tetrafluoroethane

Wed Feb 22 2006, 07:52PM

Registered Member #127 Joined: Fri Feb 10 2006, 03:36PM
Location: Cincinnati, OH - USA
Posts: 44

This morning I sat down with a calculator and actually figured up the cutoff frequency of the filter (I probably should have done it before building the thing) designed by Terry Fritz (of which I am running a version without the MOVs). In the default configuration the cutoff frequency is almost 290kHz. When I complete the new topload for my coil the secondary resonance looks like it will be somewhere in the 265kHz region making the filter quite a bit less effective.

Now I know a bunch of coilers don't use filters at all, but as I currently only have one NST I would like to take every precaution. Has anybody had any experience increasing the values on the snubber caps to obtain a lower cutoff? This seems like it would stop some noise from getting back to the NST. Or is there something else going on here that I am missing? I couldn't find squat on Google about modifying the filter, just stuff about basic RC filter theory (of which I am under the impression I have a working understanding).

Hazmatt_(The Underdog)

Wed Feb 22 2006, 08:35PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

I've been trying to look into this for safety reasons. My capacitence per leg would be 500pF out of convienece, and for now I have been running with just protection resistors of 1500R at 50W. Terry demonstrates that with the resistors alone you can reduce current spikes, but I am not so sure how effective this is because the spike happens so fast. I think its sort of like turning on and off a light before it has a chance to get to its designed brightness.
What I can say is that with the resistors, and my old configuration of 1.6xCres, I didn't notice any drop in performance with the resistors shorted.

You could do a frequency sweep in PSpice to confirm your findings.

What I am concerned about is terminal current. So when I make the filter with the capacitors I will be looking at the current through that filter capacitor to gnd.

more to come later.
Matt

HV Enthusiast

Wed Feb 22 2006, 08:45PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

wrote ...

In the default configuration the cutoff frequency is almost 290kHz. When I complete the new topload for my coil the secondary resonance looks like it will be somewhere in the 265kHz region making the filter quite a bit less effective.

I wouldn't expect the cut-off frequency of this filter to correlate in any way with the resonant frequency of the tesla coil. The filter is designed to attenuate mostly transients kicked back from the tesla coil which will be reflected back rich in harmonics. Every time the spark gap fires, its going to kick back stuff well in to the 10's of MHz and perhaps even more.

Tetrafluoroethane

Wed Feb 22 2006, 09:48PM

Registered Member #127 Joined: Fri Feb 10 2006, 03:36PM
Location: Cincinnati, OH - USA
Posts: 44

EastVoltResearch wrote ...

I wouldn't expect the cut-off frequency of this filter to correlate in any way with the resonant frequency of the tesla coil. The filter is designed to attenuate mostly transients kicked back from the tesla coil which will be reflected back rich in harmonics. Every time the spark gap fires, its going to kick back stuff well in to the 10's of MHz and perhaps even more.

That makes sense.

I was thinking there would be some energy from the secondary getting dumped back into the tank (and thus the NST) once the main gap quenches. Won't that be at resonant frequency? I am sure the tank cap helps to absorb some of that, but it should still capacitively couple noise back onto the line, right? Or am I just thinking under a faulty assumption?

magnetomotive

Wed Mar 01 2006, 06:23PM

Registered Member #267 Joined: Mon Feb 27 2006, 09:44PM
Location:
Posts: 46

I am also looking into making a filter to protect my NST. I was going to try one like this:

It also does not have MOVs. Would not having MOVs make much of a difference in the protection? Since you already have a safety gap, why would you even need the MOVS?

When you use an NST filter, are you supposed to connect the mains ground and the RF ground together? This does not make much sense to me, but I have seen it done on the schematic for the Terry filter.

Hazmatt_(The Underdog)

Wed Mar 01 2006, 07:08PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

I usually ground everything together EXCEPT the counterpoise. I just leave that as a conductive sheet on the concrete. The transformer and potential divider are grounded and go back to the line conditioner.

magnetomotive

Wed Mar 01 2006, 07:44PM

Registered Member #267 Joined: Mon Feb 27 2006, 09:44PM
Location:
Posts: 46

Hazmatt_(The Underdog) wrote ...

I usually ground everything together EXCEPT the counterpoise. I just leave that as a conductive sheet on the concrete. The transformer and potential divider are grounded and go back to the line conditioner.

So, you don't have the RF ground connected in any way to the NST ground? I am asking because I would think that having the two connected would be just like using the NST mains ground as an RF ground, so you would risk damaging things that are plugged into your wall outlets.

Hazmatt_(The Underdog)

Wed Mar 01 2006, 10:56PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Well, I'm more worried about backfeeding high current through the NST then what's in the house. Remember that at moderate power levels like what I'm doing, I can push 10A at the bottom of the Tesla Coil. So if a chance streamer hits the primary, that will couple through the transformer back to ground, and possibly kill the NST.

I have had ALL grounds connected before with no big deal. BUT! I am going to try to scope out the differences in ground potentials and ground currents. I think this would be a good idea before giving the final verdict.

I will say that even with the RF line filter and counterpoise isolated I was transmitting enough RF to mess with the light dimmers in the house 30' away, and I was in the garage. I would think that grounding everything would have intensified the surges, but I will investigate just the same.

HV Enthusiast

Thu Mar 02 2006, 12:19AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Your RF ground should NOT be connected to the NST ground!!!!

Steve Conner

Thu Mar 02 2006, 11:08AM

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

I say that you want the terry filter ground, and the NST case ground, and the ground of any line filter you put on the primary side, all connected together to the green wire ground. This should give the best protection to the NST, against common-mode voltage spikes from primary hits and flashovers.

It doesn't seem to matter much what you connect the secondary base and strike rail to: the coil will pump craploads of interference into all apparatus around it by capacitive coupling if it doesn't do it by direct connection through the ground.

1 2 next

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