Causes of NST Failure, possibility to troubleshoot? / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Causes of NST Failure, possibility to troubleshoot?

Move Thread

LAN_403

Energyandfrequency

Mon Feb 11 2013, 03:24AM

Registered Member #9130 Joined: Sat Dec 29 2012, 03:31PM
Location:
Posts: 17

Hello everyone!

My 15kV 30mA Franceformer NST has failed, and I am trying to figure out exactly what caused the failure and if it is possible to fix it.

I was messing around with my saltwater capacitor array (6nF, unknown voltage rating, guessing 25kV, for beer bottles) and hooked my cap bank in series with a spark gap... it fired across the spark gap and then after about 30 seconds, it stopped arcing and I cannot get an arc out of the secondary anymore.

I am having a hard time figuring out why it did this. Overcurrent/overvoltage wouldn't really be possible with this setup, would it? The NST is current limited... also to my knowledge there is no ground fault protection on it. I have run a jacobs ladder multiple times off this transformer without a problem.

Also, I popped the lid off to check the insides, and there are cracks in the resin. If I reheated the resin could it heal any shorting in the secondary? Maybe? ;)

Thanks!/Haley

Graham Armitage

Mon Feb 11 2013, 03:51PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

I have had this problem number of times and it can be very frustrating. Most of the time I only see one side of the ST fail. my understanding of this problem is that when connecting a cap (especially impedance matched with the NST) would will get voltage spikes across the secondary that exceeds the limits of the secondary insulation. Running a Jacobs ladder does not create these kinds of voltage spike as there is no resonating signal. Here is an example - http://www.hvtesla.com/safety_gap.html

The way to protect against this is to create a 3 point safety spark gap with the center electrode grounded. That way when voltage rises above safe level on either side of the secondary it arcs across the safety gap and not inside your NST. You will have to keep the tank circuit spark gap set lower enough to prevent safety discharges. Means your tesla output maybe a little lower, but your NST is safe.

I have heard about heating them to try and recover a blown one, but have never tried.

Steve Conner

Tue Feb 12 2013, 03:05PM

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

Did you have the NST connected across the cap bank or the spark gap? Either way, it was probably overvoltage or high RF voltage that killed it. A couple of large resistors in series with the NST output leads, as in the Terry Filter, can help. They damp the RF oscillations without absorbing too much of the line frequency power.

...

Tue Feb 12 2013, 03:37PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

As far a heating the NST to revive it, I had success with this once (I set it on a hotplate until the tar was liquid, then let it cool) although afterward it seemed more prone to failure than before the initial failure.

Like others have said, it is critical to have a safety gap!

Energyandfrequency

Thu Feb 14 2013, 04:53AM

Registered Member #9130 Joined: Sat Dec 29 2012, 03:31PM
Location:
Posts: 17

Thank you very much everyone. Gonna try the re heat before I get another one. One question- if I had just my cap bank connected in series with a spark gap to my xfmr, could there have been over voltage? Would the cap bank/spark gap caused a resonance situation to develop with the xfmr secondary? I was reading in one of the replies that over voltage could have messed up the xfmr and I wanted to clarify where that could be happening. Normally a capacitor connected to an AC voltage source wouldn't ever store the peak energy because of losses, right? But adding the inductive element of the xfmr to the mix could create a series resonance between that and the cap bank and sprk gap that I tested?

GrantX

Thu Feb 14 2013, 07:09AM

Registered Member #4074 Joined: Mon Aug 29 2011, 06:58AM
Location: Australia
Posts: 335

Hello, sorry to hear about your NST. I'm trying to figure out how the secondary failed while using only a capacitor bank and a spark gap. I've attached a shoddy MSpaint diagram, which shows the different ways a cap and spark gap could be configured. In the first instance (1.) over voltage can only occur if the capacitor is resonant with the leakage inductance of the transformer, if the capacitor is smaller than resonant, then it will simply limit the current flowing across the spark gap. In the second situation (2.) then the capacitor will charge with each half-cycle, until it reaches a high enough voltage to ionise the spark gap and dump all of it's charge. These sharp discharges can create a potentially damaging RF current through the transformers secondary, which is why a Terry Filter (or similar low-pass filter) is used in Tesla Coil duty. The third example (3.) is identical to number 2, and requires some form of filtering to protect the transformer's insulation.

You said you had the capacitors and spark gap in series (like example 1.), which would only cause damage if the capacitor was resonating with the secondary. However it is quite obvious when an NST is resonating, as it produces very violent arcs and increased vibrations/humming: NST with resonant cap

If I recall correctly, a 20nF capacitor is almost resonant with my 15kV/60mA NST. If you are using a 15kV/30mA NST, then I think approximately 10nF should be near resonance (the limiting inductance is doubled, so the cap value needed to resonate is halved). It is possible that your 6nF cap bank is getting close to dangerous territory.

Steve Conner

Thu Feb 14 2013, 11:57AM

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

Values smaller than the theoretical resonant value can also be dangerous, because the NST's inductance is non-linear. The higher the secondary current, the lower the inductance, which allows it to "lock in" to resonance with a wider range of capacitances than you might expect. This is sometimes called ferroresonance.

GrantX

Fri Feb 15 2013, 03:28AM

Registered Member #4074 Joined: Mon Aug 29 2011, 06:58AM
Location: Australia
Posts: 335

Steve Conner wrote ...

Values smaller than the theoretical resonant value can also be dangerous, because the NST's inductance is non-linear. The higher the secondary current, the lower the inductance, which allows it to "lock in" to resonance with a wider range of capacitances than you might expect. This is sometimes called ferroresonance.

Thanks, I just did a bit of reading on ferroresonance, now I can say I learned something new today!

Is this why I've heard people say you should choose a 'larger than resonant' capacitor when running an NST powered SGTC? As in, a smaller than resonant capacitor could still cause damage, so it is safer to select a capacitance which is far beyond resonant value?

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