1:33:33 CT accuracy? / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

1:33:33 CT accuracy?

1 2 next

Move Thread

LAN_403

brtaman

Fri Feb 05 2010, 01:22AM

Registered Member #2161 Joined: Fri Jun 05 2009, 03:36PM
Location:
Posts: 247

Hello,

I have recently decided to finally transition from SSTC to DR-SSTC, one thing I still love about SSTCs is that they are nice and quiet,while DR-SSTCs are quite obnoxious and extremely loud. Though the beauty of the flowing sparks does (almost) make up for it.

Which brings me to the question, how accurate is the Ward style 1:33:33 style CT in primary current measuring? I am using Epcos N30 based ferrite torroids, which I have used at these frequencies with good results. The burden resistor is exactly 10.6 ohm (:-p 5% tolerance), measured with a calibrated DMM. Given the turns ratio, I should be getting very close to 1v=100amp on my o-scope?

The problem is that this coil

just a quick lash-up to test out my CTs and other equipment: registers in at ~4-4.5 V...making it 400-450AMP. It has had little tunning, but still I somehow doubt I am pumping that much through the poor little IRGP4PC50W IGBTs (BTW has anyone tested the limits of these little IGBT's?), the gates are being fed 24v.

Is it just a case of poor energy transfer due to inadequate tuning, still 450amp seems a bit high for the IGBTs with their 220 amp pulse rating, even with over-driven gates? The CTs wire couldn't be wound tighter on the ferrite so I doubt that is the problem..?

I have read the forums and seen references to 5% reliability, I am interested in what you members with calibrated commercial CT's who have compared them to the 1:33:33 think of accuracy.

Thanks,
brtaman

Dr. Dark Current

Fri Feb 05 2010, 11:23AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

I can only say that all overcurrent shutdown schemes I tried for my SSTC resulted in the shutdown being activated long before it should (I had ~10amp peak in the primary and OC shutdown set to 25A was tripping).
I never solved this problem even after putting the oc circuit into a shielded metallic case, though I still think it has something to do with interference (either from topload or from the primary circuit).

HV Enthusiast

Fri Feb 05 2010, 01:37PM

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

It depends on what you mean by accurate. Commercially available current monitors are usually rated at +/-1% for INITIAL pulse only as there is droop on each pulse.

As far as that 1:33:33 transformer you're talking about, its probably not going to be very accurate. However, it is good enough for what it is used for - current protection. These transformers all have narrow bands of operation, droop characteristics, etc... which all have to matched for a particular operation. I don't believe any of these were considered when Steve first made that design based on discussions he made on the forum years ago.

If you're looking for a steady-state level of current, these transformers are okay, but if you are looking for accurate representation of a single current pulse, than i would say no.

That said, if you are looking for alternative current transformers, there are some CST current transformers which are great for DRSSTCs (up to 500A) on Digikey for about $6.00 each. Part numbers are something like CST-2A, etc...

Dan

Luca

Fri Feb 05 2010, 02:23PM

Registered Member #2481 Joined: Mon Nov 23 2009, 03:07PM
Location: ITALY
Posts: 134

brtaman wrote ...

The problem is that this coil

I think that 450 Amps is quite realistic for a DRSSTC...
Your IGBT should probably withstand very short pulses (few us) with very low duty cycle (few %) of currents up to 500 amps...

Regards,

Luca

Arcstarter

Fri Feb 05 2010, 07:01PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Just a little addition to what EVR said, you could also get a current transducer. Another way to measure the current is with a shunt, but the entire tank circuit could have thousands of volts present while in operation, so hooking that to a scope could be bad. Not sure how to overcome that, but i am sure it is easy.

brtaman

Fri Feb 05 2010, 08:48PM

Registered Member #2161 Joined: Fri Jun 05 2009, 03:36PM
Location:
Posts: 247

Thanks for the replies guys!

Dr. Kv: I was also thinking interference, as it seems to be overrating current quite drastically. Then again, better to have an overrating CT than one that underrates. :)

EVR: Much thanks for the tip, I was hoping you would chime in on the thread. Those CT's look like a great deal. I suspect I will be placing an order soon. :) Have you done any comparisons between the DIY and the commercial CTs?

On the other hand, I have been doing a bit more DR-SSTC work lately and have managed to improve spark length greatly, yet the CT is still showing the same amount of current, perhaps it isn't as inaccurate as I initially thought? Sparks get larger towards the end, I am actively increasing PW in the video (right until arc-over :)).

What do you think EVR can it be said that those are 400 +-50A sparks? I know you have done a lot of studying and experimenting with DR coils (tuning guide is great work BTW). For size reference it is a small coil (form 19cm, winding 16.3cm), the sparks are ~3x Secondary winding length , I remember reading that efficiency goes way down when you start multiplying the secondary length in spark distance? Personally with really no DR background but a lot of videos watched I would say 300ish amps?

Luca: Should have been more clear in my post, but it was quite late and I was drowzy. I meant to say that based on the miniscule PW combined with lack of heating (delta of 5 degree C in a 5+ minute run) that I don't believe that I am pushing the IGBTs at the current, though I believe you are correct, they seem to be standing up the abuse quite well...the pulse width has increased greatly and they are still going strong.

Arc, shunt measurement is an interesting idea, but I think I will stick with the safety and error of the CT, +-100A. a few blown IGBTs < blown o-scope.

Best Regards,
brtaman

Dr Hankenstein

Sat Feb 06 2010, 01:27AM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

The problem is obvious. To accurately measure current, one should use only a single core CT with a burden resistor no more than 2-3 ohms. The CT really needs to be loaded down.

HV Enthusiast

Sat Feb 06 2010, 03:30AM

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

Dr Hankenstein wrote ...

The problem is obvious. To accurately measure current, one should use only a single core CT with a burden resistor no more than 2-3 ohms. The CT really needs to be loaded down.

2-3 ohms is pretty low. Most, if not all, commercial high accuracy current monitors (such as Ion Physics, Pearson, etc...) are all loaded to 50 ohms.

hvguy

Sat Feb 06 2010, 04:27AM

Registered Member #289 Joined: Mon Mar 06 2006, 10:45AM
Location: Conroe, TX
Posts: 154

I recommend either finding a friend or colleague with access to a Pearson (or the like) or using a low value shunt of known calibration (like a 100A 50mV shunt from a battery charger, golf cart, forklift, ect.). You could always make and calibrate your own shunt as well. Regardless, the best way to answer this questions is to compare your CT to a known good signal. My guess is it's working fine (+/-15%).

Dr Hankenstein

Sat Feb 06 2010, 05:09AM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

Dear Mr EastVoltResearch; There is a consideral difference between a "current monitor";as you point out and a Metering Class CT (0.3% accuracy at 1.8 ohms burden). Please refrence the Electric Meterman's Handbook 10th Edition for clarification. I know exactly what brtaman is talking about when it comes to the accuracy problems he was experiencing with his 10 ohm burden resistor as I experienced the exact same thing myself. By reducing the burden to less than three ohms with a single core-low turn count CT; I was able to improve the accuracy considerally. And further more, the 50 ohm output CT's you are refering to are designed for "voltage" output ...not current...Check the nameplate on your pearson CT fo verification. You will see that the 50 ohms is not the burden...but the output impedance.
Hank

1 2 next

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