Inductive or capacitive ballast for big transformer / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Inductive or capacitive ballast for big transformer

Move Thread

LAN_403

GrantX

Sun Jul 26 2015, 04:15PM

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

Hey everyone,

I'm wanting to build a new ballast for my large HV power supply project. The power supply is based around a dry-type 15kVA, 4.16kV utility transformer made by GE, with 230V 50Hz imposed on the two 120V coils in series. The final design will include three lighting ballasts (6A each at 230V, for a 6A - 18A variable ballast) for low power stuff, so now I need to build a much higher power choke. The range I had in mind is roughly 16mH - 18.5mH.

I've got a nice large MOT core, with bolt holes on all four corners, which will make it very easy to reassemble with an airgap. I've also got the bare core from a 150-250VA transformer (based on its size), the laminations are interleaved with no airgap.
Starting with the MOT, I measured the winding window and I think I can (optimistically) squeeze in about 83 turns of 2.5mm^2 insulated wire. After measuring the inductance of the primary with the secondaries shorted and counting the turns it seems my 83 turn coil would end up around 19.8mH with no airgap. The core will obviously saturate without a gap, and I'm worried the airgap may end up reducing the inductance below any useful value.
I had an idea for increasing the inductance without needing a larger core, but I'm not entirely confidant that it will work. I was thinking about reducing the number of turns of 2.5mm^2 wire and squeezing on a secondary coil consisting of a few turns of thick wire. The smaller iron E core I mentioned earlier will have a few turns of thick wire wound onto it, and will be connected across the ballast's secondary coil. Essentially, turning the ballast inductor into a transformer with a gapped core.

Is this feasible for intermittent duty? The power supply will be mainly fed by a 2.5mm^2 line and 20A circuit breaker, but operation from higher-powered connections is likely in the future.
Here's a schematic showing the two ideas:
1437926651 4074 FT1630 Inductive Ballast

1437926651 4074 FT1630 Inductive Ballast

Here's a few pics of the MOT core and its smaller brother:

And some pics of the 135kg transformer in question:

Wolfram

Sun Jul 26 2015, 05:43PM

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

Adding the secondary and the second inductor to the MOT is equivalent to increasing the air gap, so there's no real benefit to doing it. No free lunch to be had there.

A single MOT core contains way too little iron to reasonably ballast such a large transformer, at least if you want any significant power for a significant time. More iron is the first order of business, an old welder would be a step in the right direction.

GrantX

Sun Jul 26 2015, 06:57PM

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

Haha, yes. I guess I rushed through it and failed to see that error. Loading the ballast with a secondary coil and inductor will decrease the impedance seen on the primary circuit rather than increase it (just like shorting out the secondary of an MOT when using the primary as a ballast). Oops.

The other compact and affordable option is capacitive ballasting. It seems to be best to place the capacitor in series with the HV coil, rather than the primary. A string of motor run capacitors would probably be easiest. But I do worry about the chance of voltage spikes damaging the transformer's insulation, despite the relatively low voltage.

GrantX

Sun Aug 02 2015, 03:02PM

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

I had a chance to search through my workshop, and have come up with 2 solutions. The first one would involve removing the coils from a bunch more MOTs, rewinding them with 4mm^2 cable and placing them in series (with air-gaps in the cores of course).

However, it would be easier and cheaper to make use of some of the capacitors laying around. I have enough to make various rugged HV capacitor arrays, up to about 6uF max, with sufficient voltage and current headroom. My first choice would be lots of identical motor-run capacitors in series along with bleeder resistors, like these:

First of all, I'd like some advice on the do's and dont's of capacitive ballasts and safety in particular, since this is a lot of power (the capability for 24kVA+). I'll be wearing a ton of PPE along with an experienced friend and all the relevant fusing/breakers. The motor run caps are easy to mount, so I think I'll construct some sort of blast shield out of lexan sheeting. Is this a sane option, or should I continue with the inductive ballast? My main concern is the capacitors suddenly shorting out and letting through 50+ kVA for a moment, or failure of the transformer's epoxy insulation (which seems unlikely).

Another big concern is versatility. A capacitive ballast in series with the HV coils may not be suitable for all loads, but then again it may not become an issue. Is there anything in particular that could cause damage to either the transformer or caps?

Ash Small

Sun Aug 02 2015, 04:29PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Resonant rise?.....What's the inductance of the secondary?

GrantX

Mon Aug 03 2015, 02:40PM

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

Ash Small wrote ...

Resonant rise?.....What's the inductance of the secondary?

Just measured the leakage inductance of the HV coil tonight. Roughly 134 mH, with the primary shorted. So I guess something in the range of 75 uF, in series with the HV coil, would resonate (~42 Ohms @ 50 Hz). I was planning on using 2-4 uF, with a bank of microwave oven caps (or the lighting ballasts) for low power and a large bank of motor-run caps for higher power.

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