Portable HV supply, inverter question / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Portable HV supply, inverter question

Move Thread

LAN_403

Quantum Singularity

Wed Jun 28 2006, 05:50AM

Registered Member #158 Joined: Sun Feb 12 2006, 09:53PM
Location: Central Ohio
Posts: 282

Hello all, I have my first 'lytic powered coilgun finally in operation today. However I am currently limited to 200V with my supply. I am wanting to make a portable supply capable of charging my bank up ~800-1000V. My plan right now is to have a 12V battery powing an inverter (I have a 550W inverter right now). The output of that will drive a variac with a 140Vac output. The variac will connect to a 240:120 step down xformer hooked up in reverse. I will then make a voltage doubler using my bank as the caps in the inverter. That should provide about 800V overall.

Two questions; 1) I know inverters are not perfect sine waves, I assume there may be some spikes, will it be ok? And will this effect the variacs output?
2) I am making a control box to house all the controls and such. I was thinking to have the option of also having a power plug for a standard outlet if the battery dies or whatever reason. The outputs from the inverter and the standard outlet will be wired together and fed into the variac. My question is do I need to completely diconnect the inverter output from the line when its energized with standard AC (the inverter will of course be off in this situation, but not sure if having power on its outputs when its off is ok or not)?

Sulaiman

Wed Jun 28 2006, 08:13AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I"m not saying it will not work, but there are a few points i notice;
1)
It doesn't matter that the waveform isn't pure sinusoidal,
a 'squarish' waveform will pass through the Variac and Transformer
and the Doubler works on peak-to-peak voltage from the Transformer.
2)
A 'squarish' wave will stress and heat the Variac, Transformer and Doubler
more than a sinewave, but not too much.
3)
A common dc/ac invertor is designed to supply a fairly constant voltage at variable current
A capacitor charger ideally produces a constant current at variable voltage
A discharged capacitor initially 'looks' like a short-circuit to an invertor
So charging a capacitor from 'empty' will severely stress the invertor.
4)
I would definately NOT apply ac line voltage to the output of the invertor
Use a 2pole 2way suitably rated switch to connect the Variac (COMMON terminals)
to EITHER the Invertor OR the ac Line.
5)
I'd not use the 120:240 transformer with a doubler
Instead I'd use 120 (up to 140) Vac with a Quadrupler.
This would put less stress on the Invertor, Variac etc.
The capacitor values for the Quadrupler will determine charging time
The voltage rating for each capacitor in the Quadrupler is 2 x 140 x SQRT(2) = 400V
The Value of the capacitors will be 1uf (5W max) to 100uF (500W max)
Any capacitor type will do (Electrolytic, Paper/Oil, Mylar/plastic/PET, poly-anything etc.)
If cost is more important than reliability I'd go for 330V photoflash capacitors from used cameras
(because I've got lots of them).
Otherwise Electrolytic or Mylar give the best price/reliability choice.

Myself I'd go for a flyback invertor because
A) I like 555 mosfet/igbt invertors because they're simple and reliable
B) I have 555s and mosfets and igbts and the experience
c) It would be smaller, lighter and much more efficient.

Quantum Singularity

Wed Jun 28 2006, 06:28PM

Registered Member #158 Joined: Sun Feb 12 2006, 09:53PM
Location: Central Ohio
Posts: 282

Sulaiman wrote ...

4)
I would definately NOT apply ac line voltage to the output of the invertor
Use a 2pole 2way suitably rated switch to connect the Variac (COMMON terminals)
to EITHER the Invertor OR the ac Line.

Thats what I was thinking, wanted to make sure.

I too thought about making a 555 or similar powered flyback converter, but have yet to come across any easy designs that would allow one to charge a 1kV cap bank of over 3kJ. As far as the quadrupler, I thought that too, but a 2X transformer and doubler is probably just as easy. And in a quadrupler I have yet to see a design that would allow me to use my cap banks in the circuit, rather like you say I will have to buy more parts like photoflash caps. With a doubler I can seperate my bank into two sections in series, and use those in the doubler circuit itself. Perhaps there are other ways to configure a quadrupler? What I am going by is the schematic here. The dishcarge path of the caps is connected with diodes, and thats not a major concern, but C1's discharge path goes through the source so I dont think thats going to work. However, your standard doubler like this, the caps are direct in series and straight across the output, so the doubler would directly charge the caps with no extra caps and charge time required.

Sulaiman wrote ...

A discharged capacitor initially 'looks' like a short-circuit to an invertor
So charging a capacitor from 'empty' will severely stress the invertor.

Well of course I am going to use some current limiting resistors, that wasnt my question. Also with no current limiting resistors youd probably blow a fuse of trip the overcurrent protection in the inverter if the voltage was brought up too quick with the variac.

...

Wed Jun 28 2006, 07:58PM

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

A good solid state bank charger? One word; ZVS

You could use an old flyback core, or a core out of a computer supply, a pair of irfp250's (way overkill) a 3+3 turn primary, about 1ufd tank capacitor, 100mh series inductance, and about 300 turns (or so, you need to experiment to see how much it will fall under heavy load conditions)... It is self limiting, light, simple (I count about 10 components), etc... The only problem I see would be that you can't use your variac to control charging... A comparitor on the bleeder resistor connected to a fet in series with the zvs would solve that, and give you more accurate/precise voltage control

Marko

Wed Jun 28 2006, 08:42PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Two questions; 1) I know inverters are not perfect sine waves, I assume there may be some spikes, will it be ok? And will this effect the variacs output?

Inverters usually put out a bit 'deformed' sinewave (chopping is suare) wich is result of transformer's leakage inductance. It proved to be as good as normal mains and I never saw people having trouble with inverter-powered stuff (transformers, mtors, etc.).

You can safely connect variac or anything you like to the output.

2) I am making a control box to house all the controls and such. I was thinking to have the option of also having a power plug for a standard outlet if the battery dies or whatever reason. The outputs from the inverter and the standard outlet will be wired together and fed into the variac. My question is do I need to completely diconnect the inverter output from the line when its energized with standard AC (the inverter will of course be off in this situation, but not sure if having power on its outputs when its off is ok or not)?

Yeah, you should. You would basically induce current in it's transformer (reverse) and it may damage the inverter.You could use a relay wich switches inverter output to mains plug, when 12V supply is out or etc.

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