CO2 Power Supply Schematic Help

[DashApple]

First off thank you guys so much for putting up with my noobishness and it is really great to have some teachers even if on a forum.


Ok so basically stick the 10kv diodes in series to triple their rating

Yup , the diodes are all ready made from multiple series diodes inside the one package , series 3 "diodes" will give you a 30Kv diode .

It's Easyer to string lower voltage diodes together , I did the same thing with 600v diodes on my 2Kv transformer

 

[The Lightning Stalker]

For the capacitor, can't exactly find many good priced capacitors in the nf range rated for 30kv. Would making my own work and be efficient enough?

That is the problem with doing it the
"easy" way. With mains frequencies, you
need big expensive capacitors. Just be
thankful you live in an area with 60Hz
mains. Most of the world uses 50Hz, making
the problem even worse. You could get 2
sets of these and parallel them to get
enough capacitance.

015 uf 30000 WVDC Capacitors 15nF 30KV High Voltage Quantity 9 | eBay

Or keep looking. Or you could make your
own. With 2mm thick glass, it would only
take about 15x15cm area to get to 100nF,
though it would depend on the glass.

 

[Sigurthr]

^ Which is why it is a LOT easier to use a HF drive and a flyback style transformer. Either standard DC, or a custom/DIY AC one followed by a cascade multiplier. Either way you need to smooth the DC pulses, and the capacitance value at 60KHz is much easier to achieve.

 

[Cyparagon]

The doubled open circuit
voltage is twice that, which means they
will need to be tripled up.

It's not doubling the RMS voltage, it's doubling the peak voltage, so the final output will be 34kV, and you'd therefore need each rectification stage to be at least 40kV rated.

 

[The Lightning Stalker]

It's not doubling the RMS voltage, it's doubling the peak voltage, so the final output will be 34kV, and you'd therefore need each rectification stage to be at least 40kV rated.

Aha, yes, good catch. I the schematic has
been updated.

 

[upaa27]

^ Which is why it is a LOT easier to use a HF drive and a flyback style transformer. Either standard DC, or a custom/DIY AC one followed by a cascade multiplier. Either way you need to smooth the DC pulses, and the capacitance value at 60KHz is much easier to achieve.

Wouldn't the caps smooth the output?

 

[The Lightning Stalker]

Wouldn't the caps smooth the output?

Yeah, he's just saying that at higher
frequencies, the caps can be smaller
because the pulses occur more often keeping
the voltage on the caps topped off. Using
Sigurthur's example, at 60kHz you could get
away with even 500pF caps without a problem.

 

[upaa27]

I think I will use wax paper and aluminum foil (5 meters) for around 6.5uf while still being able to take the voltage load then rol it up.

Will this work without jumping plates?

 

[Sigurthr]

You have to make the plates smaller than the dielectric so that the cap won't arc over. You also have to place the lead-ins away from each other.

Do watch for dielectric heating and corona. It'l take out a cap quickly.

 

[upaa27]

You have to make the plates smaller than the dielectric so that the cap won't arc over. You also have to place the lead-ins away from each other.

Do watch for dielectric heating and corona. It'l take out a cap quickly.

Will do. I might go with glass and aluminum plates now due to the fact that they will be easier to set up and should be more resilient to corona. To bad that they will be big

 

[Sigurthr]

Glass works excellently. When I did a DIY plate cap I used glass and heavy duty aluminium foil. It's just heavy, expensive, and fragile. I tried PP too, but it eventually failed from dielectric heating.

 

[upaa27]

Glass works excellently. When I did a DIY plate cap I used glass and heavy duty aluminium foil. It's just heavy, expensive, and fragile. I tried PP too, but it eventually failed from dielectric heating.

How would an electrolytic cap work? I would basically soak paper towels in baking soda water, put plates on either side, roll and put into a jar full of the baking soda water.

Would that be able to take the voltage load without using a ridiculous amount of aluminum foil?

Only thing that I can see that might go wrong is the polarity might interfere with the circuit

 

[Sigurthr]

The problem is the varying levels of moisture in the electrolyte and induced electrolyte breakdown and hydrolysis. Just stick with a dry cap.

 

[Cyparagon]

Also, electrolytics aren't even commercially available much higher than 500V, so there's no way in hell you're going to DIY a 20kV one. I'd just buy a pair if I were you... or omit them entirely.

 

[upaa27]

I might just go with the zvs flyback circuit as it is a lot easier to hook up(not using massive cables) and is much smaller than making huge caps. Only problem is the fbt as It looks a bit harder to acquire.

Pump the flyback into the nst then voltage double it using the 2200pf caps

 

[swamidog]

i understand the joy of DIY, but don't overlook the option of getting a commercial psu:

40W~60W PWM CO2 Laser Power Supply (AC110V)

commercial psu's often have interlocks so they won't fire the laser if there are ground problems or if your water cooling isn't flowing. very important!