Author Topic: Good topology for a capacitor charging power supply  (Read 4368 times)

0 Members and 1 Guest are viewing this topic.

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Good topology for a capacitor charging power supply
« on: April 07, 2019, 06:49:02 am »
What's an efficient topology for charging capacitors.
The charging voltage would be from 100 to 400V DC.
Input voltage 12 to 15V DC.
I have seen topologies that use hard inductor switching. but I'm leaning more towards Soft switching, ZVS / ZCS or resonant mode.
I've read a few threads on EEVblog regarding these and especially https://www.eevblog.com/forum/beginners/what_s-inside-a-low-power-isolated-dc-dc-converter-brick/
Where T3sl4co1l mentions there are of 3 kinds: Free running, Flyback and resonant.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Good topology for a capacitor charging power supply
« Reply #1 on: April 07, 2019, 07:19:52 am »
For what requirement ?
Cap leakage tester ?
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #2 on: April 07, 2019, 07:24:23 am »
For what requirement ?
Cap leakage tester ?

Not really, this is a general purpose power supply would charge the capacitors to a set voltage say between 100 to 400V, the charged capacitors can then be used for various pulse discharge circuits.
Basically the aim of building this power supply is to efficiently charge a capacitor with a constant current output.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Good topology for a capacitor charging power supply
« Reply #3 on: April 07, 2019, 07:39:07 am »
For what requirement ?
Cap leakage tester ?

Not really, this is a general purpose power supply would charge the capacitors to a set voltage say between 100 to 400V, the charged capacitors can then be used for various pulse discharge circuits.
Basically the aim of building this power supply is to efficiently charge a capacitor with a constant current output.
Put the word Pulse in there and immediately you're into Pulse Grade cap territory. The discharge curve will determine what cap type is required.

Pulse frequency ?
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #4 on: April 07, 2019, 07:47:24 am »
For what requirement ?
Cap leakage tester ?

Not really, this is a general purpose power supply would charge the capacitors to a set voltage say between 100 to 400V, the charged capacitors can then be used for various pulse discharge circuits.
Basically the aim of building this power supply is to efficiently charge a capacitor with a constant current output.
Put the word Pulse in there and immediately you're into Pulse Grade cap territory. The discharge curve will determine what cap type is required.

Pulse frequency ?

Apologies I wasn't clear, the intention of this question was to build a constant current capacitor charger that can charge capacitors to various voltage levels like 100V to 400VDC.
Basically I needed a efficient capacitor charger, i used to charge my capacitor with a voltage source and a current limiting resistor.
The charged  capacitors can the be sued for various other circuits / experiments. The capacitors would by polypropylene film/foil type .
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Good topology for a capacitor charging power supply
« Reply #5 on: April 07, 2019, 07:51:54 am »
Boost flyback with inbuilt current limiting if you were to KISS.  :)

I guess now to find one with the voltage rating you need.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #6 on: April 07, 2019, 08:13:32 am »
Boost flyback with inbuilt current limiting if you were to KISS.  :)

I guess now to find one with the voltage rating you need.

Thats good enough I guess but that would be hard switched? wouldn't it?
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Good topology for a capacitor charging power supply
« Reply #7 on: April 07, 2019, 08:21:09 am »
Boost flyback with inbuilt current limiting if you were to KISS.  :)

I guess now to find one with the voltage rating you need.

Thats good enough I guess but that would be hard switched? wouldn't it?
Maybe not at the higher voltages where you're out of the realms of low ESR.
Anyways strategically placed inductors should soften things up some.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 
The following users thanked this post: ZeroResistance

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Good topology for a capacitor charging power supply
« Reply #8 on: April 07, 2019, 09:14:48 am »
Is it battery powered? If not then go old school, with an variac connected to a diode bridge. Increase the voltage slowly enough to avoid blowing the rectifier or connect an incandescent lamp or inductive fluorescent lamp ballast in series with the rectifier to limit the current.

If it's a battery powered application and it doesn't have to be lightweight, then use the above, run off a cheap modified sinewave inverter.

You may want to consider an isolation transformer, especially if it's connected directly to the mains and it's mandatory if it will have an oscilloscope or signal generator connected to the secondary.
 
The following users thanked this post: ZeroResistance

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #9 on: April 07, 2019, 09:30:44 am »
Is it battery powered? If not then go old school, with an variac connected to a diode bridge. Increase the voltage slowly enough to avoid blowing the rectifier or connect an incandescent lamp or inductive fluorescent lamp ballast in series with the rectifier to limit the current.

If it's a battery powered application and it doesn't have to be lightweight, then use the above, run off a cheap modified sinewave inverter.

You may want to consider an isolation transformer, especially if it's connected directly to the mains and it's mandatory if it will have an oscilloscope or signal generator connected to the secondary.

Is not battery powered, I was thinking on the lines of a modified Mazilli ZVS oscillator with some kind of current limiting, the current limiting part is what I'm currently stuck at?
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Good topology for a capacitor charging power supply
« Reply #10 on: April 07, 2019, 10:00:37 am »
Is it battery powered? If not then go old school, with an variac connected to a diode bridge. Increase the voltage slowly enough to avoid blowing the rectifier or connect an incandescent lamp or inductive fluorescent lamp ballast in series with the rectifier to limit the current.

If it's a battery powered application and it doesn't have to be lightweight, then use the above, run off a cheap modified sinewave inverter.

You may want to consider an isolation transformer, especially if it's connected directly to the mains and it's mandatory if it will have an oscilloscope or signal generator connected to the secondary.

Is not battery powered, I was thinking on the lines of a modified Mazilli ZVS oscillator with some kind of current limiting, the current limiting part is what I'm currently stuck at?
The currrent limiting needn't be complicated. All you need is a high enough impedance in series with the rectifier to limit the current to a safe level. You could add a separate inductor or design the transformer with a high leakage inductance to limit the current.
 
The following users thanked this post: ZeroResistance

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Good topology for a capacitor charging power supply
« Reply #11 on: April 07, 2019, 11:16:52 am »
...
Is not battery powered, I was thinking on the lines of a modified Mazilli ZVS oscillator with some kind of current limiting, the current limiting part is what I'm currently stuck at?

No such thing as a "Mazilli ZVS" - that was just some overly-arrogant person sticking their name on the well-known-for-several-decades "current-fed, parallel-resonant push-pull" converter.

A topology which, btw, is singularly awful at charging up capacitors because it is not tolerant of short circuits (in fact, no parallel resonant converters are) and a capacitor looks like a short at the beginning of charging.

In addition to the ultra-simple Variac + isolation transformer idea above, the self-oscillating flyback (aka "blocking oscillator") was used for decades to charge up the capacitor in camera flashes because it is almost ideal for the task, as long as the ultimate charging voltage isn't too high and the capacitance not too large. The useful operating range can be extended by using a current mode (PWM'ed) flyback with, optionally, secondary side current limiting (primary side current limiting - automatically provided with current-mode operation - only limits primary power, same as the blocking oscillator, but that can still result in dangerously high current in the secondary rectifier).

For really big capacitors charged to really high voltages - or when you need charging rates of 1kJ/s or faster - the series resonant full bridge converter operated at a switching frequency of less than 1/2 the resonant frequency will automatically provide a constant current output and is immune to short circuits (but does not tolerate an open circuit, like any series resonant topology). Truly no offense is intended, but you have zero chance of getting such a circuit to work reliably in anything less than 1 year based on your current level of knowledge, but it would be a helluva learning experience to try.

 
The following users thanked this post: ZeroResistance

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Good topology for a capacitor charging power supply
« Reply #12 on: April 07, 2019, 11:23:46 am »
Not really, this is a general purpose power supply would charge the capacitors to a set voltage say between 100 to 400V, the charged capacitors can then be used for various pulse discharge circuits.
Basically the aim of building this power supply is to efficiently charge a capacitor with a constant current output.
Unless I am missing something this seems extremely simple. By rectifying the 230V mains you can get up to 320 V. Using a voltage doubler up to twice that. So having a capacitor charged to any of those voltages is trivial. That is your starting out DC voltage.

Now from there you want to charge a capacitor to a fraction of that voltage? Just use a resistor and a switch that cuts off when the required voltage is reached.

What am I missing?
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline ocset

  • Super Contributor
  • ***
  • !
  • Posts: 1516
  • Country: 00
Re: Good topology for a capacitor charging power supply
« Reply #13 on: April 07, 2019, 03:37:40 pm »
Your going to struggle to find a fully resonant (as opposed to a quasi resonant converter) converter to do the 12 to 300v charging.
This is  because most fully resonant converters like to have a pretty fixed vin and vout, and with a charger, the vout is changing constantly.

I would use a dual cascaded boost converter.... for that.

..you could always make it pretty well BCM switching to reduce switching losses and reduce  diode reverse recovery

Ill send some stuff on it soon.
(I had a similar situation as you when i worked at a place in South Wales)

If you want my full free SMPS course, ill send it to you...just email 
andymassey22@gmail.com

and ill send you the google drive link....

unless you know of a way i can somehow post the google drive link on here?
« Last Edit: April 07, 2019, 03:55:38 pm by treez »
 
The following users thanked this post: ZeroResistance

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
.  That took much longer than I thought it would.
 
The following users thanked this post: ZeroResistance

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 10173
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #15 on: April 07, 2019, 04:30:21 pm »
Unless I am missing something this seems extremely simple. By rectifying the 230V mains you can get up to 320 V. Using a voltage doubler up to twice that. So having a capacitor charged to any of those voltages is trivial. That is your starting out DC voltage.

Now from there you want to charge a capacitor to a fraction of that voltage? Just use a resistor and a switch that cuts off when the required voltage is reached.

What am I missing?

Given that this is the Beginners section, a mains isolating transformer. However given the hazardous amount of energy being stored in the capacitor, an easy mistake to make.
Best Regards, Chris
 
The following users thanked this post: ZeroResistance

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Good topology for a capacitor charging power supply
« Reply #16 on: April 07, 2019, 04:42:12 pm »
Unless I am missing something this seems extremely simple. By rectifying the 230V mains you can get up to 320 V. Using a voltage doubler up to twice that. So having a capacitor charged to any of those voltages is trivial. That is your starting out DC voltage.

Now from there you want to charge a capacitor to a fraction of that voltage? Just use a resistor and a switch that cuts off when the required voltage is reached.

What am I missing?

Given that this is the Beginners section, a mains isolating transformer. However given the hazardous amount of energy being stored in the capacitor, an easy mistake to make.

Yes, I guess I should have said so.

On re-reading I see I also missed the essential spec where he wants to charge from a low voltage battery so what I was suggesting is out.

Here is a post of mine where I included a couple circuits from flash chargers.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 
The following users thanked this post: ZeroResistance

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Good topology for a capacitor charging power supply
« Reply #17 on: April 07, 2019, 04:54:52 pm »
Current mode converters are automagically current limiting although you may want to configure a specific current limit.  Many switching battery charger ASICs include a separate current limit and might be amendable to using a flyback transformer reducing the voltage demands on the switch.  I think a SEPIC controller can operate this way and I know some have separate current limiting.

The more powerful configuration that I would use when charging from a low voltage DC source is a current limited buck converter driving a push-pull output transformer.  Some off-line switching regulators work this way.  Output current is continuous and the transformer operating on a current instead of a voltage is inherently balanced.
 
The following users thanked this post: ZeroResistance

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1718
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #18 on: April 07, 2019, 06:37:24 pm »
I have built similar things for medium voltage (200-400V) cap charging, and used flyback topology.  It generally lends itself better to cap charging, as its pretty much a current source out of the transformer/coupled inductor.  It can handle shorts on the output, and allows for isolation if you use primary side sensing.  There are many dedicated devices to control it, and you can control the input current, and charge time by adjusting the peak current and off-time.  It scales up nicely by I've only ever built one up to 50W.
 
The following users thanked this post: ZeroResistance

Offline Berni

  • Super Contributor
  • ***
  • Posts: 5050
  • Country: si
Re: Good topology for a capacitor charging power supply
« Reply #19 on: April 07, 2019, 07:43:20 pm »
For capacitor charging flyback topology is usually the best because it works well over a large output voltage range and doesn't mind its output being shorted.

A ZVS oscillator circuit can be used too, but current limiting it can be quite tricky. Softening of its output is likely best done by adding a inductor in series with the transformer secondary, otherwise it will try to draw a massive current if the output is shorted. The troubles of keeping the ZVS cirucit under control are only really worth it if you need a lot of power (over 50W) due to its incredibly low switching losses, or you want the circuit to produce a very clean, noise free output.
 
The following users thanked this post: ZeroResistance

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Good topology for a capacitor charging power supply
« Reply #20 on: April 07, 2019, 08:13:28 pm »
If you want my full free SMPS course, ill send it to you...just email 
andymassey22@gmail.com

and ill send you the google drive link....

unless you know of a way i can somehow post the google drive link on here?
I guess there is a "Insert Hyperlink" buttion when you try to post a reply. You could use that.
Or just directly post the link as plain text incase you don't want to go through the hassle of doing all that.

Also thanks for offering your smps course would be delighted to hear more about that. Do you have any website with more details?
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Good topology for a capacitor charging power supply
« Reply #21 on: April 07, 2019, 09:03:15 pm »
Resonant is probably not in the cards for such a wide output voltage range.  Well, actually Idunno, resonant converters can have nice output V-I characteristics.  You wouldn't get the constant-power sort of response that a boost/flyback approaches, but this doesn't make much difference in the charging rate, just a few ten percent at the low end.

What power level?

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline ocset

  • Super Contributor
  • ***
  • !
  • Posts: 1516
  • Country: 00
Re: Good topology for a capacitor charging power supply
« Reply #22 on: April 07, 2019, 10:05:50 pm »
Quote
Or just directly post the link as plain text incase you don't want to go through the hassle of doing all that.

This is the link from my google drive

https://drive.google.com/open?id=1cmk81KxXA7-EliHHWr8ZimpEp7HCMIcY

Please tell me if it doesnt work.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Good topology for a capacitor charging power supply
« Reply #23 on: April 07, 2019, 10:08:24 pm »
I've built several of the circuits this guy designed, and my own variations of them, they've worked well.

http://uzzors2k.4hv.org/index.php?page=capcharger2
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 5050
  • Country: si
Re: Good topology for a capacitor charging power supply
« Reply #24 on: April 08, 2019, 05:55:31 am »
I've built several of the circuits this guy designed, and my own variations of them, they've worked well.

http://uzzors2k.4hv.org/index.php?page=capcharger2

That is also a good way to limit the output of a ZVS oscillator by winding the primary and secondary on opposing sides of the core. This causes extra leakage inductance that makes the output sag under heavy load rather than trying to push more and more current into it.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf