Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

help with high voltage creative idea

Status
Not open for further replies.

moshik3

Member level 2
Joined
Jun 8, 2012
Messages
47
Helped
2
Reputation
4
Reaction score
1
Trophy points
1,288
Activity points
1,788
Hi,

I have a circuit which is charging a big capacitor from 12v battery (flyback)-
the output capacitor is charged to 200-800v.
it is user selectable. sometimes it works at 220v, sometimes higher. selectable using a pot.

along with that i have a small trigger coil (with its own small capacitor) which requires the same voltage at all times to work.
it requires about 300v min to work reliably.

the basic idea would be to feed the trigger circuit from the main capacitor, and it works.
even over-powering it with 800v is acceptable.
the problems comes in when the output capacitor is at 200v, the trigger coil just doesn't get enough voltage to work reliably....

is there any really simple (and low parts count!!!) to raise that 200v to 300v just for the trigger capacitor ? (0.47uF max)

i am really desperate, and even thought of adding a stand-alone independent boost converter to make 300v out of the 12v battery -
but is is really a bad idea - more PCB space, another coil, mosfet, PWM IC, sense resistor, and some other small components which will double the parts count for me... it's really bad.

any idea, my bright friends ? :grin:
 

Dear moshik3
Hi
There are many simple ways , but it depends on how much power do you need ? if you mention it , maybe i can help you .
Best Wishes
Goldsmith
 

moshik3, you should try changing your trigger transformer coil: See if you can increase the turns ratio (perhaps it comes in a family of parts, and there is one with a higher turns ratio, or perhaps you can find a trigger coil from a different manufacturer). If this is just a one-off project, you can manually remove some of the turns on the primary (test to make sure it doesn't cause over-voltage breakdown on the secondary when you run at 800V). If this is a production product and you're really desparate, you can just take two trigger coils and place their primary coils in parallel and their secondary coils in series (be mindful of polarity to make sure they add, rather than subtract!).

There are many other solutions, so let us know if these don't work for you.

Hope this helps.
 

i am really desperate, and even thought of adding a stand-alone independent boost converter to make 300v out of the 12v battery
Can't you add a diode-capacitor voltage doubler to the existing converter, to feed the small capacitor?
 
Hi
as i wrote the current i need is minimal - just to charge a 0.1uF capacitor, every 2 seconds.

a voltage doubler will not work - the output of the flyback is not AC - it is DC pulses if i understand it right.

the trigger coil i have chosen is already the most "sensitive" (highest output at lowest input) possible.
it works good even at 200v, but for small lamps. when i use a bigger lamp it requires a higher voltage pulse. and i must have more voltage in the trigger capacitor....

i will try two coils in parallel-series, but this too, sound a bit sluggish...
the first coil will output already a few kilovolts, which is too high for returning into the primary of another coil on a PCB with 5mm spacing legs...
 

a voltage doubler will not work - the output of the flyback is not AC - it is DC pulses if i understand it right.

With flybacks, you can't do voltage doubling. However, if the flyback is configured like a normal flyback and has a buck-boost transfer function (i.e., N*D/(1-D) ), then using charge pumps it is possible to make another output which has a boost transfer function ( N*1/(1-D) ) and yet another output with a unity transfer function ( -N ).

i will try two coils in parallel-series, but this too, sound a bit sluggish...
the first coil will output already a few kilovolts, which is too high for returning into the primary of another coil on a PCB with 5mm spacing legs...

No, no, that's not what I meant! The output of the first coil should not be fed to the input of the second coil; rather, the inputs of both coils should be tied together in parallel, while the outputs of both coils should be tied together in series. (Hopefully they are built to withstand the voltages incurred; you may want to do a study on the robustness of this configuration).
 

No, no, that's not what I meant! The output of the first coil should not be fed to the input of the second coil; rather, the inputs of both coils should be tied together in parallel, while the outputs of both coils should be tied together in series. (Hopefully they are built to withstand the voltages incurred; you may want to do a study on the robustness of this configuration).

yes i got it,
but in trigger coil, those who have 4 pins and not 3 pins, the start of the secondary pin is really near the primary legs, because it should be tied to one of the primary legs. the end of the secondary is a flying wire away from the PCB,
i am afraid that if i feed the flying wire HV output of trigger coil number one into the SECONDARY start of trigger coil number two (which is located 5mm away of the primaries) it will arc with the primary, even before talking about the (doubled) output of trigger coil number two.

also, i am afraid that 800v, after all, is an overkill, which might kill the trigger coil prematurely.
so i'd really prefer a narrower voltage feed to the trigger coil. 300-600v would be great,
my probklem is how to create the 300v out of 200v....
 

but in trigger coil, those who have 4 pins and not 3 pins, the start of the secondary pin is really near the primary legs, because it should be tied to one of the primary legs. the end of the secondary is a flying wire away from the PCB,
i am afraid that if i feed the flying wire HV output of trigger coil number one into the SECONDARY start of trigger coil number two (which is located 5mm away of the primaries) it will arc with the primary, even before talking about the (doubled) output of trigger coil number two.

Ah, I get it now. They were smart about squeezing every last volt out of their transformer by connecting the primary and secondary together.

What is the turns ratio on the transformer that performs the charging? If the turns ratio is great enough, then -N*12V may be -300V. Using a charge pump, you can create an auxiliary output which provides constant voltage (or, more precisely, has voltage that varies only with the 12V power supply), and it might be enough.
 

Turns ratio on the flyback is 10....
is it possible ? how to do it ?
 

I guess you are talking about a flash lamp trigger transformer? They are designed for a specific voltage range. You also shouldn't supply the trigger transformer with a much higher voltage, because it can involve isolation breakdown on long terms.

Flyback converters can be used with voltage multiplying rectifier circuits and are quite often for HV supplies (e.g in a classical TV). But a simple doubler doesn't have considerable effect, because it only adds the most likely smaller forward voltage (about 120 V with 12 V + 1:10 transformer). More stages are needed.

But you'll need a kind of voltage regulator for the trigger circuit. Using a separate trigger converter with constant voltage may be in fact easier.
 

Flyback converters can be used with voltage multiplying rectifier circuits and are quite often for HV supplies (e.g in a classical TV). But a simple doubler doesn't have considerable effect, because it only adds the most likely smaller forward voltage (about 120 V with 12 V + 1:10 transformer). More stages are needed.

can you expand on this ? how many stages and how to wire them ?
i really prefer having 10 more diodes on my PCB than a small boost converter.

or otherwise,
do you know of any chip that can, with a low parts count, make 300v out of my 12v source ? (even integrated FET)

- - - Updated - - -

Hi i thought of the LM2577,
but i see the maximum switch voltage is 65v.
i would need to use the UC3843 PWM chip with a 400v MOSFET, right ?
no one knows of a more integrated solution ?
 

a voltage doubler will not work - the output of the flyback is not AC - it is DC pulses if i understand it right.
I was thinking of something like this. I suppose it's not exactly a doubler, but it should work as some sort of multiplier. I'd expect HV2 to be a bit more than double HV1.



You only need to add two diodes and one small capacitor to the circuit (assuming you already have the capacitor with V2 across it).
 
Last edited:
  • Like
Reactions: moshik3 and FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating

    moshik3

    Points: 2
    Helpful Answer Positive Rating
I'd expect HV2 to be a bit more than double HV1.
Yes, it's a voltage trippler, in this case generating 2*Vflyback (Vflyback = HV1) + 1*Vforward

You can design arbitary combinations if N forward and M flyback rectifier stages, with a range for M = N-1...N+1.
 

That looks great n the paper !
that means i get 400v at the trigger capacitor when my main capacitor is at 200v.
but how do i prevent it from rising further ?
(i would get 1600v when the main cap is at 800v)

can i do something with two 200v zener diodes ?
how would the wiring look like ?

Thanks !!
looking forward your next reply ! you help a lot !
 

Zener diodes can work as voltage limiter, but as a drawback, Z-diode and series resistance are creating losses, on the other hand, it most be low enough to fully charge the trigger capacitor in said 2 seconds.

A very simple solution can be found for free running periodical flashers, if you use the trigger circuit time constant to set the frequency. But for an externally triggered flasher, you preferably use a high voltage series regulator, that disconnects the charging current. It's more complex of course than a simple R-Z voltage limiter.
 

I found that one:
https://www.farnell.com/datasheets/69906.pdf
the 200v version.
i will use 2 in series to limit to 400v.
the question is, if it will handle the 1600v max ?

actually i dont need any 1600v capacitor, right ?
the small buffer capacitor of the doubler will see a maximum of 800v, and the trigger capacitor will be limited to 400v using zeners.
i just need the 2 additional diodes in godfreyl's drawing to be rated at 2kv, right ?
 

There are a few unclear points. If I understand right, the main capacitor charge process will be feedback controlled, effectively stopping the inverter when the nominal voltage is reached. At this time, the auxilary trigger circuit supply won't get any more energy. To design the dual output rectifier circuit exactly, you need to consider how flyback pulse shape (forward and reverse voltage magnitude and duration) changes during the charging process, for the different output voltage settings. This seems to be a pretty good job for LTSpice or another simulator tool.
 

well,
i have built that multiplier on the earlier schematic posted here, using a 0.047uF buffer capacitor
and it works ! i get 300v+ while the main capacitor is at ~210v.

the thing is, after the long charge of the big capacitor, the trigger capacitor gets to about 450v.,
and then the flyback is stopping, when the big capacitor reaches 210v, so the multiplier is not working anymore like during the charge,
luckily, the flyback, after reaching the desired voltage, pulses every 100ms (10Hz) to check the output voltage and re-charge it if needed (feedback is through primary current sense)
so it keeps the trigger capacitor at about 300v. (using a 100K resistor to limit charge rate)

this is a cool and cheap solution, my question is about the zener diodes,
will they hold the high voltage ?
 
Last edited:

anyone ?

how would i wire the zener diodes in the circuit schematic posted a few posts above to limit the small output trigger capacitor to 400v ?
and would it work even when the doubler supplies 1600v+ ?
 

I'm not sure what would be best, but you could try one of these:



The resistor could be something like 1M, 2Watt. Increasing the value slows the charging but reduces the power wasted. Reducing the value speeds up the charging but increases the power wasted.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top