Searching \ for 'OT 2kVolt from battery' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power/batterys.htm?key=battery
Search entire site for: 'OT 2kVolt from battery'.

Truncated match.
PICList Thread
'OT 2kVolt from battery'
1999\09\06@152126 by Matthew Ballinger

picon face
   I'm working on a little project right now that requires a 2kV source
current limited to about 3-4 ma and powered from 4 AA batteries. At the
moment I'm using a PIC for PWM to a logic level power mosfet which chops
power to the secondary a mini audio z matching  1:12 xfrmr which then feeds 4
stage cascading voltage multiplier. This setup is very compact (also
required) and  gets me around 1.6kV, but current is much to low. Then I used
2 xfrmrs in parallel, but still to little current. I'm also using a 470uF cap
to stiffen up the power supply some which helped somewhat. Anybody got ideas
of a better way to do this? Any help would be greatly appriciated.
Matt B

1999\09\06@154645 by Jason Sachs

flavicon
face
2kV * 4mA = 8W output!

I'm guessing it's the transformer that's limiting you.
Either the magnetics can't handle that much power, or
the resistance of the transformer winding is killing you,
or both. (also, I don't know how much the batteries
would sag, but I'm guessing even if you were lucky enough
to get 80% efficiency, I'm guessing the AA's wouldn't
be so happy providing 10W for long.)

There are a number of companies which sell transformers
for electroluminescent display drivers (Coiltronics comes
to mind right away). these are meant
to directly convert voltages in the range of 5-15V to
600V - 1kV or so, and are pretty small. you'd need to
pick one with the right power rating, though.

also, your switching frequency will affect power handling
capability. higher frequency = more power through magnetics
of a given size, but parasitic capacitances & MOSFET gate drive
speed would become important...

good luck....

--Jason

p.s. just curious: which diodes and caps are you using
on the output?

> {Original Message removed}

1999\09\06@160950 by Matthew Ballinger

picon face
> 2kV * 4mA = 8W output!

   Yes, I realize that is quite drain from AA batteries. The duty cycle is
very low however. Just 2-3 sec every few minutes. So I don't think this will
be a big problem.

>  I'm guessing it's the transformer that's limiting you.
>  Either the magnetics can't handle that much power, or
>  the resistance of the transformer winding is killing you.

I agree.

>  also, your switching frequency will affect power handling
>  capability. higher frequency = more power through magnetics
>  of a given size, but parasitic capacitances & MOSFET gate drive
>  speed would become important...

Using the PIC I was easily able to determine the best freq and pulse width.
Only about 1200Hz. So the mosfet shouldn't have problems coping with that.

>  p.s. just curious: which diodes and caps are you using
>   on the output?

500 volt, .01uf caps
1N4007 1Amp, 1kV PIV diodes

>  > {Original Message removed}

1999\09\06@162231 by l.allen

picon face
Jason wrote

> 2kV * 4mA = 8W output!
>
> I'm guessing it's the transformer that's limiting you.
> Either the magnetics can't handle that much power, or
> the resistance of the transformer winding is killing you,
> or both. (also, I don't know how much the batteries
> would sag, but I'm guessing even if you were lucky enough
> to get 80% efficiency, I'm guessing the AA's wouldn't
> be so happy providing 10W for long.)
>
Matt wrote
> >
> >     I'm working on a little project right now that requires a 2kV source
> > current limited to about 3-4 ma and powered from 4 AA batteries.

Hi Matt
Jason has a point!  I have been down this road myself thinking an EL
backlight must be wonderfully efficient at producing lumens per mW.
Alas it sucked....  by the time there were multiple efficiency drops
between inverter, EL luminarie and tripler chain combined.
If you're app isn't for an EL backlight I guess this is a waste of time
but if it is... find a low voltage alternative.. like leds.
Oh by the way... the EMI generation from the inverter was difficult
to suppress, resulting in more suppression components added to an
already high(ish) component count.
_____________________________

Lance Allen
Technical Officer
Uni of Auckland
Psych Dept
New Zealand
_____________________________

1999\09\06@172611 by Richard Prosser

flavicon
face
Also you might want to look at the caps.
Presumably you're using a doubler or tripler arrangement as you are only
using 1kV diodes.

8W output = 8J/sec.
switch frequency is 1200Hz - period = 833uS
ie cap energy storage requirement is 8/1200 = 6.67mJ per cycle. = 1/2CV^2

Using 10nF caps, this represents a voltage of 1.15kV - with a multiplier you
may be charging to less than 1/2 of this.

Doublers and triplers usually act like 1/2 wave rectifiers - charge caps in
parallel and discharge in series - so even if charged to a kV or so it is
likely to produce a lot of ripple.

Depending on configuration you may need diodes with a higher voltage rating
- or connect them in series with low track leakage. 1Megohm at 1kV is 1mA -
1/3 of your output current - and dirty tracks can exhibit this amount of
leakage.

Just my 2c

Richard
> {Original Message removed}

1999\09\06@175635 by paulb

flavicon
face
Matthew Ballinger wrote:

>> 2kV * 4mA = 8W output!
>   Yes, I realize that is quite drain from AA batteries.  The duty
> cycle is very low however.  Just 2-3 sec every few minutes. So I don't
> think this will be a big problem.

 Using Alkalines.

>>  I'm guessing it's the transformer that's limiting you.

 You need a purpose-wound transformer on a ferrite core with an air gap
if you are using single-ended drive.  You'll be using 1 turn per volt or
less and the primary will be quite substantial wire.  Look around
http://www.oatleyelectronics.com/catindex.html for ideas.

>>  but parasitic capacitances & MOSFET gate drive speed would become
>> important...

 Involves matters of MOSFET turn-on voltage.  Frankly, I can't see what
use a PIC would be in a power converter, other than to turn it on.

> 500 volt, .01uf caps
> 1N4007 1Amp, 1kV PIV diodes

 Beware!  You mentioned 1200 Hz.  1N4007 diodes do NOT work well at
that frequency.  They do not work *at all* at 10 kHz.
--
 Cheers,
       Paul B.

1999\09\06@181553 by Reginald Neale

flavicon
face
>    I'm working on a little project right now that requires a 2kV source
>current limited to about 3-4 ma and powered from 4 AA batteries. At the
>moment I'm using a PIC for PWM to a logic level power mosfet which chops
>power to the secondary a mini audio z matching  1:12 xfrmr which then feeds 4
>stage cascading voltage multiplier. This setup is very compact (also
>required) and  gets me around 1.6kV, but current is much to low. Then I used
>2 xfrmrs in parallel, but still to little current. I'm also using a 470uF cap
>to stiffen up the power supply some which helped somewhat. Anybody got ideas
>of a better way to do this? Any help would be greatly appriciated.
>Matt B

Matt:

You didn't say whether this is a commercial project and what your budget
is. Companies like Venus Scientific make subcompact kV level power supplies
for use with multiplier phototubes, where the current demand is similar to
yours.

Reg Neale

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...