Re: Flyback Driver PCBs

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Arpit Thomas" <arpit-at-inzo-dot-org> 

Hi kreso. I took a 145 watt psu apart and without my solder wick i couldnt 
desolder teh transformer so i soldered wires onto 2 of the 4 lines into the 
transformer.  I then took a monitor flyback and found some input pins with 
a resistance between them of 1.5 ohms and connected the wires to them. I 
had a 150 mercury vapour light connected toa power board ( it was dark). 
WHen I plugged the psu into the power board the mercury vapour lamp turned 
off then restarted from the slowly getting brighter cycle. the psu made a 
quiet buzzing like they do when overloaded, and I got no sparks whatever. 
I'm giving up on your method, and going to try a halogen download smps. :)

*********** REPLY SEPARATOR  ***********

On 17/04/2004 at 8:30 AM Tesla list wrote:

 >Original poster: Matthew Smith <matt-at-kbc-dot-net.au>
 >
 >Kreso wrote:
 > > I invite anybody to try my setup up. It will bring HUGE sparks.
 > >
 > > 1. Get yourself a PC SMPS i used AT 200 W
 > > 2. Toke the board out.
 > > 3. Remove ferite cored transformer
 > > 4. Connect flyback  primary to the points where primary of transformer
 > > 5. Apply power
 >
 >Jim Mitchell replied:
 >>Kreso I've tried that and the PSU just goes into protection and makes a
 >5mm
 >>spark at 2hz.
 >
 >Matthew speculates (as an SMPS novice):
 >
 >I'd guess that Jim's attempt may have been with a supply, the controller
 >of
 >which has switch current limiting.  To reproduce Kreso's results, you
 >would
 >need need a supply which relies on monitoring secondary voltage (normally
 >feedback through opto-coupler) and no limiting of the switch current.  If
 >no feedback were received (increase in secondary voltage regulates through
 >decrease in switch pulse width), the unit could run "flat out", giving
 >results as reported by Kreso.
 >
 >I think that it's already been pointed out that most TV line-out
 >transformers normally only see 115V or so.  A Computer supply is generally
 >fed by a bridge rectifier, making the voltage available to the switch
 >Vline
 >/ (SQRT(2)/2).  I'm guessing that Kreso is in "220 Land", so his switch
 >would be pumping:
 >(220/(sqrt(2)/2)-Vswitch_loss =roughly= 310V into transformer.  In "240
 >Land", you're closer to 340V.
 >
 >If considering this experiment, therefore, I would suggest using
 >transformers that you don't mind getting "fried".
 >
 >Cheers
 >
 >M
 >
 >--
 >Matthew Smith
 >Kadina Business Consultancy
 >South Australia
 >http://www.kbc-dot-net.au