Hi,
I'm looking for a fairly efficient variable regulated power supply for my digital amplifier projekt. Have anyone tried the variable power supply with the SCR pre-regulator suggested in th LT1083 datasheet?
/Jesper
I'm looking for a fairly efficient variable regulated power supply for my digital amplifier projekt. Have anyone tried the variable power supply with the SCR pre-regulator suggested in th LT1083 datasheet?
/Jesper
The PCR series from Soundcraftsman used this topology.
This is essentially the same as sticking a light dimmer on the primary like Carver did in most of his amplifiers.
If you want to make it quieter use a bridge rectifier with a MOSFET inside of it. The triac and scr make noise because they turn on half way through the cycle. With an FET you can have it turn on at the zero crossing and then turn off at any point in the waveform.
This is essentially the same as sticking a light dimmer on the primary like Carver did in most of his amplifiers.
If you want to make it quieter use a bridge rectifier with a MOSFET inside of it. The triac and scr make noise because they turn on half way through the cycle. With an FET you can have it turn on at the zero crossing and then turn off at any point in the waveform.
Thanks for the input.
Why would it be quiter to turn off early rather than turning on late? Do you know where I can find more information and perhaps a reference design using a MOSFET as you describe.
/Jesper
Why would it be quiter to turn off early rather than turning on late? Do you know where I can find more information and perhaps a reference design using a MOSFET as you describe.
/Jesper
If you put a cap of about 1uF per ampere that you are switching, across the drain/source of the fet it will be super quiet. Did this several years ago just messing around on the bench. Make sure you use a choke-input fiiter. Also, if you about a 10k to 47k or so resistor in series with the fet gate it will limit how fast it turns on, so that when it short cct's the cap the discharge current won't be too high. OK for mains frequencies, big no-no for high switching frequencies. This is a really good method of power supply regulation for higher power setups, IMHO.djk said:
Thanks Circlotron,
I'll try the resistor approach to reduce on and off slope. I intend to use snubber circuits to dampen oscillation between transformer and rectifier bridge induced by the switching transients. I belive this give similar result as a cap across the FET. Anyway, I'll test both. I'll work out a schematic and post it here.
/Jesper
I'll try the resistor approach to reduce on and off slope. I intend to use snubber circuits to dampen oscillation between transformer and rectifier bridge induced by the switching transients. I belive this give similar result as a cap across the FET. Anyway, I'll test both. I'll work out a schematic and post it here.
/Jesper
I found this slogging around:
http://164.195.100.11/netacgi/nph-P...,708,577.WKU.&OS=PN/5,708,577&RS=PN/5,708,577
You need the free TIFF viewer, then click on images.
http://164.195.100.11/netacgi/nph-P...,708,577.WKU.&OS=PN/5,708,577&RS=PN/5,708,577
You need the free TIFF viewer, then click on images.
This is more what I had in mind:
http://164.195.100.11/netacgi/nph-P...,510,679.WKU.&OS=PN/5,510,679&RS=PN/5,510,679
It uses back-to-back FETs or IGBTs, a single device inside a bridge is what I had in mind. Whatever gives the lowest voltage drop for your load current is best.
Remember the reverse diode in the FET is conducting in the back-to-back circuits and has basically the same resistance as the device RDS on specification.
If running 240V line at 20A load an IRFP22N50A would drop 4.6V vs a bridge rectifier loss of 2V or so. An IGBT may be a better choice here.
http://164.195.100.11/netacgi/nph-P...,510,679.WKU.&OS=PN/5,510,679&RS=PN/5,510,679
It uses back-to-back FETs or IGBTs, a single device inside a bridge is what I had in mind. Whatever gives the lowest voltage drop for your load current is best.
Remember the reverse diode in the FET is conducting in the back-to-back circuits and has basically the same resistance as the device RDS on specification.
If running 240V line at 20A load an IRFP22N50A would drop 4.6V vs a bridge rectifier loss of 2V or so. An IGBT may be a better choice here.
- Status
- Not open for further replies.