Hi all
A silly question:
If I have a trafo 0-46 VAC and I want a +/- 31VDC, is correct the arrangement in the schematic, or I just do the big bang?
Thank You
A silly question:
If I have a trafo 0-46 VAC and I want a +/- 31VDC, is correct the arrangement in the schematic, or I just do the big bang?
Thank You
An externally hosted image should be here but it was not working when we last tested it.
Your proposal does not work . You can use the one of post 4290. I use it with my F5 and some other class AB amp.
One 10k resistor has to be tweaked under load to get accurate symetrical voltages. If no tweak, there is a slow drift that equilibrates at a few volts difference between the two halves.
You can google Quad 606 ( and 909) amplifier which uses this PSU.
> You might check out post 4290 in the Pass F5 thread it's suposed to work.
http://www.diyaudio.com/forums/showthread.php?postid=1864574#post1864574
http://www.diyaudio.com/forums/attachment.php?s=&postid=1864574&stamp=1245948705
I have built this about 5 years ago, but with some slight modifications.
Firstly I put a resistor between the two base of the transistor pair, instead of shorting them. The value of this resistor is chosen such that it creates about 1.5V across the two base connections, and hence bias them to Class A. I also added low value emitter degenerating resistors (depending on your current consumption) to create bias stability.
A yet simpler solution is to use a high voltage power opamp (say capable of 200mA and +/-40V rails) connected as a follower and connect the inpout to the mid-point of the 10k resistors. The output of the opamp is your virtual ground, and the opamp is to be powered by the two rails.
But you have to bear in mind that such a system cannot sustain constant current draw from one rail to ground continuously, unless you use very high power devices capable of delivering the continuous working current.
Patrick
http://www.diyaudio.com/forums/showthread.php?postid=1864574#post1864574
http://www.diyaudio.com/forums/attachment.php?s=&postid=1864574&stamp=1245948705
I have built this about 5 years ago, but with some slight modifications.
Firstly I put a resistor between the two base of the transistor pair, instead of shorting them. The value of this resistor is chosen such that it creates about 1.5V across the two base connections, and hence bias them to Class A. I also added low value emitter degenerating resistors (depending on your current consumption) to create bias stability.
A yet simpler solution is to use a high voltage power opamp (say capable of 200mA and +/-40V rails) connected as a follower and connect the inpout to the mid-point of the 10k resistors. The output of the opamp is your virtual ground, and the opamp is to be powered by the two rails.
But you have to bear in mind that such a system cannot sustain constant current draw from one rail to ground continuously, unless you use very high power devices capable of delivering the continuous working current.
Patrick
Couple of diodes will work. 
Imagine 2 electrolytics in series.
One output leg of the trafo goes in the middle of them.
From the other one goes 2 diodes: one rectifies positive polarity for the positive rail, another rectifies negative polarity for a negative rail.
It is called a "voltage doubler".
Imagine 2 electrolytics in series.
One output leg of the trafo goes in the middle of them.
From the other one goes 2 diodes: one rectifies positive polarity for the positive rail, another rectifies negative polarity for a negative rail.
It is called a "voltage doubler".
Wavebourn's "voltage doubler" : see the circuit in the first part of this article (the one with the filter capacitors):
http://www.play-hookey.com/ac_theory/ps_v_multipliers.html
You would get about 46 * 1.414 = +/-65VDC from your transformer with no load.
http://www.play-hookey.com/ac_theory/ps_v_multipliers.html
You would get about 46 * 1.414 = +/-65VDC from your transformer with no load.
You get +/-65V unloaded which drops like a stone from the first milliamp of load. 60 Hz ripple frequency instead of 120 and worse power factor than you would with a bridge.
All it takes is a standard full wave bridge, and two equal valued filter caps and 10k bleeder/equalizer resistors. Let the ground float at the half way point. It doesn't perform quite as well as it would if you had the center tap but it's better than any of the other crazy alternatives. Older QSC amps didn't have the center tap either, and the only ill effect was slightly worse bass performance when clipping. That's the price you pay for using what's on hand, but it will work. If the caps are big enough (20k uf) you'd never notice a problem with 20 Hz or higher frequency material.
All it takes is a standard full wave bridge, and two equal valued filter caps and 10k bleeder/equalizer resistors. Let the ground float at the half way point. It doesn't perform quite as well as it would if you had the center tap but it's better than any of the other crazy alternatives. Older QSC amps didn't have the center tap either, and the only ill effect was slightly worse bass performance when clipping. That's the price you pay for using what's on hand, but it will work. If the caps are big enough (20k uf) you'd never notice a problem with 20 Hz or higher frequency material.
bobodioulasso said:
They would if the circuit had all its internal bias references from rail to rail like op-amps do. Or if any current from one rail to "ground" were balanced with an opposing current to the other rail. There's almost always a way to make it work. With pre-built modules you may not know exactly what's going on inside, but if you actually DIY the amplifier.....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.