Alright, I'm working on a class A SE mosfet amplifier. The psu transformer i have is center tapped, and capable of 8 amps at 24v ac. The amplifier is going to use a dual supply, with each channel drawing around 1.75A each constant. What would be a suitable configuration to boost my rails? I'm shooting for about 20~30v per rail. Could I use a conventional doubler with just the center point of my caps as ground? For smoothing I was going to use a capacitance multiplier if that makes a difference.
Do you mean this circuit?
That would be the easiest option in your case, it gives ~33V peak.
You need plenty of capacity for filtering however, since in effect you have two half-wave rectifiers. See attached screenshot - with 47000uF (47mF), the ripple is about 2V p-p.
Your capacitance multiplier can regulate that down to ~30V nicely.
An externally hosted image should be here but it was not working when we last tested it.
That would be the easiest option in your case, it gives ~33V peak.
You need plenty of capacity for filtering however, since in effect you have two half-wave rectifiers. See attached screenshot - with 47000uF (47mF), the ripple is about 2V p-p.
Your capacitance multiplier can regulate that down to ~30V nicely.
Attachments
your SE mosfet amplifier is very unlikely to draw constant current from it's dual polarity supply rails. Almost all topologies of ClassA modulate the current in the supply rails to follow the output current.Lingwendil said:
High currents do not suit voltage doublers. A voltage doubler is ideally suited to the tiny currents required to power the voltage amplifier stage of a power amp or for pre-amps, not the high current demands of power amps. The ripple on the supply after the doubler will be as bad as a half wave rectified PSU and the voltage droop as load current increases will be worse than a half wave rectified supply.
Re: Re: Need help with voltage doubler
Since his Class-A amp will draw <2A, we are not talking about *high* currents here. The dual half-wave "doubler" is definitely not only for "tiny" currents - you have one of these in the input of every PC power supply that has a 115/230V switch, whether it handles 250 or 600 Watts (5.2A @ 115V!)
Of course, switchers work much better with the high ripple than linears. No question, two transformers would be better, but with enough capacity and a regulator, a half wave rectifier can also do the job. You just need to calculate the effects and decide whether the result is acceptable.
AndrewT said:
Since his Class-A amp will draw <2A, we are not talking about *high* currents here. The dual half-wave "doubler" is definitely not only for "tiny" currents - you have one of these in the input of every PC power supply that has a 115/230V switch, whether it handles 250 or 600 Watts (5.2A @ 115V!)
Of course, switchers work much better with the high ripple than linears. No question, two transformers would be better, but with enough capacity and a regulator, a half wave rectifier can also do the job. You just need to calculate the effects and decide whether the result is acceptable.
Well, since I have the parts on hand I figure I may as well give it a try. This is for a junk-box/ found parts project anyway, so if it has small flaws I don't mind too much. I just would like to get a bigger swing from my amp, and figure'd it'd be worth a try.
The circuit is basically a beefed up oversized Szekeres buffer. No gain, no problem, as I have a nice tube preamp. As is, the psu is unregulated or filtered, just a pair of 33000uF across my rails. Hum is just barely audible. my rails sit at 14.65v each under load, with or without signal. As of this writing, idle current is 1.25A per channel via a LM317 CCS. Would using the doubler give me al least 24v rails under these conditions, all else remaining equal?
The circuit is basically a beefed up oversized Szekeres buffer. No gain, no problem, as I have a nice tube preamp. As is, the psu is unregulated or filtered, just a pair of 33000uF across my rails. Hum is just barely audible. my rails sit at 14.65v each under load, with or without signal. As of this writing, idle current is 1.25A per channel via a LM317 CCS. Would using the doubler give me al least 24v rails under these conditions, all else remaining equal?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.