How much ripple is exceptable on a GOOD powersupply?
I am bussy to repair 2 x Mullard 5-20's, which was totally modified (for the worse). On the existing powersupply a 2H inductor and 2 x 30uF caps are used. (PI filter) I replaced the caps with 220Uf ones and now get a relative smooth DC output with a 50mVpp ripple - under load.
1) Is it possible to improve this more while still using the 2H choke?
2) Just for a matter of interest, how much ripple to you have on your PSU?
I am bussy to repair 2 x Mullard 5-20's, which was totally modified (for the worse). On the existing powersupply a 2H inductor and 2 x 30uF caps are used. (PI filter) I replaced the caps with 220Uf ones and now get a relative smooth DC output with a 50mVpp ripple - under load.
1) Is it possible to improve this more while still using the 2H choke?
2) Just for a matter of interest, how much ripple to you have on your PSU?
Depends on topology. Triodes will be very sensitive to hum due to their low Rp, whereas pentodes will be rather insensitive for their high Rp (though screen still has to be well filtered).
PP operation cancels common-mode effects such as hum, so far more ripple can be tolerated, triode or pentode.
You could even analyze this based on a desired amount of hum on the output. Say 10mW hum on the output, if the primary is 2.5kohms that's 5V across the OPT primary. If the tube is a 6L6 in pentode, about 25kohm Rp, 10 times the Rl. So approxametly 50V ripple could be tolerated, theoretically. Of course, since this is a whopping 20% of the supply voltage, IMD should be pretty bad. And much more filtering is easy anyway.
If it were a triode, only maybe 6V of ripple would be tolerable given the 10mW spec above. Of course, a perfectly balanced PP amp would tolerate 100% (i.e. no filtering) hum, but we're not in a perfect world.. especially not when there's a signal applied.
Tim
PP operation cancels common-mode effects such as hum, so far more ripple can be tolerated, triode or pentode.
You could even analyze this based on a desired amount of hum on the output. Say 10mW hum on the output, if the primary is 2.5kohms that's 5V across the OPT primary. If the tube is a 6L6 in pentode, about 25kohm Rp, 10 times the Rl. So approxametly 50V ripple could be tolerated, theoretically. Of course, since this is a whopping 20% of the supply voltage, IMD should be pretty bad. And much more filtering is easy anyway.
If it were a triode, only maybe 6V of ripple would be tolerable given the 10mW spec above. Of course, a perfectly balanced PP amp would tolerate 100% (i.e. no filtering) hum, but we're not in a perfect world.. especially not when there's a signal applied.
Tim
<1V
For a push-pull amplifier like yours, anything less than 1V peak to peak ripple at the output transformer centre tap is fine since it will be rejected by push-pull action. Single-ended amplifiers need far less ripple because they have virtually no rejection.
The maximum allowable capacitance for a GZ34 is 60uF. All other valve rectifiers are even more fragile.
For a push-pull amplifier like yours, anything less than 1V peak to peak ripple at the output transformer centre tap is fine since it will be rejected by push-pull action. Single-ended amplifiers need far less ripple because they have virtually no rejection.
The maximum allowable capacitance for a GZ34 is 60uF. All other valve rectifiers are even more fragile.
Re: <1V
Should have said, my amps get DC from silicone rectifiers, not org. GZ34 - part of some of the weird previous modifications, but at least gives me more volts now.
EC8010 said:The maximum allowable capacitance for a GZ34 is 60uF. All other valve rectifiers are even more fragile.
Should have said, my amps get DC from silicone rectifiers, not org. GZ34 - part of some of the weird previous modifications, but at least gives me more volts now.
JDeV said:silicone rectifiers,
Silicone: soft, squidgy polymer used for breast implants.
Silicon: hard, brittle, metallic element used for semiconductors.
EC8010 said:
Silicone: soft, squidgy polymer used for breast implants.
Silicon: hard, brittle, metallic element used for semiconductors.
But the former describes the supply rail regulation best ..soft..squidgy..
Hi,
But the latter describes the sound of diodes perfectly well ...hard, brittle...
Silicon: hard, brittle, metallic element used for semiconductors.
But the latter describes the sound of diodes perfectly well ...hard, brittle...
dhaen said:
But the former describes the supply rail regulation best ..soft..squidgy..
Not in class A PP.
Silicon rectifier = hashy, grey sound.
Re: <1V
Perhaps my CLCLC - LCRCRC power supply is a bit of overkill for a class A PP? (the last section suited what I had on hand to get the right voltage for the driver)
<5mVp-p ripple at the OPT c/t, and not visible on my scope at the driver.
EC8010 said:For a push-pull amplifier like yours, anything less than 1V peak to peak ripple at the output transformer centre tap is fine since it will be rejected by push-pull action.
Perhaps my CLCLC - LCRCRC power supply is a bit of overkill for a class A PP? (the last section suited what I had on hand to get the right voltage for the driver)
<5mVp-p ripple at the OPT c/t, and not visible on my scope at the driver.
RE:That's just showing off!
Hi,
For one, the old QIIs sound...err well really old when not modded.
Once modified they sound decent...say acceptable, but never really great.
Most modifications help to reduce the ripple voltage to acceptable levels amongst other shortcomings.
Most of the Leak amps sound way better after an update and much better than the QIIs IMHO.
Cheers,
Hi,
For one, the old QIIs sound...err well really old when not modded.
Once modified they sound decent...say acceptable, but never really great.
Most modifications help to reduce the ripple voltage to acceptable levels amongst other shortcomings.
Most of the Leak amps sound way better after an update and much better than the QIIs IMHO.
Cheers,
Re: RE:That's just showing off!
I agree. I always thought a Quad II sounded slightly woolly compared to a Leak.
fdegrove said:Most of the Leak amps sound way better after an update and much better than the QIIs IMHO.
I agree. I always thought a Quad II sounded slightly woolly compared to a Leak.
Ripple in a single ended power amplifier can be higher then in a single ended preamp. My poweramp has about 80mV of ripple with 2*47uF around a 10H choke. That's enough to get the poweramp quiet.
The preamp needed much more filtering. This one is also a single ended transformer coupled amp. I use 22uF - 15H - 150uF - 15H - 150uF there and have a ripple of 6 micro Volt. Anything above 0,5mV ripple was audible because the hum is also amplified again by the poweramp.
My push pull amp is with a ripple of 2,5 Volt deadly quiet.
The preamp needed much more filtering. This one is also a single ended transformer coupled amp. I use 22uF - 15H - 150uF - 15H - 150uF there and have a ripple of 6 micro Volt. Anything above 0,5mV ripple was audible because the hum is also amplified again by the poweramp.
My push pull amp is with a ripple of 2,5 Volt deadly quiet.
Indeed, my preamp (now listening: Metallica - Master Of Puppets on WJJO) has FWB > 47uF > 330 ohms > 330uF > 680 ohms > 100uF, output (12AU7 CF) is running from the 330uF I think, 12AX7 from the 100uF. Kinda funny on startup, it hums for a moment then goes to baseline.
I should probably up the filter resistors some day..
Tim
I should probably up the filter resistors some day..
Tim
RIPPERS...
Hi,
Could be the bleeders aren't low enough in value too.
Cheers,
Hi,
I should probably up the filter resistors some day..
Could be the bleeders aren't low enough in value too.
Cheers,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Tubes / Valves
- How much ripple??