I built mine as a dual-mono chassis with completely
independent power supplies. I like to think it improves
power supply regulation and crosstalk. At the time
I was able to pick up a batch of 30,000 uF @ 60V caps
and used four on each channel for a total of a quarter-
farad(!).
I did find it necessary to improvise an inrush current
limiter.
My Ultimate Leach Amplifier might use >four< independent
power supplies, one for each supply rail and each channel.
And I used the leftover parts from my first project and
upgrades to build a backup amplifier with the power
supply done exactly as you want to do. Works fine.
The more the merrier, at least up to a practical point.
If you've got the parts, use 'em.
independent power supplies. I like to think it improves
power supply regulation and crosstalk. At the time
I was able to pick up a batch of 30,000 uF @ 60V caps
and used four on each channel for a total of a quarter-
farad(!).
I did find it necessary to improvise an inrush current
limiter.
My Ultimate Leach Amplifier might use >four< independent
power supplies, one for each supply rail and each channel.
And I used the leftover parts from my first project and
upgrades to build a backup amplifier with the power
supply done exactly as you want to do. Works fine.
The more the merrier, at least up to a practical point.
If you've got the parts, use 'em.
HT-Configuration - same question
I'm currently designing a HT-Amp, based on the Opti-MOS amp and
I'm wondering what is the best configuration for the power supply, too.
I have two really huge torodial transformers, 1200VA each, center tapped,
so one would be surely enough for 5 (maybe 6) channels with a more
conservative design.
But how to configure the power-supply?
My current plan is to use only one transformer, one rectifier, 80.000uF shared
cap and another 20.000uF on eachs channels pcb.
Another solution would be to use one transformer, with dedicated rectifiers
and 40.000uF for each channel. Low crosstalk but low capacity for each channel.
Or should I really use both transformers and use two independant
power-supplies, one for stereo-L/R and another one for the surround
channels, with the added benefit that I can switch off the surround channels
in stereo mode?
Or maybe use two PS and divide them Left-Front/Rear + Center-F and
Right-Front/Rear + Center-B to minimize L/R-crosstalk in stereo??
I'm currently designing a HT-Amp, based on the Opti-MOS amp and
I'm wondering what is the best configuration for the power supply, too.
I have two really huge torodial transformers, 1200VA each, center tapped,
so one would be surely enough for 5 (maybe 6) channels with a more
conservative design.
But how to configure the power-supply?
My current plan is to use only one transformer, one rectifier, 80.000uF shared
cap and another 20.000uF on eachs channels pcb.
Another solution would be to use one transformer, with dedicated rectifiers
and 40.000uF for each channel. Low crosstalk but low capacity for each channel.
Or should I really use both transformers and use two independant
power-supplies, one for stereo-L/R and another one for the surround
channels, with the added benefit that I can switch off the surround channels
in stereo mode?
Or maybe use two PS and divide them Left-Front/Rear + Center-F and
Right-Front/Rear + Center-B to minimize L/R-crosstalk in stereo??
I to am debating, if I should use dual rectifiers in my power supply. I have done masses searching on this site, I found 2 other threads (sorry I lost the link). One tells the story of how someone who has built an amp fixed his ground loop problem by going from dual rectifiers to using just one. In the other thread it’s clamed that the dual rectifiers have sonic improvements. I just don’t know. On the TNT site that guy is wrong about dual rectifiers “sharing” the work load, they actually dissipate twice the heat (power) and drop the rail voltage down twice as much. It’s also only half as reliable, because the probability of something braking is doubled because the are twice as many parts. I just can’t make up my mind. I guess I’ll just have to prototype it both ways and see if I can hear a difference.
Leve
Leve
acenovelty said:
'First, it is well known, and can be proved very easily, that large capacitors will filter well lower frequencies, such as 50/60 Hz, but will filter high frequencies rather poorly. This means you are likely to encounter RF problems, either directly (as, for example, a sound reminding one of bird chirping), or indirectly, as an elevated level of intermodulation.'
Fair enough.
' Next, large capacitors do filter better than smaller ones, but are also slower than smaller ones (all other things being equal), so the subjective speed of your amplifier in particular greatly depends on the quality of those capacitors'
Uh?
Hi Roger & Michaelb,
I recommend separate rectifier and smoothing to each channel.
The speaker impedance times the capacitance, on each rail, should aim for an RC time constant of about 100mS to 150mS (8ohm and 12mF to 20mF meets this). You can go with lower RC but only the extreme low bass seems to benefit.
This generally ensures good bass with ClassAB designs and meets the requirement of the fast brigade that do not want excessive smoothing slowing down the mid response.
I recommend separate rectifier and smoothing to each channel.
The speaker impedance times the capacitance, on each rail, should aim for an RC time constant of about 100mS to 150mS (8ohm and 12mF to 20mF meets this). You can go with lower RC but only the extreme low bass seems to benefit.
This generally ensures good bass with ClassAB designs and meets the requirement of the fast brigade that do not want excessive smoothing slowing down the mid response.
I'm with Andrew on this; the Leach I completed yesterday (yay!) has separate bridges for each rail and separate 650VA transformers for each channel, with 30mF per rail per channel made from 2 12mF and a 6.8mF. I can't hear any hum or white noise with my ear inside my Tannoy HPD315, and I have no issue with speed or depth of bass.
Compared to the P3A it seems somehow clearer; easier to hear the pitch of deep bass notes. I'm very happy with it, it has the ability to make the house shake without hurting my ears. What could be finer?
Compared to the P3A it seems somehow clearer; easier to hear the pitch of deep bass notes. I'm very happy with it, it has the ability to make the house shake without hurting my ears. What could be finer?
bremen nacht said:
Out of curiosity, did you mix screw terminal caps with snap-in types? I've been looking for caps to parallel in my Leach supply, but it seems that 10,000 uf are the smallest with screw terminals (at least that I could find in computer grade). Mixing these with snap-ins has got me wondering how to best wire in the dissimilar style terminals, especially when I was considering a copper ground plate screwed into the capacitor ground terminals.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.