Aleph 5 PSU
- Thread starter icceman
- Start date
- Status
The voltage is much too high for an aleph 5, and even the aleph 2. 40-0-40 will give you around +/- 56.4v dc (not accounting for diode losses) after rectifying. I would use a lower voltage transformer. 28-0-28VAC should give you the 34v rails that you need for the aleph 5, unless you want to use some kind of active power supply regulation.
Mark Finnis has good notes on the Aleph 5 ps here:
http://www.users.bigpond.net.au/mefinnis/passlabs/aleph5_construction.htm
On a side note, where are you getting your 40-0-40 from. I am looking for a good deal on a 1kVA 40-0-40 for my new Leach amp project.
--
Brian
Mark Finnis has good notes on the Aleph 5 ps here:
http://www.users.bigpond.net.au/mefinnis/passlabs/aleph5_construction.htm
On a side note, where are you getting your 40-0-40 from. I am looking for a good deal on a 1kVA 40-0-40 for my new Leach amp project.
--
Brian
The Aleph 5 will work ok with 30-0-30 transformer voltage.
You will get slightly more output power and heat dissipation, but the rest will be the same.
The 500 VA transformer will be ok for the one channel.
For two channels use minimum 600 VA transformer.
Best regards,
Kristijan Kljucaric
http://web.vip.hr/pcb-design.vip
I disagree with this idea
I am a big fan of Pi filters, actually just added one to my Aleph5, however you are not going to pull a 40-0-40VAC transformer down to 36VDC rails .... not unless you have chokes the size of the toroid itself
If you have the 40VAC toroid and want to use it, why not look at a regulated supply as Nelson has designed with the latest Zen variations. This will work well with the 50V DC rail you are going to have after regulation.
My 650VA toroid is as low as I would possibly go for a stereo Aleph5, 500VA would be unwise (IMHO).
cheers, mark
PS: My toroid was 28-0-28VAC and my rails are 36VDC
PPS: Only just noticed the last question, sorry - you can use transformers in parallel but have to use "paralleling chokes" to balance the load, otherwise you will have stray currents as no 2 transformers are identical. If you can find 14-0-14VAC/500VA toroids, then use the secondaries in series, with the new "center-tap" being the connection of the two 2'y windings (the 2 old "centers" become redundant)
Thanks, I think it will be easier for me if I just use 2 x 500VA 30-0-30 transformer, 1 transformer for each channel of the ALEPH 5. But this will more expensive than just using a 1000VA transformer (too bad I can't find one with a 28-0-28 output), as I will need 2 separate banks of filter capacitors (1 x 1000VA transformer is also cheaper than 2 x 500VA).
Not necessarily !
If you think about it, since each channel draws only 1/2 the current of the stereo version, you do not need as much filter "uF".
For example, I use 4x 80,000uF caps in a common supply. If I split the supply for each channel, then using 2x for each would be identical. I would not need to buy another 4 caps !!!
Where this "falls to ground" is if you wish to use Pi filters .... then just buy smaller caps. Alas, this will likely increase the overall price.
cheers, mark
I think Rodd was referring to a choke input supply, not a pi type. If you put the choke directly connected to the rectifiers, your output voltage will be significantly lowered.
You can also exercise a certain amount of control over the output voltage of a Pi filtered supply by changing the value of the first set of filter caps. That is, a Pi filter with the first set of caps entirely removed is the same as a choke-input filter. As you start to add capacitance to the front end, the output voltage will rise. At some point with very large capcitance in the first stage, the output voltage will be as high as a C-only supply, minus the resistive losses in the inductor.
But, bear in mind that as you increase this first stage capacitance, the charging current pulses in the rectifiers become shorter and shorter, and the current gets higher and higher. This has two side effects: it decreases the effective VA rating of the transformer, since the peak currents are much higher, and it also causes more radiated noise, since the rectifying currents are so much higher, and the pulses are shorter. You must also make sure that the rectifier diodes can handle the higher charging pulse currents.
Personally, I think a choke-input supply is a great way to go...
You can also exercise a certain amount of control over the output voltage of a Pi filtered supply by changing the value of the first set of filter caps. That is, a Pi filter with the first set of caps entirely removed is the same as a choke-input filter. As you start to add capacitance to the front end, the output voltage will rise. At some point with very large capcitance in the first stage, the output voltage will be as high as a C-only supply, minus the resistive losses in the inductor.
But, bear in mind that as you increase this first stage capacitance, the charging current pulses in the rectifiers become shorter and shorter, and the current gets higher and higher. This has two side effects: it decreases the effective VA rating of the transformer, since the peak currents are much higher, and it also causes more radiated noise, since the rectifying currents are so much higher, and the pulses are shorter. You must also make sure that the rectifier diodes can handle the higher charging pulse currents.
Personally, I think a choke-input supply is a great way to go...
Ummm ..... OK, here again my limited EE theory sinks me like a stone
I did not appreciate there would be this much difference. Are there any simple formulae governing the voltage drop across a choke-input / Pi filter (choke input being where C1 -> 0uF)?
This might help a simpleton like me
cheers, mark
I did not appreciate there would be this much difference. Are there any simple formulae governing the voltage drop across a choke-input / Pi filter (choke input being where C1 -> 0uF)?
This might help a simpleton like me
cheers, mark
Hi Mark,
For an amp like the Aleph or other class A designs where there is a significant current draw (even at idle) there has to be voltage drops associated with the series DC resistance in the PSU. The inductor in the PI type PSU (Under load) will limit the peak AC voltage causing the voltage across the output supply cap to be about RMS voltage minus the voltage drops due to series DC resistance.
It would be interesting to see what the PSU voltage is in your Aleph 5 if you measure:
1.)DC rails disconnected (0 current draw)
2.)At normal idle (8ohm load, no signal)
3.)At near full output (1kHz, same 8ohm load)
I think you might be supprised.
Rodd Yamas***a
For an amp like the Aleph or other class A designs where there is a significant current draw (even at idle) there has to be voltage drops associated with the series DC resistance in the PSU. The inductor in the PI type PSU (Under load) will limit the peak AC voltage causing the voltage across the output supply cap to be about RMS voltage minus the voltage drops due to series DC resistance.
It would be interesting to see what the PSU voltage is in your Aleph 5 if you measure:
1.)DC rails disconnected (0 current draw)
2.)At normal idle (8ohm load, no signal)
3.)At near full output (1kHz, same 8ohm load)
I think you might be supprised.
Rodd Yamas***a
- Status