LM3886 PSU , need

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
after testing with one travo center tap and one rectifier brdidge
the sound is better than using two rectifier bridgs ( somthing like snubber )
on the two rectifier bridges I used two travo .
still more willing to hear ideas and suggestions about PSU .
thanks for all and this Great Site
 
Hi, I've made Alex88's BPA300, and I would like to have a great snubbed PSU 4 them

Can I use Carlosfm's unregulated PSU to drive BPA 300 or this one is good only for a single (or double) Lm3886???

What shall I change on it to work properly with 2 (or 4) BPA150 Board??

Thanks 4 helping me out
 
madam said:
What shall I change on it to work properly with 2 (or 4) BPA150 Board??

  • - The transformer must be bigger. CarlosFM's version uses 250 VA, if I remember correctly. BPA 200 uses 2 x 385 VA, which should be your benchmark for BPA 300 as well.
    - The MUR860 must be replaced with diodes or rectifier bridges that are capable of the increased current demands. The new rectifiers may need snubbers of their own to be determined.
    - Depending on the version you want to build, you will need more smoothing capacity. Again BPA 200 shoud be your benchmark with 1.200 µF + 4 * (10.000 µF + 470 µF + 0,1 µF) per rail.
 
The easy way is one Carlos FM snubberised to each chipamp.

I suspect that an inverting and non-inverting pair could be run from a single Carlos FM snubberised. But, don't run a parallel pair from one PSU.

Powering the opposite pairs this way will require just three PSUs for a BPA300.

If we follow the usual recommendation of VA = max output power to twice max output power then a 300VA to 600VA should be sufficient to power one BPA300. I would use 500VA or 625VA for a 360W amplifier.
 
AndrewT said:
The easy way is one Carlos FM snubberised to each chipamp.
That would be six snubberised power supplies for one amplifier. That is the easy way as opposed to ...?

AndrewT said:
Powering the opposite pairs this way will require just three PSUs for a BPA300.
On these working conditions the MUR860s would probably welcome some heatsinking.
What is wrong with making a single big power supply with a decent high current bridge rectifier? For those, who hear a difference, snubbers can be refitted to make it perform just as well as the MURs above the audio range.
 
pacificblue said:

That would be six snubberised power supplies for one amplifier. That is the easy way as opposed to ...?


I meeeen 1 PSU board per 1 or 2 of the PBA150 board (if 2 boards, only in the same signal-polarity) .... with some higher caps/more 10.000uF in the rail.
I know I would need more power(and caps) I have a 800VA toroid to support the unit, and 2X47.000uF at the moment.
But if I make an upgrade I would make a good-one and not an "intermadiate" one. (would make a goood PCB for it)

On these working conditions the MUR860s would probably welcome some heatsinking.
What is wrong with making a single big power supply with a decent high current bridge rectifier? For those, who hear a difference, snubbers can be refitted to make it perform just as well as the MURs above the audio range.

There is nothin' wrong with the bridges..... but if I make a great unit (and I did) and would like to upgrade it..... than why not upgrade to some serious diodes too? It should worth the money and effort if they are so widely used.... with the great results we all know. There must be other diodes, fast and relyable like the MUR in higher Ampers...... And I would be O.K. to have one BIIIIIIG unit of PSU, if it's works just fine with the 4 BPA boards I must drive with.
 
madam said:
than why not upgrade to some serious diodes too?
That is what the high current bridge rectifiers are.

madam said:
There must be other diodes, fast and relyable like the MUR in higher Ampers
MUR3060 exist as well. That is 15 A average, 30 A repetitive peak and 150 A non-repetitive peak up to 600 V. Heatsinking is of course required for reliable operation.
Then there are the RURG3060 with 30 A average, 70 A repetitive and 325 A non-repetitive also up to 600 V.
Even faster switching with 23 ns is the state-of-the-art HEXFred HFA25PB60 (MUR860 -> 50 ns) with 25 A average, 100 A repetitive and 225 A non-repetitive up to 600 V.
The point with all of them is, the fast switching is supposed to bring ringing down, which is the same you do with a snubber across "normal" rectifier diodes much cheaper.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.