First Lm3886

[quadtech]

PJPro,
The first schematic you have has an error.
Here is the correct snubberised PSU schematic

- Prasad

[PJPro]

Originally posted by Redshift187
I suspect it would work ok. My thoughts are:
a) is the 5A limit of the LM338 enough for the chip and voltage you are using? The LM3886 can peak at a little more than 6A.
b) the snubbering is probably not necessary. Voltage regulators offer some smoothing of their own.

Interesting you should mention peak current, I've been having a go at calculating it for 50W power output driving an 8 ohm load using the equations on the data sheet. I have included the calculations in my build thread.

Clicky

I make the peak current 3.5A.

If my calculations are correct, why do people go for much transformers capable of delivering 12A?

[PJPro]

Originally posted by quadtech
PJPro,
The first schematic you have has an error.
Here is the correct snubberised PSU schematic

- Prasad

Thanks for the revision.....but I can't see the difference between the two.

[Redshift187]

I got around 6.16A (with 4 ohm speakers) using the Overture Design Guide, you should check it out. It does a lot of the calculations for you.

As for transformers, I think you typically go for higher current ratings because some of the time it's idling, and some of the time it's charging the caps, which has to have higher current than if it were just a constant draw. Also, higher current transformers have better regulation, which means the unloaded voltage doesn't go too far above the load voltage. The opposite way of saying that being that the voltage doesn't "sag" as much under heavy load. In other words, you get a more consistent voltage without going to a regulated supply.

[Redshift187]

Originally posted by PJPro


Thanks for the revision.....but I can't see the difference between the two.

Look at where the ground lines come out of the bridge rectifiers... in the revised one, the +PGND comes from the "left" side of the bridge, and -PGND comes from the "right" side of the bridge. You're diagram has both PGNDs coming from the "left" side of the bridge.

[PJPro]

Originally posted by Redshift187
I got around 6.16A (with 4 ohm speakers) using the Overture Design Guide, you should check it out. It does a lot of the calculations for you......[snip]

Hey, that's cool. Thanks.

Using the spreadsheet, I got 3.9A for +/- 35VDC at 8 ohm....a shade more than I calculated using the equations.

[PJPro]

Originally posted by Redshift187



Look at where the ground lines come out of the bridge rectifiers... in the revised one, the +PGND comes from the "left" side of the bridge, and -PGND comes from the "right" side of the bridge. You're diagram has both PGNDs coming from the "left" side of the bridge.

Gotcha. It's my schematic that's wrong. I was comparing the schematic quadtech posted with the chipamp.cpm schematic.

I'll revise mine and repost.

Thanks

[AndrewT]

you need more current capability because a speaker does not behave like an equivalent resistor.

[PJPro]

Originally posted by AndrewT
you need more current capability because a speaker does not behave like an equivalent resistor.

Do you think that my two 160VA transformers will be under powered?

[AndrewT]

if your smoothing caps and decoupling caps are sized appropriately then the peak current requirement comes from the capacitors not the transformer.