JLH Updated Power Supply
 

Hi,

I am planning on making a 15W JLH Updated version.
I'm still thinking on how to create the PCB for the amp itself, I can't seem to find a good layout.

So, since I was tired of trying to create a layout for the amp, yesterday I tried one for the power supply.

Ok, so, first, the power supply schematic (this isn't on Geoff's site, he send it to me):
http://users.skynet.be/devil_hack/el...-schematic.gif

The parts layout (96dpi):
http://users.skynet.be/devil_hack/el...pply-parts.gif

The traces (96dpi):
http://users.skynet.be/devil_hack/el...ply-traces.gif

Both (96dpi):
http://users.skynet.be/devil_hack/el...supply-all.gif

Would this work, or will I encounter problems?
Any traces that should be rerouted?

have you built the amp yet? you may or may not have any problems with a conventional ps. if so, why go through all the trouble? I built one recently, powered by a very simple ps (just two 3300uf filter caps), and i have no problem whatsoever.

maybe a simple cap multiplier will do for you as well.

Good idea to start a separate thread.
Geoff has reported that the power supply makes a big difference, and others (e.g. Rod Elliott) seem to feel that a Class-A amp needs a very quiet power supply. Makes sense. Geoff has also reported that the TL431 reference is much quieter than a zener, although he also suggested a simple constant current source / resistor combination is even better.
I suggested in the other thread that you build up a prototype first so as not to get surprised. Another thing that occurs to me is that the diodes might get hot, carrying a continuous load of several amps, so check that out.

Ok, thanks for the info.

I'm going to build this circuit anyway, since I already have all the things I need for it. And it's cheaper than putting +- 8 big caps in parallel.

So the PCB layout should be fine, right?
I'm going to make some minor changes to it and then I'll make it this afternoon. (To bad my 0.8mm drill bit is broken :().

I haven't made the amp itself yet, it's the first amp I'll be making. I'm still looking for a relatively cheap source for good heatsinks.

Oh, btw, I have two types of diodes. The first one are MUR410 (http://www.onsemi.com/pub/Collateral/MUR420-D.PDF) ultra fast 4A 50-600V. The second are MR856 (http://www.onsemi.com/pub/Collateral/MR850-D.PDF) fast recovery 3A 50-600V. The first one are the best, right?

Since you don't have an amp yet, you need some way to test this. Have you got a dummy load of some kind?

Not really. I've been bussy with electronics as a hobby for a year and a half now, but only recently I started getting really interested. So except for some basic tools, I don't really have anything.

How would you test this? Could I use some simple 11 ohm 5W resistors? (So that I'd get 2 amp max @ 22v)

No they would smoke, 2 amp max @ 22v=44watts.

Test at first with 100 ohm 5W, then you'll need something much bigger.

:) sreten.

Aye, there's the rub. You almost need to build the amp to test the power supply, and vice versa.

Hm, ok. Well, I'm quite certain the power supply will work. Then again, even if I want to test it I can't anyway, since I haven't got the transformer yet :).

Ok, I test fitted my parts on a printed version of the PCB and noticed some flaws. If you should be interested, here's the improved PCB. It's 96 dpi.

use car-light bulbs for load
 

Hi,

pcb for a regulated powersupply is available on my site amp

amp site .

although it assumes a bridge and main caps are somewhere else.

A decent load to test a PSU can be made out of car-head-light bulbs.
They are usually 12V 50W so two in series makes a bright 24V 50W load.