A. Holton's AV800 amplifier

Situation: I will tomorrow go and get those many IRF240/9240 of mine, and I'm not quite sure how soon I can start with the X circuit. I also have some problems with deciding what kind of case to use for it.

But since I have two big subwoofers here and I've decided to skip any Leach SuperAmp project, I also thought about building a usual mosfet amplifier, since I will now have those IRFs.

Housing. I've thought about Fischer SKE 3 400. That is 400mm (16 in) deep and 3U high 19" case with 4cm of fins covering both sides. By evenly placing the power devices on the heatsink, I think I can achieve just a bit better than 0.2K/W or 5W/1deg. What bothers me is that Anthony recommends 0.1K/W.

A 30 degrees temperature rise is considered acceptable. Leach amp has recommended heatsink of 0.67K/W and so 45W/30deg. Anthony's AV800 recommendation is 300W/30deg. In AV800, that is 3W for every 8W of output. With Leach Amp driving 4 Ohms (200W RMS) that is 3W for every 12..15W of output.

Yet my Leach Amps with very badly placed heatsinks (see another thread) have done well (in 30 deg C room temperature and partly exposed to sunlight) with only 1K/W a channel (actually a not so well placed 0.4K/W), and that is 3W for every 20W of output. This one had no fan and was run as a subwoofer amp at full power for at least 15 minutes while testing that the temperature rise was not alarming.

I have been testing power supplies with www.duncanamps.com's free PSU Designer II. First I tried to simulate two 700VA 30+30V transformers for two AV800 channels, combined as one 1400VA 60+60V, having 6% regulation and 63.6V no-load voltage. I got 88V unloaded rails that went to 73V minimum under load. The software simulates only one side of the supply, so I used a guesstimate of 12.5A into 4 Ohms, that is 50V RMS and 70V peak output. Two channels 4 Ohm is really 2 Ohm, but since this is only one side of the supply, only half of the power is taken from it. Of course this approach averages much. Yet the point is, this way I got only 620W of output into 4 Ohms. And I had 60 000uF of capacitance in one rail (for the very crude simulation of 2 channels).

1. Do you know any cases with integrated fins available in Europe that could house two AV800 channels?
2. What kind of transformers did Anthony use to get 800W out of an AV800? That 90V is, I guess, the nominal peak voltage of the transformer under load minus 0.5V for diodes. Or are unloaded rails 90V in his amplifier?
3. How much transformers would you recommend for 2xAV800? Is 1400VA enough? How about 1800VA?
4. I'm not going to email A. Holton, since I have done that twice earlier and both times either my mail or his reply got lost somewhere.
5. How long leads are allowed from drivers to final stage before I need to be afraid of oscillations?
6. Is the gate resistor better placed near the driving transistor or near the power fet's gate?

The heatsink favored by Holton looks like Fischer SK144 non-anodized (on page A70 in Fischer catalog). I don't know how to build a case with those heatsinks. I really dislike anything wider than 19" and the height should also be kept as low as possible. The transformers that have caught my eye need the 3U.

And of course, more later, at least everythin I forgot to ask now. Hopefully Anthony will come by and see this.

-Kimmo Sundqvist
 
Musher said:

1. Do you know any cases with integrated fins available in Europe that could house two AV800 channels?

As far as I know Anthony recommends 0.5K/W with fan cooling or 0.2 without.
The only case I know is the one from Fischer you mentioned above.
I built an amplifier with 4 AV800's
and used 2 fan cooling agregates(LA 21, 200mm length) from Fischer (one for two AV800's).
The amplifier is running very cool and the temp. rise is 10K at 130% fan power at full amp power.


3. How much transformers would you recommend for 2xAV800? Is 1400VA enough? How about 1800VA?

I used 2x 2.5 kVA for 4 PCBs.


4. I'm not going to email A. Holton, since I have done that twice earlier and both times either my mail or his reply got lost somewhere.

This is very annoying!!!!
My mails seem to get lost too...


The heatsink favored by Holton looks like Fischer SK144 non-anodized (on page A70 in Fischer catalog). I don't know how to build a case with those heatsinks. I really dislike anything wider than 19" and the height should also be kept as low as possible. The transformers that have caught my eye need the 3U.

I managed to get all four into a 19" case with 5U, depth 550mm by using the fan agregates.


Some of my experiences:
1. The MOSFETs are very very sensitive to static voltages

2. When you adjust the bias of the amp be very careful and better test using a current meter in one supply as you can blow the MOSFETS very easily when adjusting to fast.