Burning Amplifier BA-2

[tinitus]

Originally Posted by Backbones

Hi Reyntz,
Do you think you need that much heatsink for the diode bridges?
Chris

cant say for sure, but I dont think theres such big need for heatsink on bridge
I have never experienced a bridge even getting just a little hot
mounting on bottom plate have always been good enough
You will notice Nelson doing it that way too
double bridge ought to help too

[reytnz]

Originally Posted by tinitus

cant say for sure, but I dont think theres such big need for heatsink on bridge
I have never experienced a bridge even getting just a little hot
mounting on bottom plate have always been good enough
You will notice Nelson doing it that way too
double bridge ought to help too

I agree with you, this is generally true for people who have adequate chassis size, mounting the bridge diodes on bottom plate is effectively making the whole chassis the heatsink for the bridges. Those bridges do get hot hence the metal backplate. In my case, I ended up with a compact chassis; 30cm conrad. From other people experiences a much bigger conrad than mine do get hot from an F5 with much less amps than BA2. As much as possible I would like to take heat away from the metal chassis, will also elevate the trafo a bit,

[Variac]

The diode bridges on my F-4's do get quite warm.. and they have heatsinks in open air..

[AndrewT]

I have dual bridges in my Krell Klone monoblock.
I was surprised, without heatsinks they became too hot to touch when biased to 2.6A.
A single 7C/W for the dual bridges makes them run cool.

[Backbones]

Andrew,
How do you wire up the bridges in a dual configuration?
Also, how do you calculate the power dissipation per diode? I take it you don't use P=VI because the voltage phase is changing or is that bunk?
Let's say I have 300VA transformer into four diodes (rated each at 20A, 200V as it happens)from two 0-18V secondary windings?
That's 300/36 for the current from both secondaries isn't it, making 8.33A.
What would the voltage be across each of the diodes?

[AndrewT]

measure the voltage across one of the diodes in a bridge rectifier and you'll get close to 700mVdc.

If the diode is passing 1Aac or 1Adc then the diode must dissipate 700mW.
If the bridge has 4 diodes, half of which are passing at any one time then the bridge dissipation for 1Aac and 1Adc is 1.4W.

But the current is not AC. It is pulsed. This changes the dissipation.
Assuming sinewave current flow, my bridges passing 2.6Adc of bias should dissipate 3.64W.
Judging by how hot they got, I suspect actual dissipation was higher.

[Backbones]

Thanks Andrew,
I'm going to use these Schottkys that I have (they fit neatly into the pcb) and I'll see if they run warm before I add a heatsink. I also have some higher rated HexFreds (which don't fit on the pcb). Ideally I'd like to listen to them both to see if they sound that different before settling on one or the other. I have used lower current devices before and I do prefer the HexFred design. They give a more analogue sound to my ears. Just to make things more complicated I have two 35A 200V bridges. The final decision will be ease of fitting I think.
Regards,
Chris

[nightrush]

Originally Posted by Backbones

Andrew,
How do you wire up the bridges in a dual configuration?

Take a look at the F5 PSU recommended by Nelson, it uses a dual bridge.

[Backbones]

Thanks nightrush,
Any advantage of this design over the BA2?
Regards,
Chris

[tinitus]

Originally Posted by Backbones

How do you wire up the bridges in a dual configuration?

some will like the ground to "join" at the last cap pair, and some will "join" the ground at first cap pair