[Sycraft]

I notice that after some play time, my ground buffers are much warmer than my L/R buffers. I'd guess the L/R buffers are about 35 degrees C and the ground buffers 45 degrees (27 degrees ambient temperature).

Is this something I should be concerned about, and if so what should I do to attempt to find the problem?

This is 4x 5002 buffers per channel.

 

[Glassman]

it's logical, ground channel deals with twice the current of L/R channels..

 

[Sycraft]

Thanks. Please excuse my parnoia regarding this amp. As I said, first precision analogue device, I worry. I am very happy with ti and don't want to find out at some point that it burns out because I screwed something up. I'm considering finalizing it and putting it in a case, so I'm worrying since that's just in my nature. If it passess a scope test and I can't find anything to be neurotic about, I'll probably box it up and hope it's good.

 

[Sycraft]

And actually, nevermind. I don't think there is a perceptable difference, upon furthers poking. The perception of temperature is subjective and I am not doing it right. Damn. Alcohol and empiricism do not mix. I need to grab a laser thermometre and check this for sure. At this point my post is more or less worhtless since I now feel the same temperature (within the limits of impared human perception) with a different biasing method.

I am truly sorry if I have been annoying those that better understand this.

 

[tangent]

Don't bother with the laser thermometer. They've got a wide angle of measurement, and the temperature will vary surprisingly widely over the surface of a chip. Your aim will affect the measurement.

Since you can find an oscilloscope, you can probably also fnd a DMM with a thermocouple. Then you can measure at the same place on all the chips.

If you indeed find that you're getting 45C in the worst case, I wouldn't worry about it. You could drop some DIP-16 heat sinks on top of them, if you were still feeling paranoid.

 

[DCameronMauch]

According to Nelson Pass, audio God, semiconductors should never go above 50 C, if you want them to last a while. If you are at 45 C, there shouldn't be any problems.

 

[tangent]

Quote:

semiconductors should never go above 50 C

Hear that, Glassman? I guess your next revision of the diamond buffers should have room for heat sinks.

 

[dreamnid]

Nothing wrong with burning transistors to start up your day. Nobody would want to use it except yourself - perfect insurance that your amp will not be broken by others

 

[Sycraft]

Quote:

Since you can find an oscilloscope, you can probably also fnd a DMM with a thermocouple.

About 300 oscilloscopes, actually. I work for Electrical and Computer Engineering at the University of Arizona. We do not lack for electronics hardware, as one might imagine.

I'm sure we have a thermocouple for a DMM, so I'll ask the shop for one and give that a try. Or maybe just skip it and put heatsinks on. I'm actually not that worried about them damaging themselves, it's more general worry that their warmth is a sign of other problems. As I said, I though I percieved a significant difference on the ground channel from the other two.

I'm just being neurotic about this. I honestly expected to screw this up in some fasion or another since electronics isn't my field and my knowledge isn't that great. I was quite happy to find it working perfectly and I don't want to screw it up now with something stupid.


As a side note: Tangent, you have any kind of ETA on the PPA power supply you are working on? Right now I'm using my lab variable supply, which works fine, but probably introduces some noise and I'd like to use it for otherthigns anyhow, so I'm thinking of building a powersupply.

Again, thanks for all the support.