Aleph output resistors

[Stabist]

I'm gathering stuff for my A5 ...

I wonder about those output resistors - 4 0R47 in parallel - 3W each ... If I understand it right - there are four in parallel from two reasons:
- no need for bigger power factor/resisitor
- 0R47 / 4 is more precise than single 0R117R one ...

But I wonder - I intend to buy Mills resistor - but they are quite expensive - but also allready 5W ones - I gues it won't hurt to use just two 0R23 ones?

Thanks

[GRollins]

As long as the resistance is right and you have the ability to dissipate sufficient heat you can use any combination of resistors you like. The precision aspect isn't really much of a factor--sure it's nice if you can hit the resistance on the button, but it's nothing to get sweaty about. You might want to check the balance between the gain device and the CCS when you're done, just to make sure that you haven't changed that part of the equation. But then, that's something you'll want to do anyway, as the thermal environment plays a part in the interaction of the output devices.

Grey

[jonk]

I love this place. I’m was just wiring up my boards when I thought I’d look up output resistors and the topic is right at the top of the list.
Gray: When you say precision isn’t much of an issue, to what reasonable limits you mean. I’m making mono blocks with 30-volt rails and 8 fets per amp (4 gain and 4 CCS). My source resistors are 0.5R. My understanding is that the output resistor array should be close to the combine parallel source resistors (8 x 0.5 = 0.0625R). I have a nice 0.1R Caddock 25W resistor in my parts drawer I’m wanting to use…do you think that’s too far off?
TIA

[Stabist]

Aha - thanks ...

Btw - how can I estimate power dissipation for those fellows ---

And what is with this interconnection with source resisitors?? Didn't now that they are connected ...
I intend to use 0R82/3W resistors instead of 1R (R40-42 and 64-66 in original schematic) in the schematic - that's a value from experiences with friend's A5 assures 60W/8R ...

Oh and yep - pass section of this forum IS a a GREAT place

[GRollins]

jonk,
Speaking for myself, I'm spoiled...I regard 5% as the outer limit for resistors and prefer to stay less than 2%. Caps and such are another matter, as it's difficult to find tight tolerances, though I'm using 1% caps in the phono stage that I'm working on.
Stabist,
Easy. Just figure out how much current is likely to pass through them and do the I-squared-R thing. For instance, 60W into 8 ohms is roughly 2.75A. Get out your trusty calculator and run that (squared) across whatever R you end up with. Then double that, just because.
Mind you, it's not likely that you're going to run the amp at full power for any length of time. Maybe a few minutes of testing or something, but not with music. So there's a bit of leeway right there. Another factor is that resistors are pretty hardy beasts and can take some punishment. That's more leeway. Still, it's best to be polite to your parts. Run them at half power or less.
The relation of the Source resistors and the output resistor array is described in the patent.
Granted, it's been a while since I scrutinized the 60 schematic, but the 5 should do 60W into 8 ohms without a problem. If I recall from back when I was studying the Aleph series, Nelson set most of them up with sufficient current to go down to about 5-5.5 ohms. The 30W was the exception, as it would double into a 4 ohm load. Not that I'm trying to discourage you from increasing the bias--just make sure you do it for the right reasons.
It sounds better.

Grey

[Stabist]

Grey thanks - I'm a bit for asking for basic formulas and for showing I didn't thoughtfull read the patent

About increasing the bias - practical experinces with Kirc's A5 are that with source resistors 0R82 and that current ratio of the current source around 62% I can get around 90-100W/5R - and that's what interests me - since my speakers are 5R ones ... Without that - it has almost same power at 4R than at 8 ...

THANKS again!