Krell KSA 50 PCB

Hi lgreen

I agree with AndrewT thoughts:
Hang on, 30uV is 114db below maximum output.
Why do we not see a very significant noise on your scope?

What is it you have displayed? It cannot be amplifier output, otherwise noise would be showing!

To me it more looks likes to be a HF pickup from your test leads or something not related to your amp ( HF at mains etc.... ). Try to measure with coupling at "HF-reject" What do you see? Any differences. Also try to alter your ground connection for your test-probe (you could use a separate ground connection to your scope). Also you could add a 100 ohms resistor between your test probes tip and measuring point to see if the oscillation is still visible (if still visible, well it could be from amp but not for sure)

Just my thoughts.

Regards 😎
 
ok

Ok, i'll try what Flodstroem suggested to see if I'm just picking up RF, probably a lot of rf around here.

The changes to wiring were- (a) add new AC inlet, new wires run in same path as old, (b) add 1uF caps across electrlytics, (c) add 10K resistors across electroytics, (d) change Cap grounding- before all 4 caps connected to a terminal on one cap (ground) which went to star on chassis. Now L chan and R channel caps have own local ground and each then has a connection to star (chassis ground). Also, (e) reconnected L and R inputs inverted, so swapped connections at the boards to compensate, now black is - and white is +, but don't think color of wire should make a difference.

I thought the 465M was the same as 465B except ruggedized for military use? I've not had a problem with it other than the stand/handle on one side shearing off. I say "not good" because its specs are on the weaker end of the Tek line.

In my "travels" I've had occassion to "visit" one of their facilties in Beaverton, nice place, nice people. I've also been to LeCroy (NY) and the scope R&D plant at HP (now Agilent) in Colo. Springs. Same can be said of them as well, great places with great people.
 
Yes that is Brian GT PSU boards with 60000uF for each channel
It is 5$ capacitors, i personally don't believe in the more expensive the better - and i'll rather "blow up " cheap caps in the testing fase than expensive ones.
If you look at the right heatsink there is a metalplate "sticking out" my plan was to mount 2 or 3 psu coolers on that one (fore each side) but i don't really think its necessary - might be a little overkill.
Output fets will be mounted at a copper bar and then at the heatsinks.
Comparing my heatsinks with those mounted at the original, i think they are big enough - and coolingfan might not be necessary - psu fans don't give much air anyway.
I am trying very hard to make a WFD friendly design - and i hope i will succed on this one 😀
As an alternative i might mount 2 80 mm vans at the back of the amplifier - but that will ruin my design:bawling:
 
Well actually one side of the heatsinks will be in the chassis, on the top of the heatsinks there is a slice in wich i will put a 4 mm ALU plate as the top of the chassis - with ventilationholes - maybe i could mount two 80 mm fans inside the chassis to pull the heat out of there - that wouldn't ruin the design that much - and now back to the drawing board....:scratch:
 
The amount of heat that is radiated from a surface of a heatsink is determined by:
- the temperature difference between sink and ambient.
- the surface area.
- the thermal factor of the heatsink.

If your heatsinks can thermally be divided into three equal parts, the side that is inside the chassis will radiate 1/3d of the powerhandling, unless the temperature inside the chassis rises.
Means that if you wish to keep the temperature inside the chassis low and retain the nominal thermal factor of the heatsink you'll need to push a lot of heat out.
With a covering lid on top with holes, normal convection is not going to cut it.
 
Thanks fore the advice Jacco
I don''t know the data of the heatsinks somone was so kind to give me 4 of those, if i understand you right fans will be a good idea - i'm going fore that and to make it even better ill make the inner sides of the "channels" in the bottom of the heatsink open - that will make the wiring easyer too, and then mount a pulling and a pushing fan on the top cover of the chassis with a small difference in the voltage supply - eventually adjustable to get the right balance.. that will give a lot of turbulense inside the chassis - thats good i suppose...
 
Hi AndrewT

No matter how you place a heatsink heat goes up thats why you burn your fingers when you turn a lighter on upside down 😀
There is a lot of different opinions on this but what happens if you put the same heatsource on the bottom of a heatsink with vertical fins and on a heatsink with horisontal fins - both heatsinks with the same dissipationarea ?
Will there be any measureable difference ?
If you can prove to me why vertical fins should be better thats fine but i think this will do anyway, i got a gainclone running at home, chips are mounted on a massive ALU bar when i touch that on the opposite of the chip - it's coold when i touch it at the top it's hot.
Actually these heatsinks got "tree" form, you can see that at the top. I'm willing to take the risc.
 
2000VA transformer

HI
I would like to ask a question from those people who orderd transformer fro the audiohobby guy if everibody receved it .
I orderd a transformer a couple weeks a go for the Krell clone but until now I did not receved yet .
I wrote some email to the him but he does not answer to me .
Just to be clear why I'm worried not just no answer , and no transformer , he closed his site and he close at the Audiogon site to .
Please let me know if you are in the same shue like me .
Thanks
 
Hi rov,
look up any data sheet from the manufacturer and you will find that at least 25% of the dissipation capacity is lost if you mount the fins horizontal.

The effect of this is that the devices WILL run hotter.

But you can alleviate that by blowing air through the sink.
 
Hi AndrewT

Well this heatsink is not only with horisontal fins, i haven't read any data on any heatsinks, do they tell anything about how much dissipation will be improoved if the heatsink are painted black ?
I will mount fans just to be sure, but the heatsinks are pretty big
360*120*80 mm.