L'Amp: A simple SIT Amp

[Roscoe Primrose]

Looking at the drawing attached to post 989, -ref is nothing more than a label.

Roscoe

[juanitox]

@juanitox did you do your vote ......

no Merkel has done it for us

i like your " brainstorming" for the CCS version, but at last what have done N.P to prefer a Res load to CCS for his bigger Sit AMP ?

[generg]

Originally Posted by juanitox

no Merkel has done it for us

i like your " brainstorming" for the CCS version, but at last what have done N.P to prefer a Res load to CCS for his bigger Sit AMP ?

Oh, we have Cohn here, should we send Angela to you..... but will Carla say...?


the brainstorming is simply because SIT-2 was published earlier and I supposed it to be a Mu-follower like J2, so I was fascinated by the idea to drive a Mu with two Semisouth alone without J-Fets. Indeed I succeeded to get it running, but the gain of the two "pentodes" was too high about 25db. Much hum and hissing was the consequence. But the sound was good and then Mike started with the SIT and it has a lower gain so I saw the chance to build a better Mu with less gain......
That is the reason I was fixed a bit on the Mu....
But for sure the resistor solution has the best details and attack at least with the thick film resistors.

I am sure Nelson saw small advantages for the resistor solution and maybe the cap he chooses is a bit more warmish and can compensate a bit the directness of the thick films.

[juanitox]

something i don"t understand in the Sit 1 pic , i can see the Two white cap for inpout and output coupling, the power supply in the center but what is for the seven big caps near the loading res , additional filtering?

[PSz.]

I've been staring at those big caps in that picture too. Additional PS filtering, output coupling, slow start, and source R bypass are the options that come to mind.

I want to know how the gate bias voltage is generated and how the slow start is implemented in the SIT1. Knowing this would help sort those caps out.

The gate biasing circuit of the SIT2 prototype isn't obvious in the SIT1. Perhaps they are doing it differently.

The SIT1 has a nice slow turn on (~10 seconds until the bias settles). Nelson accomplished this same thing on the F2 with BIG caps, so I wouldn't be surprised to see brute force used here too.

[dantwomey]

I've never mounted a TO-3 before. Are these two pieces good candidates or are there better choices?

Insulator
Digi-Key - BER166-ND (Manufacturer - HF115AC-0.0055-AC-05)

Socket
Digi-Key - 4601K-ND (Manufacturer - 4601)

Regards,
Dan

[Roscoe Primrose]

You're probably better off with mica & grease, but the socket looks fine (or, you could do without a socket)...

Roscoe

[vdi_nenna]

I picked this up because it was a complete kit.

http://search.digikey.com/scripts/Dk...5K-ND&x=15&y=5

4725K-ND

[dantwomey]

Heatsinks Revisited. I'm wondering if the Wakefield 433K would work.
433K - WAKEFIELD SOLUTIONS - Heat Sink | Newark/element14 Canada

It seems to have similar thermal efficiency to the aforementioned Aavid D180-20.
D180-20 - AAVID THERMALLOY - HEAT SINK, T SLOTS | Newark/element14 Canada

Regards,
Dan

[Michael Rothacher]

If that spec sheet rating means 42C above ambient, then no.

The forced air rating for the 433K is .28C/W, so .84C/W for natural convection sounds feasible, so again, no.

The D180-20 has a natural convection rating of .33C/W according to the catalog.