L'Amp: A simple SIT Amp

[Salas]

With 2800pF Ciss and no cascode that IXYS solution isn't as good as those we discussed here.

 

[ra7]

Well, maybe not. It works with an IRFP part also. And is much simpler than those suggested here; simplicity being the motto of this design.

If you want to make it complex there are readymade circuits at hand. The ZV9 aleph ccs would be the obvious choice.

 

[juanitox]

nice much more simple ! do you think we can use the same CCS value for the Sit AMp (with lower B+)

 

[Franz Gysi]

simplicity being the motto of this design.

It is also about building a triode like SE amp with semiconductor and without output transformator.

I agree: the De-Lite CCS circuit seems to be the way to go.

About bulbs: yes, this are only resistors with dissipation bulb. What I dont like is the thermal reaction of bulbs. As long as they are cold, the have nearby zero resistance. The operation resistance is depending on the temperature.

Franz

 

[BTW]

Just like to add.... My friend has built quite a few pass amps .. we have compared them and .... IMHO the Light Bulbs somehow will give the amps a more "tube" like signature which I prefer.

About bulbs: yes, this are only resistors with dissipation bulb. What I dont like is the thermal reaction of bulbs. As long as they are cold, the have nearby zero resistance. The operation resistance is depending on the temperature.

Franz

 

[Salas]

Just one fat gate Mosfet is a slow capacitive solution. I would prefer the light bulb if to adhere to the minimum component motto.

 

[ra7]

You know Salas, I did hear a lot of top end distortion on the delite ccs circuit. I chalked it up to poor layout coz I was not careful in building the circuit. But maybe it was the high ciss of the ccs MOSFET. It did have more power and lower distortion overall though.

Maybe Mike has something similar brewing for part duex

 

[Franz Gysi]

O.K., thinking about, the Salas CCS could be a very interesting variant.

What will be the drain-source voltage for the fet circuit you have shown in post 132, Salas?

Can you show us the BJT variant, mentionned in post 139?

Kind regards
Franz

 

[juanitox]

i have begining mine , just have to wait the power transfo... ( french custo) first 11ohm resitor and source resistor.. is there any drawback to bypass it ?

 

[Salas]

Franz:

~6V on Q1, ~4-5V on Q2 ~0.7V on R1, all at ~2.5A and 15V across the SIT with a 20VAC Tx (about 27VDC rectified under load, so can duck at 25VDC maybe). About BJT I should simulate something.

 

[PSz.]

Here are self and fixed bias versions with an F2 CCS.

 

[generg]

on the way.....

either Salas CCS or the F2 solution........

 

[Salas]

About to try both and evaluate?

 

[juanitox]

my " precious" is also growing , but resistor load , or perhaps CCS if someone sell me a mounted one later ..

my wife just give me the garage to play... cold these days

An externally hosted image should be here but it was not working when we last tested it.

 

[Nelson Pass]

With 2800pF Ciss and no cascode that IXYS solution isn't as good as those we discussed here.

You have to watch the input capacitance, particularly with SITs. I have
some of the 2SK77B's which must be driven by less than 600 ohms to
make flat audio band. This is not ordinarily much of an issue for people
using ss/tube preamps with low output impedances, but is a real problem
for passive preamps and tubes demanding loads of 47K. Unfortunately
this comprises a sizable segment of the target audience for this type of
power amp.

 

[Franz Gysi]

Nelson

Unfortunately
this comprises a sizable segment of the target audience for this type of
power amp.

A step towards studio standards, 600 Ohm, isn't it?

Good news, at least for me

We should convice people to use B1's (or JC-2's) as line amp, getting also rid of cable problems.

BTW, Nelson: Salas was speaking about the Ciss from the CCS circuit, not the input Ciss from the SIT amp.

Franz

 

[juanitox]

Figure 11 is pretty straight forward.

With figure 10, the thing that could be tricky is remembering that the gate bias is negative. If you let it go too positive, Id goes up, and eventually it goes poof.

I tested all my power supplies out of circuit first. Then, I adjusted the bias supply for -12V, powered up the circuit, and adjusted the pot so I had around 13.4V from Drain to Source.

hi michael , 13,4V on the drain for 2.7amp is your sweet spot for the sit with no source degeneration .

so does it mean 16V for Fig11 , or the sweet spot will not the same with a 1R source resistor ?

 

[Michael Rothacher]

All my samples (KD-33, Fig. 11, 50V supply) operate at around 2.45-2.55A, with 15-16V from Drain to Source.

If your SIT runs at significantly higher current under those conditions, then it may not be a good candidate for the self-bias version. (And, IMO, running a larger source resistor with a bypass cap doesn't cut it.)

Other grades such as KE-33 won't even work in figure 11 and will probably go poof if you plug them in there.

 

[juanitox]

thanks michael , great information as i have no distorsion meter i will search now 17,5V /18,5V 2.5amp from drain to ground with my variable load resistor.

 

[generg]

maybe very important if you order SITs acronman sells here

VFET Sony's Rare 2SK82 (KD-33) N-CH Static Induction Transistor x 1 pcs | eBay

the KD-33 Mike suggested


and here

http://www.circuitdiy.com/product/2sj28-2sk82-sony-vfet-for-ta-n88-class-d-amp-ke-33-1-pairs

the KE-33 he doesn't recommend

because I wanted complementary pair for fun

meanwhile I ordered three pairs and one pair was KE-33!

because I googled the internetshop and did not remark the difference to the ebay offer!


be aware!!!!!!!!!!!!!!!!!

you see the wrong one even on my picture... hu, hu!