SIT-Preamp

Behind this schematics stands an idea to try using SIT transistors with triode-like output characteristics for simple preamp design. The common source stage based on KP926A Russian SIT transistor is used for amplification (nearly 20 dB) at 100V power supply voltage and 175mA quiescent current, class A operation. Unfortunately, KP926A has 1600pF of Ciss (instead of this transistor much better would be SONY 2SK60 which have Ciss = 190pF). To handle the 1600pF of input capacitance of KP926A one needs a buffer stage, which I decided to make like well known j-fet follower with 25mA quiescent current and plus-minus 15V supply voltages. One needs a decent j-fets for such a buffer, like for instance CP650. In my case I used Russian KP601A, which is close to CP650. The Ciss for KP601A is 6pF. Before this buffer a simple volume control based on 10kOhm ALPS resistor stands more than comfortably. In fact, I used a voltage divider constituted by 7,8kOhm low-noise constant resistor and the variable part of ALPS below it. Capacitors used include Elna Cerafine, Mundorf polypropylene etc. This schematics has no NFB, its output resistance (measured by immitance meter) is 200 Ohms. I compared it sonically with the good Parasound JC2, which has only 60 Ohms output resistance. Power amp was 50W version of Pass Zen9, Shanling CD T-1500 player, speakers are Paradigm Studio 100 v4, cables Stereovox (speaker), Nordost (interconnects), power is provided by PS Audio Power Plant Premier. I can say that I love my Parasound JC2 very much, but in this comparison SIT-preamp sounded more musical. It gave more detailed and pleasant low-mids and the bottom end. At the beginning I thought that SIT-preamp has lower output resistance than JC2, but it is not. So, I do not know what a magic stands behind it. Speaking about transparency of highs the two preamps are very close, in some cases maybe JC2 is a bit better (difference on the level of tiny cabling effects), but JC2 is connected to XLR outs of Shanling (known of being better) on contrary to RCA outs for SIT-preamp. I would be grateful for help in understanding what a “devil” is behind this SIT-sound.
 

Attachments

  • SIT-preamp.jpg
    SIT-preamp.jpg
    88.2 KB · Views: 3,321
Very nice. What kind of resistor did you use for R7? How about an active load or a choke? Output impedance will go up but still be reasonable.

I did try CCS load and local NFB, it gives no plus to sound compared to simple resistor load. For R7 I used 11 pcs of 3K3 2W resistors in parallel, but even better would be using of 3K6 resistors in order to make them less hot. A choke I believe would be a best solution, but I did not try it.
 
Hey Vladimir,
Those SITs look promising. Do you have a spec-sheet? If searching for a low Ciss, triodelike FET the one below looks great. Hard to find though and expensive on eBay though. A pity they don´t make tham any longer

Another way to get a FET like 2SK170 to get triode-characteristics is to use localfeedback.
 

Attachments

  • 2SK79.pdf
    56.3 KB · Views: 668
Hey Vladimir,
Those SITs look promising. Do you have a spec-sheet? If searching for a low Ciss, triodelike FET the one below looks great. Hard to find though and expensive on eBay though. A pity they don´t make tham any longer

Another way to get a FET like 2SK170 to get triode-characteristics is to use localfeedback.

Lars, thanks a lot for the interesting datasheet. As of KP926A, I did some measurements by myself, enclose it. Enclose also datasheet in russian just for some minor information.
 

Attachments

  • SIT-preamp Measurements of KP926A.JPG
    SIT-preamp Measurements of KP926A.JPG
    172.2 KB · Views: 2,955
  • SIT-preamp KP926 russian datasheet.pdf
    42.4 KB · Views: 414
Hi Vladimir,
I really like your preamp and I would like to build one myself but I have a problem finding a source for KP926A (google wasn't my friend this time). Can you recommend the on-line store or some other way to buy these SITs ?

I suppose this is the device you used:
 

Attachments

  • kp926a.jpg
    kp926a.jpg
    45.3 KB · Views: 2,653
Hi Vladimir,
I really like your preamp and I would like to build one myself but I have a problem finding a source for KP926A (google wasn't my friend this time). Can you recommend the on-line store or some other way to buy these SITs ?
I suppose this is the device you used:

Yes, the foto is correct. I made a search at russian site www.ChipFind.ru and it gave several possible suppliers

I bought them at http://www.anion.ru
 
Last edited:

Attachments

  • kp.jpg
    kp.jpg
    97.4 KB · Views: 2,676
Last edited:
Have you seen this: ‚’ep

I know this site for a long time, but the simple ciclotron schematic with 2SK183 is very interesting, thanks. However, I see no hint whether SITs (not like the huge 2SK183) are good for preamps.

A friend of mine says that it is important for preamp to have an idle current 150mA or more, but can't explain why.
 
Last edited:
... How about an active load ...

Theoretically an active load is better than resistor, since current consumption becomes fixed and signal current does not go through the power supply. But in this case one must employ NFB for achievind desirable gain of the stage. I tried an arrangement like a combination of Zen2 and Zen9, with 2SJ162 lateral transistor in the current source, without input buffer, and the resulting sound was not very detailed.
The present simplest resistive load, though impose tuff requirements to power supply quality, sounds good. I used Elna Cerafine 470uF + 2,2uF polypropilene at the end of regulated PS (from 25V to 15V) for the buffer stage, and 10000uF Samwha + 15uF polypropilene + 2,2uF polypropilene at the end of double (connected in series) PS (150V to 125V to 100V) for the SIT amplification stage.
Rectifiers in the PS are Shottky diodes.
Of course, PS as a part of signal chain becomes complex, but it is a payment for the gain stage simplicity and NO NFB.
 
Thinking about further possible improvemens of simple SIT Preamp schematics, I plan to make parallel stabilizer with current source on the top of it for the amplification stage. Could somebody comment it, especially the choice of active components ?
 

Attachments

  • SIT Preamp With Parallel Supply.JPG
    SIT Preamp With Parallel Supply.JPG
    68.6 KB · Views: 1,460
Active device 2SJ200 are very good as old Hitachi L-mos. I don't like the zener diode as voltage reference. Try Norton equivalent as J. Curl. However, i think is better to put a big capacitor in parallel at the output of the 2sj200 to close the AC path of the output stage.

Also sincerely i think your amp is almost a SE Power amp than a preamp.

Are there other little russian SIT devices ( about 1-10 W) that you can employ?

Otherwise i am finding for those little Sit :D
I not understand russian language: can you tell me about?

Francesco
 
Active device 2SJ200 are very good as old Hitachi L-mos. I don't like the zener diode as voltage reference. Try Norton equivalent as J. Curl. However, i think is better to put a big capacitor in parallel at the output of the 2sj200 to close the AC path of the output stage.

Also sincerely i think your amp is almost a SE Power amp than a preamp.

Are there other little russian SIT devices ( about 1-10 W) that you can employ?

Otherwise i am finding for those little Sit :D
I not understand russian language: can you tell me about?

Francesco

Francesco, thanks for your comments. Indeed, this schematics could be considered as a current amplifier (Rout=200 ohms) for speakers like Fostex etc. It sounds good, but some doubt is remaining that with shunt type power supply it could be even better. Now I escape from the power supply modulation only due to big polypropylen cap. I should also check the Norton equivalent.
As of medium power SITs, there are nothing among russian parts. I guess the only one in the world is 2SK79, indicated in the poat #4.
There is very good medium power russian j-FET, KP903A, N-channel, P=6W, Io=450...700mA, S=85mA/V, Ciss=15pF, Uds max = 20V. They look excellent in cascoded mode for driver stages of Pass Aleph like schematics, and for capacitorless buffer preamps.
 
Francesco, thanks for your comments. Indeed, this schematics could be considered as a current amplifier (Rout=200 ohms) for speakers like Fostex etc. It sounds good, but some doubt is remaining that with shunt type power supply it could be even better. Now I escape from the power supply modulation only due to big polypropylen cap. I should also check the Norton equivalent.
As of medium power SITs, there are nothing among russian parts. I guess the only one in the world is 2SK79, indicated in the poat #4.
There is very good medium power russian j-FET, KP903A, N-channel, P=6W, Io=450...700mA, S=85mA/V, Ciss=15pF, Uds max = 20V. They look excellent in cascoded mode for driver stages of Pass Aleph like schematics, and for capacitorless buffer preamps.

T4 act as a current generator and T5 as a voltage generator.
Yes i like your amp, it is a current amplifier or as i prefer to call it "transconductance amplifier". (see my site also) :)

Other little V-FETs are 2SK63 (equiv. to 2SK79) 2SK69/2SJ19, 2SK78/2SJ24 and 2SK75 as medium power, but all they are very scarce (i have 2sk79 only).

Thanks for your advice about KP903A but at this moment i need a P-CH type.
Is there a similar jfet as KP 903A in a P-channel type?

Francesco