SK386 DIY Project ....

[Carl_Huff]

Hello fellas,

I've got a box of Toshiba 3K386 FETs that I'd like to consume doing channels if a nice DIY mic preamp project. Can anybody suggest a good circuit?? The Cohen circuit is on FF's website but the design is incomplete and seems overly complex.

If not a mic preamp then how'sa bout a DAO project that fits in a 990 socket?

Best Regards,
Carl Huff

 

[Guest]

[quote author="Carl_Huff"]I've got a box of Toshiba 3K386 FETs that I'd like to consume doing channels if a nice DIY mic preamp project. Can anybody suggest a good circuit?? The Cohen circuit is on FF's website but the design is incomplete and seems overly complex.
[/quote]
2SK386 N-FET 450V 10A 120W

Nice idea, maybe transformerless ribbon preamp. To use several in paralel?
Starter lead battery power?
There was russian schemos adaptabile for you, some two fets
differential at input and nice common base amp as second stage with lower self current. For transformerless ribbon single fet as input and the
some second stage would be better.

xvlk

 

[Carl_Huff]

Well ...

As in dual super matched N Channel FET by Toshiba.

Should make a great low noise transformerless mic preamp or an interesting discrete FET op amp gain block.

Take a peek ....

http://www.ampslab.com/PDF/2sk389.pdf

Best Regards,
Calr Huff

 

[Guest]

[quote author="Carl_Huff"]
http://www.ampslab.com/PDF/2sk389.pdf
[/quote]
Yes, complement to 2SJ109, I know it. Good parts.
Look at some normal transformerless FET preamp.
What mark have you? BL ?


xvlk

 

[Carl_Huff]

What mark??

Mine are all marked Toshiba 2SK386-BL

I am surprised that this part is not used more often. It looks like Toshiba is going to 'end of life' this part soon. Does anybody make an equivalent or better part?? I can't seem to find one. Bummer if it goes away ....

Best Reagrds,
Carl Huff

 

[Bo Hansen]

Hi Carl,

You have talk about two different FET transistors, the 2SK386 that is a high volt power MOS-FET transistor (450V 10A 120W ) and the 2SK-389 that is a dual JFET small signal transistor for audio.
(I hope you have got a box of Toshiba 2SK-389 because this is more exciting than the high volt power MOS-FET)

The 2SK-389(BL) is a great audio FET that you can have much pleasure with if you build pre-amps or simular gear.
It is a dual low noise N-channel JFET with characteristics as: VGDS= 50 volt, IDSS up to 20 mA, PD=200 mW, NF= 0,5 dB/1 kohm.
The BL stands for selected IDSS/drain current between 6 to 12 mA.
(and ther are also a complementary P-channel 2SJ-109)

You can use it as first input pair in differential amplifiers, (op-amps)
or on half as amp transistor and the other half as a cascade or constant current source with good thermal stability.

--Bo

 

[Carl_Huff]

No high voltage here!!!

Small signal matched FETs is what I have ....

I have been looking at the Fred Forsell op amp 'version 1' circuit on his web site and have been toying with the idea of using them as the input pair as shown in his schematic.

I also suspect that they would work well in a transformerless mic preamp. Anybody out there aware of a proven design to offer up?

Best Regards,
Carl Huff

 

[rafafredd]

Fred Forssell has also a schemaic for a transformerless preamp in his site. Take a look. It uses two sk389 at the front, folowed by BB opamps.

 

[Consul]

In fact, it's right here:

http://www.forsselltech.com/JFETMP1.PDF

I hope Fred doesn't mind if I link straight to one of his PDFs... I'll remove the link if you want me to, Fred. :grin:

I'm curious about two resistors marked TBD, which I assume means "to be determined." Might that be because they might be different depending on the tolerance of other parts in the circuit, or because the circuit just needs some more experimentation before values can be decided upon?

 

[SeventhCircle]

Has anybody built Mr. Forssell's version of this circuit? The original circuit, from a 1984 AES paper, used a pair of LM394s on the front end and recommended using dual op-amps for better offset matching. The more I look at it the better I like it. It's simple, elegant and has great performance. One thing I'd like to see is a beefier output stage. I'm working on a prototype with a pair of simple diamond buffers on the output. I was thinking of using a JFET front end, but buying 2SK389s is too much of a pain.

 

[Carl_Huff]

The mic pre circuit on Fred's site ....

I looked at the circuit on Fred's site some time ago and remembered it as being incomplete. I'll look again but this time with an eye to prototype something.

Best Reagrds,
Carl Huff

 

[Carl_Huff]

Hey Seventh Circle ...

I'll send you a pair of SK389's if you share your thoughts and design.

Best Regards,
Carl Huff

 

[SeventhCircle]

Thanks, but I've already decided to dump the JFETs. I just have no patience for chasing after obsolete parts, even if they're great. National still has the LM394 in full production, and it's used in so many other applications I doubt it's going to disappear anytime soon. There's no reason you couldn't use a pair of those K389s on my board, though.

 

[Carl_Huff]

Hey Seventh Circle,

Have you had the time to spend any time on this? I was curious as to your current thoughts and status. I was toying with the idea of starting a project of my own inspired by the circuit on FF's site.

Best Regards,
Carl Huff

 

[CJ]

Check the Borbely site also.

 

[Carl_Huff]

Good advice CJ. Thank you.

 

[tk@halmi]

[quote author="SeventhCircle"]Has anybody built Mr. Forssell's version of this circuit? The original circuit, from a 1984 AES paper, used a pair of LM394s on the front end and recommended using dual op-amps for better offset matching. The more I look at it the better I like it. It's simple, elegant and has great performance. One thing I'd like to see is a beefier output stage. I'm working on a prototype with a pair of simple diamond buffers on the output.[/quote]

Tim,

I have breadboarded the orginal AES circuit. It measures outstanding, but I have not had the chance to listen to anything through it (too many projects). It looks less prone to distortion in the 20KHz range than other discrete front-end topographies. People keep passing up this circuit becuase it has monolithic opamps that are not so hip these days. But the beauty of the circuit is that it uses components readily available, requires no measuring and sorting of components, and you can have a channel for about $30 including decent input caps and rlog pot.
I have started a PCB design that I should finish in the Xmas break.

Cheers,
Tamas

 

[Carl_Huff]

Tamas,

What are using to control the gain?

 

[tk@halmi]

[quote author="Carl_Huff"]Tamas, What are using to control the gain?[/quote]

About a year back I ordered 25 pieces of a 1k custom rlog pot from potentiometers.com.
The original Cohen circuit works well with a 1k rlog pot, but the Forssell circuit needs a 5K to get around the 6dB minumum setting. With a 1k pot you would get a 14dB minimum setting.

Gain for Cohen circuit:
(600/Rg)+1

Gain for Forssell circuit:
(4000/Rg)+1

Tamas