Thanks, I thought it might have been already mentioned in some form.

I didnt think about it until I saw the word "vendetta".

Thanks, I thought it might have been already mentioned in some form.

I didnt think about it until I saw the word "vendetta".

Oops my mistake, thought you meant the two huge Blowtorch threads.

Only way to get the noise down

is with discrete parts.

I agree if were talking about the MC input stage.. But I'm not convinced the Klyne is 100% discrete and would love to get a closer look at one. Looks like the same gain modules are used in the line pre section.

I'm not sure I'd bother with the Vendetta. It has to be hand-built unit by unit and to do this you really need some decent test equipment. I tried to breadboard one and it was very unstable and drifted all over the place.

I'm almost positive the glued devices are violet grade. I can give you some part numbers for very nice SOT JFETs that have lower noise than any of the Toshiba parts, but you probably aleady are looking at those.

Kevin what is your experience matching the 2SC3421 or 2SA1358? Just curious to get a rough idea how many to order of these to get suitable values...

Check out the old Dynahi threads on HF where people posted a long list of hfe values to get an idea...

I have slightly different values for things like R30-31 (100K) but that doesn't matter.

Brown Black Black Yellow Brown Red ==== 1 Meg 1% 50ppm ??

I believe that's 100K 1% 50 ppm.

try again

6 Band Resistor Color Codes

or

5 Band Resistor Color Code Calculator

You might consider putting a resistor in parallel with C3 in case someone wants to run it with a MM input. Also, a trimmer on the input is lame. Switched resistors (and capacitors) is better. Is Q16 flipped (drain and source flipped that is)? Same with Q14? Is C11 hooked up right? D1, D2 are 15V.

try again

6 Band Resistor Color Codes

or

5 Band Resistor Color Code Calculator

I was thinking it was 10.0 x 10^4

i'm sure c11 and c14 are correct. Q16 and Q14 may be in backwards, will look at that soon.

D3 and D4 are 1n4746 which are 18v zeners

D1 and D2 are 1n4737 which are 7.5v zeners

This version has to run as a moving coil unit. Reason, the .1uf 1% cap and 750 ohm resistor at the output of the first part are

one of the 3 riaa filters. Another version has the input resistor and pot replaced with a 47k resistor, and then the

gain of the input stage reduced by 20db or so by changing around a bunch of resistors.

a couple of other things fixed too.

Edited December 25, 201015 yr by kevin gilmore

So here is a little historical fun.

Back in 1975 and 1976 while in college i was building and selling

world class phono sections and preamps. (sound familiar...)

Virtually all of which are still in use today.

So while some things like good circuit design seem to be timeless,

other things like double sided circuit boards with heavy ground coverage

are new.

These days i would add a servo to the thing to make sure there was no DC on the

output, something that really did not matter back then, as there were no fully DC coupled

power amplifiers.

I don't actually remember which of the riaa sections i ended up with, there were at least 3,

so i showed the one from Robert Linsey Hood

The original fets were ct134, which i replaced with 2sk170 in this schematic.

http://gilmore.chem.northwestern.edu/kgphono.pdf

this was one of the first boards without a solder mask.

http://gilmore.chem.northwestern.edu/opampfront.jpg

http://gilmore.chem.northwestern.edu/opampback.jpg

Primitive elegance...

Some things never change (like the dual FET front end).

Love the KG logo too!

there seems to be a growing interest in the phono section...

http://gilmore.chem.northwestern.edu/kgphonoservo.pdf

not counting the power supply, this is no more than $50 in parts if you

use the lsk389, a lot less if you use 2sk170's... With the servo, matching

is virtually unnecessary.

I'm trying to find the moving coil schematic i did way back when...

This phono section would be quite interesting, there are very few discrete diy options out there. Maybe we should start a new thread about it?

Its time to get back to work on this, now that the other 2 major projects

are completed.

I've been told to stay away from OLED's because power supply failures can

cause burned up lines. So unless someone has a better idea, i'm going to

go with this one.

http://www.4dsystems.com.au/prod.php?id=114

I want a touch screen, and i want the embedded proc to be able to do

i2s to drive the switching etc.

Edit: screw it, i ordered both the oled and lcd versions.

Also ordered enough stuff to make the thing programmable from

wifi. You could do a version with no touchpanel or knobs at all.

Edited April 2, 201115 yr by kevin gilmore

You could do a version with no touchpanel or knobs at all.

Would this be using the linuxworks/AMB attenuator and an infrared sensor?

The digital attenuator part is similar to, but drives different from AMB's thing.

Its still an 8 bit relay thing, with latching relays. I used AMB's calculator

for the resistor values. My version drives twice as fast as linuxworks thing.

I originally was going to use AMB/Linuxworks thing as is, and give credit but

linuxworks turned into such a jerk... And they still have not actually released the

thing.

Mine may be out before theirs

The version with no front panel, i was thinking an ipad app. wifi...

infrared? that is SO last century...

Edited April 2, 201115 yr by kevin gilmore

RF then...??

Well you clearly know there's an app for that

So i've been working on stuffing the house with lutron RA2.

And there is in fact an ipad app for it. But its too combersome.

So i wrote my own. Runs native on a HP slate 500, or any

windows box. And even on an ipad inside firefox with my

local web server box with ASP.

http://gilmore.chem.northwestern.edu/lutronapp.jpg

Next week i'm going to automate the 3 bathrooms

Full up macro capabilities...

Then integrate in the TV's, dvd players, music server...

This could go on forever.

The preamp app will be just as easy.

Source code for this thing (all of my new things) will be

available, and for the usual and customary price.

Edited April 2, 201115 yr by kevin gilmore

The digital attenuator part is similar to, but drives different from AMB's thing.

Its still an 8 bit relay thing, with latching relays. I used AMB's calculator

for the resistor values. My version drives twice as fast as linuxworks thing.

I originally was going to use AMB/Linuxworks thing as is, and give credit but

linuxworks turned into such a jerk... And they still have not actually released the

thing.

Mine may be out before theirs

The version with no front panel, i was thinking an ipad app. wifi...

infrared? that is SO last century...

yes a bit embarrased by the IR remark. ipad functionality with no screen/knobs seems cool to me.

I'm in for whatever volume control you feel is the most transparent. I intend to use this preamp in a high end speaker setup.

Would the digital attenuator be a separate module (which means I can use it with other builds) or 'fixed' into the KGITSOJC?

I suppose i could yank that part out and do it as a seperate board.

It would be nice to have a board that would would fit into the standard balanced POTs used for the T2, etc using some sort of pins and would mount above the amp board. A simple plug in. The board would also have a ribbon cable for the I2C control stuff. Not sure if the size would workout, but it would be nice.

This is possible, but need to find the absolutely smallest latching DPDT

relay that run on 5 or 12 volts.

With the current relays, the board would be 7.4 x 2.6 inches and

you would need 2 of them.