[milosz]

A while back I bought a used Cornet 2, it has 12AX7 gain stages and a 6SN7 cathode follower.

It sounds very thin, and I checked the frequency response and the RIAA curve isn't right,  it's -9 dB at 20 Hz and +3 dB at 20 kHz.


The builder I got it from substituted a lot of Russian caps- some teflon, some P-I-O.  All the values measure OK except C202L & C202R measure 55~57 nF instead of 47.  So I'm going to change that- that may be the source of the RIAA problem, or maybe there's also more going on, I'm not sure what the original builder changed other than wiring an octal 6SN7 socket into the circuit- don't know if he changed any resistors over the original noval design, maybe the 12AX7's have a low plate voltage, or a high plate voltage over the original design.  (Preamp is torn apart right now, can't  quickly sweep it back together to measure B+ for 6SN7 and 12AX7's.  I'll check that in a day or two.)

I haven't checked all the resistors around the 6SN7 cathode follower, I assume this octal tube needs different voltages from the original noval tube- should R220/R221 (each 1k8, in series in the original all-noval design) be changed to 3k3 like in the "Octal Cornet" diagram?

Should R211 be changed to 6k8 from 10k?

Are any changes needed in the CCS that the 6SN7's cathode is standing on?

??

[hagtech]

R211 is not that critical of a value.  The only "octal" design I did was with all tubes, not just the output stage.  But it should work ok.  Measure the EQ again after the 47nF is fixed.  Also check C203, as it seems the LF rolloff is too high.

jh

[poty]

If you have the first two stages of the same design and tubes complement as noval - then you can use the marked values for voltages and currents. The main "tune element" - R211.
As far as I can say - the only changes in the schematic should be the CCS resistor R218 - to bump the current for the last stage - and tweaking the drop resistors R220-221 to save the 330V on the anode of the V3. Also it seems you should tweak the R223 to have 6.2V for heating and CCS.

[milosz]

Also check C203, as it seems the LF rolloff is too high.

jh

Thanks, for C202 I ordered some 47 nF @630v  caps, 3% film from Kemet.  Also a pair of 400v 47 nF Russian PIO's  we'll see if those are close to an actual 47 nF.  Of course my C meter could be off a little, too, but it's supposed to be +/- 2%.

C203R  is 108.4 nF  and C203L is 107.6 nF    -  within 10%, would + ~8 nF  have much impact on the LF rolloff?  If anything, wouldn't a slight increase in C203 give a LOWER LF rolloff?

[hagtech]

You 100nF are ok.  Let's see what the new EQ looks like. 

jh

[milosz]

OK  replaced the Russian PIO's that were 55~57 nF for C202  with some Kemet film caps said to be 47 nF +/- 3%, and hand-selected out the closest two of ten at 46 nF each.

RIAA is MUCH better now-  see the graph-


From 40 Hz to 15 kHz +0.52 -0.81 dB

From 20 Hz to 20 kHz -1.88, +0.84 dB

Thanks for the help

[milosz]

Here's what it looks like....