Simple discrete Sziklai pre

[rhythmsandy]

Specs here are:
Simple enough to fab point to point, no ICs what so ever, works with any semiconductor, programmable gain, low 0db input distortion and lots of headroom.


Plots shown are with 7.5dB gain.

The overall personality is of course set with bias pot. You can go for 0.04%thd symmetrical clipping 18v or more PP output, lowest thd for hi-fi, or anything in between.
Typical THD is .01-.02% no global FB. No 5ths observed.

The circuit does not model correctly with any version of Spice unless empirically tweaked.

Test load for plots is 3.3k, simulating a low-z tone stack.

The real beauty of old 60's Sziklai CFP is it is amazingly indifference to semiconductors used. I have mixed Soviet Germanium with modern mosfets etc.

Bruce

which software have you used for the FFT?

 

[jfetter]

which software have you used for the FFT?

Here is software link. Site has excellent white papers too.
RA Functions (DSSF3/RAE/RAD/RAL)

Hardware is Steinberg UR22 . I have an HP 331A ,Leader LDM-170 and Fluke 97 w/software for reality checks.

Cheers

 

[jfetter]

update

The circuit has been upgraded to hybrid.
Here is topic link:
http://www.diyaudio.com/forums/tubes-valves/291660-hybrid-output-6.html#post4744962



Further required ip reading, the fet can of course be replaced with a valve.
The idea with patent searches is to get a broarder view of the subject.

https://www.google.com/patents/US78...ved=0ahUKEwjer7Ks-6TNAhUIQlIKHZxNDKoQ6AEIHjAA

 

[Bigun]

R7 is very high given that you'll have some grid current at these low plate voltages.

 

[jfetter]

R7 is very high given that you'll have some grid current at these low plate voltages.

thanks for heads up.
what would be the symptom?
i did not see anything , mv to 10V input.
the thd plots and response are at several different line levels

cheers

 

[jfetter]

Here are plots:


http://www.diyaudio.com/forums/tubes-valves/291660-hybrid-output-5.html#post4732973


-

 

[jfetter]

Lots of data out there on using 12au7 and 6sn7 low voltage.


Here is one running on 12V.

NP-100v12: DIY 12AU7 (ECC82) Tube / IRF510 MOSFET Headphone Amplifier

But a lot more data everywhere , also not sure if noticed , there is not a capacitor on the bias divider output. I discovered better to let it float.

cheers
bruce


-

 

[jerluwoo]

thanks for heads up.
what would be the symptom?
i did not see anything , mv to 10V input.
the thd plots and response are at several different line levels

cheers

The symptom would be a rapid decrease in input impedance. The grid and cathode take on the character of a forward biased diode when the grid is positive in respect to cathode. It may not be noticeable when driven from a very low impedance source such as a signal generator, but, becomes a real issue for a source with a not so insignificant impedance. This also creates a state where the input capacitor will change the average dc bias on the grid, have a look at what is known as blocking distortion in tube amps. The datasheets for most tubes also give a maximum value for the grid resistor for a given bias type, eg. self,fixed,etc. I believe 12au7 is 1meg for self bias and 500k for fixed. Your 2 meg with both grids would constitute a grid leak bias, so you have a combination of all three going on here.

 

[jfetter]

ok thanks, i will test it by inserting a 1M in series between signal source and input.

-

 

[jfetter]

Grid leak test

Thanks Bigun for asking about this. I had not tested for grid leak.

The results are in hybrid thread since this is SS section.
http://www.diyaudio.com/forums/tubes-valves/291660-hybrid-output-6.html#post4747266

Cheers
Bruce

-

 

[jfetter]

Final

not sure if i shared the final, here is zip link with tests.

http://www.diyaudio.com/forums/inst...ment-speakers-almost-extinct.html#post4854659

-

 

[jfetter]

So here is a real-time compressor using non-linear resistors with Sziklai CFP, conventional op-amps or valve/ss Sziklai hybrid. No adjustments. The design can select a single stage or cascaded two stage.