4in1 tube mic preamp SE/PP TRI/PEN

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

solkatten

Well-known member
Joined
Apr 3, 2021
Messages
316
Location
sweden
Hi everyone....
I would like to share a tube mic preamp I design. I have not seen a mic preamp that can switch between SE/PP and TRI-UL-PEN. It sounds nice och has a lot of diffrent flavours. It has no NFB. The idea was also to design the preamp around cheap but good performing tubes. The russian/soviet 6J32P, 6N1P and 6N6P fit that bill.

I used the UTM-3580 PP or Carnhill WTB2291 (even if it not design as a PP transfomer it does a nice job), I guess you could use a PP transfomer from edcore if you live in US. As input transfomer anything from 1:4 to 1:10 should work fine depending on taste... headroom, s/n etc...

My build also has DI, HP, input and output PAD, phase reverse and 48V... The B+ is around 250VdC....

Picture in Post 16 shows how implement a DI that breaks the MIC connection, and how to integrate a -20dB PAD and Phantom power and polarity reverse.

At the output there can be a output PAD and a load resistor that forces the line stage to work hard and produce harmonic distortion.

I omitted the gain control switch in the input stage, because it interacts with the tri/pen controll and because the triode mode makes the input stage clean enough, more feedback makes the sound sterile in my opinion. This is a coloured pre amp...

Fre. response -3dB ca. 20Hz-30KHz.


Clean setting: TRI + PP= THD+N: 0,18% at 1KHz 0 dBu (without NFB).
Coloured setting: PEN + SE + -15dB OUTPUT PAD: THD+N: 2,7 % (without NFB).
The H2 and H3 content varies depending on mode. PP cancels H2, PEN increases H3.
The TRI/PEN and PP/SE modes can alter the preamps gain by about 20 dB.


The 1K5 dropping resistor in the PSU high voltage rail needs to be in the 7-10 W range otherwise there will be smoke...

Audio sample post #24: 4in1 tube mic preamp SE/PP TRI/PEN
 

Attachments

  • 16A8701D-8667-49C8-B8BF-B431D1A10603.jpeg
    16A8701D-8667-49C8-B8BF-B431D1A10603.jpeg
    163.6 KB · Views: 301
  • 8EB5A2BB-F6EC-42BE-9C76-84018A880691.jpeg
    8EB5A2BB-F6EC-42BE-9C76-84018A880691.jpeg
    100.5 KB · Views: 233
  • BBFE7EA8-AD9D-4872-B8CF-C329D5233B31.jpeg
    BBFE7EA8-AD9D-4872-B8CF-C329D5233B31.jpeg
    137 KB · Views: 262
  • 4in1 PSU.jpeg
    4in1 PSU.jpeg
    168.3 KB · Views: 261
  • Mic-DI-input.jpeg
    Mic-DI-input.jpeg
    115.9 KB · Views: 83
  • Inputstage.jpeg
    Inputstage.jpeg
    187.5 KB · Views: 84
  • Splitter+Line stage.jpeg
    Splitter+Line stage.jpeg
    175 KB · Views: 81
  • Load + output PAD.jpeg
    Load + output PAD.jpeg
    138.7 KB · Views: 95
Last edited:
HI I have not measured the preamp for optimum clean perforamance instead its build as a coloured preamp that goes from clean enough to light saturation. SE got more second harmonics. PP is crisper with cleaner bass. TRI is clean and pentode adds more gain and odd harmonics. I see it as a creative tool. I will post some more build pictures soon...
 
Looks like a lot of options there for different colours, cool :)

Ultra linear mode on the input pentode would be cool too but probably quite a bit more hassle to include and, maybe after all the effort it wouldn't add much more to the palette you have already anyway?
 
Last edited:
Looks like a lot of options there for different colours, cool :)

Ultra linear mode on the input pentode would be cool too but probably quite a bit more hassle to include and, maybe after all the effort it wouldn't add much more to the palette you have already anyway?
Hi you actually get "UL" if the pot is in middle position.
 
I am surprised you got away with using the ungapped VTB2291 transformer - maybe that was intentional - instead of the VTB2290 gapped version.

Cheers

Ian
 
My understanding of UL is that we have a percentage of feedback happening. I forget what David Hafler used as the "magic" number, but 25% would be in the ballpark.
On your schematic, unless I'm mistaken, UL would take a signal from the anode, and present a certain proportion of it to the screen grid. Along with supplying the screen grid's current.

In a pentode, this'd probably require a buffer. If you made a pentode out of two triodes in cascode configuration, then you could do it without a buffer as the artificial screen grid would require no current.
 
Hi Ian a actually have both 2290 and 2291 at hand. I liked the higher primary inductance of the ungapped version, because the PP-stage can be well balanced with the pot it works fine... I connected it 8:1...
 
One thing I forgot to say is 'Welcome' 🙂

Your first post here was a schematic of your own design which is great.

Any suggestions here are part of what you might go through in a typical 'design review' at a company meeting.

Don't take them as criticisms.
 
Thanxs for the welcome. I think groupdiy is great forum and i have found many great projects here over the years. In all practical seens I think you could consider the stage as morphable from triode to UL and pentode depending, the amount of feedback between anode and screen is adjustble from 100-0 procent. The pot acts both as gain control and a harmonic content controll.

Do you mean that the pot only supply ac voltage and no dc current to the screen and that makes it no true UL?
 
Last edited:
On thinking this through a bit more, I think for all practical purposes of your goals you're right.

I honestly don't know that there'd be any sonic colour differences doing it along the lines I suggested.
It might just end up as a whole lot more effort and parts, to end up with, basically, the same sonic options and flavours.

Cool.
 
Hi Ian a actually have both 2290 and 2291 at hand. I liked the higher primary inductance of the ungapped version, because the PP-stage can be well balanced with the pot it works fine... I connected it 8:1...
Yes you definitely get a lot more primary inductance with the un-gapped version. I understand Rupert's 'silk' enhancement is achieved by adding some dc to his output transformer so maybe there is another sonic effect ot be had by tweaking the balance a little bit off centre.

Cheers

Ian
 
Yes you definitely get a lot more primary inductance with the un-gapped version. I understand Rupert's 'silk' enhancement is achieved by adding some dc to his output transformer so maybe there is another sonic effect ot be had by tweaking the balance a little bit off centre.

Cheers

Ian
Yes that is definitely somthing you could explore. I’ve heard some people role their PP primary slightly off center for the second harmonic effect as well... My build has a switch between PP/SE but if you do it with a pot you get asymmertical PP in the middle position, but no DC saturation.
 
Last edited:
SE got more second harmonics. PP is crisper with cleaner bass.
Have you made measurements/analysis of distortion in the different positions?
I would think the fact that the second triode is still connected in "SE" mode actually tends to produce less waveform assymetry than a true SE.
I understand you can't disconnect the 2nd half since it would result in very heavy DC unbalance in the xfmr pri.
 
HI I have not made any measurments maybe its time to get some software. To my knowlegde (just a audio hobbyist) the lower triodes grid is grounded it gets no ac-signal. but act as a constant current sink and balance the DC in the output transfomer in SE mode. The pot only effects the ac-signal. There should be no even order cancelation in SE but full even order cancelation in PP mode and somewhere inbetween in the asymmertic PP range...
 
Last edited:
solkatten, hi! I assembled the triode-pentode according to this scheme.
There is a Hungarian transformer at the entrance. Output transformerless.
Nice very similar to my build, it does not show in my schematic but its exactly as I have impIemented the DI, the input PAD and phantom power...
 
HI I have not made any measurments maybe its time to get some software. To my knowlegde (just a audio hobbyist) the lower triodes grid is grounded it gets no ac-signal. but act as a constant current sink
No. A triode is not a constant-current sink. When the upper triode receives a positive on its grid, current increases, as well as Gm, and Rp decreases. At teh same time, the lower triode sees its plate voltage increase, as do current and Gm, for a decrease in Rp; this tends to linearize operation. Not as much as in a true push-pull, though.
 
Back
Top