Pre amp therapy / therapy pre

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Oops, I haven't followed discussions for a long time.

I use toroidal power transformers made to my specks. They have electrostatic and magnetic shields. The trannie was designed for my commercial products and didn't have optimal secondaries for this project, but it is fine.

Now I can tell something about the sound. It has been in intense use at my friends studio. So far he has felt that pentode mode has the most to offer but other modes have been used too. He uses it as a line amp and mic pre both with minimal distortion and with different amounts of distortion up to fuzz-sounds. Good on drums if he is to be believed. Very versatile amp, which indeed was the whole idea.

-Jonte
 
Original thread by Jonte Knif:
http://www.groupdiy.com/index.php?topic=38252.msg471585#msg471585
some tips from the drawing board
http://www.groupdiy.com/index.php?topic=42919.msg554436#msg554436

Jonte, I'd like to thank you very much indeed for making this great studio tool. It is a pre amp but in reality it is much more than that.
I have several other preamps but this one and my version of the 9K see the most use.
I use it for:
Gain make up after my 26ch passive summing mixer.
To process individual instruments (like software synths which really come to life through this thing).
Record various sources.
Re-record dead and sterile mixes through it.
Due to the I/O transformer configuration that I used, I have slight  extreme bass and  extreme highs roll off .
Hi pass filter is very useful as a creative tool not only as a band aid. I expanded original filter to 8 frequencies
(30-60-90-150-250-330-440-750Hz)
I'm planning the low pass as well,  have to wind  the inductor, soon I hope.
The preamp is very quiet in any setting and has a lot of gain.
I've added several features to the original design and my PSU is a bit different to the original design  (6.3V- 338 regulated, 783 for the 48V, HV is the same as on Jonte's schematic, and 317 regulated for the relays and VU buffer).
I have the input transformer (OEP) relay switchable primaries and secondaries for series parallel winding configuration.
Everything apart from UL,P & T choices is relay switchable.
Build inside dead M1R case, TRS for ins-outs, was lazy to make bigger holes for XLRs.
PCB is huge, double sided.
Filter caps soldered on switches.
Russian pio in the sound path.
Russian 6n23p on the input
Mullard NOS E81L on the output.
I'll be building at least 4 more channels, this time with line input option only and different input and output transformers.
I'm having it for a couple months now and I'm using it every time I do music.
Cheers
1.jpg

2.jpg

3.jpg

4.jpg

6.jpg
 
That is an extrememly nice build project. You even made your own PCBs..impressive. Carnhills on the outputs, wierd as steppdowns?
 
Thank you. Yes I made therapy more complex than it was originally, but kept Jonte's audio engine and HV supply.
I love chokes for the HV, I love what they do for the sound.
At first I didn't see the need to go upper market transformers (like Jonte did) but now that I have it build,
I can only imagine benefits from using a better iron. I used Carnhill VTB2290, 9600:600 gapped transformer on the output and
OEP A262A3E for the input. Carnhill rolls off before 20kHz and OEP  starts to roll of around 30-40Hz.
However I didn't feel lack of bass or highs at all, ever. The funny thing is that I do mostly electronic music with plenty of sub bass.
Ok, there is some energy taken away at the very bottom end of the spectrum but I would filter it out any way.
The hi roll of takes away digital hash at the very top. I processed some rock  (digital masters, I.T.B. mixes) material and they sounded more "record like"
with therapy than without it. It is a very nice processor. I would not call it a preamp, even though one can certainly us it in that way.
Better iron would show more of the harmonics generated by the tubes and the circuit and that is what I'm looking for in my next therapy build.
Best
 
syn said:
I used Carnhill VTB2290, 9600:600 gapped transformer on the output

So it actually works? You're probably the first one who ever used it in a true SE output configuration. I admit I have been skeptical about this transformer.
 
@ tommypiper
                  thank you, yes, I hope Jonte will show up again, too. This is so trouble free project. He is truly a great designer.
@ desol
                  Cheers.
@ Kingston
                    no, actually it does not work. It rocks and for the money it rolls as well :) Seriously it is an inexpensive, cool sounding transformer.
The other problem is that therapy has to offer much more, than I/O transformers that I used can deliver (and cost). For the money I can not think of better transformers for the job.
 
Hi,
My 2 cents on, this project : here's a P2P layout I've made, heavily inspired by Jonte's layout but with some minor mods and with a 12 positions Hi Pass added. The HP board is a little piece of perfboard bolted on the main turret board, and tubes and PSU filter caps are mounted on a 90° piece of aluminium, like on Knif's pic.
Therapy%20Pre%20P2P%20layout_zpsnw0dnmze.png

I prefered using labels instead of wires for more clarity.
Please feel free to comment.
Best regards.
Eric
 
Hi,

I finally decided to make a PCB for this project, in order to be able to use the Lundahl LL1528 I have in stock, and also to use relays for some switching, to make the audio path as short as possible. Note that the input switching  (phase, PAD...) are done externally. The tube sockets are ceramic that go through the PCB and are wired from the bottom. That's in order to be able to fit the PCB in a 2U rack, else the height would be excessive.

Any opinion on the routing ?

For now I have used Eagle's "polygon" function to make the ground trave as beefy as possible. Would it be a good idea to fill all blanks on the PCB with ground ?

Thanx in advance for your advices.

Best regards.

Eric
 

Attachments

  • therapy_PCB.png
    therapy_PCB.png
    153.7 KB · Views: 84
Here comes the schematic image, for the names correspondance.

BTW, the huge caps are russian K75-10 PIO, and the high-pass caps are on a separate board.
 

Attachments

  • therapy_schem.png
    therapy_schem.png
    63.9 KB · Views: 99
Hi !

I'm pleased to say that I finally found some time to build a first channel of this preamp and it's working fine ! A second channel will be coming next, I just wanted to test my PCB routing before teching and building the second one. Was actually a good idea as I spotted quite a few mistakes on the one I built, from too small room for certain electrolytics to wrong heaters wiring ;)

I used Lundahl LL1528 mic trafo because I had it in stock, and Carnhill VTB2290 because those Lundahl output trafos (2x200€) are just off-budget for the moment. Thanx to Syn for the tip BTW.

To have the XLR and jack on the front panel, and also avoid taking too much space for them, I used a combo XLR/jack. The DI relay is actually switched when the jack is inserted. Not a very "vintage" approach but whatever ;)

Anyway it works fine, is dead silent even when all is set for max gain. The only annoying thing the HUGE output level, that makes my soundcard's input (RME Multiface) saturate very easily.

I must admit that for the moment I mostly played with a guitar in it to try squeeze out some overdriven tones and it is really nice, especially in penthode mode ! Now I'm thinking of building the second channel with some extra options dedicated to get an even better guitar tone.

First thing I thought is to add a switchable grid stopper on the pentode, because at very high gain and distortion I have some blocking distortion (http://www.aikenamps.com/index.php/what-is-blocking-distortion). Sounds unpleasing and can be seen on the scope at the grid. 100k-470k should be fine, but has to be switchable not to create an undesirable filtering with the grid capacitance while using it as a mic preamp.

Maybe switching along the 220uF cathode bypass cap for 10-22uF could help as well to get a cleaner distortion sound.

Any other ideas for this purpose ?

To be continued...

Best regards.

Eric

PS : on the picture obviously some wires need to be cleaned and made shorter, it'll be done in the end. The 2 tubes in the middle only have their heater wired. If I had to do that again I'd better go for regulated 6,3VDC not 12,6 with heaters in series.  The two 15R also get quite hot (had to bodge two 100R across them to get 12,6V, will be replaced by a clean 11R later)
https://drive.google.com/file/d/0B6K-0QVHN7zUekFkUTM5TUxqNkE/view?usp=sharing
(for some reason I couldn't attach my pic the regular way, was refused....)
 
First measures with HOLM Impulse / RME Multiface :

MIC/LINE = LINE
GAIN = min
LEVEL = min
Feedback = OFF
Load = 150R
Low Pass = OFF
-> Comparison of Pentode/Triode/UL

Curves have been matched in software, pentode normally has the most gain, then UL, then triode.
Funny how the triode mode is unexpectedly cleaner than the UL. This is the same if I reduce the input signal instead of matching curves in software.
 

Attachments

  • Line Gain min Level min FB OFF Load 150 Mode comparison.png
    Line Gain min Level min FB OFF Load 150 Mode comparison.png
    57.1 KB · Views: 36
Load comparison  (BTW I removed the 1k resistor, replaced it with 330R and replaced 330R by 160R, as 150R is the "optimal" load according to Carnhill) :

MIC/LINE = LINE
GAIN = min
LEVEL = min
Feedback = OFF
Mode = Pentode
Low Pass = OFF

When load increases the gain increases and so the bandwidth decreases...
 

Attachments

  • Line Gain min Level min FB OFF Mode Pentode load comparison.png
    Line Gain min Level min FB OFF Mode Pentode load comparison.png
    54.2 KB · Views: 27
Mode comparison with feedback ON :
MIC/LINE = LINE
GAIN = min
LEVEL = min
Feedback = ON
Load = 150R
Low Pass = OFF

Feedback results in a cleaner THD and more linear frequency response at the expense of less gain.
 

Attachments

  • Line Gain min Level min Load 150 FB ON Mode Comparison.png
    Line Gain min Level min Load 150 FB ON Mode Comparison.png
    53.3 KB · Views: 28
Hi,

Second channel is up and running ! I've added the "guitar" switch I talked about earlier, and it make a huge difference on the distortion quality.

At the two highest gain settings and in "normal" mode (grid stoppers = 1k), I noticed that I have some oscillations at about 166 kHz. Those oscillations are killed when switching to "guitar" mode (grid stoppers = 100k), probably because of the grid capacitance which makes a RC filter with the grid stoppers.

BTW I added 1 gain step to Jonte's schematic, which totally disconnects the feedback to have a configuration similar to a guitar preamp, so it's actually the very last step on Jonte's schematic that's problematic.

What would be the most clever way to kill those oscillations ? Increasing the "normal" mode grid stoppers slightly ? Adding a small cap across V1A or V1B anode resistor ?

I must say I don't hear much noise rise when switching from 1k to 100k, at about 83dB gain there's a slight buzz domainating, but the high frequency get rolled off a bit more, as you can see on the joined image (pentode-no feedback-load=150R)

Thanx in advance for your advices. Pics of the final build will follow soon, as well as some audio files ;)

Best regards.

Eric
 

Attachments

  • Guit-Norm_Comp.png
    Guit-Norm_Comp.png
    33.5 KB · Views: 19

Latest posts

Back
Top