Results 1 to 35 of 35
Like Tree4Likes
  • 1 Post By Merlinb
  • 1 Post By loudthud
  • 1 Post By nickb
  • 1 Post By twiggybush

Thread: Tube Compressor

  1. #1
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276

    Tube Compressor

    So I heard Come Together on the radio the other day... Beatles original version. It reminded me that I absolutely love the sound of the lead guitar. It sounds "clean" but quite compressed and "tubey." Has anyone played around with diy tube compressors for guitar? I'm now on a mission to build a guitar amp with built-in tube compression, but without "distortion." I know, technically compression is distortion but ya know what I mean. A cleanish sound, but compressed. I've done some surface research on google. Seems there are two main types of tube compressors: Variable-mu and opto-coupled. I think for now I'd like to stay as simple as I can with this, but not without getting the right sound. It has to have that spongy feel, or it's all for naught. I certainly don't want a hard limiter feel. I think a custom built 5E3 modded to be loud and clean would be my ideal amp circuit to incorporate this into. So I guess my question is does anyone have any schematics they would share? Or any thoughts on this.

  2. #2
    Member
    Join Date
    Mar 2011
    Location
    Ulm, Germany
    Posts
    86
    Search for the schematic of the Trace Elliot V8, there you will find a tube compressor inside the preamp.

  3. #3
    Senior Member
    Join Date
    Dec 2007
    Posts
    858
    Could you use a large value screen resistor on the output tubes to give you that compression? Or maybe use a rheostat instead of the screen resistors?

  4. #4
    Senior Member
    Join Date
    May 2010
    Posts
    221
    Google gates sta level schematic, Teletronix LA2A, Fairchild 670, Analogue Tube AT101

    try gearslutz.com or groupdiy.com

    MAXSON CA-1589 Tube Limiter and Tubes

  5. #5
    Senior Member
    Join Date
    Aug 2006
    Location
    Lancashire
    Posts
    579
    Quote Originally Posted by lowell View Post
    Seems there are two main types of tube compressors: Variable-mu and opto-coupled.
    All valve compressors are variable-mu (short for variable mutual conductance, or variable-gm).
    Opto compressors are quite separate, and built from light-dependent resistors. You can bolt a valve preamp or side chain onto an opto compressor, but it's a bit grandiose to call the end result a "tube compressor". For example, the LA-2A is just an opto compressor with some valves in it, not a valve compressor. Same goes for the EH blackfinger.

    You could get the same effect from an SS opto compressor (or indeed any SS compressor) and a separate valve preamp.
    J M Fahey likes this.

  6. #6
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Quote Originally Posted by Diablo View Post
    Could you use a large value screen resistor on the output tubes to give you that compression? Or maybe use a rheostat instead of the screen resistors?
    While a largish Rs provides some squish it's definitely not a "compressor." There is not envelope, attack etc.. controls. Also, as Rs goes up, bias voltage needs to go more positive, thus reducing the maximum clean signal one can apply to the grid. Same goes for a self-biased amp and Vgk. That all being said one could still run this setup "clean" and have some gain makeup afterwards. It's something to consider.

  7. #7
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Thanks Merlin so I'm thinking I'll stay away from an opto-coupled compressor. That seems to me like "faking it." It's akin to those low voltage "tube" stompboxes.

  8. #8
    Noodle of Reality Steve Conner's Avatar
    Join Date
    May 2006
    Location
    Glasgow, Scotland
    Posts
    7,246
    Quote Originally Posted by lowell View Post
    While a largish Rs provides some squish it's definitely not a "compressor." There is not envelope, attack etc.. controls.
    If you use a large shared screen resistor with a capacitor, then you have a time constant that gives a more "compressory" feel.
    "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

  9. #9
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Quote Originally Posted by Steve Conner View Post
    If you use a large shared screen resistor with a capacitor, then you have a time constant that gives a more "compressory" feel.
    A capacitor in parallel with Rscreen?

  10. #10
    Senior Member
    Join Date
    Aug 2006
    Location
    Lancashire
    Posts
    579
    Quote Originally Posted by Merlinb View Post
    All valve compressors are variable-mu (short for variable mutual conductance, or variable-gm).
    Actually it just occurred to me that there must be some diode-bridge compressors around, using valve diodes. Those would also qualify as proper valve compressors!

  11. #11
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Ok so that's also something I've seen. A diode rectifier on the OT secondary, which feeds back to the input tube's grid/s.

  12. #12
    Senior Member
    Join Date
    May 2010
    Posts
    221
    I would think adding a couple of tube diodes in the side chain would not impart too much tubeyness to the dc signal.

  13. #13
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    would NOT? Hmmm... not sure but I'd initially think the internal resistance of a tube rectifier might play a part here. However, that is only the case when current demand is high so in this particular setup it wouldn't impart "sag". correct?

  14. #14
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Found something online "suggesting" that the Fairchild 670 may have been used on the recording. It's an interesting schematic with all that transformer coupling going on. There are said to be some mistakes on this forum:

    http://recording.org/diy-pro-audio-forum/40934-fairchild-670-readable-schematic.html


    http://www.vintagedesign.halmstad.ne...s/670schem.pdf
    Last edited by tboy; 10-25-2012 at 09:45 AM. Reason: link repair

  15. #15
    Supporting Member loudthud's Avatar
    Join Date
    May 2006
    Location
    Near Dallas Texas
    Posts
    2,533
    Did anybody notice that the bridge rectifiers are drawn wrong?
    lowell likes this.
    WARNING! Musical Instrument amplifiers contain lethal voltages and can retain them even when unplugged. Refer service to qualified personel.

  16. #16
    Old Timer
    Join Date
    Oct 2007
    Location
    Buenos Aires, Argentina
    Posts
    5,550
    [facepalm] I've seen worse !!
    Juan Manuel Fahey

  17. #17
    Senior Member
    Join Date
    May 2006
    Location
    Cornelius, Oregon
    Posts
    1,155
    I'd recommend to get the "Recording the Beatles" book as they go into extreme detail about the recording setup and equipment in use at Abbey Road at the time. By that time they were using the solid state mixing desk, but still using lots of glorious tube mics and the Fairchild compressors were used on just about everything, but EMI modified them so who knows what the real schematics of them are. Anyway the book is super excellent and will probably help you on your quest to build a tube compressor.

    Harrison most likely used his 1957 Les Paul with PAF's known as Lucy on that recording for the solos, and by then they were largely using Fender amplifiers, probably the '68 Twin Reverb or maybe the '63 Bassman. They were recording to an 8 track tape machine too by then.

    Greg

  18. #18
    Senior Member Austin's Avatar
    Join Date
    Jun 2012
    Location
    A van down by the RIVER!
    Posts
    316
    compressor.gif

    tube_compressor.gif

    From here: Circuits


    Saw these yesterday and reminded me of this post.

  19. #19
    Senior Member
    Join Date
    Jan 2011
    Location
    Australia
    Posts
    281
    I think I accidentally made a tube compressor whilst trying to work a Heptode a long time ago... I didn't actually know what I was doing, but it sounded pretty neat!

  20. #20
    Senior Member Enzo's Avatar
    Join Date
    May 2006
    Location
    Lansing, Michigan, USA
    Posts
    21,087
    Heptode - a beatnik amphibian?
    Education is what you're left with after you have forgotten what you have learned.

  21. #21
    Supporting Member Jazz P Bass's Avatar
    Join Date
    Dec 2008
    Location
    Wernersville, PA
    Posts
    7,876
    The Heptode.
    What a wild application specific tube.
    (the way it was intended to be used by the designers.)
    5 grids.
    Yeaha.
    The Heptode
    Attached Thumbnails Attached Thumbnails heptode.jpg  

  22. #22
    Senior Member Enzo's Avatar
    Join Date
    May 2006
    Location
    Lansing, Michigan, USA
    Posts
    21,087
    We used to call them pentagrid converter tubes, as they were used mainly in things like IF circuits .
    Education is what you're left with after you have forgotten what you have learned.

  23. #23
    Supporting Member
    Join Date
    Dec 2009
    Location
    Devon, UK
    Posts
    478
    Out of interest, and because I'm too lazy to calculate it by hand, I did a simulation of a 12AX7 Voltage Gain vs Grid Voltage (Plate Load= 470k, Vsupply=250v). Interesting wide range. I had no idea it was so big. Hope I got it right...

    av-vs-vg.png

    Makes you think, hmm...
    J M Fahey likes this.

  24. #24
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Whoa wild!

    Austin, I do not for the life of me get how the 2nd stage in that "electronic compressor limiter" schem works at all. There is no DC supply on the anode pin1 !? How can that tube conduct at all with that being the case? Or am I missing something.. most likely.

  25. #25
    Senior Member
    Join Date
    Jan 2011
    Location
    Australia
    Posts
    281
    Quote Originally Posted by lowell View Post
    Whoa wild!

    Austin, I do not for the life of me get how the 2nd stage in that "electronic compressor limiter" schem works at all. There is no DC supply on the anode pin1 !? How can that tube conduct at all with that being the case? Or am I missing something.. most likely.
    I think the 1Meg cathode resistor is probably indicative that the currents involved are really quite low. The HT supply appears to be an AC voltage that is coupled from the first triodes cathode (it's a cathode follower, though it doesn't appear to be biased at all! - it probably doesn't need to be since it should have adequate headroom for low level signals, <2v). The second triode then sees an AC voltage on it's plate and grid that are identical, except that the plate source actually has some current driving capacity. I think the first triode is purely some sort of current buffer, though it may have some compressive (clipping?) qualities due to the lack of bias. As to what happens when you have an identical signal on plate AND grid? I don't actually know... Furthermore, the signal on the plate of the second triode would only be perhaps one or two volts, which probably necessitates that 1 meg cathode resistor. It then goes on to another cathode follower, DC coupled to gain stage (that again has no bias!), that mixes the signal with the original signal. Weird design for sure.

  26. #26
    Senior Member
    Join Date
    Mar 2010
    Posts
    133
    Quote Originally Posted by Enzo View Post
    We used to call them pentagrid converter tubes, as they were used mainly in things like IF circuits .
    Ooh. I picked up a box of valves a while ago to get some 'useful' amp building ones - and there were several heptodes and pentagrid converters. They're in a box under the stairs right now. But I'm interested to see where this thread goes....

    cheers
    It's not microphonic - it's undocumented reverb.

  27. #27
    Supporting Member
    Join Date
    Dec 2009
    Location
    Devon, UK
    Posts
    478
    I think the second section of the first triode acts as a diode. The second section of the second triode acts as a variable resistance.

  28. #28
    Old Timer
    Join Date
    Oct 2007
    Location
    Buenos Aires, Argentina
    Posts
    5,550
    Out of interest, and because I'm too lazy to calculate it by hand, I did a simulation of a 12AX7 Voltage Gain vs Grid Voltage (Plate Load= 470k, Vsupply=250v). Interesting wide range. I had no idea it was so big. Hope I got it right...
    You did !!
    Simply, tubes compress on their own, even if you don't want to.
    Juan Manuel Fahey

  29. #29
    Member
    Join Date
    Mar 2010
    Location
    Wisconsin
    Posts
    38
    Ever put a photo resistor across your input jack attenuated from a light bulb driven by the speaker output? It ain't La2a but it works quite well and it's clean.

  30. #30
    Old Timer
    Join Date
    Jun 2007
    Location
    Chicago
    Posts
    2,276
    Huh? Wha? Sounds interesting. Question, do you use a opto-coupler as in fender vibrato circuits? And can explain in a bit more detail the connections? Is the bulb in parallel with the speaker? Or series?

  31. #31
    Member
    Join Date
    Mar 2010
    Location
    Wisconsin
    Posts
    38
    Well, I have not messed with this circuit in many many years so I forget the specifics. It's super simple. The light bulb is the tricky part. Basically the louder the amp, the brighter the bulb, which opens the LDR, shunting the input to ground. To get a taste, put a LDR across the sleeve of a guitar cable. Use your hand to cover it from the light then back off. The trick with the bulb is to find a sweet spot. Suppose your amps on 5 and you've used a potentiometer connected to the bulb to find a sweet spot. If you turn your amp up or down, you've lost the sweet spot. I've not messed with this in years, but if I was, I'd look for a flashlight bulb and put a resistor on it. Perhaps a 1 meg pot. Just a guess, but a 10K resistor feeding a 10K pot gives you line level off a speaker. Find a flashlight bulb. Another source of parts is a 120V night light. It has a neon bulb and the LDR. I'm not sure if the speaker can kick a neon, maybe. Another obstacle is creating the LDR/bulb package sealed from out side light. For kicks I suppose black electrical tape will get ya goin. In the old days I used 35mm plastic film cans. For as simple as this circuit is, it is no less legitimate than any commercial LDR compressor using the same principle. To answer your Fender opto coupler question, a bulb and LDR package is an opto coupler. Commercial versions exist for easy pcb integration. The bulb is in paralell to the speaker. The voltage it taps is so small it will not affect sound. Taken to the limits a pot could be mounted on the amp. It works best for clean compression. Once an amp is driven to distortion it is compressed and has less of a voltage swing so the circuit becomes useless. So that's the long and short of it as I recall. It's a fun and super easy experiment. It probably never became popular because people wanted a compressed, distorted guitar sound which made compressor pedals more practical. Compared to an La2a, it does have iron and tubes. After this experiment consider a volume pedal. How about a stereo tremolo with a tiny motor driving a cardboard strobe between 2 LDRs. Have fun!
    Austin likes this.

  32. #32
    Senior Member Austin's Avatar
    Join Date
    Jun 2012
    Location
    A van down by the RIVER!
    Posts
    316
    Quote Originally Posted by twiggybush View Post
    Well, I have not messed with this circuit in many many years so I forget the specifics. It's super simple. The light bulb is the tricky part. Basically the louder the amp, the brighter the bulb, which opens the LDR, shunting the input to ground. To get a taste, put a LDR across the sleeve of a guitar cable. Use your hand to cover it from the light then back off. The trick with the bulb is to find a sweet spot. Suppose your amps on 5 and you've used a potentiometer connected to the bulb to find a sweet spot. If you turn your amp up or down, you've lost the sweet spot. I've not messed with this in years, but if I was, I'd look for a flashlight bulb and put a resistor on it. Perhaps a 1 meg pot. Just a guess, but a 10K resistor feeding a 10K pot gives you line level off a speaker. Find a flashlight bulb. Another source of parts is a 120V night light. It has a neon bulb and the LDR. I'm not sure if the speaker can kick a neon, maybe. Another obstacle is creating the LDR/bulb package sealed from out side light. For kicks I suppose black electrical tape will get ya goin. In the old days I used 35mm plastic film cans. For as simple as this circuit is, it is no less legitimate than any commercial LDR compressor using the same principle. To answer your Fender opto coupler question, a bulb and LDR package is an opto coupler. Commercial versions exist for easy pcb integration. The bulb is in paralell to the speaker. The voltage it taps is so small it will not affect sound. Taken to the limits a pot could be mounted on the amp. It works best for clean compression. Once an amp is driven to distortion it is compressed and has less of a voltage swing so the circuit becomes useless. So that's the long and short of it as I recall. It's a fun and super easy experiment. It probably never became popular because people wanted a compressed, distorted guitar sound which made compressor pedals more practical. Compared to an La2a, it does have iron and tubes. After this experiment consider a volume pedal. How about a stereo tremolo with a tiny motor driving a cardboard strobe between 2 LDRs. Have fun!
    So would the neon bulb sound different than the flashlight bulb? Would an LED work?

  33. #33
    Senior Member
    Join Date
    Jan 2011
    Location
    Australia
    Posts
    281
    People use LED's all the time in commercial opto-compressors, well at least the photocoupled sort you find for channel switching amps. They'd definitely all give you different results, mainly because they require 3 different ways of driving them! (and their luminous intensity vs voltage/current curves being all different, as well as having different time constants). Ignoring this though, I'd think the LDR is the limiting factor in an optical compressor, as it has a sloooooow response time (though people seem to love the sound of the things). I've used things like photodiodes to get really fast attack/release times with LED's.

  34. #34
    Member
    Join Date
    Mar 2010
    Location
    Wisconsin
    Posts
    38
    Austin, a Led is dc. The speaker is ac. I've never tried a neon. I just mentioned it as there is one inside a night light. I have to laugh because back in the day LDR's were critisized for slow response. Today they are as popular as ever. Actually the slow response can be advantageous to letting the initial attack pass. Lowell was looking for a clean tube sound. This is such an old school approach I thought it needed to be mentioned. Photodiodes require dc and support parts. Might as well buy a pedal if going that route. For anyone wanting to learn more on this subject, I'd suggest looking up Craig Andertons 'Electronics Projects fo Musicians'. It's from the 70's and has a easily built compressor surrounding 741 type chips.

  35. #35
    Senior Member Enzo's Avatar
    Join Date
    May 2006
    Location
    Lansing, Michigan, USA
    Posts
    21,087
    Neon needs a lot of voltage to fire, your speaker would have to have that much across it for neon to even try to work. Neon is not linear, it is dark until the threshold is met, then it comes on. From that point, increasing the voltage through the tube can increase brightness a little, but nothing at all like controlling an incandescent or LED.

    LEDs run on DC, and typically do not like much reverse voltage, but simply adding a regular diode in series will block reverse, and then the LED can respond to the AC of the speaker. Incandescents care not about polarity or AC/DC.
    Education is what you're left with after you have forgotten what you have learned.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Behringer T1952 Tube Compressor
    By Tejaus in forum Music Electronics
    Replies: 12
    Last Post: 04-05-2012, 05:19 PM
  2. Behringer T1952 Tube Compressor
    By Tejaus in forum Schematic Requests
    Replies: 2
    Last Post: 11-01-2011, 07:45 PM
  3. Teletronix LA-2A tube compressor
    By Wombaticus in forum Other Amps
    Replies: 6
    Last Post: 05-04-2011, 05:08 AM
  4. Compressor build
    By Artspoke in forum Music Electronics
    Replies: 6
    Last Post: 02-14-2009, 11:19 PM
  5. Compressor or Expander?
    By tbryanh in forum Theory & Design
    Replies: 2
    Last Post: 08-31-2007, 03:44 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •