Hello,
I recently built my first tube power amp. It's a small, 8-watt design sold as a kit from a small company in Arizona. This is my first real listening experience with tube equipment & I'm really excited about the sound - especially when the volume is turned up & the amp is straining, just a bit, to push my rather inefficient speakers. Everything sounds good - much less hard, smoother, more pleasant! I've listened to my entire music collection, twice!
So,
I want to build a "special" tube preamp. One that has a particular, and a rather unusual function. I am a recording engineer - mostly acoustic instruments & primarily classical chamber music. I do this professionally for an NPR affiliate & also as a freelance thing for various musicians & groups. I originally built this little tube amp to use as a nearfield monitor amp for recording ... but, it makes everything sound TOO good. I need to hear all the warts, clams, etc. while engineering a recording. I'll go back to my usual solid state amp for recording. However, it would be absolutely wonderful to be able to bring some (or a lot) of what I've been hearing with this little tube amp, into the actual recordings I engineer. One could say "Just build or buy a couple of really fine tube microphone preamps." and indeed, that would be one way to do it. What I hope to do is build a preamp-like box that can be inserted at various points within the signal chain ... between mixer & recorder, between outboard mic preamp & mixer input, between playback recorder and record deck during dubbing, etc.
Here's the general plan:
two channels, unbalanced in & out, variable input level, variable output level, enough gain (adjustable?) in the tube circuit to go from very clean to mild saturation to heavy saturation and on into clipping (never to be used like a guitar distortion device, though). Input levels would range from quiet music at -10dbv (CD player average level) up to +4 dbm (Pro mixer output average level). I've designed & built a few devices & power supplies in the past - mostly IC-based for audio use. I can understand and plan opamp gain staging etc. But tube electronics are new to me & I need some help in conceiving this thing.
I've been lurking around the net, reading all the "tube primers" & other info. I can find, checking out EVERYBODY'S web site. I checked out Morgan Jones's Valve Amp book from the library, ordered a copy of Tube Cad, and am very enthusiastic about getting the design phase fully underway. Hopefully, you good folk here at DIY will help this newbie out just a bit.
Would a single triode gain stage, with a relatively low mu tube such as a 6SN7 or a 12au7 in a grounded cathode design be able to be pushed fairly hard into saturation by the standard input levels as described above?
Would lowering the B+ voltage to around 150 v. also lower the tube's saturation point? If so, would the lower voltage cause the tube to sound less lovely and wonderful than it could?
If using a low mu (15-20) tube with a B+ of 250 - 280 v., would two gain stages with a 500K pot in between be a better approach? Would that result be just too much gain & too much additional noise?
Right now I plan to use a unity IC buffer (Burr-Brown BUF634) at the gain stage output so the tube never has to face difficult loads. The BUF634 has a very very high input impedance, is absolutely impossible to hear working & can pump out huge amounts of current. I know some people don't approve of mixing tubes & silicon but this thing works really well & I believe in using the best tools for the job. Of course a completely separate power supply would be included for the IC's. I might also include an IC differential opamp configuration to optionally accomodate balanced inputs.
In trying to come up with ways to effect the tube's saturation point, I was wondering how this might work: put a 100K pot on the tube stage output wired as a variable resistor to ground. Varying the pot's level would raise or lower the load seen by the gain stage. Maybe put a 100 ohm resistor in series with the pot to ground to keep the load at least 100 ohms at the pot's minimum position. Remember, the IC buffer would still be the link with the outside world. Could this approach be used to adjust the tube's saturation point or is it just going to screw up the sound?
I would prefer to avoid negative feedback as I understand this can muck with the balance of low & high order harmonic distortion and the 2nd & some 3rd HD is what supposedly sounds so attractive. The IC buffer would preclude the use of a cathode follower. What else needs to be considered?
This could be a really interesting & fun project/experiment. Any and all comments, questions, and especially, suggestions are most welcome.
I recently built my first tube power amp. It's a small, 8-watt design sold as a kit from a small company in Arizona. This is my first real listening experience with tube equipment & I'm really excited about the sound - especially when the volume is turned up & the amp is straining, just a bit, to push my rather inefficient speakers. Everything sounds good - much less hard, smoother, more pleasant! I've listened to my entire music collection, twice!
So,
I want to build a "special" tube preamp. One that has a particular, and a rather unusual function. I am a recording engineer - mostly acoustic instruments & primarily classical chamber music. I do this professionally for an NPR affiliate & also as a freelance thing for various musicians & groups. I originally built this little tube amp to use as a nearfield monitor amp for recording ... but, it makes everything sound TOO good. I need to hear all the warts, clams, etc. while engineering a recording. I'll go back to my usual solid state amp for recording. However, it would be absolutely wonderful to be able to bring some (or a lot) of what I've been hearing with this little tube amp, into the actual recordings I engineer. One could say "Just build or buy a couple of really fine tube microphone preamps." and indeed, that would be one way to do it. What I hope to do is build a preamp-like box that can be inserted at various points within the signal chain ... between mixer & recorder, between outboard mic preamp & mixer input, between playback recorder and record deck during dubbing, etc.
Here's the general plan:
two channels, unbalanced in & out, variable input level, variable output level, enough gain (adjustable?) in the tube circuit to go from very clean to mild saturation to heavy saturation and on into clipping (never to be used like a guitar distortion device, though). Input levels would range from quiet music at -10dbv (CD player average level) up to +4 dbm (Pro mixer output average level). I've designed & built a few devices & power supplies in the past - mostly IC-based for audio use. I can understand and plan opamp gain staging etc. But tube electronics are new to me & I need some help in conceiving this thing.
I've been lurking around the net, reading all the "tube primers" & other info. I can find, checking out EVERYBODY'S web site. I checked out Morgan Jones's Valve Amp book from the library, ordered a copy of Tube Cad, and am very enthusiastic about getting the design phase fully underway. Hopefully, you good folk here at DIY will help this newbie out just a bit.
Would a single triode gain stage, with a relatively low mu tube such as a 6SN7 or a 12au7 in a grounded cathode design be able to be pushed fairly hard into saturation by the standard input levels as described above?
Would lowering the B+ voltage to around 150 v. also lower the tube's saturation point? If so, would the lower voltage cause the tube to sound less lovely and wonderful than it could?
If using a low mu (15-20) tube with a B+ of 250 - 280 v., would two gain stages with a 500K pot in between be a better approach? Would that result be just too much gain & too much additional noise?
Right now I plan to use a unity IC buffer (Burr-Brown BUF634) at the gain stage output so the tube never has to face difficult loads. The BUF634 has a very very high input impedance, is absolutely impossible to hear working & can pump out huge amounts of current. I know some people don't approve of mixing tubes & silicon but this thing works really well & I believe in using the best tools for the job. Of course a completely separate power supply would be included for the IC's. I might also include an IC differential opamp configuration to optionally accomodate balanced inputs.
In trying to come up with ways to effect the tube's saturation point, I was wondering how this might work: put a 100K pot on the tube stage output wired as a variable resistor to ground. Varying the pot's level would raise or lower the load seen by the gain stage. Maybe put a 100 ohm resistor in series with the pot to ground to keep the load at least 100 ohms at the pot's minimum position. Remember, the IC buffer would still be the link with the outside world. Could this approach be used to adjust the tube's saturation point or is it just going to screw up the sound?
I would prefer to avoid negative feedback as I understand this can muck with the balance of low & high order harmonic distortion and the 2nd & some 3rd HD is what supposedly sounds so attractive. The IC buffer would preclude the use of a cathode follower. What else needs to be considered?
This could be a really interesting & fun project/experiment. Any and all comments, questions, and especially, suggestions are most welcome.
Give it a go
To do what you want, you probably need a valve that produces more even harmonic distortion than a 6SN7; a 6SL7 might do nicely.
If you had a ganged log volume control at the input ganged with an antilog at the output, you could effectively vary the amount of overdrive without changing apparent level. I've never seen such a pot, so you will have to make a switched attenuator, one wafer having the resistors in one order, the other wafer having the resistors in the other order. The gain of the 6SL7 is likely to be about 40 (32dB), so if the attenuator is designed for 32dB loss, your box will have unity gain.
You could also make the value of cathode bias resistor variable to adjust the operating point, which will affect distortion. You may find that a 250V HT works better than the more usual 400V. Your buffer chip would go after the final volume control.
To do what you want, you probably need a valve that produces more even harmonic distortion than a 6SN7; a 6SL7 might do nicely.
If you had a ganged log volume control at the input ganged with an antilog at the output, you could effectively vary the amount of overdrive without changing apparent level. I've never seen such a pot, so you will have to make a switched attenuator, one wafer having the resistors in one order, the other wafer having the resistors in the other order. The gain of the 6SL7 is likely to be about 40 (32dB), so if the attenuator is designed for 32dB loss, your box will have unity gain.
You could also make the value of cathode bias resistor variable to adjust the operating point, which will affect distortion. You may find that a 250V HT works better than the more usual 400V. Your buffer chip would go after the final volume control.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.