Hi Bas, wow someone is actually trying this out! A few hints on this design: the maximum input voltage without causing distortion varies with the current through the tube (ie the volume setting) , so the lower the volume setting the lower the maximum input signal you can have. Thus to maximise the input level you want the current flow at max level setting to be as high as possible. You have to ballance that with the plate dissipation and the maximum through the tube. If I remember correctly the tube is rated for 20mA cathode current, try to get close to that.
Of course the greater the current the greater the voltage drop across the anode resistors, you have to take that into account when designing the PS and specing the resistors.
The resuslt is that I found that when I had the volume very low (say for listening late at night) a high output CD player would start distorting, at higher volume levels it was fine. The DDDAC with its somewhat lower levels worked fine at all settings.
On my tweaked version of this design I put transformers on the output rather than the resistors. An IT with tapped primary is what is needed here. I tried the cheap Hammond 124B here and it worked VERY well. The only problem is you have to make sure you get the plates balanced so there is no DC voltage on the transformer, I did this with a pot on the non driven deflector, wiper to the defelctor and +-3V on the ends. With this I could tweak the balance. Of course you could go parafeed if you wanted but that starts getting pretty expensive.
The other main problem is that the tubes are highly microphonic so tube dampers etc can be very useful. But they do get HOT when run at close to full current so be careful to get dampers that can withstand that.
The other thing to remember is that the mu is low, about 6, so you will not be able to get much gain out of it, but that is not usually that big a deal with a line stage these days. In practical use a gain of 4 was about it.
Its very easy to alter the overall gain, just change the anode resistors, what you use is going to take a little playing around so you get enough gain for a soft source but don't have to turn the volume down too far for loud sources.
BTW at 20mA on the cathode it will handle a 90V signal on the input without significant distorion, its only when the volume is very low and the current is in the microamp range that you have problems.
Have fun!
John S.
Of course the greater the current the greater the voltage drop across the anode resistors, you have to take that into account when designing the PS and specing the resistors.
The resuslt is that I found that when I had the volume very low (say for listening late at night) a high output CD player would start distorting, at higher volume levels it was fine. The DDDAC with its somewhat lower levels worked fine at all settings.
On my tweaked version of this design I put transformers on the output rather than the resistors. An IT with tapped primary is what is needed here. I tried the cheap Hammond 124B here and it worked VERY well. The only problem is you have to make sure you get the plates balanced so there is no DC voltage on the transformer, I did this with a pot on the non driven deflector, wiper to the defelctor and +-3V on the ends. With this I could tweak the balance. Of course you could go parafeed if you wanted but that starts getting pretty expensive.
The other main problem is that the tubes are highly microphonic so tube dampers etc can be very useful. But they do get HOT when run at close to full current so be careful to get dampers that can withstand that.
The other thing to remember is that the mu is low, about 6, so you will not be able to get much gain out of it, but that is not usually that big a deal with a line stage these days. In practical use a gain of 4 was about it.
Its very easy to alter the overall gain, just change the anode resistors, what you use is going to take a little playing around so you get enough gain for a soft source but don't have to turn the volume down too far for loud sources.
BTW at 20mA on the cathode it will handle a 90V signal on the input without significant distorion, its only when the volume is very low and the current is in the microamp range that you have problems.
Have fun!
John S.
Hi Paul,
actually I'm not using it anymore. After I built a single ended parafeed amp with 0.03% distortion (yep thats right) the residual distortions and primarily the microphonics were too much. I started realizing the preamp was the limiting factor. Its not that it sounds BAD, but the rest of the system had improved so much (including my DIY USB DAC) I was starting to hear the little problems with this. It was the best bang for the buck I had ever come up with but it wasn't cutting it with the rest of the system.
What I wound up with was doesn't even use tubes (oh horror!) its a modification of Gary Pimm's CCS circuit into an amplifier and buffer, the result is really incredible. Gary has been playing with these concepts as well and likes to call them the solid state pentode, the circuits look and behave vary similar to tube circuits but aren't. They even run at high voltages and use tranformers and such just like equivalent tube circuits.
One of the things I REALLY like about the BDT preamp is the volume control does not cary signal, thus you don't need expensive controlls or switched attenuators etc. Also it makes remote controls really easy. On the one mentioned above I had to spend 4 times as much money on the volume control as the whole BDT preamp cost in order to get to the same quality level.
John S.
actually I'm not using it anymore. After I built a single ended parafeed amp with 0.03% distortion (yep thats right) the residual distortions and primarily the microphonics were too much. I started realizing the preamp was the limiting factor. Its not that it sounds BAD, but the rest of the system had improved so much (including my DIY USB DAC) I was starting to hear the little problems with this. It was the best bang for the buck I had ever come up with but it wasn't cutting it with the rest of the system.
What I wound up with was doesn't even use tubes (oh horror!) its a modification of Gary Pimm's CCS circuit into an amplifier and buffer, the result is really incredible. Gary has been playing with these concepts as well and likes to call them the solid state pentode, the circuits look and behave vary similar to tube circuits but aren't. They even run at high voltages and use tranformers and such just like equivalent tube circuits.
One of the things I REALLY like about the BDT preamp is the volume control does not cary signal, thus you don't need expensive controlls or switched attenuators etc. Also it makes remote controls really easy. On the one mentioned above I had to spend 4 times as much money on the volume control as the whole BDT preamp cost in order to get to the same quality level.
John S.
Solid State Triode
In order to not dilute this post with a different project which I'm sure most of us are also very interested in, I started a new post for the Solid State Triode http://www.diyaudio.com/forums/showthread.php?postid=887866#post887866
In order to not dilute this post with a different project which I'm sure most of us are also very interested in, I started a new post for the Solid State Triode http://www.diyaudio.com/forums/showthread.php?postid=887866#post887866
6AR8, 6JH8, 6ME8
Glad to see that many people are interested in this fantastic tubes. To continue this brain-storming about beam deflection tubes, I would just add another unconventionnal use for them. As I always thought that common phase inverters used before push-pull stages were not defect free, I decided to try BDT's for that use. And it works really well! The only condition is to keep the anode current high enough to ensure good linearity. The only drawbacks of this method are: 1/ a very low voltage gain (about 2) 2/ a low output swing (20 volts peak max) (depends on the tube) For example, have a look at this page: Push-pull 6S33S, 6N23P, 6JH8, 6N1P-EV, IN13 The lack of voltage to drive the huges 6S33S has been ovecome whith an SRPP low Z gain stage. The sound is really terrific! Sorry guys, it's not in English yet, but the schematics are downloadable. I gave up whith 6AR8, because species were far too different one another. 6JH8 or 6ME8 are far better tubes for that purpose. Both of them can be found around 5$ each when brand new. Hope this will be of some interest for someone... Greetings from France. Alain
Glad to see that many people are interested in this fantastic tubes. To continue this brain-storming about beam deflection tubes, I would just add another unconventionnal use for them. As I always thought that common phase inverters used before push-pull stages were not defect free, I decided to try BDT's for that use. And it works really well! The only condition is to keep the anode current high enough to ensure good linearity. The only drawbacks of this method are: 1/ a very low voltage gain (about 2) 2/ a low output swing (20 volts peak max) (depends on the tube) For example, have a look at this page: Push-pull 6S33S, 6N23P, 6JH8, 6N1P-EV, IN13 The lack of voltage to drive the huges 6S33S has been ovecome whith an SRPP low Z gain stage. The sound is really terrific! Sorry guys, it's not in English yet, but the schematics are downloadable. I gave up whith 6AR8, because species were far too different one another. 6JH8 or 6ME8 are far better tubes for that purpose. Both of them can be found around 5$ each when brand new. Hope this will be of some interest for someone... Greetings from France. Alain
You can get a lot more gain out of these BDTs if you put a Mosfet cascode stage(s) above the plates. When the plates are resistor loaded, the plates act like triodes or beam deflecting plates due to the back voltage effects. With the cascodes above, the plate voltages are frozen, so no internal feedback. You will likely loose some linearity this way, but local feedback can still be used to restore that.
Also, some of the non-uniformity between BDT tubes can be explained by magnetized internals from long storage on the shelf in the earths field. An initial processing with an AC demagnetizing coil can restore much of the uniformity, and lower distortion as well. These tubes are usually used with a magnetic shield (steel tube shield) over them as well. Keeps out hum and interference. Electronic Goldmine .com sells big 8 x 12 inch sheets of Mu metal for around $7 I think.
Also, some of the non-uniformity between BDT tubes can be explained by magnetized internals from long storage on the shelf in the earths field. An initial processing with an AC demagnetizing coil can restore much of the uniformity, and lower distortion as well. These tubes are usually used with a magnetic shield (steel tube shield) over them as well. Keeps out hum and interference. Electronic Goldmine .com sells big 8 x 12 inch sheets of Mu metal for around $7 I think.
Last edited:
"I wonder if distortion improvements could be made by using a mu-follower type of approach. "
That should help, just like for a triode with CCS loading. You should get a bit more gain too, I think around 6 is the differential Mu for 6JH8. Not given on the data sheet or derivable from the graphs unfortunatly. The BDT tubes are essentially equivalent to a two triode differential pair on a CCS tail (CCS tail current set by g1). So the CCS loading on the plates will require a gyrator(s) setup to get a stable plate voltage.
That should help, just like for a triode with CCS loading. You should get a bit more gain too, I think around 6 is the differential Mu for 6JH8. Not given on the data sheet or derivable from the graphs unfortunatly. The BDT tubes are essentially equivalent to a two triode differential pair on a CCS tail (CCS tail current set by g1). So the CCS loading on the plates will require a gyrator(s) setup to get a stable plate voltage.
Last edited:
On second thought, two gyrators or Mu followers for BDT plate loads may be unstable or bi-stable if they don't have some reasonable resistance. (the BDT acts like a CCS in the cathode already) See this thread:
http://www.diyaudio.com/forums/tubes-valves/191784-long-tails-current-sources.html
http://www.diyaudio.com/forums/tubes-valves/191784-long-tails-current-sources.html
One Fully Tested NOS Vacuum Tube 6AR8 | eBay 6AR8 are not difficult to find. But you can easily replace them with 6JH8 or 6ME8 that are really better tubes.
Cheers - got a couple on order now
I'm going to use them as a front end to an integrated PP amp, I was going to use a LTP phase splitter but this will do the volume and phase split in one go!
So the plan is to use a BDT front end, into two 6N2p SRPP tubes into GU50 output tubes (pentode mode). It will use 6 tubes but it should give a nice dynamic sound!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.