There is also the 6LE8 (and some smaller dual control relatives too: 6BU8, 6KF8, 6MK8, 6GS8, 6HS8) which operates somewhat similarly to a BDT, but is actually a pentode with two highly effective (fine mesh) suppressor grids and two plates. Their functional operation is almost drop-in for the BDT tubes (but different pin-out unfortunately). These often go for $0.50 to $1 when on sale. They were used for color TV demodulators. You can also make them operate as a normal pentode by putting about +12 V on g3 and tying the plates together. Should work for volume controls at least.
The BDTs and the 6LE8 should be useful for harmonic generation for say guitar amplifier effects, since they can be used to multiply signals. Square a signal and get 2nd harmonics, square it again and get 4th harmonic. Lots of 2nd H and some 4th harmonic make everything sound like guitar music. I plan to try this out for the cheap guitar amplifier thread.
The BDTs and the 6LE8 should be useful for harmonic generation for say guitar amplifier effects, since they can be used to multiply signals. Square a signal and get 2nd harmonics, square it again and get 4th harmonic. Lots of 2nd H and some 4th harmonic make everything sound like guitar music. I plan to try this out for the cheap guitar amplifier thread.
Last edited:
About the distortion
I interpreted from Pete Millett's post (here: Some quick measurements... - Pete Millett - Tube DIY Asylum) that the tube produces some harmonic distortion.
But also it seems that 1) reducing signal level reduces the distortion relatively even more, so this should not be bad in small signal levels. 2) However decreasing gain of the device seems to be more problematic because it increases distortion.
I interpreted from Pete Millett's post (here: Some quick measurements... - Pete Millett - Tube DIY Asylum) that the tube produces some harmonic distortion.
But also it seems that 1) reducing signal level reduces the distortion relatively even more, so this should not be bad in small signal levels. 2) However decreasing gain of the device seems to be more problematic because it increases distortion.
Distorsion remains very low if both anodes are loaded with enough iddle current. Used as a phase shifter, with grounded grid and audio signal sent to one of the deflection plates, the gain is only 2 with proper anode current settings. Both output signals are really 180° out of phase. Till today, it's the best symetrical tube phase inverter I have ever seen...
So I have a couple of these tubes now, and am going to use this schematic to drive a PP amp, one phase from each anode. I can balance the anode currents with a pot on the dead deflector to null out any stray static magnetism, and will put the tubes in the furthest place from the (shielded) power transformers.
Now these anodes will be driving some SRPP tubes which (hum permitting) will be sitting at about 500V (250V per tube, 6N2P-EV tubes - will the swings exceed their max anode voltage ratings?), so I'll really want to hang the BDTs off about 400V, so I have the questions:
How much current should each anode idle at (min volume and max volume)?
I was thinking of perhaps having a switchable gain for 'quiet' and 'loud' on these - but there is risk of a switchover thump of some magnitude.. don't really want the switch sitting at 400V either so perhaps there is a way ising the grid?
Also planning of using a 7809 for the 9V volume voltage - do you think that is good enough?
I was thinking that if hum was a big issue with these tubes is the approach of mu-metal wraps or injecting an opposing hum (perhaps from the heater AC via a pot) - any thoughts?
I don't understand the 2k anode resistors and the 250V supply rail, with the anode dissipation of 10mA total (from the datasheet).
With 2k x 5mA = 10V, so does the other 240V sit across the tube?
Usually the tube would seem to have 220k anode resistors, so could someone explain the 2k ones to me? Thanks in advance!
With 2k x 5mA = 10V, so does the other 240V sit across the tube?
Usually the tube would seem to have 220k anode resistors, so could someone explain the 2k ones to me? Thanks in advance!
Just wanted to be sure that you are on a thread about John's older design that he has refined with more that 1 revision IIRC.
Here's another thread on a newer version: http://www.diyaudio.com/forums/tubes-valves/123402-john-swensons-new-bdt-preamp.html
It might be worth it to google AA to see if there's something newer. If you actually have made a conscious decision to build this version that's fine, I just wanted to be sure you were aware.
Here's another thread on a newer version: http://www.diyaudio.com/forums/tubes-valves/123402-john-swensons-new-bdt-preamp.html
It might be worth it to google AA to see if there's something newer. If you actually have made a conscious decision to build this version that's fine, I just wanted to be sure you were aware.
Thanks leadbelly, I had seen that thread but was rather confused by it.
It seems to end with the construction of power supplies and power amps, which isn't what I want to do at all!
Also I don't want any CCS stuff (silicon) either, I like the nice simplicity of the first diagram
Plus I have limited space, I'd be tempted to put an SRPP top on the anodes but that needs two more envelopes!
I guess it all depends upon getting a nice operating volume with G1 balancing the tube nicely at about 3mA, with max level at about 5mA per anode - turning down all the way to 0mA at zero level (I'm guessing at -9V at G1 is cutoff - must check the curves again!).
It seems to end with the construction of power supplies and power amps, which isn't what I want to do at all!
Also I don't want any CCS stuff (silicon) either, I like the nice simplicity of the first diagram
Plus I have limited space, I'd be tempted to put an SRPP top on the anodes but that needs two more envelopes!
I guess it all depends upon getting a nice operating volume with G1 balancing the tube nicely at about 3mA, with max level at about 5mA per anode - turning down all the way to 0mA at zero level (I'm guessing at -9V at G1 is cutoff - must check the curves again!).
Should be able to make a parafeed phase splitter (or floating parafeed) with a BDT easily. Something like 430K and 560K from the left plate and right plate respectively over to the right deflector. Then 510K from that deflector down to opposite B- (or whatever needed) to get the deflector DC level down near ground level (or use a cap between the deflector and the above R divider, with a bias resistor from the deflector to a bias level adjust pot for the deflector). Input signal on the left deflector.
The 6ME8 will work with around +75V on the deflectors, so might be more convenient to set up.
Using some Gyrators for the B+ plate loads would get the gain up around 6 instead of 2. (higher value plate loads are more linear and higher gain, the BDT is just like two common cathode triodes setting on a CCS) The resistive divider network needs to be set up to attenuate the output signal level down to 1/gain for the right deflector to get equal outputs.
The 6ME8 will work with around +75V on the deflectors, so might be more convenient to set up.
Using some Gyrators for the B+ plate loads would get the gain up around 6 instead of 2. (higher value plate loads are more linear and higher gain, the BDT is just like two common cathode triodes setting on a CCS) The resistive divider network needs to be set up to attenuate the output signal level down to 1/gain for the right deflector to get equal outputs.
Last edited:
To me, the configuration of the tube suggests "tunable LTP". Set up properly, could the tube be set up to offer some advantage over an LTP made with two triodes, maybe something like being able to fine tune for simultaneous AC and DC balance at the plates?
The paraphase (OOPs, not parafeed, my typo earlier) local feedback network would unfortunately make the DC balance worse if it is DC coupled (still trimmable though), no effect for the AC coupled case. De-magnetizing (AC) the BDT will usually help the DC balance.
Also, my earlier description was off on another point. The paraphase network is being used to balance the deflector drives on each side, which will improve the overall gain and possibly the distortion profile. (the plate outputs are already AC balanced for the BDT, just like for a CCS tailed LTP)
-----------------------------------------
An even further stretch would be to put separate AC coupled paraphase networks on both sides (deflectors) of the BDT. These would have to be set up for less than unity local loop gain around the figure eight feedbacks to avoid making an oscillator. This would lower the deflector input impedance though. But the closer the local loop gain is to unity, the higher the input Z gets. I don't see this as simplifying the operation over the triode LTP on a CCS tail though, since equivalent can be done for either. Mainly the hope for this approach would be for reducing odd harmonic distortion (for either a BDT or a triode LTP).
Putting gyrator plate loads on the BDT (or an LTP) should also lower odd harmonic distortion.
Also, my earlier description was off on another point. The paraphase network is being used to balance the deflector drives on each side, which will improve the overall gain and possibly the distortion profile. (the plate outputs are already AC balanced for the BDT, just like for a CCS tailed LTP)
-----------------------------------------
An even further stretch would be to put separate AC coupled paraphase networks on both sides (deflectors) of the BDT. These would have to be set up for less than unity local loop gain around the figure eight feedbacks to avoid making an oscillator. This would lower the deflector input impedance though. But the closer the local loop gain is to unity, the higher the input Z gets. I don't see this as simplifying the operation over the triode LTP on a CCS tail though, since equivalent can be done for either. Mainly the hope for this approach would be for reducing odd harmonic distortion (for either a BDT or a triode LTP).
Putting gyrator plate loads on the BDT (or an LTP) should also lower odd harmonic distortion.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.