John Swenson's BDT preamp - finally!!

[jkeny]

I have just done a quick noise calc. Assume 100k equivalent noise resistance (guess, but probably OK within a factor of 2) you get 5.5uV in a 20kHz bandwidth. If your input signal is 100's of mV then this won't be noticeable, athough it is more noise than a well-designed conventional preamp.

Well at these levels it probably is not heard? Don't overlook the factor that really makes this shine as a preamp - the signal is not passing through a pot!! So whatever noise/distortion this generates is obviated! And it would appear from the Lightspeed thread that there are benefits to this approach.

This type of circuit should sound clean on signal peaks as the deflector can handle 10V of signal. A typical line stage will struggle with more than a few volts, and some poorly designed ones will fall over before then due to poor biassing.

That's what JS seems to suggest by "effortless presentation"

I hope this design does not become too popular, as we radio amateurs want to keep the price of BDT down so we can use them in RF equipment!

Oops, but don't worry, this design was made public back in 2007, I think & there have been maybe 3 people who have built one? I don't know I can't find anybody else!

 

[jkeny]

JS:

The maximum signal it can take at the input is dependant on the current through the tube (hence the "volume setting"). At max (about 8 mA per plate) I can get about 20V P-P at .03% distortion on the output. As the volume goes down the maximum input signal goes down as well. A traditional 2V RMS CD output starts distorting at around -45 to -50 db. If you like listening at that level this is probably not a good choice for volume control. I run into a fairly insensitive tube amp and never hit the distortion limit. At the point where it kicks in its so soft I can barely hear it.

 

[DF96]

Yes, you would get soft limiting on signal peaks.

Incidentally, cascode operation of a BDT was suggested in the 1950's for use in an SSB transmitter.

 

[jkeny]

I'm getting there 🙂
A couple of issues to sort out:
- I have 5K resistors across the secondaries but need more gain so I'll change these to 10K which should give pain inducing sound at max through these headphones. Later in the week I should be able to try AKG701s
- I need to work out my vol control connections - something is messed up here - I'm not getting the correct control

I have sound on my cheap Sony headphones form a SOny portable DAC & it sounds very good - about medium volume at the moment is the max I can get. But no hiss, hum or noise! Let's see tomorrow how it sounds when I double the gain.

I will feed it from a better source (a modified Hiface) & use better headphones (balanced Superlux 6681s). Unfortunately my amplifier is out being reviewed so I might not be able to hear it through my Jordan JX92S speakers unless I might be able to dust off a TA2020 amp & have a listen?

 

[jkeny]

It turned out that my Slagle output transformers were not suitable to drive a 32ohm headphones. So I tried the 8K:16ohm O/p trafos that are on the Rogers Cadet III chassis that I'm building this into. I'm replacing the existing preamplifier in that chassis with this one! (I've already changed the amplifier circuit to a "Baby Huey" configuration to great effect. SO this should make for a great integrated amplifier.

With the 8K primary the sound is a lot louder & loud enough for me but not ear-splitting. I wonder what ir will be like driving AKG701s - I might get a chance to hear this combo next week.

I was hoping that it would also double as a great headphone amplifier but this was really a side goal! All is not lost yet, I'm wondering if an external trafo could sit between the output & the headphones to better match the impedances. Possibly a 40K to 16ohm one. Is there such a beast? What about PS toroids?

Anyway, I can hear some slight hum also which I have to track down - possibly the PS trafo's magnetic field effecting the tubes. the tubes are 6" & 7" away.