Tube vs. Solid State Preamp

All that attitude reflects is no knowledge of how to properly design around BJT's. Admittedly, I've been doing it for 55 years, so _maybe_ I've acquired some skill at it :-) ...Jim Thompson

The TL598 has dead-time control but it's usually just too weak to drive really fat MOSFETs. Got to follow it with some real gate drivers, or use a better chip from LTC.

But a tantalum? I can already picture it, a sharp popping sound, some hissing, an orange puff of smoke wafting through the room ...

I have one transformer here where the winding is square copper, wrapped in lacquer-drenched cloth, then hammered in place around the bobbin. Lots of amps.

I once did a 5V/200A transformer... I used insulated copper foil that was the width of the bobbin, 90° fold at the ends for output :-) ...Jim Thompson

Well, that can most certainly be arranged:

I've actually made a very HV transistor by

Yup, used them. In fact, one of my RF tools here (a dipmeter) has one in there except they called them acorn tube in them's days.

That I don't remember. I'm not _that_ old :-))

Just wanted to show that it is indeed possible. The nuvistor preamps from the olden days where really not inferior in noise figure, even when compared to super low noise FETs. I think at first it was RCA FETs (40673?) and then I believe the first NEC devices came out.

The good thing was that the nuvistor preamp wasn't dead if serious weather static hit the input or if you accidentally transmitted into it. FET amps instantly became doorknobs ... tssst ... phut.

Tubes are sort of big for that.

That's just plain dumb audiphoollery.

I needed a 1000V device with characteristic curves much like an ideal CMOS transistor... no ability to get feedback in there. Piece-a-cake with an NPN in the cathode circuit of a pentode (with local feedback).

Yeah, You are! You just don't admit it :-) ...Jim Thompson

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Way ahead of 'ya:

Think I'm going to dip them in epoxy. The windings aren't touching, but there isn't anything to keep them apart. Note that insulated copper strap is useless, because it scratches right through when winding. The core edges are already a bit nicked, and it's got two layers (yellow on white).

Tim

Ah, but these *are* the gate drivers.

Ya, that was before I realized these things are a little bigger than your average CMOS logic. ;-o

To be fair, despite the rated ESR, I think the transistors were burning most of the heat, not the tantalum. Maybe its ESR is way less than the datasheet claims. Power supply rail basically just looked like deltas, no ringing or recovery inbetween, which corresponds to my experience (i.e., a tantalum can be quite excellently modeled as ideal C + ESR + a few milimeters of wire, no spooky higher order elements).

Tim

I'm going to have to make one, if I ever get access to the equipment (and glassblowing expertise!) to do it.

There's something fundamentally wrong about making a tube with gas, though. It can work, but it can't work very well. It's very difficult to pump the electrode section low enough so that it's only ions, no electrons and definitely no secondary ionization taking place. And as soon as those ions hit the plate (= cathode), they become hydrogen again. The plate material has to resist atomic hydrogen, which is rather acidic (literally, 100% Bronstead-Lowry acid). Plus, ions carry enough momentum to sputter the cathode, which is even worse.

Even if your P-tube has a tranconductance ~2000 times lower than a 12AX7 (i.e., about 1umho), it's worth considering because hey, a current mirror is a current mirror.

Tim

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Well, the man said FET, which is distinctly not a BJT.

Tim

But it'll sure turn some heads during a party :-)

But it was you who called me young buck and yourself old buck :-)

That's MOSFET pairs, not gate drivers, that won't improve transitional things all that much.

TC4420, now that's what I'm talking about ;-)

My experience with tantalums is that they are excellent in terms of electrical behavior. Until one fine day ... *PHOOMP*

My disagreement wasn't with the design around the devices, it was with a statement made about the devices themselves, which I believed was inaccurate.

I don't advocate designing valve or FET circuits without feedback, but it is often possible to get away without feedback in low end audio applications - however, BJTs could never be used that way*. It is not 'good' design practice seen from the quality viewpoint, but it might be appropriate design practice if space, component count or cost considerations over-rode the desire for quality.

• I've just remembered a circuit in which I did use transistors to handle audio without feedback - it amplified the sound of a telephone bell, as picked up by a crystal microphone, and fed it to a loudspeaker to warn a partially-deaf person another room. Everything was biassed off to save battery power and the output from the microphone, which was slightly greater than 0.6 volts, turned it on during the tips of the waveform. The sound was a horrible teeth-grinding squawk, but it served its purpose (except when a visitor with normal hearing dropped a cup of tea in shock).

My understanding is they only go Chernobyl when overheated. So maybe with a combination of particularly harsh operating conditions in the middle of Indian summer, a few might pop. As long as you keep RMS amps and operating temperature within limits, they'll work.

I haven't had one explode on me yet, but I haven't used many, either. I've seen a few pieces of equipment that cooked them, but far more that haven't failed, and haven't had one go off in front of me. I must admit the horror stories have me scared...

Tim

Acorn tubes and Nuvistors were very different things. Acron tubes were used in a direction finder loop pre-amp that I used in the U. S. Army Signal Corps in 1943-1945 in North Africa and Italy. They were 954 and

A Nuvister was a small ceramic vacuum tube. The last one I saw was used in my garage door opener receiver. When it failed, I couldn't find a replacement and had to replace the whole receiver with a transistorized one.

I was at Bell Labs when transistors were young, and have taught transistor electronics at UCLA Extension, but most of my career has included vacuum tubes from magnetrons to traveling wave tubes.

I'm older than either of you! :-(

Here's one:

ftp://jjlarkin.lmi.net/Acorn.JPG

I have some nuvistors around here somewhere too. They're not as pretty... metal cans that look a lot like an overgrown TO-5 transistor.

John

Nuvistors are cool. I remember as a little kid reading the 1966 ARRL handbook that my brother gave me--iirc it had a 432 MHz transverter based on 6CW4 Nuvistors. I've salvaged some from old HP counters too.

Cheers

Phil Hobbs

It was a HV shunt regulator. I'm trying to find the original paperwork, but the tube was quite happy.

I'm just yanking your chain :-) ...Jim Thompson

Not by much! IIRC you said something like 74?? I'm 70. ...Jim Thompson