7
\$\begingroup\$

If you encounter an old text which uses tubes instead of transistors, and if you are not familiar with the tubes, which one of the transistor from the following transistor tree would you think of to understand the circuit?

enter image description here

I sometimes encounter an ancient text but when I see tubes in the diagram I used to skip reading. Should I instead think of it as an NPN transistor or a particular type of MOSFET? Which particular transistor in the above tree makes most sense for a tube circuit?

\$\endgroup\$
14
  • 1
    \$\begingroup\$ Which type of vacuum tube? A N type Jfet is a fairly good match for a triode, you can use the same biasing schemes. \$\endgroup\$ Jan 11, 2018 at 22:07
  • 3
    \$\begingroup\$ A JFET has similar transfer characteristics to a triode, but the voltages are usually much lower. They also don't come with heaters. You'll have to add your own nichrome wire for that, or simulate it with a orange LED. \$\endgroup\$ Jan 11, 2018 at 22:27
  • 1
    \$\begingroup\$ As long as the grid is negative, the electrons are pushed away from it, like in a diode. \$\endgroup\$
    – τεκ
    Jan 11, 2018 at 23:29
  • 2
    \$\begingroup\$ @user16307 the grid is negative relative to the cathode. If the grid voltage becomes positive, current does flow from the grid (hm, very much like a JFET) \$\endgroup\$
    – τεκ
    Jan 12, 2018 at 0:00
  • 1
    \$\begingroup\$ No. Grid is usually negative w.r.t. cathode - as in a (depletion mode) JFET which is why the same self-biasing schemes work. But you can pull grid positive to turn the tube on harder - and then it does collect a few electrons, causing "grid current". This is rarely done except in battery radios (where the 1.5V heater supply is only capable of weak emission) or the highest power circuits like RF transmitters, both of which need as much help to make anode current as they can get. \$\endgroup\$ Jan 13, 2018 at 13:55

5 Answers 5

5
\$\begingroup\$

None of the above. They are voltage-controlled, like FETs, but their characteristics are not so similar. Perhaps JFETs are closest, since a more negative voltage on the grid turns them 'off'.

You should just look at the I-V curves of vacuum tubes for a basic understanding.

Consider that if the anode current is fixed at 1.5mA and the grid voltage goes from -1V to -2V, the anode voltage goes from 160V to 260V - a gain of -100!

\$\endgroup\$
2
\$\begingroup\$

Bipolars with high Vearly (rather flat Iout curves, for Vce > 2 volts) are good replacements for pentodes, ignoring the input base current and other biasing differences.

\$\endgroup\$
1
\$\begingroup\$

I wonder if IGBT’s could be adapted for good linearity. As VCCS devices they have a narrow voltage input range but a huge output voltage and current range.

STGP19NC60HD IGBT 600V 40A 130W Through Hole TO-220

\$\endgroup\$
1
\$\begingroup\$

For your use-case of trying to get a gross understanding of a tube circuit without understanding tubes, you can think of a vacuum tube as an N-channel JFET. This applies to circuits using either triodes or pentodes.

Note well the "circuits" part of that last sentence: the analogy only goes so far of course, and if you look at the characteristic curves, you'll see that a triode's curves look nothing like an N-channel JFET's. When put in an actual amplifying circuit however, the resulting behavior of the two is very similar. (For what it's worth on the flip side though, a pentode's curves do look very similar to a JFET's).

This correspondence actually translates to real-world uses. In vacuum tube circuit circles, it's not uncommon to use e.g. the LND150, a high-voltage N-channel depletion-mode MOSFET, as an almost drop-in solid-state replacement for a 12AX7. (A depletion-mode MOSFET can be thought of as a JFET which never draws gate current.)

\$\endgroup\$
1
\$\begingroup\$

The diagram is incomplete: the direct analog of the triode is the static induction transistor.

By further analogy, I suppose one could make all the other poly-odes with the same method, but a much more available equivalent would be the JFET or MOSFET, but only in terms of grid/plate characteristics.

A perfect analogy does not exist, as no devices have the same sequence of electrodes, work potentials (tendency for electrons to emit/absorb at the surface, given some energy), fields, and phenomena like secondary emission, leading to such quirks as negative resistance in the plate voltage saturation region, or enabling the phantastron oscillator circuit. Or of transport phenomena, like nonreciprocal admittance from lower grids to higher but not the other way around. (6BE6 for example, say at ~1MHz, has very small but nonzero admittance Yg1g3 (some ~µS), but not in reverse (Yg3g1 ≈ 0), and not due merely to stray interelectrode capacitance, but independent of it!).

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.