tube current sources

Everything is a "source" when viewed from the right location O:-)

Files renamed:

Wonder how one would actually source down from B+ with a toob ?? ...Jim Thompson

As long as you are breaking all the rules and using semiconductors, why not cascode a depletion fet into a tube?

Jim, did you look at my circuit? Its bias control is floating. It's a genuine 2-terminal source ckt, usable either as a current source or current sink.

BTW, to answer John's comment, the reason for an IC instead of a depletion FET, etc., is to obtain close to 1% accuracy.

BTW, both circuits are prob gonna need comp caps.

To get a low "knee" VG has to be slightly above VK. Simulate yours with a tube model (DuncanAmps.com) and see where the knee is.

Yours may also need a "kick-start", though leakage may be enough. ...Jim Thompson

Like this...

The Pentode needs less headroom, but breaks down the LM317L :-( ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

The requested URL was not found on this server.

For a milliamp, with 300V B+, it takes a 300k 1% resistor and gives you several volts of compliance. That's gotta be a half-watt resistor, though.

Bwahahahahaha! The master of electronics provides the ultimate solution >:-} ...Jim Thompson

Connecting to OUT of LM317L instead of ADJ gains 24V of headroom...

...Jim Thompson

The irony is that:

- For 100% tube state,

- Pull-up CCS are harder,

- Pull-down CCS are almost trivial (e.g., ground/Vneg referenced pentode),

- And for the places you'd ever need a pull-down CCS (e.g., diff pair), a resistor is several times good enough, anyway!

Tim

Yes, talking about breaking rules, replace the tube with a depletion MOSFET and be done with it. . . Two-terminal floating current source/sink . ----------------------------------------- . o CS high WH 12-31-2015 . | . '--| depletion FET . ,->| |---, Table 3.6 . +--| | . | | . +-----, | . \_|__ | | . /_\ R2 | . | +--' Iz = -Vgs / R2 . +-||--+ . | _|_ . | | | . +---|fb | IC ** . | |___| Table 9.7 . R1 | . +-----' Ics = Vfb / R1 + Iz . | . o CS low

One could simply use the depletion FET with a self-biasing resistor, but adding a reference lets us have a precision current. The zener voltage must be high enough to allow for the FET's negative bias, plus reference headroom.

Isn't there a problem with "precision"? Iz (actually current in R2) is dependent on repeatability and TC characteristics of VGS. ...Jim Thompson

Maybe like this...

...Jim Thompson

Right, Jim, good point. A high-current version with a low-Iz reference might not look so bad. But high currents and tube voltages don't mix.

Higher zener voltages reduce the change ratio.

But we can fix things with a BJT, greatly reducing the variation in Iz, and adding R2's varying current to R1's measurement.

. Two-terminal floating current source/sink . ----------------------------------------- . o CS high WH 12-31-2015 . | . '--| depletion FET . ,->| |--, Table 3.6 . +--| | . | | . +---, | . | R2 | . | | | . | +---+-, . \_|__ | | . /_\ | R3 . | >| | . | |---+ Iz = Vbe / R3 . | /| | . | | | . | +--||-+ . | | _|_ . | | | | . +---+---|fb | IC ** . | |___| Table 9.7 . R1 | . +---------' Ics = Vfb / R1 + Iz . | . o CS low

Helper added to improve TC...

...Jim Thompson

The challenge was to do it with tubes.

Tubes!

MOSFETs!