Talk:Constant-current diode

Diode-Connected Transistor
I was under the impression that a diode-connected transistor is either a BJT with base connected to collector, or a JFET with the Drain and Source tied together. Can someone who is more of an expert than I confirm this? Miles — Preceding unsigned comment added by 128.32.13.9 (talk) 20:12, 13 March 2012 (UTC)


 * I'm under the impression that constant-current diodes should be possible with any unidirectional analog (or at least "multi-step") "resistor" (including e.g. triodes, or a diode + a potentiometer controlled by a servo & circuit, or certain configurations of charge-pumps, or...), as long as the resistance can be controlled via the state of one of the two "diode" connections. Never heard of it being done in those ways, though. - 99.23.95.33 (talk) 02:05, 19 August 2014 (UTC)


 * It's been a long time since this question was posted, but for anyone who happens to read the talk page: a "diode-connected transistor" is a transistor connected so it behaves like an ordinary diode, and that is totally different from a "constant-current diode," the subject of this article.  Diode-connected transistors have an exponential current/voltage curve and no upper limit on the current until they are destroyed; constant-current diodes limit current to a maximum, safely.  It was a mistake to ever include that phrase in the current article.  188.182.238.181 (talk) 17:19, 31 May 2015 (UTC)

Too specific?
The CALY Technologies products for example use the acronym "Current Limiting Devices". And they have bidirectional ones, and even the unidirectional ones seem to be described by a distinctly non-diodic current-voltage curve. Perhaps defining it as a "shorted JFET" or even a "diode" at all is too narrow these days? --RProgrammer (talk) 17:36, 26 April 2020 (UTC)