2N2646 UniJunction transistor - what is it and how does it work?

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,852
I'm researching bio-feedback devices that measure galvanic responses in human flesh - USING A 9 VOLT BATTERY! Don't nobody go getting upset with me. The circuit is very simple but I'd like to know exactly how this thing works; and aside from my question, what one would use such a device for.2N2646.jpg
[edit] This circuit
https://www.electroschematics.com/6315/biofeedback-by-skin-resistance/
uses two of these injunction transistors, one fed to a headphone that responds to your state of stress while the other generates a tone you can set. The lower the tone the more relaxed you are, and the more you are able to relax the more you can make one side tone match the other side.
 
Last edited:

crutschow

Joined Mar 14, 2008
34,280
It's commonly used as a trigger for an SCR or as a relaxation oscillator (which is it's purpose in your application).
Here's an explanation of the device.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,852
Those things seem rare. As rare as Hen's Teeth. What would be a good substitute? Or what's the alternative for the circuit I've found?
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,852
I was just looking at PUT's. But what has me confused is the "Programmable" part in the description. And I'm assuming the lead pin-out is the same as the UJT.

OR maybe I could replace it with an SCR. What'cha all think?
 

crutschow

Joined Mar 14, 2008
34,280
The PUT pinout is different from the original unijunction.
The G terminal has resistors connected to it to determine the peak voltage and current characteristics.
Terminal A is the same as the original Unijunction E terminal.
The K terminal is the same as the original B2 terminal and would go to the headphone.
Here's a typical circuit connection:
upload_2018-4-16_0-46-21.png
It's programmable in the sense that you can change resistor values to affect the peak point voltage and current.
You possibly can do that to match the characteristics of the 2n2646 in the original circuit.

No, an SCR is an entirely different animal and won't work in that circuit.

Reading this data sheet may help.
 

Attachments

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,852
@Bordodynov I struggle with LT Spice. Can you show me Vb2 as well? It might help me understand how this device is programmed.

@crutschow Thanks for the help with the 2N6027. Still rare, and a lot of obsolete parts, but also a lot less expensive; around 96¢ per unit. Amazon has 5 for $4.95.
 

ebp

Joined Feb 8, 2018
2,332
You can approximate the behavior of a PUT by using an NPN and a PNP transistor (anything common, like 2N3904 & 3906). If you use surface mount you can get both devices in a tiny 6-pin package. The circuit is not identical in behavior, but nonetheless useful/

Danko used this in a very nifty circuit for pulsing the input to an optocoupler in a thread not too long ago (January?). I can't remember what the thread was about, other than driving an optocoupler with very low average input current.
 

OBW0549

Joined Mar 2, 2015
3,566
Holy Schmitt that's a big book.
Yup. And it's worth thumbing through, because there's a lot of good information in it. I still have my copy; it was my first electronics book back when I was a kid. Some of the information is only of historical interest since some of the devices it discusses, such a unijunction transistors, silicon controlled switches, germanium transistors and tunnel diodes, are rarely if ever used anymore. But that's how things were done back in 1964.

If you'd rather not play around with ancient technology like the 2N2646, note that the functionality of your biofeedback circuit could easily be replicated with a pair of 555 timers-- or by a single TLC556 (dual CMOS 555). Much more modern!
 

RichardO

Joined May 4, 2013
2,270
Top