help me with UJT - should i use it?

Thread Starter

slackguy

Joined Feb 11, 2016
76
I'm just a hobbyist. I bought UJT-2N6027 but find it "difficult to use". Should I just throw it away or is there an "easy" circuit I can try to use it? Any advice (other than IC chips - which i think i'll move on to) for what is easier to use than UJT?

BACKGROUND

I bought the UJT because I "heard" from sources such as

http://www.learningaboutelectronics.com/Articles/What-are-unijunction-transistors

that UJT are used as "digital switches". HOWEVER - in one night sitting I could not get it to switch anything ! :) I read

https://www.electronics-tutorials.ws/power/unijunction-transistor.html

but in that night could not get that circuit working (the examples provided in that article are not for blinking led or something quickly built and my changes didn't work and exact values aren't given in the example)

My new understanding is (correct me please), UJT is just like a normal transistor which has a slightly exaggerated response curve. (chips made with them inside were likely not easy to design at first, but are easy from a legacy perspective). If I'm right: I need input to exactly hit a threshhold (of a curve, rather than transisitor more linear response), I ALSO have to have an input circuit that can deal with the issue "it's not 0 or 1, 5V or 0V", the voltage output from a UJT is actually "full" or "limited" (subject to a curve that can sway quickly if the input to UJT is not clean). In one sitting I could design both sides and hit the right curves without an oscilliscope. (btw I have a moderate oscope now, didn't then).
 

Thread Starter

slackguy

Joined Feb 11, 2016
76
project wise, my only plan was to make a blinking LED by a (simple) RLC oscillator and a UJT which gave or cut power. I had no idea I would be in for a "long lesson" when I bought the UJT? (btw i also got a cheap signal generator with the oscope)
 

Thread Starter

slackguy

Joined Feb 11, 2016
76
Pitch it -- nobody uses them much
thank you I will

this has taught me to appreciate the effort of all those hard working predecessors who made things like the 555-timer or other IC chips that are so simple to use, the effort it takes on the inside of IC where there no clean "on and off" without allot of design and insuring of signal levels
 

Thread Starter

slackguy

Joined Feb 11, 2016
76
thanks albert i will try that. unfortunately it looks like the diagram i already tried that doesn't work. (i forgot to say that there is a schematic floating on the inet many websites copy and i tried it and it didn't work, in that night's sitting. could be i got a bad ujt. could be the schem. you found will work - isn't that "wrong" schem that's been floating around on the inet, don't know for sure yet. thanks)
 

Bernard

Joined Aug 7, 2008
5,784
With a PUT oscillator, a LED is not a great load as it does not allow C to discharge fully to ground. Adding a resistor from cathode to common
usually corrects the problem, maybe 1k or less. Just a few weeks ago I pulled some out for a frequency divider. I was wondering if a common gate bias could be used for several PUTs & discovered that when triggered the gate fell to ground with a nice clean neg. square pulse with same width as K + pulse.
 

AlbertHall

Joined Jun 4, 2014
12,345
thanks albert i will try that. unfortunately it looks like the diagram i already tried that doesn't work. (i forgot to say that there is a schematic floating on the inet many websites copy and i tried it and it didn't work, in that night's sitting. could be i got a bad ujt. could be the schem. you found will work - isn't that "wrong" schem that's been floating around on the inet, don't know for sure yet. thanks)
Make sure you have the connections right.
 

Bernard

Joined Aug 7, 2008
5,784
From post #4. " Load current approx300 uA " That is load on power supply, load on circuit cathode about 1A, poor LED; what happened to current limiting?
 

Bernard

Joined Aug 7, 2008
5,784
GE SCR MANUAL, 1972, " in the saturation region is approximately 3 ohms." 6 V on gate - 2.5 V for red LED = 3.5 V / 3 ohms = 1.16 A.
In reality just pulled it out of the blue sky with fore knowledge that PUT could drive high current SCRs with gate requirements in amp range.
 
Last edited:

Tonyr1084

Joined Sep 24, 2015
7,853
@Bordodynov I see an additional resistor (R5) not in @AlbertHall's link. That's the only difference I'm seeing in the two diagrams, aside from a few value changes. 10K & 22K are changed to 100K each. Not going to spend time comparing every aspect of the two circuits. But it looks like this makes a nice negative edge trigger (clock).

Gotta go to work now. See y'all later.
 

Bordodynov

Joined May 20, 2015
3,177
I didn't see post #4, unfortunately. And I drew the scheme myself. I did not aim to get a period of one second. Without an additional R5 resistor, I got a high impulse current through the LED. The additional resistor R5 reduced the current by about 10 times.
There's another difference you haven't noticed is where the voltage divider is on.
 
Top