Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Having fun with discrete transistors

Status
Not open for further replies.

schmitt trigger

Advanced Member level 5
Joined
Apr 17, 2013
Messages
2,645
Helped
843
Reputation
1,690
Reaction score
819
Trophy points
1,393
Activity points
17,751
It was cold and raining outside, a miserable Sunday afternoon. I had just finished a book, and did not want to binge-watch another Netflix series. Bored to death, I headed to my lab to figure out something, anything, to do.

Rummaging thru my parts bin, I noticed some 2N2646 unijunction transistors. When was the last time I actually built something employing a metal-case transistor? Eons ago. So I decided that this was it.
I quickly downloaded the 2N2646 datasheet, and quickly breadboarded a circuit, and in no time I had a oscillator running consisting of only 3 resistors, 1 capacitor and a single transistor.

Could this simple circuit be enhanced? The Base1 pulse is kind of wimpy, and the emitter waveform is the classic non linear RC-charging ramp.

The first thing is to replace the capacitor charging resistor with a CC source. There are dozens of CC schematics, but in the spirit of discrete transistors, it was replaced with a J175 P-chan JFET found in my parts bin. This transistor has as very healthy IDSS of up to -60 mA. Therefore it has to be throttled down with a source resistor. If a trimpot is inserted in series with the resistor, the current and therefore the frequency can be adjusted.

To amplify the output pulse to a rail to rail level, the B1 pulse was applied to the gate of an 2N7000 Mosfet, also from the parts bin.
Now that I have a nice looking pulse, what should I do with it? Drive a transistor bi-stable multivibrator, of course. Classic circuitry, I initially simulated in LTSpice to find workable values. Once that I became satisfied, I breadboarded the circuit, and connected it to the pulse.

The final schematic is attached, and the results are shown in the waveforms image. Amazing that with such simple circuitry, using only discrete semiconductors, many things can be achieved.
BONUS POINTS: And it was done with the four main transistor families: UJT, JFET, MOSFET and BJT. Hurray!

By now, my wife is calling me. Supper is ready
 

Attachments


d123

Advanced Member level 5
Joined
Jun 7, 2015
Messages
1,984
Helped
444
Reputation
889
Reaction score
436
Trophy points
83
Activity points
20,283
Hi,

Nice circuit, nice results. :thumbsup: I like metal can BJTs. Never used a unijunction so nice to see one in a circuit. I think it took me 11 transistors + passives (6 MOSFETS (oscillator) and five BJTs (current source pair, ramp generator and rail-to-rail output pair)) to achieve a similar sawtooth result.

Thanks for sharing this.
 

schmitt trigger

Advanced Member level 5
Joined
Apr 17, 2013
Messages
2,645
Helped
843
Reputation
1,690
Reaction score
819
Trophy points
1,393
Activity points
17,751
If my memory serves me well, in the late 80s to early 90s, Popular Electronics magazine had a a section about DANDY circuits. DANDY an acronym for Discretes Are Not Dead Yet.

Plenty of very ingenious circuits.
 

betwixt

Super Moderator
Staff member
Joined
Jul 4, 2009
Messages
14,179
Helped
4,684
Reputation
9,383
Reaction score
4,437
Trophy points
1,393
Location
Aberdyfi, West Wales, UK
Activity points
121,096
Somewhat reminds me of a 'beeper' I can across a few months ago. It was a tone at about 1KHz pulsing at about 2Hz rate with apparently no active components. I reverse engineered it and found a resistor, a capacitor, a diac and a self contained beeper. The cap charged through the resistor until the diac conducted then it discharged into the beeper. Simple and very neat.

Going back to the unijunction, bear in mind the same PNPN (or is it NPNP?) junction can be made by connecting two normal bipolar transistors across each other and it works equally well.

Brian.
 

schmitt trigger

Advanced Member level 5
Joined
Apr 17, 2013
Messages
2,645
Helped
843
Reputation
1,690
Reaction score
819
Trophy points
1,393
Activity points
17,751
You can go one further....remember neon lamp ring counters?
Nothing but resistors, capacitors and NE-2 lamps.
 

betwixt

Super Moderator
Staff member
Joined
Jul 4, 2009
Messages
14,179
Helped
4,684
Reputation
9,383
Reaction score
4,437
Trophy points
1,393
Location
Aberdyfi, West Wales, UK
Activity points
121,096
I'm pretty sure I've got some NE-2 here in my box of ever ageing junk. Yes, I do remember them in counters and I think there was a commercial frequency counter made by RACAL that used them. I seem to recollect you had to read the numbers beside the highest columns of lit-up neons.

Brian.
 

schmitt trigger

Advanced Member level 5
Joined
Apr 17, 2013
Messages
2,645
Helped
843
Reputation
1,690
Reaction score
819
Trophy points
1,393
Activity points
17,751
We are showing our age Brian....
Of course, multivibrators predate semiconductors. Most functions were realized using vacuum tubes.
 

Status
Not open for further replies.
Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top