I started building radios in 1963. At some point I got acquainted with tunnel diodes and purchased some. I still have them. That period of time was most exciting as the semiconductors were pushing out at all fronts and many exotic devices came around. Many of them are now unknown. I still have data books of that time and it is both nostalgic and exciting to read them. I guess now it is extremely difficult to buy even tunnel diodes.
Back in the early 1970s, Tektronix had a little device which was a tunnel diode “pulser”. You fed it with a pulse generator and it generated a really fast rise time pulse used to test rise-time and therefore, bandwidth of the scope and plug in units. I ran a “calibration” shop on a carrier and our set of “standards” was pretty limited due to space, environment, and limited training of most sailors (and money, of course).
05:16 IMSAI Guy discovered a chaotic non-linear "butterfly effect" with the tunnel diode and the higher inductor in series. Now we need someone to write-down the equations and come up with what the strange attractor looks like.
The tunnel diode motor or pendulum is a fun device to play with. It is described in the Scientific American magazine Amateur Scientist column by C. L. Strong, October 1965, pg. 112. The invention is attributed to Harry E. Stockman. It consists of a tunnel diode, two resistors, a coil of wire, a 1.5 volt battery, and a magnet. No commutator is needed due to the action of the tunnel diode. I built one back in that era and it worked fine. I suspect you could leave out the resistors and power it with a solar cell instead of the battery, since the resistors form a voltage divider that biases the diode to about 0.42 volts. It would be a simple series circuit of the diode, coil, and solar cell.
I remember in the old tunnel diode papers where they talked about a negative-resistance region. Operating in this region allows the diode to be used as an oscillator or amplifier. Weird stuff.
Your diode has parasitic capacitance, so if you put large inductor the oscillation frequency goes down so your curve tracer bandwidth make it able to see the oscillation. Too small an inductor make oscillate so fast the you see the end points only. The right amount of inductor make the poles and zeros ( of your curve tracer and diode system) line up correctly and cancel oscillation One way to get ride of oscillation is to put shunt resistor with your diode as I mentioned in part 1 Let R1 be the resistance of your diode and R2 is the shunt resistor you put in parallel with your diode: Here is the Math: For parallel resistors we have: Req= (R1*R2)/(R1+R2) Say now R1 ABS(R2) (i.e. ABS means Absolute Value) In that case both numerator and denominator become negative and therefore Req becomes positive (division of two negative numbers is positive) So you can measure Req and since your R2 is known positive value which you shunt R1 ( your diode ) with it. you can calculate R1(your diode) equivalent resistor.
Curve tracer loveliness perhaps a small project we could build to use with our X/Y setting on a basic oscilloscope maybe something to consider going forwards love the octopussy!
That's some funky solid-state physics in action. I think I recall seeing someone use one in a noise generator device, intended to generate "strong" random numbers for crypto applications. Waving my hands, seems like "quantum tunneling" and "random" go together like chocolate and peanut butter.
A discreet version of the tunnel diode called a lambda diode can be made with a BJT and a FET. Much easier to play with for demonstrating negative resistance . Just google lambda diode.
wikipedia has a tunnel diode article and a pic from a Tektronix 571 curve tracer. I thought that picture showed a better view of the negative region. It is not jumping around.
If I recall correctly, .35/rise time=bandwidth. 10ns rt=35 mhz bw. The rt of the pulse was so fast, you only see the rt of the scope. Why I remember, I don’t know.
I'm sorry yours is not working. It is an intriguing device. I have a Heathkit Tunnel Dipper that oscillates just fine, but it doesn't make the best dip meter. The Tunnel Dipper got a bad reputation because the meter is quite touchy to adjust with the gain knob. Unlike a grid dip oscillator where the oscillation amplitude is constrained by the negative feedback of the grid bias, the Tunnel Dipper oscillation amplitude is clamped by the "walls" of the diode I-V characteristic. Thus, it is not very sensitive to outside influences such as the nearby tuned circuits it is supposed to sense. The designer tried to make up for the lack of sensitivity by adding a lot of dc gain in the meter circuit. The touchy adjustment was the result. Had the design incorporated another means of negative feedback to limit the amplitude, it could have been an excellent instrument. If your tunnel diode is good, maybe the trouble is wrong value bias resistors or the contacts to the plug-in coils are out of whack or dirty.
Hey, there are some hints at how you might improve your setup in the art of Electronics: the x chapters (which came in 2020). I believe they used a super low inductance in addition to a really fast oscilloscope.
A tunnel diode or Esaki diode is a type of semiconductor diode that has effectively "negative resistance" due to the quantum mechanical effect called tunneling. It was invented in August 1957 by Leo Esaki when working at Tokyo Tsushin Kogyo, now known as Sony.
That second curve is not what you saw with the East Block devices. Ipso facto the Kiethly I/E forcing means there is no oscillation. The devices are just doped poorly. Get some real tunnel diodes. And why not tell your viewers where they can get a copy of the GE book. I know you didn't make a Xerox copy of my original.
@@andymouse I don't need the link. I have an original. The creator of this video should have read the whole book from cover to cover before creating another totally misleading video full of incorrect conclusions.
You said " And why not tell your viewers where they can get a copy of the GE book" and I pointed out that he did. It would help you with conclusions if you read the book also I think.@@williamogilvie6909
Fun stuff. When I did my tunnel diode video, I didn't have the 576. Now, I think I need to play with TDs on the 576....
I started building radios in 1963. At some point I got acquainted with tunnel diodes and purchased some. I still have them. That period of time was most exciting as the semiconductors were pushing out at all fronts and many exotic devices came around. Many of them are now unknown. I still have data books of that time and it is both nostalgic and exciting to read them. I guess now it is extremely difficult to buy even tunnel diodes.
This is a great little series... I love the idea of exploring obscure components that I, for one, will probably never actually use. :)
Back in the early 1970s, Tektronix had a little device which was a tunnel diode “pulser”. You fed it with a pulse generator and it generated a really fast rise time pulse used to test rise-time and therefore, bandwidth of the scope and plug in units. I ran a “calibration” shop on a carrier and our set of “standards” was pretty limited due to space, environment, and limited training of most sailors (and money, of course).
05:16 IMSAI Guy discovered a chaotic non-linear "butterfly effect" with the tunnel diode and the higher inductor in series. Now we need someone to write-down the equations and come up with what the strange attractor looks like.
The tunnel diode motor or pendulum is a fun device to play with. It is described in the Scientific American magazine Amateur Scientist column by C. L. Strong, October 1965, pg. 112. The invention is attributed to Harry E. Stockman. It consists of a tunnel diode, two resistors, a coil of wire, a 1.5 volt battery, and a magnet. No commutator is needed due to the action of the tunnel diode. I built one back in that era and it worked fine. I suspect you could leave out the resistors and power it with a solar cell instead of the battery, since the resistors form a voltage divider that biases the diode to about 0.42 volts. It would be a simple series circuit of the diode, coil, and solar cell.
I remember in the old tunnel diode papers where they talked about a negative-resistance region. Operating in this region allows the diode to be used as an oscillator or amplifier. Weird stuff.
Put a scope probe across the diode, and use delayed trigger, with AC mains sync, to get the negative resistance region only on screen.
Your diode has parasitic capacitance, so if you put large inductor the oscillation frequency goes down so your curve tracer bandwidth make it able to see the oscillation.
Too small an inductor make oscillate so fast the you see the end points only.
The right amount of inductor make the poles and zeros ( of your curve tracer and diode system) line up correctly and cancel oscillation
One way to get ride of oscillation is to put shunt resistor with your diode as I mentioned in part 1
Let R1 be the resistance of your diode and R2 is the shunt resistor you put in parallel with your diode:
Here is the Math:
For parallel resistors we have:
Req= (R1*R2)/(R1+R2)
Say now R1 ABS(R2)
(i.e. ABS means Absolute Value)
In that case both numerator and denominator become negative and therefore Req becomes positive (division of two negative numbers is positive)
So you can measure Req and since your R2 is known positive value which you shunt R1 ( your diode ) with it. you can calculate R1(your diode) equivalent resistor.
Curve tracer loveliness perhaps a small project we could build to use with our X/Y setting on a basic oscilloscope maybe something to consider going forwards love the octopussy!
I'd like to use these as clippers in a guitar overdrive circuit to hear what they sound like.
That's some funky solid-state physics in action. I think I recall seeing someone use one in a noise generator device, intended to generate "strong" random numbers for crypto applications. Waving my hands, seems like "quantum tunneling" and "random" go together like chocolate and peanut butter.
A discreet version of the tunnel diode called a lambda diode can be made with a BJT and a FET. Much easier to play with for demonstrating negative resistance . Just google lambda diode.
wikipedia has a tunnel diode article and a pic from a Tektronix 571 curve tracer. I thought that picture showed a better view of the negative region. It is not jumping around.
Thank you that answered my question!
If I recall correctly, .35/rise time=bandwidth. 10ns rt=35 mhz bw. The rt of the pulse was so fast, you only see the rt of the scope. Why I remember, I don’t know.
I got a heathkit "tunnel dipper" dipmeter at a hamfest, was never able to get it to oscillate.
I'm sorry yours is not working. It is an intriguing device. I have a Heathkit Tunnel Dipper that oscillates just fine, but it doesn't make the best dip meter. The Tunnel Dipper got a bad reputation because the meter is quite touchy to adjust with the gain knob. Unlike a grid dip oscillator where the oscillation amplitude is constrained by the negative feedback of the grid bias, the Tunnel Dipper oscillation amplitude is clamped by the "walls" of the diode I-V characteristic. Thus, it is not very sensitive to outside influences such as the nearby tuned circuits it is supposed to sense. The designer tried to make up for the lack of sensitivity by adding a lot of dc gain in the meter circuit. The touchy adjustment was the result. Had the design incorporated another means of negative feedback to limit the amplitude, it could have been an excellent instrument.
If your tunnel diode is good, maybe the trouble is wrong value bias resistors or the contacts to the plug-in coils are out of whack or dirty.
Hey, there are some hints at how you might improve your setup in the art of Electronics: the x chapters (which came in 2020).
I believe they used a super low inductance in addition to a really fast oscilloscope.
The cover on that manual is fantastic... you just don't get graphic design like that any more.
What do those microcontroller + LCD universal component testers show for a tunnel diode?
It oscillates back and forth--through time! Lol
That tunnel diode manual would make for good toilet reading. 😮
You'd need your meals brought in!
Is it oscillating because of the depletion region?
There's definitely some black magic involved there... Did you try to put a VooDoo doll in series with the diode? :P
really an interesting component. Was it actually a coincidence that we “invented” this diode?
A tunnel diode or Esaki diode is a type of semiconductor diode that has effectively "negative resistance" due to the quantum mechanical effect called tunneling. It was invented in August 1957 by Leo Esaki when working at Tokyo Tsushin Kogyo, now known as Sony.
understand, thank you for reply !
RTFM
That second curve is not what you saw with the East Block devices. Ipso facto the Kiethly I/E forcing means there is no oscillation. The devices are just doped poorly. Get some real tunnel diodes. And why not tell your viewers where they can get a copy of the GE book. I know you didn't make a Xerox copy of my original.
The link to the manual is right there in the description.
@@andymouse I don't need the link. I have an original. The creator of this video should have read the whole book from cover to cover before creating another totally misleading video full of incorrect conclusions.
You said " And why not tell your viewers where they can get a copy of the GE book" and I pointed out that he did. It would help you with conclusions if you read the book also I think.@@williamogilvie6909
Such a fresh episode (joking as I'm first comment)