History of Transistors. Volume 1. Your historic transistors, photos, descriptive text and storage envelopes are contained in an expandable three-ring report binder. The display envelopes are securely stored in plastic sheet holders at the rear section of the booklet.
|Published (Last):||1 August 2016|
|PDF File Size:||13.60 Mb|
|ePub File Size:||12.95 Mb|
|Price:||Free* [*Free Regsitration Required]|
The concept of negative resistance has always fascinated me. Of course, a true negative resistance is not possible, and what is meant is a negative differential resistance NDR. Negative resistance sounds like an unusual effect, but it turns out to be relatively common, showing up in places like neon lamps and a number of semiconductor structures. The best-known semiconductor device exhibiting negative resistance is the tunnel diode, also known as the Esaki diode after one of the Nobel-Prize-winning discoverers of the quantum tunneling effect responsible for its operation.
These diodes can perform at tremendous speeds; the fastest oscilloscope designs relied on them for many years. But, all hope is not lost. Rummaging through some old notebooks, I rediscovered an NDR design I came up with in using two common NPN transistors and a handful of resistors; many readers will already have the components necessary to experiment with similar circuits.
Tunnel diodes are relatively rare beasts — although these days, like most things, you can find them on eBay — so I wanted to come up with a circuit that allowed experiments with negative resistance using common parts.
With this in mind, back in , I came up with a two-transistor circuit where the negative resistance region is somewhat controlled by selection of resistors. The circuit emulates a two-terminal device like a tunnel diode, and for the purposes of this article, I made a shorthand symbol for it that resembles a tunnel diode inside a box.
Referring to the schematic, Q1 acts as an inverter, with its collector initially rising with the supply, but subsequently falling as it is biased on with increasing voltage. The result is a circuit whose current increases with applied voltage up until a point, but then decreases relatively linearly for a while — the negative resistance region — before increasing again. R6 is a key part of this circuit; it affects the width and linearity of the NDR section of the curve.
In any, case, if you find that some experimentation is in order, changing R6 is a good starting point; smaller values stretch the NDR region wider. LTspice simulations work well for this circuit — at least qualitatively. Due to its dependence on transistor parameters, the results you see on the bench may vary from simulation quite a bit, but I found the simulation can be used to gain intuition about the behavior and inform experiments on the real thing.
Actually, this pretty much sums up most of my experience with electronic simulation. The original prototype from was assembled on a small piece of perfboard, while I chose some copper clad and ugly construction for the most recent version.
I subsequently discovered that Chua, et al. In , I collected curves point-by-point with a pair of DMMs — this works, but consumes a lot of your day. This time around, I hacked together a curve tracer with a function generator and an oscilloscope. As shown in the figure, a Ohm resistor is used as a shunt to detect the device current on channel 2 of the scope, while channel 1 measures the voltage across the device-under-test and the sense resistor.
The function generator outputs a ramp waveform — even a sine will do in a pinch — at around Hz. Based on the two points shown with the cursors, the region has an average differential resistance of Ohms.
As you can see from the plot, the region is a decently linear from around 1. After that, the current ramps back up. In this case, the device is exhibiting memory or stateful behavior.
This property of NDR devices has been used in digital circuitry. What can you do with a negative resistance like this? One of the classic uses for an NDR device is in oscillators. Consider a simple parallel LC tank circuit. Normally, any resistance across the tank or even that in the inductor or capacitor themselves is enough to damp any initial oscillation to zero over time.
Supplying current through a negative resistance, however, will have just the opposite effect, causing the oscillations to build and continue indefinitely. Of course, there is no free lunch; we have to use a power supply to bias the NDR device into the negative resistance region.
Shown in the schematic is a 5 kHz oscillator I made this way using a 1 uF capacitor and a 1 mH inductor. For zero-tolerance parts — which I did not happen to have in stock — this should result in a 5. The measured frequency was 5.
I found that the circuit oscillated for supplies between 1. The output amplitude is around 4. Keen eyes might notice a component missing in the simple schematic. For audio-frequency oscillators like this one, it might be OK to have the return current flow back through the power supply if the leads are short.
In the classic tunnel diode amplifier circuit a resistor is used to bias the diode and set the gain. With a two-terminal amplifier circuit, multiplexing the input and output can be tricky. Probably the best way is to AC-couple the signals, but I chose to stay close to the canonical circuit using the offset control of the signal generator to adjust the DC bias on the diode. A second problem is that the gain of such an amplifier depends on the magnitude of the external resistor relative to the negative resistance value.
To make measurement a bit easier, I added a 51 Ohm resistor across the output of the signal generator, making the output impedance around 25 Ohms. This allows adding a 30 Ohm gain-setting resistor without blowing our budget of 64 Ohms. The input signal to this amplifier is at the top of the 30 Ohm resistor, while the output is taken at the bottom.
By adjusting the signal generator offset, I was able to find a bias point around which the device showed some voltage amplification. In the oscilloscope image, you can see that the output signal cyan trace has a higher amplitude 1. The 30 Ohm resistor and the NDR work like the familiar resistive divider, except they increase the input voltage! In the mean time, you can dig out some transistors and have some fun at the bench, or check out another of our articles on negative resistance.
Risetime was 10s of picoseconds, so well into the GHz region! Actually the Tunnel Diodes that GE were making are still around. I also have two. Then they were not cheap. But do they still work? Tunnel diodes were infamous for failing with time because of the high doping required to make them function. This site uses Akismet to reduce spam. Learn how your comment data is processed. By using our website and services, you expressly agree to the placement of our performance, functionality and advertising cookies.
Learn more. Faking It Tunnel diodes are relatively rare beasts — although these days, like most things, you can find them on eBay — so I wanted to come up with a circuit that allowed experiments with negative resistance using common parts.
Negative differential resistance circuit built with BJTs and corresponding circuit symbol I just made up With this in mind, back in , I came up with a two-transistor circuit where the negative resistance region is somewhat controlled by selection of resistors. NDR oscillator output at 5. NDR amplifier input yellow vs output cyan. Offsets adjusted to facilitate comparison.
Report comment. Very interesting article, thank you! You can also search for GaAs gunn diodes they have the same U I curve. Leave a Reply Cancel reply. Search Search for:.
A lambda diode is an electronic circuit that combines a complementary pair of junction gated field effect transistors into a two-terminal device that exhibits an area of differential negative resistance much like a tunnel diode. Lambda diodes work at higher voltage than tunnel diodes. Whereas a typical tunnel diode  may exhibit negative differential resistance approximately between 70 mV and mV, this region occurs approximately between 1. A lambda diode therefore cannot replace a tunnel diode directly. The lambda diode current approaches zero as voltage increases, before rising quickly again at a voltage high enough to cause gate—source Zener breakdown in the FETs. This has the advantage that its properties can be fine tuned with a simple bias driver and used for high sensitivity radio applications, sometimes a modified open can PNP transistor with IR LED can be used instead.
Fun With Negative Resistance: Jellybean Transistors
Not so these days, when most people have forgotten about them; the odd few that come up on ebay are usually very expensive. A precaution to bear in mind when de-soldering and replacing TDs is their sensitivity to heat; a wise precaution with any soldered semiconductor replacement is to clamp the leads first using a pair of pliers to wick away the heat. First, a small list of tunnel diodes used in Tektronix instruments:. It is used in the Tek Is the simply a selected ? The is listed in the as being manufactured by Tektronix. The is also unobtanium.
1N3720 डेटा पत्रक PDF( Datasheet डाउनलोड )
1N3712 Datasheet PDF - New Jersey Semiconductor