2
\$\begingroup\$

I am trying to do some experiments with the following oscillator circuit extracted from http://jlnlabs.online.fr/cnr/negosc.htm based on Esaki oscillator. enter image description here

I am using 2N2222 and a 10mH inductor instead due to the availability of components, but not oscillation can be observed across the inductor. If someone can give me a head start to understand the circuit operation, it would be greatly appreciated.

In addition,can someone suggest an approach to simulate this behavior? I tried with LTspice, but their models does support to simulate breakdown

\$\endgroup\$

2 Answers 2

Reset to default
2
\$\begingroup\$

The circuit relies on the B-E-diode breakdown. The breakdown voltage is probably a very sloppy controlled parameter for a BJT, i.e. as long as it is above a specified minimum value it may vary very much among different samples and maybe it's too high in your case.

Maybe it helps to try different transitor samples or types or to use a higher operating voltage.

\$\endgroup\$
9
  • \$\begingroup\$ Thank you for the reply. But in this circuit, Base is floating. Should it be connected to a reference to make the circuit more deterministic behavior. I will also try different BJT samples. In addition, would your suggest any approach to simulate this behavior? I tried with LTspise, but their models does support to simulate breakdown. \$\endgroup\$
    – Pojj
    Commented Feb 22, 2018 at 9:04
  • \$\begingroup\$ No, that's not neccessary. Base is at collector level + 0.6V (because B-C Diode is forward biased). \$\endgroup\$
    – Curd
    Commented Feb 22, 2018 at 9:14
  • \$\begingroup\$ I'm not sure if the Spice BJT model models the breakdown behaviour of the BE diode at all; and if it does whether it models the negative resistance... pobably it doesn't. So I guess simulation doesn't help here. \$\endgroup\$
    – Curd
    Commented Feb 22, 2018 at 9:17
  • \$\begingroup\$ Thank you, I managed to grab few 2N2222A and it worked for one of them. Now I 'd like to study this effect bit further, If you can, please suggest some readings. \$\endgroup\$
    – Pojj
    Commented Feb 22, 2018 at 9:54
  • \$\begingroup\$ Good to know that you got it working! I have no particular reading suggestion just a hint that should help to find plenty information: search for "tunnel diode oscillator" (the BE diode of the 2N2222A here works as tunnel diode but a 2N2222A is much more available than a dedicated tunnel diode) \$\endgroup\$
    – Curd
    Commented Feb 22, 2018 at 10:17
2
\$\begingroup\$

This Esaki oscillator relies on some undocumented properties of the 2N2222A transistor, namely that the 2N2222A will have a negative resistor behavior under certain conditions.

See the first plot in the article from your link, note how the current decreases while the voltage increases. This behavior is not mentioned in the datasheets I found so that leads me to conclude that this behavior is not guaranteed by the manufacturer.

So what I see as a possibility is that some 2N2222A transistors might have this behavior and some don't or have a different behavior. Although any 2N2222A should behave the same, when they originate from different manufacturers, there's no guarantee, especially for undocumented behavior.

So what I would do if I were you is determine if your 2N2222A really shows the negative resistance behavior by measuring I(V) (current as function of voltage) of your transistor. It it doesn't show this behavior then your oscillator can never work with this particular transistor.

\$\endgroup\$
4
  • \$\begingroup\$ Thank you for your reply! Please note that I used 2N2222 as I didn't have any 2N2222A in my stock. Do you think this negistor effect is only available with 2N2222A? Anyway, I will try some experiments to see V-I characteristics. \$\endgroup\$
    – Pojj
    Commented Feb 22, 2018 at 9:10
  • \$\begingroup\$ Do you think this negistor effect is only available with 2N2222A? No, I don't think so. The A in 2N2222A indicates the current gain beta. So for example a 2N2222B would typically have a higher beta. A 2N2222 (no A, no B) is very likely simply not "binned" (categorized depending on value) meaning that beta is measured and an A, B or even a C might be printed on it. \$\endgroup\$
    – Bimpelrekkie
    Commented Feb 22, 2018 at 9:29
  • \$\begingroup\$ Thank you, I managed to grab few 2N2222A and it worked for one of them. Now I 'd like to study this effect bit further, If you can, please suggest some readings. \$\endgroup\$
    – Pojj
    Commented Feb 22, 2018 at 9:54
  • 2
    \$\begingroup\$ A couple of years ago I tested the negative resistance region in BJT's and almost everyone has shown negative resistance region around 6.4V...8.7V. But only NPN. PNP does not show any negative resistance region. \$\endgroup\$
    – G36
    Commented Feb 22, 2018 at 15:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.