Varistors

[RICH535]

I have been given the assignment of finding a replacement varistor in a circuit where it is not used as a protection device. Rather it is being used as a voltage-dependent resistor to start an op-amp oscillator. The varistor is in parallel with an inductor and capacitor and is in series with a fixed resistor of 2.7K ohms. The varistor itself measures 75K ohms out-of-circuit. We used to get the varistor from a company called Kanthal Globar. About the only spec I have on it is that its voltage @ 1 mA is 3.3-4.7V. Is there any thing else I could use as a substitute for the varistor?

 

[itsmoked]

What shape is it?
Got a picture?
Have you measured its trip voltage?
Is the one you have blown and you have no other?
How would THAT be used to start an oscillator?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Hi Keith,

Guess what!?

I done work for Kantahl (name derived from Kantzow and Hallstahammar). It is now a Sandvik company. This group bought the Globar (Glow Bar) heating resistor operation from (I think) a German company.

I am not sure if this is a varistor or some other highly non-linear voltage dependent resistor. But obviously it is used parallel to the feed-back resonant circuit in an oscillator so that gain is reduced when amplitude gets above the "opening voltage" of this device. I would call it an amplitude limiting device rather that a starting device.

I would try a zener in that voltage range instead. A zener is not bidirectional like your "varistor" is, but that usually is of little importance if the Q of the oscillator is high, which it probably is. So, when the zener starts conducting, the amplitude will stabilize and since your Q is high, the next (not limited) half-wave will not be much higher. You might get some even harmonics. If that is unacceptable, use two zeners anti-parallel.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[VEBill]

"...to start an op-amp oscillator..."

Starting and maintaining an oscillation doesn't normally need anything special (simply positive feedback, etc.). That's normally the easy part, provided you don't care about the waveform.

The tricky bit that calls for these sorts of unusually-applied components is regulating the output amplitude without introducing distortion, and without adding a dozen additional components.

Are you sure the varistor is used for starting the oscillation, as opposed to gently limiting the amplitude? Is the circuit required to produce a pure sine wave? If so, that's probably a clue to the purpose of the varistor.

I've seen little 'grain-of-wheat' light bulbs used in audio pure sine wave oscillators in years gone by to perform this function. People were always asking, "Why does this Morse code practice oscillator circuit have a light bulb?"

If all this supposition happens to be true, then you could look at small incandescent light bulbs as a last resort.

But you should be able to find a direct replacement if you're lucky.

 

[RICH535]

Thanks to both Skogsgurra and Ve1BLL for their informative replies. I will try the suggestions and let you know. For itsmoked, I have a limited amount of varistors but not enough for production. I have not measured its trip voltage and I can only assume that the varistor is a key component in the operation of the oscillator.

 

[RICH535]

I installed (2) 3.0 volt zener diodes in anti-parallel fashion and the circuit worked! I have contacted the manufacturer, Kanthal Globar, in the interim and the lead-time is prohibitive (18 weeks) along with the price. The zeners are cost-efficient and provide the same amplitude limiting as the varistor. Thanks Skogsgurra!

 

[ScottyUK]

Zeners in antiparallel behave much like a pair of standard diodes in antiparallel. You should have put them in anti-series if you want the zener voltage instead of the forward voltage.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[Skogsgurra]

That's what I meant. And I think that Rich understood that.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[VEBill]

Depending on the application, you might want to 'scope out the waveform to make sure it meets the requirements. Odd harmonics might be higher than before.

 

[ScottyUK]

skogs,

I know that you know!


----------------------------------

Sometimes I only open my mouth to swap feet...