Is there such a thing as an 'inverter varistor'?

Linear and non linear

Is there such a thing as an 'inverter varistor'?

Postby chrismb » Sat Oct 08, 2011 8:45 am

I'm looking for a component, or circuit, in which the resistance increases with increase in voltage across it. I could adapt a constant current circuit, but I was just wondering if such devices already exists. Or otherwise, what is the simplest way to achieve this in a circuit.

Specifically, my immediate inquiry is for a circuit element that provides approx 1kOhm at 10V, to 200kOhm at 200V.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Is there such a thing as an 'inverter varistor'?

Postby Doug Coulter » Sun Oct 09, 2011 2:21 pm

Well, you'll need something fancy to do that -- 10 ma at 10v and only 1 at 200 implies more than a 1/r change, and even more "compliant" than a constant current source. You might be able to create some positive feedback to a constant current source design, but you'd be in a very weird place trying to make a negative resistance not oscillate...

Usually if you think you need something like that, there's an issue of misconception elsewhere that can be fixed over there easier. What do you think you need this for?
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Is there such a thing as an 'inverter varistor'?

Postby chrismb » Mon Oct 10, 2011 4:37 pm

The application is that I have bought a batch of FET drivers that run straight off the mains (no additional voltage required for the FET drive) via a big resistor. These are used in things like the plug-in/screw-in mains CCFL bulbs. This is great if that is exactly what you want to do, but I was thinking about using them for a variable voltage supply at lower voltages (hence, need a lower resistace for the lower voltages) and it would've been very convenient not to have had to worry about creating additional voltages to feed the FET driver power-up.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Is there such a thing as an 'inverter varistor'?

Postby Doug Coulter » Mon Oct 10, 2011 5:56 pm

Got a part number/data sheet? Such things as negative resistances can be built of course, they are the basis of "foldback current limiting" in power supplies and audio amplifiers. They're a bit tricky. But don't you want constant current for supplying this chip? Not sure I understand your problem. A sneaky trick used when you can is to use a capacitor, instead of a resistor, for voltage dropping. You can't drive a half-wave diode that way, the cap just charges up, but I've seen this trick used to get a little current at low voltage from the mains with low power losses (since the cap has only imaginary part in the reactive impedance).

Here's a horrible hack Dan Meyer used to try to keep the output transistors in the "tiger" inside their SOA's. Q2 senses the voltage at the emitter resistor of the main output, and the collector voltage, and if the sum through the resistive divider is greater than a Vbe, it shorts the drive to the power transistor....
SOA.gif
SOA.gif (4.08 KiB) Viewed 5381 times




For the other thing, you use a capacitor to the mains, to a series resistor to ground. You get a voltage divider due to the impedances of each at mains frequency, and that smaller AC signal can now be rectified. The regulation is lousy of course, but this trick is often used to power chips that have a built-in zener across their supply terminals. The resistor wastes some power, about equal to the delivered power best case, but for something small and cheap, it works. There are some variations that substitute the resistor with a zener, then a rectifying diode that conducts when the zener is reverse biased (the way it has a normal zener volts across it). I rarely use this trick, but it's out there in a lot of off-mains switchers. Mostly in stuff where the last penny counts, like microwave ovens, other cheap consumer kit, etc. The reason it's not used more is that the series cap has an impedance that is lower the higher the frequency - and line spikes have a lot of HF content, which makes the current through the cap (and therefore the volts across the load impedance) go very high - and things emit the magic smoke.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA


Return to Analog

Who is online

Users browsing this forum: No registered users and 4 guests

cron