Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Current sensing switch

  1. Jan 24, 2010 #1
    I'm working on (building) a small induction generator setup. Its still well into the testing phase and i have a, what i think is anyways, rather complex question.

    I know that the output characteristics (voltage and line frequency) of the generator at a given load is proportional to the capacitance on the system. What i would like to do is setup a system were more capacitance is switched into the circuit as the load on the system increases.

    My current idea is to use a set of cheap-ish AC ammeters with variable scales, say a 0-5 A, 0-10 A and 0-15 A. I would modify each ammeter so that when each hits a certain value it would trip a small switch/close a circuit. For example, put a little stop contact on the scale plate, and use a metal needle on the meter, when they touch it closes a circut, also the stop contact will prevent the needle from moving further, so the circuit is closed as long as the current exceeded whatever arbitrary value i put the stop at.

    I can use this low voltage/current from the closed circuit to signal a TRIAC which will bring a new capacitor into the system at whatever current i set it to. I have not looked into any particular TRIAC, but i assume you could find one that has a holding current of a certain value so that it will remove the capacitor if the load on the system drops

    Ok, now that this lengthy idea is out of the way, what i REALLY want to know is A)is the above even feasible...or safe.... and B) is there a solid state alternative that is economically viable....
    Last edited: Jan 24, 2010
  2. jcsd
  3. Jan 24, 2010 #2
    Why use an electromechanical system to detect current levels? Quite apart from the physical difficulty of modifying ammeters etc., the response would be very slow and inaccurate.

    Think of the sensitivity to vibration and movement, and the difficulty of obtaining a reliable consistent contact given the tiny forces available from a meter pointer. This sort of solution would have been necessary before the advent of electronics, but not nowadays.

    Also, note that in an AC application TRIACS will drop out of conduction every half-cycle, whenever the current goes below the holding current. Don't rely on a TRIAC holding current to define when a capacitor gets switched out in terms of RMS current.
  4. Jan 25, 2010 #3
    Well, thanks for the input, i guess you really just summarized why i posted in the first place. I wanted to know if there is a solid state method of doing what i want to do. To be able to switch capacitors in and out of a circuit at preset current levels.

    Also, correct me if im wrong, but are TRIAC's not capable of conducting in either direction? and will remain open in an AC circuit?
  5. Jan 25, 2010 #4
    Yes, of course TRIACS can conduct in both directions, that's what they are made for, as opposed to unidirectional devices like SCRs.

    Perhaps I was misled by something you said in your earlier post:
    but it seemed as if you wanted the TRIACs to act as latching devices in AC circuits selected for holding currents so as to define the current at which the capacitors are disconnected.

    Having dealt with that point, do I understand correctly that you plan to switch capacitors in to a live AC circuit? If so, have you made arrangements to minimise (or at least manage) the inrush currents? You have not said what voltages or capacitor values would be involved, but suddenly connecting a capacitor at any old phase angle could result in huge current transients, probably damaging TRIACs etc.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook