How to have "something" happen at a certain voltage in a decreasing voltage situation

Discussion in 'General Electronics Chat' started by GCR, Mar 30, 2015.

  1. GCR

    Thread Starter New Member

    Mar 30, 2015
    6
    0
    Hello! New kid here. I would like to have a circuit that would allow something to turn off or on (ex low power 24 vdc relay) when the voltage across a thermistor decreases to 1.5vdc from 3.5v. Could some of you smart people here draw a simple circuit that would do this? Not sure how to pick a transitor that will turn on when the base voltage decreases to 1.5v.
    Thank you in advance.
     
  2. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    TL431 used in voltage comparator mode. See para 9.4.1 of the data sheet.

    How much coil current does your relay require? The TL431 might be able to switch it directly.

    Show me the circuit of how the thermistor is used in the circuit that makes it produce 3.5V decreasing to 1.5V.
     
    GCR likes this.
  3. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,986
    745
    how are you using the thermistor , as a voltage divider across a supply, can you provide more info, you could use an op amp as a comparator?
     
  4. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    You need a comparator with hysteresis. Followed by an edge-triggered device (a flip-flop or monostable) that only triggers on the negative edge.
    So that it won't trigger until the voltage is falling. It would probably need a time-out (monostable) or a reset signal (flip-flop) so that it can be ready for the next falling edge event.
     
    GCR likes this.
  5. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    Here is simple circuit that should work: It has hysteresis, snap-action, reasonable temperature stability, very high input impedance, reasonable power supply rejection ratio. No exotic components. Resistor values should be 1%; if not, some tweaking may be necessary.

    42.gif
     
    cmartinez and GCR like this.
  6. GCR

    Thread Starter New Member

    Mar 30, 2015
    6
    0
    Thankx for all the replies - very much appreciated! The load is 150mw to 200mw depending if a 5v or 24v relay is used. A resistor is placed across a thermistor to lower the total resistance and fool the machine into a ready condition. Once ready, a seperate timer isolates the thermistor, and the machine then continues to normal temperature. Why do that? Because the "time to ready" time window is too narrow, and the machine sometimes faults out. The time is not adjustable (at least by me). Once in a "fooled" ready state, the fault will not occur, and the remaining warm up can continue for about 5 seconds at most. The 1.5v number was derived by monitoring the thermistor voltage/temperature. 1.8k across the thermistor does the trick.
    Thankx again for your help. The replies came faster than anticipated!
    Glenn
     
  7. GCR

    Thread Starter New Member

    Mar 30, 2015
    6
    0

    Thankx Mikeml! Circuit works perfectly!
     
  8. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    As someone else mentioned the TL431, its internal Vref is 2.5V so the device will stop conducting if the sense voltage drops below that point - if you want the relay to energise at that point, you'll nedd to follow it with a transistor or MOSFET to invert it.

    You can use a resistor network to scale from your 1.5V to the 2.5V the TL431 needs to switch, or if you don't mind a bit of error, just shift it a bit with a diode Vf.
     
    GCR likes this.
  9. GCR

    Thread Starter New Member

    Mar 30, 2015
    6
    0
    Ian, Thankx for your reply. The circuit seems to work perfectly on the bench, in the machine not so well. I think because I am crossing the thermistor common with the machines 5vdc common using Mike's circuit (using a 5v relay instead of 24), and the machine does not like that. Is there a way to isolate the thermistor voltage from Mike's circuit?
     
  10. GCR

    Thread Starter New Member

    Mar 30, 2015
    6
    0
    Actually, I believe that it is a pulsing dc through the thermistor that is causing the issue -- relay vibrates on off. I can see it with a led I rigged across the thermistor.
    Back to the drawing board.
     
  11. tindel

    Active Member

    Sep 16, 2012
    568
    193
    If the relay is turning on and off, try lowering the resistance of R7 to increase the feedback current. A 10-turn 1Meg trim pot would probably work nicely there. Don't lower the resistance too much because that circuit can probably latch.
     
  12. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    A lot depends on what pulsating amounts to, a standard solution is to use a Schmitt trigger on the input, the main feature of this is hysteresis - the signal has to traverse a no action band from the on to the off switching condition.

    If the input is actual full blown pulses or large amplitude spikes, a Schmitt trigger will faithfully reproduce these with nice sharp edges. Whether or not you use the Schitt, it sounds like you may need a C/R LPF on the input to smooth out whatever arrives there.

    You already have part of the makings of a Schitt trigger, the TL431 is actually a comparator with a built in 2.5V reference - unfortunately there's no external access to the + input on the comparator to add the positive feedback for hysteresis. The next best thing is to invert the output and feed a small proprtion of that back to the minus input, you may already be inverting the output to drive the relay, but you may have to also introduce some negative feedback, otherwise the hysteresis will be difficult to control.

    If you do need a Schmitt, it might be better to add as a separate circuit block - unless you have a lot of time on your hands for trial & error.
     
  13. GCR

    Thread Starter New Member

    Mar 30, 2015
    6
    0
    Tindel & Ian, Thankx for your replys. Lowering R7 stopped the cycling but also stopped the circuit. Using a TL431 sounds a bit "above my pay grade" because I don't know how to do what you are suggesting. So I discovered a much simpler way to accomplish the same thing with a timer. I have searched for this particular timer and have not found it. The objective is to have a double pole relay turn on at power on, then denergize 5 sec later (with power still on). Any help will be greatly appreciated.
     
Loading...