Maker Pro
Maker Pro

Missing feature - voltage level monitor

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The problem with a capacitor across Q1 is that it will get shorted by Q1, and that's bad for Q1.

You could try placing a capacitor between Q2's base and emitter, with a schottky diode reverse biased across R2. This will prevent Q2 turning on immediately the power is applied (giving an initial pulse), but allow it to turn off reasonably quickly (assuming there is a low impedance across the power supply when it's off.

The issue here s that while you will get a pulse on power on and power off, there is no simple way to ensure that your relay stays correctly in sync (i.e. what if it gets into the wrong state?)
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
Yes I can see that might be a problem. Maybe not a simple way to prevent that, but certainly something to think about :)

What would the schottky diode do?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The schottky diode would allow the capacitor between the base and emitter to discharge quickly. Note that this assumes the input has a low impedance when the power is removed. A schottky diode is used to discharge the capacitor below 0.6V.
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
Would it be safe to connect my digital multimeter to the PWM's load output, allow the output to drop, then switch the multimeter to it's resistance ranges? To try and check the impedance?

edit.. hmm, just looked at the PWM's load output on the multimeter's voltage range, toggling the output using the manual load on / load off button. When the LCD displays 'load on', I'm getting 12.5V & the output will light a car indicator bulb brightly. When the LCD displays 'load off', the voltage drops to about 10V, and further to about 0.3V when the bulb is connected. The bulb will not light. On disconnecting the bulb, the voltage then immediately recovers to about 10V if the load terminals are left open circuit apart from the multimeter.

Makes me think the PWM's load output doesn't go to low impedance when switched off?
 
Last edited:

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
If I used a dual-coil latching relay like this;

Schneider Electric / Magnecraft 755XBXCD-12D
http://datasheet.octopart.com/755XBXCD-12D-Magnecraft-datasheet-8469450.pdf

Would a circuit like the attached, do the trick do you think?

Hopefully that kind of setup would remove any ambiguity about which state the relay was supposed to be in at any given time..

I included the resistor across the whole supply rail to pull down the PWM's load output voltage when it's allegedly 'off'. I don't know why it seems to want to float at ten-odd volts, but has virtually no oomph to keep it there under the slightest of loads. That resistor will have a very high value to keep the wasted milliamps down, and to try not to discharge C1 too much above what it needs to supply sufficient operation

I'm not at all experienced at designing circuits as you've probably guessed, so keep up the constructive criticism :)
 

Attachments

  • dual-coil-design.jpg
    dual-coil-design.jpg
    57.1 KB · Views: 124
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
With that circuit the relay "L1" will never get activated.

Note that there is a diode in my circuit that isolates the capacitor from the incoming supply rail. This is required to separate the sensing of loss of voltage from the voltage stored across the capacitor which will be dumped into the relay.
 

flippineck

Sep 8, 2013
358
Joined
Sep 8, 2013
Messages
358
Thanks Steve, I missed that diode, aye. High time I ordered a new breadboard and a stock of parts to experiment with..

Anyone suggest transistor part numbers that would be likely fits for the circuits discussed? I'll order a good selection of resistors and capacitors anyway, but the range of transistors out there is huge I could do with ideas there.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
A lot depends on the load you're switching. Test it first with LEDs (and a resistor) as the load, and use any small signal transistor (2N2222, BC548, etc).

For the real thing, you'll need to know the current required (you may still be able to use the same transistors, but check) but if they're relays you will need a catch diode across each load.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,760
Joined
Nov 17, 2011
Messages
13,760
Have a look at this thread.
A unipolar pulse (on-off) is used to generate either a positive or a negative edge which in turn can be used to control a single coil latching relay to switch between on and off positions.
 
Top