Reed Switch Application

I am using a reed switch that is activated by the end vehicle of a train to send a pulse to a relay that then activates a number of point motors.  To stop a slow train from sitting on top of the reed switch, which cause relay chatter, I would like to delay the pulse to the relay by about 3 to 5 secs.  Would anyone have a circuit that a novice could build or suggest a source for a ready built one.  The I-rdot system does not appear to do what I need

Hi, welcome to the forum.

If the real problem is "relay chatter" as you put it, then it might be easier to stop the relay rapidly switching on and off, than adding a delay of several seconds.

You can do this by connecting an elecrolytic capacitor directly across the relay coil. When the reed switch closes, current will be fed to both the relay and the capacitor. The relay will operate and at the same time, the capacitor will charge up to the voltage of your power supply. (what are you using for the relay; 12v? or maybe 5v). When the reed switch briefly opens, the capacitor holds its charge, so the relay does not release.

The value of capacitor needed will depend on the length of time the switch is open and the current taken by the relay, but as I guess it doesn't matter how long the relay takes to release under normal operation, you could just go for a fairly high value capacitor. As it is electrolytic, it needs to be rated for the voltage of the power supply you are using.

A pack like this might be useful:- http://www.thesource.ca/estore/Product.aspx?language=en-CA&catalog=Online&category=Capacitors&product=2720802

By the way, you can always add more than one capacitor to increase the effect, but remember to only use ones that are rated higher than the supply voltage you are using for the relay.

Hi Geoff,
Relay chatter is one effect rather than the underlying issue.  Basically my system is automated so that the magnet on the last vehicle through the reed switch activates a 12V automobile relay that in turn activates a series of points (through a CD unit).  The movement of the points isolate that train and activates the next train in a sequence to leave the hidden sidings.  If a vehicle does not clear the reed switch and sits on top of it then in continuously tries to activate the point sequence and the entire system is effectively blocked.  For the most part  there is enough train momentum that the last vehicle simply over-runs the reed switch but in slower trains (I try to run at realistic speeds) there is not enough momentum.  Thus a potential solution is to delay activation of the pulse so that the loco runs a short distance before the pulse goes out to the relay.   From what I have read, one should be able to use a 555 timer chip but I know nothing about electronics so i need help to come up with a circuit.  I also think, from my reading again, that if i knew how to program certain logic chips I could also build a delay in the output.  Hope this does not make it more confusing.
By the way, I did try a capacitor system to try to delay the output pulse thinking that the capacitor would take time to charge before allowing the pulse to go out.  However, I found that the pulse to the relay took preference over the capacitor charging circuit, presumably on the basis that the electrons like to take the easy path.

Okay, I understand. Here is a simple circuit I have just knocked up:-


As you mention 555 timers, then perhaps you feel confident enough to solder up a circuit like this? The 556 is a dual 555. I could let you have a board diagram if you like. R1 and C1 determine the length of delay. I can let you have the calculation. Then R2 and C2 determine the length of time the relay stays on for.

Basically, the reed switch triggers the first timer within the 556 whose output on pin 5 immediately goes high. After the R1/C1 delay, its output drops low again to 0v. This sends a short pulse through the 0.1uF capacitor into the input of the second timer at pin 8.

The output of the second timer , pin 9, then goes high and provides power to the relay. (The output of the CMOS 556 can supply 200mA, so should be plenty for your relay I would think, but it may be worth checking) This output will stay high (at 12v) for the time set by R2/C2.

Nothing more will happen until the reed switch closes again.

Let me know if you wish to continue with this.

 

Geoff,
This is exactly what I am looking for.  Many Thanks. Once I can source all the parts (next trip to Ottawa in about 3 weeks) I will have no problem building your circuit.   And yes if you could send me the calculations for the resistances and capacitors.  I am also assuming that if i put in a variable resistor pot at R1 (and R2) I can actually change the time settings.  Would this be a correct assumption?
Thks again
Jim

Hi Jim,

Got your PM. Sorry I overlooked this. The timing is very straight forward. Basically the output delay for the first timer is the time taken for the capacitor to charge up to two thirds of the power supply. This roughly equates to the value of the capacitor (C) multiplied by the value of the resistor (R).

You mentioned making the resistor (R1) a variable potentiometer, which is fine, but you should include 10k in series with it so that the input to the timer is never connected directly to the power rail.

So to calculate the values for C1/R1 and C2/R2.

Start with the capacitors - note that electrolytics are usually only an approximate value, sometimes being as much as 20% different from their marked value. A 1uF capacitor (which is 10 to -6 Farads) with a 1M resistor (which is 10 to +6 Ohms) will give a delay of 1 second.

You want 5 seconds, so I would start with a 100uF capacitor and a 100k variable potentiometer in series with a 10k resistor. That should give you about 10 seconds delay at the top of the pot, 5 seconds delay in the middle, and 1 second delay at the bottom of the pot.

One thing to be aware of is that the timer will start as soon as the reed switch first closes. Any further closing of the reed during the delay period will be ignored. In other words the delay does NOT restart everytime the reed closes; the output of the timer stays high until the capacitor as charged. Only then will a new reed switch closure start things again. Also remember that the output of the second timer only goes high (and so providing power to the relay) when the output of the first timer falls back to 0V. It is the "edge" or transition from 12v to 0v which starts the timers.

If you want a 1 second pulse from the second timer, then I would stick with another 100uF capacitor and another 10k resistor.

It is easy to try different resistor values if you want a longer pulse.

Hope this helps

Cheers

Thanks Geoff