Does anyone know of a latching reed switch part number for electronic circuit? Intention is to close the switch with a short pulse generated by vibration sensor to connect the battery to the ckt, but remain closed until a clear/reset button is pushed.

 

Hi shawn
Welcome to AAC.
Have you considered a latching relay.?
E

 

Hello,

You can use a contact from the relays to latch.
See this page for more info:
https://www.electrical4u.com/latching-relay/

Bertus

 

As you can see, no one is addressing directly a "latching reed switch". This is because we are not familiar with the term. Do you mean relay, or reed relay?

ak

 

Thanks, guys for your reply and correction. To clarify, yes what I am hoping to find is a mini relay that connects the battery to a ckt upon detecting vibration. It is part of an IoT circuit board of approx 1" sq, runs on 3V, which includes a vibration sensor, a 555 timer, led, and buzzer. They need to have less than 10 ma current draw to save battery. Once a vibration sensor, such as SW-18020P, detects any movement it wakes up the circuit and keeps LED flashing until is manually reset. I wished there was a latching vibration sensor, but those I know only give milliseconds pulses (just a rod dangling inside a spring), hence the need for closing the relay to maintain the connection to the 3V coin battery.

 

Certainly a multipole reed relay can be wired as a latching device. There can even be multiple coils to make the circuit simpler. It may not be a commercial product, but relays using reed switches are simple to fabricate.

 

Does anyone know of a latching reed switch part number for electronic circuit?

Rather than standard reed relays, which I have never come across a latching one.
There is the SS400 devices by Honeywell, these are magnetically operated devices also, they include momentary, latching and bipolar versions.
Maybe something you could use in this application?

 

Certainly a multipole reed relay can be wired as a latching device. There can even be multiple coils to make the circuit simpler. It may not be a commercial product, but relays using reed switches are simple to fabricate.

Hello,

You can use a contact from the relays to latch.
See this page for more info:
https://www.electrical4u.com/latching-relay/

Bertus

Bertus,
Thank you. This diagram that you sent in the link above, using buttons A and B, should work and is totally doable on my small circuit board (1.5" square). By any chance, do you have a part number in mind for such a small relay that can switch with 3V coin battery?
I also wonder how much current the coil draws while sitting in the closed state (which could be for days), and if I should put a resistor before the coil to limit the current.
Thanks

 

You talk about the relay contacts supplying power to the circuit. If this is true, then you need only a SPST relay because when the relay latches power to itself, it also is latching power to the load. Because the relay coil and the load are in parallel, you do not need a second contact set for the latch function. The circuit would be like the last one at the bottom of the page, with your circuit connected between A1 and A2.

A relay coil draws current, maybe enough to matter. If the above statements are correct, then the relay can be replaced by a 2-transistor latch circuit with a much lower holding current.

1 - NPN transistor
1 - PNP transistor
2 - resistors

Depending on the current draw of the alarm circuit, the holding current for the latch could be less than 1 mA.

ak

 

Here's my take on your project needs:

V+ (whatever voltage you're using) supplies power to both U1 and VS1. U1B holds one leg of the 2 IN NOR gate (U1A) to ground, keeping it from oscillating. R1 holds the other lead of U1A to ground. U1A output remains high. U1B inputs are held high which keeps U1B output low. When vibration is sensed (VS1) supplies a momentary high to one leg of U1A, triggering U1A output to go low which in turn drives U1B output high. This latches U1A output low and U1B output high. U1C & D are buffer inverters. You can use just the output from U1C or U1D, depending on what input you want to your 555 circuitry.

S1 resets the input at U1A to ground, and as long as there is no vibration sensed, U1A output remains stable high until the next vibration event occurs. 10KΩ resistance should be sufficient to hold U1A at ground until triggered. U1B provides the latching while S1 provides the reset command. Some small SMD components should make for a small package. The only external parts would be the power supply and the reset switch, which is obviously a momentary push button.

 

Q and NOT-Q are always in opposite conditions. When one is high the other is low.

ANY gates not used should have their inputs held either high or low. Never should be allowed to float because they will oscillate and cause all kinds of problems.

 

Here's an SMD CMOS chip for a buck on eBay. They want $4 for shipping - quite expensive if you ask me, but you could probably find one far cheaper elsewhere.
https://www.ebay.com/itm/383551906679

 

You talk about the relay contacts supplying power to the circuit. If this is true, then you need only a SPST relay because when the relay latches power to itself, it also is latching power to the load. Because the relay coil and the load are in parallel, you do not need a second contact set for the latch function. The circuit would be like the last one at the bottom of the page, with your circuit connected between A1 and A2.

A relay coil draws current, maybe enough to matter. If the above statements are correct, then the relay can be replaced by a 2-transistor latch circuit with a much lower holding current.

1 - NPN transistor
1 - PNP transistor
2 - resistors

Depending on the current draw of the alarm circuit, the holding current for the latch could be less than 1 mA.

ak

Are you referring to the last circuit in the link? What is the circuit if replaced with two transistors? Bcs having below 1 ma holding current is very important. Thanks.

 

Thanks to everyone for all the options and circuit solutions. I believe I found my best option.
Such a great forum.

 

I believe I found my best option.

So please tell us what you decided.

 

I will be using the 2-transistor NPN and PNP latch circuit. I think it should draw much less holding current than a coil and should be a cheaper ckt too. I also found the ckt diagram and will give it a try.
Thanks

 

Here is an alternate approach. This has an extremely low holding current, basically the leakage current of three ceramic capacitors. The CD4093 is a quad NAND gate with Schmitt Trigger inputs. This lets a single gate function as a gated oscillator, and eliminates the 555.

U1A and U1B form a latch that remembers when the circuit is triggered momentarily.

U1C plus R1 and C1 form an oscillator that replaces the 555 in your description. The values shown are for approximately 1 Hz.

U1D corrects the off-state logic polarity for driving the output stage.

C2 adds a power-on-reset (POR). As shown, the circuit powers up in the off (untriggered) state. If you want the circuit to power up in the on state as a circuit test, move C2 to U1A pin 1.

ak

 

Does anyone know of a latching reed switch part number for electronic circuit? Intention is to close the switch with a short pulse generated by vibration sensor to connect the battery to the ckt, but remain closed until a clear/reset button is pushed.

Simple mount small magnet at certain distance from reed switch. That's it.
Of course reed switch must have coil on it for reset pulse.

 

The Honeywell SS481 will latch on one magnetic pole and stay latched until unlatched by opposite pole..
Simple and compact.

 

Can't use a magnet in this application. We have a specific ckt board, and the latch is for connecting 3V battery power to the circuit and keeping it on in response to multiple events, one of which is a vibration sensor. The vibration sensor, such as SW-18010 is just a spring with a metal pole dangling inside. It produces multiple pulses and it doesn't give enough time for the circuit to turn on. Hence the need for latching.

I have not tried it, but planning to come up with a 2 transistor npn and pnp ckt? Does anyone see a problem with that for latching on the power?