TQ2-L-5V Relay questions...

[Earthscum]

So I have been sitting on this relay for a couple months now, and still not sure about it's operation. I'm at work right now, or I'd just breadboard and experiment.

So, the datasheet says that there are 2 voltages to apply for the switching: >3.6V and <3.6V. Would that mean that to latch the switch to one side I could use 2.5V trigger, and I can keep applying that 2.5V and it won't flop over to the other side until I apply a 5V trigger signal?

Now, a question of current consumption. Say I switch this thing 50 times on/off (25 full cycles) in an hour, can I figure a current consumption based off of that? I'm looking to use the smallest amount of current possible for switching this thing.

I know it's gonna come up... "Why don't you just use a stomp switch?" Don't have the room. I gotta keep it flat as possible, and trying to keep power use down to as close to nothing as possible, so I'm weighing my options for switching. I am grabbing a DPDT stomp from my local store (like the SPDT pushbuttons from RS). Hopefully it will surprise me and fit, but if not I have about 1.125" square to use for switching.

[R.G.]

What the datasheet meant to say is that you have to put >75% of the rated coil voltage (which is 5V, so 75% is 3.75V) across the coil to make it flip. The direction of the voltage across the coil is what makes it go set or reset. Once it's stable on one side or the other, another pulse to put it where it already is won't change anything. Only a pulse for the other direction will change it. Once it's stable one side or the other, you can turn off all current to the coil and it will stay where it is. That's the beauty of latching relays.

So what you have to do to set the relay is to force pin 10 more than 3.75V positive with respect to pin 1. It will then go to the reset connection. You can turn the current off or leave it on, and it will stay reset. It will stay there until you make pin 1 more than 3.75V positive with respect to pin 10, at which time it will go to the set connection, and it will stay there until you force it the other way again.

It's the ability to turn the current off entirely and have it stay in position that makes latching relays power saving. So what you want to do to save current is to design a pulsing circuit to flip it the right way, then turn the current off entirely. That way the long term current use approaches zero. The datasheet says it will flip in 3mS or less on either a set or reset once it gets its full flipping voltage applied, so you could design a circuit that will apply current for, say, 5mS just to be sure, then turn the current off and the relay will stay where you left it until it gets further instructions.

On a more general note, you have to think about what you're averaging as to whether you can average over a day, year, second, or nanosecond. To electronics, an hour is almost an infinite time, as is a minute. To an iceberg, you'd have to average over a year or two. To a star, you'd have to average over a million years. If you're looking at battery consumption, yes, averaging over an hour is OK. But using 5mS pulses 50 times an hour approaches zero power. Leaving it on means the average is "always on".

[Govmnt_Lacky]

So, the datasheet says that there are 2 voltages to apply for the switching: >3.6V and <3.6V. Would that mean that to latch the switch to one side I could use 2.5V trigger, and I can keep applying that 2.5V and it won't flop over to the other side until I apply a 5V trigger signal?

Unless there is an absoltue maximum/minimum... it looks to me like you could just use a common positive voltage (5V, 9V, 12V, etc.) on one side and just plain old GND on the other. Ground (or V-) is definitely less than 3.6V right?

Your theory looks sound as far as operation. I am merely pointing out that, with the info you have provided, you could use the voltages I have listed.

Good Luck  

[R.G.]

Unless there is an absoltue maximum/minimum... it looks to me like you could just use a common positive voltage (5V, 9V, 12V, etc.) on one side and just plain old GND on the other. Ground (or V-) is definitely less than 3.6V right?

Datasheet says absolute max is 150% of rated voltage, or 7.5V.

Another trick that works is to use twice the voltage, like maybe 9V, tie one side of the relay to the middle through a capacitor, then pull the other end of the relay coil to either high or low. One way sets it and one resets it, and the capacitor is sized to give enough juice to flip it, but then charge up and shut off the current.

I personally prefer using double coil latching relays and pulling one side or the other instead of reversing the polarity on the coil. That's why my latching relay driver works that way. (see the latching relay driver at geofex.com)

[Earthscum]

so if I'm using a 12v battery I could feasibly use half supply to switch, which would let me run the battery down to probably about 8v before it would start to glitch on the switching. Is that correct (pending a circuit to do this, of course)?

Thanks a bunch for the info. The datasheet was kind of vague to me, being new to latching relays. I understand now that the polarity illustrated shows which way it will latch.

The power consumption gets me excited. The circuit I'm making will still power off a 100u 25v cap for about 20 seconds after the battery is disconnected. Then the led finally goes out and it just amplifies for another 40 secs or so. I was shooting for a minimum of 100 hours from a (a23?)  battery, which is rated at ~35AH. I think I surpassed that mark in my design goal.

Thanks again. Hope to have this done by the end of the weekend.

[cpm]

even with the half voltage trick, you need an active source (or sink) of current to drive the coil... that adds more external and not so trivial citcuit and logic.
i prefer, as RG says, a dual coil, i'd still use a pair of transistors to drive the coils, but its a simpler arrangement