Your logic escapes me, but let's not worry about that right now. I'm a pragmatist. I mostly care about getting it working.

You can try that relay in that circuit as is, try reducing the capacitors to 68 uf, try a 9 volt battery because it has twice the voltage as the relay rating, and add a 680 uf capacitor from +9V to ground to supply the current surge that a 9V battery can't do very well...in that order.

 

Uggh scratching head.

I thought I had it. 4.5v times 22.2ma equals 100 milliwats.

I think I'm getting mixed up with the milliseconds and the capacitors.

I'll search more for a 3volt 100millisecond latched relay.

 

True. 4.5 volts times 22.2 ma = 100 mw.
That means the same thing as: the voltage is higher than 3 volts and the current is lower than 33 ma. The current being lower, the capacitors don't need to be so large. They will probably work just fine at 100 uf but they will be more economical of power if you adjust them to 68 uf.

The voltage rating of the relay being higher would seem to require a higher driving voltage. You started with a 9V battery so I suggested you continue with it because it is twice the voltage of a 4.5 volt relay coil as in the design by Alec t.

9 volt batteries are not very good at supplying current surges, so adding a capacitor across the battery will move some of the energy from the battery to the capacitor where it can be had very quickly. This has zero effect on idle current or total power used up.

 

Ok thanks. I'll try those capacitors.

I actually plan on using 4 AA batteries. I was only using a 9 volt battery in my proof of concept circuit operating a motor at low current.

I will be using a metal geared motor using more current.

In test circuit the 9volt battery was switching on 12 volt coil relays. I learned you use a 9v battery to test if your 12v automotive relays are working.

 

If 9 volts is not available, try the 4.5V relay in the circuit just as Alec t designed it. It might work in spite of the fact that you are not delivering twice the voltage that the relay was designed for.

 

What does the 100m mean on the relay?

It means 100 milliHenries; a guestimated value for the inductance of a 90 Ohm dummy relay coil I used in my LTspice simulation. Sorry about any confusion. Coincidentally, 3V latching relays with a 100mW coil energy rating (and a 90 Ohm coil resistance) are available and would be my choice. These have a typical switching time of ~3mS. Depending on the load you want to drive, you could use the latching relay as a 'pilot' to switch a more powerful heavy-current relay.

Since it is 6v, couldn't I use a 6v or 4.5v or 3v relay if I wanted? What is the function of the capacitors on the relay?

The circuit works by giving the relay coil a ~10mS pulse of current of appropriate polarity to set/reset it, drawn from either of the two caps. There is some voltage drop in the transistors, so a 6V relay might not switch state unless the caps were much higher values. 4.5V (e.g the one you linked to) or 3V rated relay coils should be fine. The peak voltage across the coil will be nearly 6V, but for only a couple of mS. If you were concerned about a 3V coil being over-driven (albeit for 2mS, which would be extremely unlikely to cause significant coil heating) you could connect a 100 Ohm resistor in series with the coil.

Edit:
The CD40106 is static-sensitive: take due precautions. The IC contains 6 gates. The inputs of the 5 unused gates should be grounded. A o.1uF ceramic cap should be connected directly between pins 7 and 14 of the IC.

 

Thanks Alec_t.

I'm having a hard time finding that 3v relay. Could you point me in the right direction?

Since the IC is static sensitive, how robust would it be in an enclosed outside environment? Once you get it in the circuit what would be the chances of it frying?

Thanks for your help.

 

I'm having a hard time finding that 3v relay. Could you point me in the right direction?

I can't suggest any local supplier since you don't say where you are. Try any of the main online distributors (Mouser, Digikey, Farnell, RadioSpares ....).
The IC will be fine (discounting a lightning strike!) once it's in circuit and enclosed (preferably in a screened box). Is this project for a chicken coop door?

 

I'm in USA. I found a 3v one from Farnell but is only for 2amps so I'm probably going with the 4.5 volt one from DigiKey that is 5 amps.

Screened box?

Yes, how did you know it was for a CD?.

I looked at the other threads and nothing seemed straightforward and simple for such a simple action in my opinion. I didn't want to use manual switches, complicated circuitry, AC voltage, or timers.

It was easy to use the photocell and dpdt relay to get the motor to reverse directions, but nothing latched. So then I looked into a latched relay, but needed to get it to switch directions so then I looked at 555 Schmitt triggers and they seemed to fit the bill except for the idle current draw, which I never could figure out.

So do you think it will work for this application?

 

Anything that can be sensed can be amplified. The only difference is in how many stages from input to output. Right now, you are set up to sense "daytime" and "not daytime". I don't care whether you want it to run an LED or a forklift. Just be clear about your goal and the method can be had.

 

Yes that is what I mentioned earlier - dark or light, day or night. I know there would be degrees of "light" in the day and at night.

I'm curious if this can be set for especially dark cloudy days and bright full moon on snow conditions at night. One setting that would work for both. Maybe I'm dreaming.

 

Look at post #14. The adjuster is labeled, "trim".

 

Yes, I know.

I'm just hoping dark cloudy day doesn't equal bright moonlit night
Still the degrees of "darkness" of those 2 are probably extreme.

I'm thinking closing it much more important at night than opening on a cloudy day. So if the door is open on a bright night, then adjust the pot until it closes.

 

Cloudy is usually a lot brighter than moonlight.

 

I'm keeping my fingers crossed! I'll throw a manual override switch in there.

 

Alec,
How did you calculate the capacitor values given the relay information?
Thanks,
Matt

 

How did you calculate the capacitor values

I looked at online specs for latching relays, found one with a 3ms switching time (which seems typical for small relays) and 90 Ohm coil, then ran a Spice simulation with different cap values and a couple of different guestimated inductance values until I got a charge/discharge pulse remaining above 3V for a time comfortably greater than 3ms.
If the relay you buy has longer switching times you may need to increase the cap values (say, by a factor of 2).

 

So for 10ms switching time would you go with 200u as opposed to 100u or would you go three times since 10ms is about 3 times 3ms?

Is this something that just requires some experimentation?

 

Is this something that just requires some experimentation?

Yes. If you know there's a 10mS switch time then I'd start with 220u and increase if necessary for reliable switching even with partially depleted batteries.
Battery standby current is mainly due to the light sensor and its trim resistor to ground. A couple of relay flip ~30mA pulses per day will make virtually no difference to the average current draw of just a few tens of uA.

 

Great thanks Alec and to #12 and everyone else who contributed on this thread.

It is exciting to learn more even if it comes veryyy slowly for me