Solid state equivalent of Low voltage switch for Hen house

[NivagSwerdna]

This might be an incredibly dumb question....

I have a controller that operates the door to the hen house, the controller itself is "VSB Electronic doorkeeper" https://gb.axt-electronic.org/uk_shop/...

... now the door opener itself is powered by 4 AA batteries i.e. around 6V

... and the controller has two pins.. pin 3 and pin 4

Pin 3 and Pin 4 go to a timer which is a 24hr clock where some segments are ON and others are OFF.  The timer either is open circuit or close circuit for the corresponding ON or OFF state.

I would like to create a gadget that goes between the timer and the controller such that I can have a manual OPEN and CLOSE button.  OPEN would make Pin 3 & 4 open circuit until the timer next changes state and then it would follow the timer, CLOSE would make Pin 3 and 4 connected until the timer changes state and then it would follow the timer.  I would power the gadget from the controller 6V.

Now... So Pin 3 and Pin 4 are open or closed... so there must be a voltage across these two pins and the current that flows determines whether the mode is OPEN or CLOSE.

How can I add a solid state circuit to Pin 3 and Pin 4 such that polarity isn't important and such that I don't insert any un-necessary voltage bias?  (I don't want to use a mechanical relay due to the power requirement.)

I read a bit about Triacs but these seem more suited to controlling big AC loads.  If I use a MOSFET then polarity will be significant?

Any guidance appreciated

Regards
Nivag

[Seekonk]

I don't see why it has to be polarity insensitive.

[NivagSwerdna]

I want it to work without knowledge of the circuit it is being attached to.

[tpowell1830]

I looked at the information on the website and it did not provide any technical information on how the circuit operated or any schematic diagram.

If you could provide a schematic diagram of the circuit that closes/opens the door, perhaps someone could help.

[mwalker]

Replace the timer with one that has a manual override, say:

http://www.ebay.com.au/itm/171033344357

[NivagSwerdna]

I don't have the schematics but I took pictures of the PCB and it appears pretty simple.... two CD4093BM quad 2-input schmitt NANDs form the basis of the circuit.

The pins where the timer attaches are either side of the unpopulated E3 reed switch.

I am not familiar with opto isolators but perhaps a 4N25 and (as mentioned by a previous poster) try and get the polarity correct?

[NivagSwerdna]

Power from the battery pack (4xAA) is applied to the pins on the left hand side.  One of the two bottom two pins (and then routes via D2 or D5) so that is +ve and then the other side to one of the two top pins.

[Kleinstein]

One should be able to measure polarity and voltage levels. I won't expect the polarity to change. It is more like two digital signal going between 0 and 6 V. It could also be high impedance in one case.
If you want manual control one has to measure the desired signal levels anyway.

The circuit does no look really complicated. It is still within reach for reverse engineering. At least get an idea where the signal at pins 3 and 4 are connected to, until they get to an IC or stage one knows how it works.

[NivagSwerdna]

The circuit does no look really complicated. It is still within reach for reverse engineering. At least get an idea where the signal at pins 3 and 4 are connected to, until they get to an IC or stage one knows how it works.

I think it looks like T2 is a position for a transistor that would effectively close the door. (The actual PCB is covered in clear resin, presumably for water proofing so the pads aren't directly accessible.
In fact it looks like the unpopulated area is for more closing options controlled magnetically via the E3 reed switch.

I can't quite work it out but I know there is a variant of this device (intended for ducks!) that can be closed with a magnet.  It looks like E3 is to detect the magnet and T1 and T2 form some kind of latch to keep the door closed until the next open (i.e. when pin 3 and 4 are no longer connected by the timer). 

I tested continuity and Pin 4 is the same as the -ve side of the battery... i.e. ground.

[NivagSwerdna]

I think I have decided to use a latching relay... so that for the 99.99% of the time the manual override is not required it will consume zero power...