Relay question

[Brent Hilpert]

On 2014-Nov-28, at 2:49 PM, Jim Brain wrote:
> On 11/28/2014 4:09 PM, Peter Coghlan wrote:
>> I have come across a relay like this controlling the door of a bus. One push to make switch, push once to open, push once to close. The bus schematics called it an "impulse relay". This is East of the Atlantic - who knows if they they call it the same thing where you are. The one I came across was 24V rather that 12V too. (There were no classic computer systems on the bus, unless you count the radio system which allegedly tracked the bus location and radioed it back to the controller. It counted pulses from the speedometer/odometer tachogenerator and did 4 bit BCD arithmatic on them using 4000 series CMOS.) Regards, Peter Coghlan. 
> I'll see if that set of words helps me.  Google was not all that useful last night.
> 
> I'd like to find something and get it here within the week, as they are coming back, and I'm hoping to test it out before I give it to him to install in the unit.

I do have a design that sounds like what you want using 3 standard relays: one D/3PDT & two 4PDT (or, one D/3PDT, one 3PDT, one 4PDT), or probably best to just call it three 4PDT relays.
A single NO pushbutton would control the relays, wired to produce a net divide-by-2 effect.
Contacts on one of the relays would be used for the ON/OFF, in series with DT contacts on one of the others to direct the UP/DOWN.

It's actually on-topic, as it's from a design exercise I did for the Simon relay computer.
The (whole) schematic is linked on this page:
	http://www.cs.ubc.ca/~hilpert/e/simon/imp.html

See the top-left corner of page 2 of the schematic, the Ø, Øa & Øb relays.
There's a timing diagram in the timing section on the web page, the Ø signal would correspond to the button pushes.

It might be possible to reduce the number of contacts if one stared at the circuit long enough.

It's been a few years since I worked on it, but I did prototype/test that part of the circuit and had it clocking around 18Hz.
Relay logic is fun.

Perhaps it's more complex than what you'd like to implement, but if it's of interest, or it's not clear, I could put together a more suitable diagram.

Perhaps the solenoid-triggerred 'mechanical flip-flop' relay Brian pointed to is closer to what you want. One thing to note about that type is when the system is turned off it can be left in - and hence will power-up in - either of the two states, you wouldn't know initially whether it was going to go up or down, unless you knew from the last time it was operated.