state change pulse generator

Hey smarties, I have a simple one for you. Short problem statement:

I have a control signal that goes high or low. It must first set two outputs to the opposite of eachother, IE
IN=1 OUTA=1 OUTB=0
IN=0 OUTA=0 OUTB=1

it also must set a third output HIGH but only for a brief pulse right after state change and then sit and wait for the next change of state.

LONG VERSION
I have a a RC reciever and need it to turn a latching relay on and off. I have a full bridge motor controller to reverse polarity on the relay. I have a pololu RC to digital converter, but need the make the three play nice together. the latching relay draws a huge current, 3 amps, so it can't sit powerd. the H bridge requires two inputs to set polarity and has a third enable input. I am trying ot keep parts count down as much as possible so I am avoiding 555 circuits, but if no one has a better idea I'll take it.

ATtiny10 or other small micro?

Aiming to avoid putting another uC into this project, and I prefere the reliablilty of discreeete components with very high ESD resistance. This circuit I am describing is predominantly used as an emergency kill for a pretty dang big robot.

IN=1 OUTA=1 OUTB=0
IN=0 OUTA=0 OUTB=1

Inverter, transistor?

third output HIGH but only for a brief pulse right after state change

XOR gate with an RC delay on one input?


Rob

A Gate package, some thinking and the use of a half monostable. A 14 pin package, 1 High intensity 5mm led, I have some I bought on Ebay that are 20 candela @ 30mA... so they will produce a lot of light @ 5 to 15 mA which is available from 2 paralleled gate packages all could be done with one 6 gate Schmidt trigger gate package ex. CD40106 2 1 1 to 4M7 ohms and one 100R ohms resistors and a small capacitor 100 to 470 nF. and it will work from 3 to 18V dc. I've used it in many variants for designs I've done over the years where I needed a small timer W/O great accuracy...

Doc
From a circuit I remember from lancasters Cmos Cookbook... Lo, these many years gone.

Any discrete transistor / logic circuit will probably be bigger and have more components than a 555 monostable. There may be other timer circuits out there, but I don't know about them. I've scribbled a bit with inverters and NOR/NAND logic and I think there's a good solution there, but you still end up with 1 or 2 IC's and at least 1 RC, not any simpler than a 555.

You're right sdturner, the smallest I think you could get is 2x SOT23 packs (inverter and XOR gate) and 1xR 1xC. If you don't want to use SMDs then of course your logic becomes 2x14pin DIPs which is a huge waste.

I haven't used a 555 for about 20 years so I'm a bit rusty on them, but I don't think a single 555 will give you the inversion and the pulse, and can it trigger on both edges?

The smallest way I can think of is a 6-pin processor, the next is an 8-pin processor (ATtiny85) which is well supported by various tools. I think that's a reasonable way to go.


Rob

There are circuits for controlling servo's in abundance... Mainly because that's all there was... before Microprocessors.
Both 555's and 556's and all dating back to about 1960 to ... 2000?
There is in abundance gate logic stuff too, although other than a lesson in theory, I can't for the life of me think of a reason why Anyone would want to duplicate them.
A PIC10FXXX would be Perfect too for this task.

Doc