Spurious triggering of 4013 flip-flop

True, but yours requires a ground free switch contact as trigger instead of a ground referenced voltage, needs both to be multi-pole relays, and needs a resistor to tweak the pull-in/release voltages.

OP says he hasn't room in the box for more relays so I guess neither circuit will get used. As a hobby project it is all fun so who cares :)

piglet

Thanks for those. As mentioned in the last of my posts to Tom (10:55 today) I did originally try a mono. maybe I didn't filter it hard enough.

I reckon the favourite will be as in that discussion with Tom, namely to use the flip-flop already available at the lounge end. But first I'll see what improvements can be made with some of the other suggestions, starting with reconsidering my earthing. It's all a great learning exercise ;-)

Terry, East Grinstead, UK

Yes, I wonder if those monostables you didn't have luck with were edge triggered (triggers the clock input of a F/F) so you were essentially getting the same spike sensitivity you have been seeing that prompted the original question. Using a level sensitive input like Set or Reset preceded by heavy RC filtering ought to be much better.

piglet

This NG tends to have 19th century politics so why not 19th century circuits? Hiyoooooooo

If anyone actually builds these though they have to sign a contract to wear a stove pipe hat during construction and say "Zounds! Forsooth whatever is the matter" if it doesn't work properly

Pin 11 was at 10mV with no signal.

Investigating possible ground loop(s) is further handicapped because my scope probes are themselves earthed.

Yeah, we all know that Karl Marx was a 19th century economist but his theories had been well debunked by the end of the 20th, by anyone with an ounce of sanity. Unfortunately, Democrats don't have an ounce of sanity.

A rather neat way of solving mechanical switch contact bounce is to use a single pole break before make changeover switch (or even a centre off biassed c/o switch) with a non inverting buffer (a pair of not gates cascaded will do the job, nor nand or schmit triggered inverters or whatever is handy for the task).

You need to wire the side contacts to the Vss and Vdd supply level voltages and connect the common pole contact to both the input and output of your non inverting buffer which drives the relay switching transistor.

The short circuit on the buffer's output won't cause any harm during the brief 10 to 20 ns transition to the input logic state as it latches itself to match the switch state (or the last state if using a centre off biassed switch), neatly eliminating the effect of switch contact bounce.

If you're using seperate PSUs at each end, make sure to use the lower voltage one for the switch connections and definitely only have a single point of connection to your local ground reference.

If you're relying on a remote receiver to send a toggling pulse, you can relocate the existing 4013 bistable to the remote Rx unit, add a 74HC14 with all six inverters paralleled up as a stand in for the switch and use two of the inverters in another 74HC14 now taking the place of the bistable latch to debounce and self latch the on/off signal now being issued from the remote controller end of your circuit.

However, I suspect all you really need in this case is simply to relocate the the bistable to the remote controller Rx and forget the debounce self latching non inverter buffer idea altogether. The low output impedance of the bistable chip will provide considerably more immunity against interference on that 60 feet of control line (but only if there's just a single connection point to ground in your circuit).

add 100K between Q and S for extra crispness.

I almost overlooked that fact in my previous reply. It seems to me to be a case of a 'self inflicted problem'. The mistakes being the use of two separated earth connections and bad placement of the flip-flop which should have been co-located to the remote controller Rx unit.

The main benefit of such hobby projects lies in the 'learning from your mistakes' aspect. Provided the mistakes aren't expensive or fatal, it can be a fun way to learn.

Generally speaking with such projects, you don't learn anything without making a mistake or two. At best, if it works perfectly first time round you're either simply proving your understanding or else just benefiting from 'pure dumb luck'.

A 'real engineer' won't need to sign such a contract since such utterances are an innate response regardless of the wearing of a stove pipe hat (which is entirely optional headgear anyway). :-)

Added crunch always good

piglet

UK mains earthing schemes are not completely straightforward either, e.g.

You most likely have TN-C-S (PME as was) to your residence, which may also be required to be locally earthed to metallic water and gas supplies if it is recently modified and updated to be in compliance.

Depending on how far your shed is from your consumer unit, it may be expected to be earthed using a TT scheme (live and neutral only, with local earth rod/mat nearby) to be compliant. This improves safety local to the shed in case of faults but may introduce potential difference between locations which you have connected using other non-power cabling. Physically isolating them is best, as already suggested.

I'm going to bet you've got a bit of twin-and-earth cable strung up on poles or buried like it's 1959 though.

North America uses a TT-like scheme generally because power distribution is often medium voltage (12.5kV/7200Y) to the street with pole transformers for the 120V local loop for domestic consumers, so no substations to provide a protective conductor with the live(/line) and neutral.