Old Flip Flop circuit

[@rt]

Hi

I went to make another of these, and it occurred to me that it’s more than a flip-flop.
The bottom two transistors don’t even have to be there, and I have built the circuit without them.

I’ve simply pulled the bases of the 2N3904 low with the switches.
Does anyone know why they are there? The capacitor coupling would prevent the 2N3904 being left on,
and maybe being burned out, but the switches could be done the same straight to the bases via series capacitors.
Cheers.

[@rt]

Messy drawing I know (modified original), but the one I just did looks more like this:

[sahko123]

whats the point if the iron core inductors? filtering?

[@rt]

It’s non volatile if square loop ferrite cores are used, but you could ignore them. If they are omitted, it’s still a flip-flop.

[sahko123]

may i ask their purpose?

also i think the transistors are there to increase switching bandwidth by rapidly drawing on the current of the alternate side(just a wild guess).

[@rt]

The purpose of what? The cores? If they are present, the flip-flop remembers it’s state when powered off.
If they are not used, the emitters would just be tied straight to ground.

[sahko123]

sorry i meant the purpose of the flip flops in general like what is it in(appliance/circuit wise) or does it just switch those lights for fun

now that i look closer i think those pnps are there to make sure that the outputs arent on at the same time or to make sure that the switchover time between the two outputs is shorter

[@rt]

The purpose for me is to make a non-volatile shift register, for which I suppose I have to add two more transistors to make an AND gate for each flip-flop
(placed in front of one of the inputs) to produce a clock input for each flip-flop.
The Clear signal might as well be connected to all of the J inputs, if I’ve thought it out correctly.

[Andy Watson]

Just a guess: The J-K normally toggles if both inputs are operated simultaneously. Wth the switches wired directly to the base it becomes a bit random as to which state the flip-flip toggles into. Look at the action of "steering diodes" for toggling the plain version of this flip-flop. My guess is that the extra transistors are there to perform the same action as the steering diodes but with tweaked time-constants to allow for the flipping of the feritte cores.

Or perhaps it's just a fancy button-debounce.

[@rt]

What I have a the moment is the modified version in the second post. If both inputs are pulled low, then both inputs are high.
So we could call it a negative edge triggered SR flip-flop from now on.

I don’t think it makes a difference for the shift register, at least so long as I can accommodate it by ensuring no other inputs are active while it’s cleared.

[@rt]

Any guesses as to why there are two cores in the original circuit, when it works fine with just one?
I just used two windings in opposite directions.

[@rt]

One more board to go, and it ends up with the same transistor count as if I’d just made D type flip-flops
... but then I’ll use a decade counter anyway to cycle the switching from the carry end to the input end so as not to consume the current required to switch 8 cores close together (in duration).