A simple negative edge one-shot circuit.

[Blueteeth]

Hey,

I'm designing a xenon flash charger based on the linear technology app notes/datasheets. It's essentially a 300v flyback converter driven from a 555 and a mosfet, but with a few little mods to keep the efficiency above 70%. This isn't for work, just a sort of challenge to see if I could make a current-mode flyback circuit with a 555 and a few extra parts, pointless I know, but given the current economic climate, I don't have much work...

Anyways, there are many one-shot circuits using flipflops, and the 555 itself, but I can't seem to find one that triggers on the falling edge. That is not to say, when the input falls below a threshold, it should be purely a high-to-low transition, with a short output pulse.

I have been tinkering with various 'edge detectors' with diodes, caps and transistors, but most seem to require quite a sharp fall time of the input (the input isn't a button to ground.....its the fall of the flyback pulse on the primary of the transformer) or simply do not operate fast enough...my converter runs at 10Khz.

So, odd question I know, but any nice simple (transistors, passives, gates) components?

Blueteeth

 

[k7elp60]

Hey,

I'm designing a xenon flash charger based on the linear technology app notes/datasheets. It's essentially a 300v flyback converter driven from a 555 and a mosfet, but with a few little mods to keep the efficiency above 70%. This isn't for work, just a sort of challenge to see if I could make a current-mode flyback circuit with a 555 and a few extra parts, pointless I know, but given the current economic climate, I don't have much work...

Anyways, there are many one-shot circuits using flipflops, and the 555 itself, but I can't seem to find one that triggers on the falling edge. That is not to say, when the input falls below a threshold, it should be purely a high-to-low transition, with a short output pulse.

I have been tinkering with various 'edge detectors' with diodes, caps and transistors, but most seem to require quite a sharp fall time of the input (the input isn't a button to ground.....its the fall of the flyback pulse on the primary of the transformer) or simply do not operate fast enough...my converter runs at 10Khz.

So, odd question I know, but any nice simple (transistors, passives, gates) components?

Blueteeth

As far as I know the 555 configured as a one-shot triggers on the negative going trigger.

 

[Blueteeth]

As far as I know the 555 configured as a one-shot triggers on the negative going trigger.

Yep, its pretty much what they were deisgned for, and the first thing I thought of, but alas, the input to the 'trigger' (pin2) is the input to a comparator, wwhose threshold is 1/3rd VCC. The input to the circuit I need goes from 0V to VCC+0.6V, and so it would be 'active' on a rising edge too. Thats said there must be a way to inhibit this in some way. I shall post a diagram tomorrow too see if anyone can help with it.

 

[k7elp60]

Yep, its pretty much what they were deisgned for, and the first thing I thought of, but alas, the input to the 'trigger' (pin2) is the input to a comparator, wwhose threshold is 1/3rd VCC. The input to the circuit I need goes from 0V to VCC+0.6V, and so it would be 'active' on a rising edge too. Thats said there must be a way to inhibit this in some way. I shall post a diagram tomorrow too see if anyone can help with it.

How about using a pull up resistor about 10K from pin 2 to Vcc, then capacitive coupling the trigger pulse to pin 2.

 

[Mikebits]

Have you looked at the 74LS123?

 

[Dean Huster]

I agree. The 74LS123 is a wonderful one-shot, able to trigger on the positive or the negative edge. Be forewarned that the 74LS123 and the 74123 have different truth tables, so you can't necessarily swap out a '123 with an 'LS123, depending upon the circuit design. As I recall, it has to do with the fact that the reset line can become a clock input on one of the two.

I know that the NSC LS/S/TTL Logic databook published in 1989 contradicts itself on page 4-124 (74123) where in the descriptive detail, it says, "National's '123 device features a unique logic realization not implemented by other manufacturers. The "Clear" input will not trigger the device, a design tailored for applications where it shall only terminate or reduce a timing pulse." Then in the truth table on that same page, there's a positive-going step on the reset input and a note of "trigger" (pulse generated) in the response column. The wiring diagram shows the reset internally connected to a third input on the triggering AND gate -- I don't think that's true in the case of the '123. As this paragraph indicates, TI, Signetics, Mitsubishi, Motorola, etc. '123s don't necessarily do this.

Under the 'LS123 description on page 2-133, it states, "The clear input also serves as a trigger input when it is pulsed with a low level pulse transistion."

So, be careful of the difference between these two and the difference between chips of different manufacturing sources!

Dean

 

[Blueteeth]

Wow, thanks guys!

k7elp60: I had already tried the capacitive coupling...however, I tihnk I over engineered it the first time lol, adding diodes and what-not to control the spikes..tihnking about it, I really don't tihnk I need any of those. I shall try your idea as it really does keep the part count low (I'm looking for falling edge, and the 555's pin2 is also 'active low'). Thankyou for reitterating it, it made me think in more practical terms.

Mike and Dean,

I shall order some tomorrow! Although its a larger IC than the 555, it still could be a one-chip solution, which is always a plus when I'm planning to build a few of these little things. Essentially, HV cap chargers, with current limit and high efficiency for portable apps. I do have some CMOS monostables somewhere, CD4538, which appear to be similar (although not identical). Your replies reminded me I had them...


It's times like these I really wish I had a scope, its why analogue scares me.

Thanks again, I'll keep you up to date. Might also be a good simple building block for efficient low power flyback DC/DC converters, as adding voltage feedback shouldn't be too hard. And it will be way more eligant than most crude 555 drivers with no current limiting at all.

Blueteeth