The 2 ways to make mark & space equal on a 555 astable circuit

Second version only works correctly with CMOS 555s it comes comes close-ish with regular 555s but comes out a bit low. If the supply voltage is fixed this can be corrected by tweaking with an additional resistor.

The way I remember it, the diode scheme came from Signetics, so that appeared first.

The one where the capacitor is charged and discharged via pin 3 and a single resistor, that was in "Engineer's Notebook" or "Designer's Casebook" (or maybe I have those jumbled) in "Electronics" magazine, just a circuit and a small description, sometime in the early seventies.

I've always used that circuit with the resistor from pin 3 since I saw that bit in "Electronics", it's just so much simpler unless you need something more complicated.

On the other hand, some have pointed out that the original 555 isn't perfect in this regard, the scheme works better with the CMOS 555. So that might factor in if you had some very specific need for exact 50% duty cycle. The original circuit in "Electronics" was for the original 555, though if I remember properly, they also had a pullup resistor on pin 3 (but I've never bothered with that).

I would point out that if someone needs an exact 50% duty cycle, it's just as easy to put the signal through a divider to get that 50%, and digital is often simpler than analog.

Michael

There's an even easier way (at least for the CMOS version). The charge/discharge resistor is driven from the output pin. ...Jim Thompson

As pointed out, the second way requires a CMOS "555".

The diode method lets you set any arbitrary duty cycle, not just 50%.

For frequency _and_ duty-cycle control see Freq_Duty_555.pdf on the S.E.D/Schematics Page of my website. ...Jim Thompson

sure. tie the trigger and threshold together with a R to common and a Cap from output to this same node. This will give you a square wave that is close enough.

Jamie

The CD4541, or ICM7242, has the oscillator and divider(s) all wired into one chip.

sure, if you don't care about the frequency. else swap C and R

And Exar or someone used to have a timer that was probably a 555 followed by about a 7 stage ripple counter, for really long time delays.

Michael

The CD4020?

CD4020 is just a ripple coubnter, CD4060 is the one with the in-built oscillator, it has a 14 stage divider but not all stages are tapped.

Among other applications it's used for the timer in cheap toasters. (not that i've ever _repaired_ a cheap toaster)

No, the xr2242 (I couldn't remember the exact number earlier, so I didn't list it, now I had to check).

It's been a while since I looked at the datasheet, but if I recall, it was a bit more than an oscillator and divider chain, there was some feedback point so the divider output could stop the oscillator or something.

Michael

Do they make expensive ones now?

I opened at least one, all I saw was an 8pin DIP, I assumed it was a microcontroller of some sort, kind of overkill, but probably simpler than the timer.

Michael

I haven't used the 4020 since the '80s, but I remembered that it was a multi stage ripple counter.

Yes it is. But some of those long ripple counters didn't have all the outputs available, though I can't remember if that was the 4020 or the

I think I missed something there. I bet the ICM7242 that somebody mentioned is just the XR timer I was thinking of but from a different manufacturer and hence a variant on the number.

The XR had some feedback, so if that output changed states, you could stop the oscillator, or something like that. So it was a more integrated concept than just an oscillator with a divider chain.

Michael

Maynard Philbrook

All it creates is high frequency squegging or nothing.

... Phil

Oh excuse me for not looking closer, the connects are correct, I just had the two components switched around.

Either way, they both produce 50% duty cycle. One rounded corners at high freq max to the chip operatin and the other square at RC freq..

I do make misakes at times. At least I can admit to it.

Jamie