Reducing On_Off ANALOG Timer Pin Count

[nickelflippr]

Looking to investigate an ANALOG timer setup that will be on for maybe 30 sec, and off for 20 or 30 min. The boiler plate solution seems to be a 555 timer for the 30 sec on time, and a CD4060 counter for the minutes off timer. The circuit will run off a nominal 12V battery voltage.

Is there another scheme, with a smaller total pin count?

This is for a solar tracker circuit with photodiodes. Late in the day, noticed that the up-down relay was going tat-tat-tat-tat every 10 to 20 seconds, kind of annoying. The sensor sensitivity can be played with physically by photodiode location/shading and electrically with comparator feedback. This approach may yet be the solution, but wanted suggestions on a timer if that doesn't go well.

 

[BrownOut]

You should be able to get both times from a single 555 timer configured in the astable multivibrator mode.

 

[nickelflippr]

Thanks, I think this will work out. First had R1 and R2 hooked up backwards, oops!

The values picked for the astable mode:

R1 = 4.7M
R2 = 100K
C = 220uf
t1 on times was about 28 minutes
t2 off time was a knda short, well under 30 secs, so will play with that.

 

[ljcox]

As you're powering it from a battery, then a pair of the 7555 CMOS ICs would consume less battery current.

Alternatively, for a single IC solution, you could use a Schmitt Trigger IC such as the 40106, 74C14, etc.

See the attachment for inspiration. It is not exactly what you want but it could be modified.

It can be significantly simplified for your purpose. You would only need 2 caps, 3 resistors. 1 diode and the IC.

 

[BrownOut]

Thanks, I think this will work out. First had R1 and R2 hooked up backwards, oops!

The values picked for the astable mode:

R1 = 4.7M
R2 = 100K
C = 220uf
t1 on times was about 28 minutes
t2 off time was a knda short, well under 30 secs, so will play with that.

I didn't take the time to do the math on your values. However, I'll just warn that the "normal" equations aren't valid for long delay times, due to leakage. Best to double the values from the calcs, and experiment. If it turns out that 28 minutes is too long to achieve ( I think it will be OK ) then you might use a dual 556 timer, or one of the lower power cousins to split the effort.

 

[MikeMl]

Because of your long delays, the approach described in this thread might be better.

 

[nickelflippr]

ljcox:

Good ideas there on the CMOS circuits.

Yea, should have said a 12V car battery. So yes I will be burning a handful of ma's for the standard LM555 drive current. Don't have the CMOS version in the parts bins.

BrownOut:

Funny you should mention the time equations! I almost tossed in the towell saying this wouldn't work, when the calc'd value for the cap was originally 470uf for 26 min. Well 52.5 min later the led came on (EDIT: meant to say led Off), and had to start a new reply to the thread after halving the cap value to 220uf.

Yes, it's been breadboarded, and watching the paint dry, confirmed.

 

[nickelflippr]

Because of your long delays, the approach described in this thread might be better.

Saw that thread yesterday during research, but didn't really sink in till now. The best I could come up with in CMOS (and in the parts bin) would be a CD4060, rc timer on pins 9, 10, 11, and AND the outputs to vary the ON time?

 

[ljcox]

This is for a solar tracker circuit with photodiodes. Late in the day, noticed that the up-down relay was going tat-tat-tat-tat every 10 to 20 seconds, kind of annoying. The sensor sensitivity can be played with physically by photodiode location/shading and electrically with comparator feedback. This approach may yet be the solution, but wanted suggestions on a timer if that doesn't go well.

I assume by the part highlighted in red you mean add hystersis by positive feedback.

This would reduce the sensitivity to small changes in light level.

As for the timer solution, it could be done with a 4060 counter with suitable decoding using diodes (as you wrote in the post above).

 

[nickelflippr]

I assume by the part highlighted in red you mean add hystersis by positive feedback.

Really poor choice of words, sorry. Currently using a 10M resistor for "hysterisis".

Meant to say the photodiodes are paired up for E-W and UP-DN tracker movement. By putting a voltage divider on diodeA to one input of the comparator, and referencing the other input to diodeB( no divider) a sensitivity or deadband control is obtained. Vice-versa for the other comparator, to drive the opposite direction. For a schematic, and description by Mneary, look at very recent solar tracker thread. Really didn't want to get into that, in this post.

 

[ljcox]

Perhaps you should reduce the 10M to a lower value to give more hystersis.

I looked at the data sheet of the 4060 counter. It is not ideal as it does not have an output from stage 11.

So the 4020 would be a better choice.

All you need is the counter with 2 R + 1 C for its oscillator and 5 diodes + 1 resistor connected to outputs 8 ~ 14 to form an OR gate, ie. the gate output is high when any of these outputs are high.