Daily On/Off circuit?

I fixed that already. See my post from 7:37 PM. It's

32768*86400 = 2**22 * 5**2 * 3**3 = 2**11 * 2**11 * 675 .

Not when they're way down the chain. You sure don't put jam-loaded ones at the front, I agree. Synchronous-reset parts such as the 74xx163 are okay anyplace.

You have to watch the min/max ripple delay. Towards the end of the chain, you can make the jam reset almost arbitrarily slow--it only has to be quicker than 86400/675 = 128 seconds.

Neither you nor I would do it that way today, but it's still a decent option for hobbyists who just want to get the gizmo working and not make a science project out of it. And I still do that sort of thing sometimes in hand-wired PLLs for RF signal processing and stuff even in customer work.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs
Loading thread data ...

You guys are over-thinking it. What you need is a rooster!

formatting link

Reply to
mpm

2^15 = 32768

15 flipflpos will give you 1 hz

Cheers

Reply to
Martin Riddle

Never mind. I rebooted and now everything is fine:

2^11 * 2^11 * 675 2,831,155,200

google now says the same thing

2^11 * 2^11 * 675 = 2 831 155 200

It must have been some wierd glitch in Windows. Sorry to bug you,

Reply to
Steve Wilson

Yes, and for a light only system the photo-diode would work too. I have something like that for colored LED strips, but it gets the time from the LAN and switches on / off, color, what have you, at programmabe times via ethernet:

formatting link

3 micros... Been working OK 24/7 for years.

It can also be used as disco lights...

formatting link
It also switches lights if you are not home to keep intruders away.

Reply to
<698839253X6D445TD

------------------^^^ This should be 675, not 625

No glitch in Windows. You were typing in a wrong number! See above.

Rick C.

Reply to
gnuarm.deletethisbit

Your simplest option is probably to cannibalise one of the cheap and cheerful radio controlled atomic clocks that has an alarm function or an old digital watch with the same functionality.

Another option would be trigger a CMOS 555 to run for a fixed period after sunset which is much easier - by changing the specification.

Various RTC ICs can do it but unless you want to design your own it may be a lot of effort for small reward. I used a PIC 1684 to do something similar to water my greenhouse morning and evening but didn't really care if the timing drifted so used crude RC timing.

--
Regards, 
Martin Brown
Reply to
Martin Brown

Thanks for all the subsequent suggestions.

PICs/microcontrollers/FPGA solutions are non-starters, as I'm not into programming.

--------------------

Phil: I was thinking along broadly similar lines, as I said in my post. But avoiding that over-complex logic by accepting less precision and simply adjusting the clock frequency to get close to 24 hours.

And also using a longer counter, such as the 4536 I mentioned. (I've ordered a couple from ebay.)

Or alternatively I've now found the CSS555C micro-power version of the

555. From the datasheet that appears capable of the task unaided by ripple counters.

However, so far the only source I've found is in USA, at $2.49 each, but with absurd postage charge to UK.

formatting link

--------------------

Jasen: Will follow-up on LM8560 thanks, sounds good. That might give me the flexibility I described, perhaps eliminating the LDR altogether.

The Garden Lamps circuit gets its power from nearby 13.3V DC output of a variable power supply I made about 30 years ago. No batteries required!

Terry, East Grinstead, UK

Reply to
Terry Pinnell

So, then, 2^15 + 2^16 + 2^14 + 2^12 + 2^8 + 2^7 should give one pulse per day from a 32768 clock.

Reply to
John S

I'd buy a timer, they have dusk till dawn built in. you can even mix and match. Turn on at dusk and turn off at 11pm.

Cheers

Reply to
Martin Riddle

ote:

t

pulse

pm.

king,

king

'd

back

d I

the

be your

ounting IC's that will do 2^16 or more. I've forgotten all the digital IC numbers.

art.

e
**2,
,

I got one for about $15, programmed over wifi which the OP says he doesn't have at the shed. It could be programmed in the house but he likely doesn' t want that bother. It works a treat.

I have it programmed to cut off during the peak billing times of ToU. Once it is programmed it seems to work just fine remembering it's settings inte rnally. I haven't tried to see how is survives power failures if it doesn' t have access to the Internet in case it needs that to restore it's ToD clo ck.

I'm not a big fan of the cell phone app that is used to program it. I thin k they put a minimal effort into it and it has bugs. Good thing it is not an important part of everyday use. Oh, the unit also measures the power co nsumption of the load.

Rick C.

Reply to
gnuarm.deletethisbit

Not that many moons ago, I did a 1 second to ~100 second timer this way. I can't recall the chips, but I think I did it synchronously... at least the early divisions to about a second.

Of course I'm also a man with found memories of the 70 - 80's... not all that many electronics related. :^)

George H.

Reply to
George Herold

Windows having a glitch ?!!??

Nahhhh... Couldn't ever happen !

Could it ?

:)

Reply to
boB

Yes, Windows caused a keyboard error that resulting in the number being typed wrong, 625 instead of 675.

Rick C.

Reply to
gnuarm.deletethisbit

I hear you. Glory days, and all that. ;)

My fractional-N synthesizer for that satellite radio (1982-3) was built like that too, with an ECL 10/11 prescaler (Motorola 12013 I think), two unit-cascaded 74S161s, and two stages of CD4027 rate multipliers for the bottom three decades. That got me a comparison frequency of 10/11 MHz instead of 10/11 kHz at the price of a bit of jitter that I could filter out.

It was a big board full of DIP-packaged MSI logic and Mini Circuits RF stuff, one of whose selling points was that you could set the microwave LO frequency directly on the BCD dip switches--the 12-14 GHz oscillator was multiplied up by 110x from my VHF output. (I think it was a snap-recovery diode followed by a quintupler and a doubler--we farmed that part out.)

I still build RF stuff like that sometimes--three or four years ago I did the signal processing for a two-axis electronically scanned heterodyne laser microscope that way, dead bug in about six die-cast Bud boxes. After a bunch of frequency doublers, mixers, and so on, the resulting 300-450 MHz LO output had enough spurs to give children nightmares, and most of them crossed the desired output.

Fortunately the scan was relatively slow because I was using a lock-in as the phase sensitive back end, so a 1:1 PLL cleaned the LO right up. Good medicine in its way.

Of course now that Simon and I are working together, we'd go straight to a DDS solution if the opportunity arose again.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs

OK that sounds very far away from the max freq (except for edges) 32kHz few second thing I did. :^)

DDS is great, I know next to nothing, but love my rigol sig gen.

Simon is your son? My son needs more big smacks... not enough life experience... which is partly my fault, but too late for that now. After much clubbing he got his application into UB, and was accepted. I fear his lackadaisical 'tude, is going to be a first year crash and burn. (I hope I'm wrong.)

George H.

Reply to
George Herold

Do you have access to the line? Count the line frequency:

50->60->60-> n x hours, just some BCD counters would be simplest.
Reply to
bloggs.fredbloggs.fred

Yeah, formerly known as Dashing Firmware Hunchback. Then he turned into a partner, so I have to be much more polite. ;)

I resemble that remark. ;)

I took a year off after my first year in university because my marks weren't great--I barely scraped an A in physics, and the rest were all B's because I cut class a lot. I entered at 16, and although I had no problems with the material, I just wasn't mature enough to supply the self-motivation.

I decided that if I wasn't going to take school seriously, I ought to go do something else for a bit. The something else included delivering pizzas, working in a warehouse, sleeping late, driving my parents half crazy, getting fired from a job selling cameras, and bicycling from Vienna to Paris via Switzerland against the prevailing wind. The wind turned out to be a much bigger deal than the hills.

I got a lifetime's worth of bumming around out of my system that year, and did a lot better subsequently.

Various family members have done vaguely similar things. So, chin up.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs

Those diagrams are almost obscene: grey on grey. Please use white brilliant paper (92 or better), and a black marker. Quality of all else is superb.

Reply to
Robert Baer

Where the heck have you been? Half the Internet is grey on grey! Does the black on YELLOW of his web pages not bother you at all? An artistic frien d who has worked as a professional decorator has pointed out there is a rea son why bees are black on yellow. It is a universal sign of aggression. I t is used for danger signs. They are colors that only a madman would use f or a web page... and you are complaining about the nearly indecipherable sc ratchings of this person because the contrast isn't high enough???

Wow! I suppose next you are going to complain about the lack of a title bl ock?

Rick C.

Reply to
gnuarm.deletethisbit

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.