Sequencing output based on Power Cycling

[Danneskjold]

Here's what I'm thinking of: imagine for example a sprinkler system being fed by a pump. There are 6 sprinkler circuits. what I want to do is sequence them such that every time it powers up, it's advanced to the next station.

kind of like an alternating relay, but with more stations. pointers on where to start?

Thanks, RD

 

[Harald Kapp]

In a steampunk way:
Build a geared wheel with 6 positions. Use a solenoid with a kind of piston to advance the wheel by 60 ° each time power is turned on. (Think of a stepper motor with 60 ° steps). Add contacts to the wheel to distribute power to one of the 6 pumps.

With today's available means:
Use a microcontroller with EEPROM or NVSRAM (both are non-volatile memory types). At power up, read the last pump number from that memory, advance by one, activate the respective pump and store the new pump number into non-volatile memory.
Wrap around at 7 so the sequence goes from 1 to 6 and then re-starts at 1.

 

[Danneskjold]

microcontrollers are slightly above my pay grade... any thoughts on an analog solution?

Thanks, Harald!

 

[bertus]

Hello,

How about an oscillator (for the switching speed) and an 4017 as sequencer.
The 4017 could drive some SSR's for switching the pumps.

Bertus

 

[Harald Kapp]

The 4017 can do the sequencing, but you'll have to add circuitry to store the last position while power is off so the counter can advance the next time power is turned on.

 

[hevans1944]

Rotary stepping switches, which used to be used in telephone exchanges to accept dialed-in numbers and connect the incoming line to the dialed line, are an obvious solution. Problem is finding and using them. @Harald Kapp's suggested steam punk solution would probably be easier to build today, although you might be able to find stepping switches on eBay. The ones I worked with back in the 1960s, used for Air Force flight-line "black box" test equipment, needed about 300 V DC on the actuator coil.

Modern microprocessors should not be above your pay grade unless you are willing to accept subsistence wages. I first got involved when Intel invented microprocessors in the 1970s and development systems back then cost thousands of dollars. In the 1980s and 1990s I developed embedded microprocessor data acquisition and control systems using the Intel 8080 and 8085 8-bit microprocessors, later moving on to embedded personal computers based on IBM PC architecture.

Today, you can buy an Arduino microprocessor for less than fifty bux, or a Microchip PIC microprocessor for less than a dollar. You need to spend about fifty bux to buy a programming pod for the PIC, and you need a personal computer with a USB port for either one to develop and download programs. The development software is free. I learned how to do this in my 70s, so what is your handicap or disability that prevents you from doing the same? Afraid of moving above your pay grade?

There are PIC models with non-volatile, program-accessible memory that will (temporarily) store the state of which water sprinkler is next in line to operate. Add a few solid-state AC power switches for the sprinkler pump and perhaps for valves directing the water to individual sprinklers, all controlled by PIC output ports, along with a few lines of code and you are in business. A wall-wart provides 5 V DC power to the PIC, and if power fails, no problem because of the non-volatile memory. You can even put the PIC to "sleep" to save power in between watering cycles. Lots of free software and online help available for ordinary folks who just want to putter around a little bit and have some fun while being productive.

 

[Danneskjold]

Thanks for the lengthy reply.

I'm quite sure I can learn Arduino and all that. I just do not have time right now....

What I decided to do is simple and stupid: A high-torque clock motor connected to a rotary make-before-break 6 position switch. Each circuit will get 10 minutes. Cost is under $20.

 

[hevans1944]

If it works for you, it ain't stupid. Simple is also good; simple and inexpensive is even better! Glad you found a solution to your irrigation problem.

Most folks who hang out here apparently are not as strapped for time as you are. Electronics may have been a significant part of our working careers, but now it is also (sometimes exclusively) a hobby to be pursued at our leisure... often with reduced monetary resources, but with lots of "free" time. I was fortunate to be able to acquire all the tools needed to pursue this hobby before I retired, plus accumulating the usual "junque box" of vintage parts to play with. I did spend a few dollars gearing up to develop projects based on Microchip PIC embedded microprocessors, but before that I was also playing around with Arduino and Raspberry Pi micros. The R-Pi is generally more than I need for simple embedded processing, and the Arduino falls somewhere in between the PIC and the R-Pi.

PICs remind me of where TTL (Transistor-Transistor-Logic) was during the 1960s and early 1970s: the go-to logic choice as medium-scale integrated circuits (MSI) and large-scale integrated circuits (LSI) seemed to spring up overnight. This tech was eventually replaced by CMOS logic, which became almost as fast, but required much less power.

Then Intel invented the microprocessor, and soon thereafter IBM built the 8-bit Intel 8080 into the widely popular IBM personal computer. Many other manufacturers, such as Motorola and Zilog, followed Intel's lead with their own versions of 8-bit microprocessors. For a short period of time it seemed like embedded 8-bit micros were everywhere. They still are today, but personal computers evolved at an even faster pace, driven perhaps by the gaming community and the widening availability of Internet services with decent bandwidth.

I am not a programmer by profession. I hack code as best I know how, freely borrowing on what others have done before. I dabble in high-level languages like C, BASIC, and FORTRAN but prefer to use Assembly because it is closest to the hardware, banging on the bits as it were. Even closer would be binary, but it is hard enough to read and interpret that without trying to program in binary. In the 1970s and 1980s microprocessors became very complicated, supporting (actually requiring) real operating systems like their mainframe cousins and becoming outfitted with high-end peripherals like magnetic tape drives, hard-disk storage, high-speed printers... if it was ever attached to a mainframe, someone figured out a way to attach it to a PC. It was about this time in the early 1980s that I quit trying to keep up and decided professional programmers were better qualified by temperament, desire, and training than I would ever be to program personal computers. Oh, I puttered along for a few more years rolling code for embedded Intel 8085 8-bit micros, but my heart wasn't in it anymore.

Now flash forward to the 21st Century and my post-retirement. With the help of some folks here on Electronics Point, I discovered Microchip PIC microprocessors. Wonderful little "toys" and very cheap, too. A programming "pod" was very affordable, and development software was free. Forums sprung up on the Internet that supported development and some even encouraged "newbies" to participate. But even better, a person of modest means and resources could build hardware from scratch, either on an etched circuit board or on a Veroboard, and be up and running in a few days or a few weeks (depending on the complexity of the PIC). There are dozens of PICs to choose from, depending on what "peripheral" devices such as analog-to-digital or digital-to-analog converters, or various timers and digital input/output ports your project needs.

And that's where it stands as I approach my 76th birthday this year. I hope to embed a PIC (or two or three) in an Amateur Radio project in the coming months. At the very least, I would like to again help someone here learn to use PIC microprocessors.