I want to build a timer that when pulsed will allow some led ribbon lights to stay on for five minutes. I have seen in your threads ones that turn on after 5 mins, but I need the opposite effect. Thanks

 

Welcome to AAC!

How much current do you need and how accurate does time need to be?

 

This is a marine night head (bathroom) light consisting of about 6' of led ribbon, so the draw has to be less than 1 watt. Time not that precise and if they need more time to finish their "business" then they could crack the door to trigger the magnetic reed switch and re set.

 

5 minutes is a bit long for an LM555 timer without adding a counter. I'd try a CD4538. In theory you can use large value resistors and/or capacitors for timing.



Pulse is Rx*Cx; 5 minutes will still require largish values; capacitor leakage will be a limiter on timing accuracy. Output can drive a BJT, MOSFET, or solid state relay to control the LEDs.

Power dissipation isn't very meaningful if you don't include voltage or current information.

 

I don't like long R-C timers, and a CD4060 counter would do this nicely. But to keep things simple...

To prevent the light being on continuously if the door is left open, can the reed switch be mounted an inch or two off the jamb so it pulses whenever the door is opened or closed? If yes, then this reduces to a CMOS 555, a small n-channel power MOSFET, and some Rs and Cs.

ak

 

Yes, the reed switch can be mounted at a distance from the jamb. However since it is a boat and a door left open will bang back and forth when it rolls, all doors and drawers tend to be firmly shut, so not worried about that, but maybe a good idea in any case. Voltage is 12vdc although can run up to 14 when engine is running and current draw is < .25 amps.
I have no claim on a 555 and if a CD4060 would work better that that is fine too. I notice that there is no pot, so figure that you have calculated the values for ≈ 5 mins. Anywhere from 4 to 6 is fine.

I'm assuming from the diagram that the output here controls the ground on the timed output/lights, and the left connection tab that runs to TR on the 555 remains open.

 

Almost. The left connection point to the TR input goes to the switch. The switch return goes to ground.
In electronics, one time constant is R (ohms) times C (farads), and is approx 63.2% charge on the cap. A 555 monostable triggers at 67.7% charge, a difference of 1.07%. This is why the timing equation is 1.1 x R x C. With perfect components that works out to 5.8 minutes, but your results may vary. 270 uF yields 4.75 minutes. A 100K pot in series with R1 gives you some adjustability.

ak

 

You could use triggers and tact them at 1 sec or 1 min to get a digital clock?

 

You could use triggers and tact them at 1 sec or 1 min to get a digital clock?

Thank you all--Parts will be arriving soon and it should be fun to see what happens. Will let you know.

 

Yes, the reed switch can be mounted at a distance from the jamb. However since it is a boat and a door left open will bang back and forth when it rolls, all doors and drawers tend to be firmly shut, so not worried about that, but maybe a good idea in any case. Voltage is 12vdc although can run up to 14 when engine is running and current draw is < .25 amps.
I have no claim on a 555 and if a CD4060 would work better that that is fine too. I notice that there is no pot, so figure that you have calculated the values for ≈ 5 mins. Anywhere from 4 to 6 is fine.

I'm assuming from the diagram that the output here controls the ground on the timed output/lights, and the left connection tab that runs to TR on the 555 remains open.

For your needs, you should find a CMOS 555 just fine. The very first paid design I did (while an undergrad and before I knew about digital logic, let alone microcontrollers) was a 15 minute "hospitality timer" for the Taco Bell I worked at. It was used to let the staff know that it was time to go out and check the seating area and clean it up if necessary, since when we were busy it was the first thing to start getting neglected. It worked great, even though the RC sizes were significantly different than what the book would say because of the capacitor leakage current -- in fact that's how I really learned about capacitor leakage as a real design factor.

 

For your needs, you should find a CMOS 555 just fine. The very first paid design I did (while an undergrad and before I knew about digital logic, let alone microcontrollers) was a 15 minute "hospitality timer" for the Taco Bell I worked at. It was used to let the staff know that it was time to go out and check the seating area and clean it up if necessary, since when we were busy it was the first thing to start getting neglected. It worked great, even though the RC sizes were significantly different than what the book would say because of the capacitor leakage current -- in fact that's how I really learned about capacitor leakage as a real design factor.

Where do you find these jobs? I don't believe that by asking randomly 100 shops I would get even 1 offer? How does this work even?

 

Where do you find these jobs? I don't believe that by asking randomly 100 shops I would get even 1 offer? How does this work even?

The key to this one was the phrase, "at the Taco Bell I worked at." So the answer of how I found the job was I applied for it. I was working as a co-op student at the National Institute of Standards and Technology (NIST) at the time (though it was still called the National Bureau of Standards back then) and the pay sucked. So I also worked part time at Taco Bell. My manager knew this and knew that I was learning about electronics, so he approached me with an idea for this hospitality timer and asked if I was interested in building one. I think he had the notion of trying to sell the corporation on it and getting a bonus if it was adopted. I got paid $50 for doing it, plus the cost of the parts.

Several of my consulting gigs, including some pretty lucrative ones, have come about this way. While waiting for my commencement ceremony for my bachelor's degree to start I was talking to one of the physics professors (who I had actually never taken a class from, but who had sat in on my Senior Design presentation) and he and a couple of others had gotten a patent related to a new twist of doing electronic radon level monitoring using dollar-scale components and he asked me if I was interested in doing a prototype. Over the next few years I did several of them as they wanted to demonstrate more capabilities. I also did several consulting jobs for the group I had worked for at NIST when they wanted to expand their servohydraulic testing capabilities or design and build a new experiment apparatus. They knew I had worked with high-performance hydraulic systems on jet fighters and had solid mechanical design skills. Later, working as an ASIC design engineer, I did a few side jobs for my company and even took over a customer project from them when we got bought by another company. In fact, I've never actually done any marketing to find jobs and don't really have any idea how to do it or how successful I would be at it. For me it's all been word of mouth and opportunities as they come up. The closest I've come to marketing is noticing something that an employer might find useful and asking, "Hey, would you be interested in having me...."

 

That cant happen in the country I am. Thanks for the explanation anyway!