metal banister idea :)

[idi]

I needed some help on a new project i was thinking of doing at home.

I have a metal banister on my staircase. i wanted to connect a circuit so when ever someone touches the banister whilst they are walking up or down, the stairs will light up, and when ever they let go, the lights go off. I dont have a problem wiring the LEDS for the lights on the stairs ( I am just going to use a 12v adapter and use a resistor) I just need a component that will realise when a the banister is touched and join the circuit, and break the circuit when release.

I dont even know if such thing exists lol. If so, where can i get it from? and would the idea even work?

Thanks

 

[dknguyen]

Capacitance is the first thing that comes to my mind but it seems rather expensive for something as large as a banister. And it doesn't sound like you can get inside the metal banister either to implement that method. Personally I'd just go with a motion detector haha. A lot easier and smaller and moveable to places other than stairs. Plus lets you run up the stairs without holding onto the banister.

But if you insist on touch, there are the obvious, but ugly, ways. Conductive compressive foam sandwiched between conductive membranes and measur ethe resistance changes for pressure. Of course, this is probably out of the question because it'd be so ugly and requires large sheets of material. It wouldn't even need the banister to be metal, though if it were the banister itself could be used as the bottom conductive layer rather than a membrane.

The easiest touch method would probably be some kind of very sensitive current detection method using the conductivity of the human body. But I'm not sure how reliably that can be made to work because it would have to be so sensitive. You might find it too sensitive or not sensitive enough and depending on how you went about it, if a person was not properly grounded (socks, shoes, etc) it might not work reliably, and it deifinately wouldn't work without gloves but then again nothing but pressure methods or motion sensing would work in that case.

If you had two separated metal strips on the railing it'd be a lot easier. Just have the person's hand complete the circuit when their hand touches both metal strips at the same time and measure the change in uA of current that flows from completion of the circuit. But your railing is probably one single metal block.

 

[idi]

Thanks for the info

I can get inside the banister actualy, there are caps on either side that can just come off. The motion dector is a good idea actually, but i have been challenged by my family who think i cant do it, so now i HAVE to make it the touch way.

The conductive foam could also work, where can i get some from? and how does it look? I dont want it to spoil the look of the banister.

And the idea of the sensitve current meter is the only idea i could think of aswell

 

[dknguyen]

I don't know where you can get the conductive foam from. I've never needed to use any. But it's black and looks like foam and would spoil it. It's unlikely you could get it to wrap as cleanly as you want it to. Your challengers might consider that cheating too though.

But you can get inside the banister so who cares about the foam. In that case it might be possible to run a long strip of conductive material all inside the banister that is very close to the metal surface but not touching (so likely insulated by a thin film). And then measure the change in capacitance from someone's conductive hand touching the conductive banister. The banister's metal might be too thick though to allow reasonable sensitivity levels. There's also the difficulty of how to lay it flat and glue it to the inside of the banister when you only have access to the ends.

But one very reasonable alternative that is much easier is to run the same strip along the outside of the banister. Just make it pretty and narrow and run it along the center. There would need to be a thin film insulating the strip from the metal banister but keeping it them very close. Another possibiliy is this same construction, but measuring small current changes when someone's conductive hand touches both the metal banister and the metal strip completing the circuit. You could test out the effectiveness of both methods this way since the implementation on the railing is the same- just the end sensing circuit is different. You'd have to test them out due to see how reliable they can be made.

You might also consider simply using a vibration type microphone inside the banister lol. Pictures of the banister might help us get more ideas or more specific ideas.

I'm going to assume you know the basics of a sensitive current meter. Measuring capacitance is more complicated. The simplest way is probably an R-C type circuit where the cap is the two pieces of metal separated by a very thin film. First, either charge it up to full voltage or discharge it to 0V (probably charge/dischrage just through a short or low value resistor just to speed things up) to get it to a known voltage. Then discharge or charge it up through an accurate high value resistor and measure the amount of time it takes to fall or rise to a certain percentage of the full voltage. From here you can calculate the RC time constant and from there capacitance or resistance if you know one or the other.

The more complicated but direct way is to pulse a high frequency AC current at an accurate frequency through the capacitor and measure the AC current. The amount of current that flows will be dependent upon the capacitor's impedance, which is determined by the frequency used and the capacitor. Basically you are just measuring the "resistance" of a frequency-dependent resistor- the so called impedance.

 

[kinarfi]

get one of those touch light modules and a timer, if someone is coming up or going down, I would want the light to stay on for me and not be shutting off every time I removed my hand. Set the timer for something like 30 seconds to drop all the power to circuit so it can reset.

 

[idi]

What if i got a normal touch switch like this:
**broken link removed**

and put it underneath the floor boards or something. and extended the touch part onto the metal banister. would that work?

 

[dknguyen]

The touch part is likely a two membranes separated by a thin film so you can't just connect wires to the banister. But never hurts to try if it's cheap, fast, and easy to do.

But kinarfi's idea is really good. That is more apt to work than the using the touch sensor because I don't think any touch lamp has a membrane running underneath the whole thing. But if the touch sensor works the same way it would also work. (I'm not entirely sure how touch lamps work so...)

EDIT:
Kinarfi's seems like the best idea so far by far. The article seems to indicate a type of capacitance measurement that is much more indirect, but also much more rugged and does not require membranes or anything like that so you can just connect wires.
https://science.howstuffworks.com/innovation/question42.htm

So do those lamps just measure the very small voltage change that happens when a person touches and transfers charge to the banister or lamp? Because if so...that's fairly easy to do. The hard part is making it very sensitive.

 

[idi]

well i have a spare one at home, so ill just quickly give it a try tonight. if not, i think am gnna have to use the motion sensor idea that you give me.

Thanks alot for you help mate. I appreciate it

 

[Jaguarjoe]

Cheap and easy Picaxe microprocessors have "touch" inputs. They can also provide whatever logic you may need as well as timing functions if needed.

 

[Sceadwian]

Good call kinarfi, one of those touch modules should work just fine as long as the banister isn't grounded, I actually have one in one of my bits bins. Could probably buy a touch lamp at wallmart for 15 dollars and rip the module out of it. One problem with them is they're VERY finicky little devices, my stepsons lamp had a cheap chinese made touch unit and if you turned a fan or florescent lighting sometimes the touch sensor would falsly trigger, so what I would suggest is using the touch sensor to trigger a timer that would trigger a relay that would actually turn the lighting on and off, that way even if it falsly triggered it'd still turn itself off. You could also buffer the power lines a bit, I'm sure the false triggerings were from the inductive load being turn on being a cheap product it doesn't contain as much filter circuitry as it could.

 

[dknguyen]

Did some very quick and brief reading and it sounds like the touch lamp doesn't work by the human body adding charge to it during a touch. BUt in the opposite matter. The banister is charged up and left to float while it's voltage is monitored. When a grounded conductive human body touches the banister they complete a circuit that drains some of the charge from the banister producing a measurable voltage change. Neat!

So a type of capacitance method, but without actually making a capacitor of two memranes separated by a film, but just using a conducting body to store charge.

 

[#12]

Much like Post#2, I'm thinking you should connect an AC signal to the bannister. Make the bannister part of the oscillator frequency tank and look for fast changes in frequency. (In this case, fast means less than a tenth of a second.)

Trying to pick up 60HZ hum from the victim might work if you based the sensing on sudden changes in amplitude. The bannister will naturally pick up plenty of hum and the amplitude will drift all over the place, but sensing fast changes might work. It depends on how much antenna a human is compared to the bannister. If you load the bannister with a resistor to keep the AC hum voltage down to a couple of volts, a human touch might work if it gives you a 10% amplitude change. It will probably pick up the vacuum cleaner, too!

Any way you do it, you must add a time constant to keep the lights on for a while when the touch is removed.

 

[dknguyen]

So many good ideas. Way simpler and better than my original ones requiring extra stuff to be all over the banister.

 

[Sceadwian]

I'd try the adapted touch circuit from a lamp first, I know you can pick those up for 15 dollars at wallmart and it wouldn't take you but 5 minutes to test it. You are dealing with mains AC power for it though so be careful, for testing it would simply be a matter of attaching a wire from the banister to the base of the touch lamp itself, then touching the banister should turn the light on and off. Obviously taking the unit out of the lamp would be trhe ideal solution testing would require no rewiring. I think the important part is that the banister must be highly isolated, if it's grounded even a little bit it probably won't work. For a 15 dollar investment that will take 30 seconds to test, I'd say it's worth it.

 

[idi]

I have a touch light switch in my bedroom, so i thought id test it on that. i got a screwdriver and touched it on the "touch" part of the switch. i touched the screwdriver and it turned the light on!

Obviously a screwdriver is tiny compared to the banister, so i am going to buy a cheap touch lamp tomorrow and test it out. i hopw it works this way. it would save me alot of work and money

 

[Sceadwian]

Good luck.

 

[shortbus=]

After reading this thread I went on to the other threads and found this; https://www.electro-tech-online.com/threads/other-uses-for-garage-door-eyes.114086/

Why couldn't you use the sensors from a garage door? Make new end caps to mount the light and sensor slightly above the top surface of the railing. The sunlight would keep it from working in the day time, which is a good thing. After dark a hand put on the railing would break the beam turning on the lights, a timer keeping them on for whatever time you need to get up the steps.

The end caps holding the sensor/light could be made decorative. These units work over a 16-18 foot garage door opening, don't know how long the railing is, so you might need two or more if its a long distance.

 

[idi]

After reading this thread I went on to the other threads and found this; https://www.electro-tech-online.com/threads/other-uses-for-garage-door-eyes.114086/

Why couldn't you use the sensors from a garage door? Make new end caps to mount the light and sensor slightly above the top surface of the railing. The sunlight would keep it from working in the day time, which is a good thing. After dark a hand put on the railing would break the beam turning on the lights, a timer keeping them on for whatever time you need to get up the steps.

The end caps holding the sensor/light could be made decorative. These units work over a 16-18 foot garage door opening, don't know how long the railing is, so you might need two or more if its a long distance.

That is a brilliant idea actually. i think this is the one i am gnna use. Thanks for that

 

[#12]

The last time I worked on a garage door safety eye, it was pulsed so it WOULD work in the daytime.

Just reporting my experience.

 

[kinarfi]

Universal Lamp Touch Switch 6523
This is what I had in mind when I suggested a touch light. A banister rail may have to much surface area or is probably grounded, don't know for sure if either would make it not work and it is also a dimmer.
Another idea would be to use a motion detector like the ones people put in the back yard, there even some that work with an on/off switch that if you turn on, off, on quickly, it turns on and stays on.
As they say, good luck
Kinarfi