Detect switch without a complete circuit?

[cfgrs]

Hello everyone! I have a bit of an engineering puzzle. I've only a passing understanding of electrical engineering, so still trying to puzzle if out.

Basically, imagine you have one black box A connected, via a single wire with a switch, to black box B. There is no returning wire. The components inside the boxes are up to you, but somehow I need them to be able to detect when the switch between them is closed. Since there is no returning wire, my basic electronics seems insufficient without a common ground to be had.

Any suggestions? Many thanks!

- Chacho

 

[sophiecentaur]

Hi
The answer is, in principle, that it can be done but the actual setup would need to be specified.

 

[cfgrs]

What more needs to be known about the setup? That's pretty much it in terms of the restrictions. Single wire between two boxes, no return feed, and try to get a light to go off or something when the switch is closed?

Any tips or possible avenues to explore would be appreciated!

 

[sophiecentaur]

The reason that the actual set-up is important is because you would need to send a very short pulse down the line and watch for the reflection at the switch. (Time Domain Reflectometry).
If the circuit consisted of nice parallel or straight wires, the sent and reflected pulse could be identified fairly easily. If the set up was only 10cm long then the accuracy would be poor - likewise, if the setup was 1km long, the pulse would disperse and the position of the switch would be indistinct.
You are limited by the shortness of pulse that you use and the power available. A good system can spot features down to a cm or so - in ideal circumstances.

If the wires were jumbled up or went in a circle, the pulse could take multiple paths and be hard to identify.

If you want to 'break the rules' then life is hard.

 

[NascentOxygen]

There is always capacitance between conductors and the earth. So if the connecting wire were made part of a tuned circuit, when the switch is closed the circuit now incorporating the length of wire will tune to a slightly different frequency.

In homes fed with AC electricity, cables in walls and ceiling cause induced voltage on lengths of wire. If the switch grounds or ungrounds the wire, the magnitude of the induced 50/60Hz voltage in it will change.

Is this a thought experiment, or do you have a practical application in mind? How long is the connecting cable?

 

[cfgrs]

Hello,

Well, basically I'm thinking of a small system for wireless sabre fencing. In a typical system, the sabre has +5V or somesuch, and makes contact with the opponent's grounded conductive vest to register a hit. To make the system wireless, I'd need to come up with a way around the fact that I no longer have a common ground.

Would it be possible to do something like sending a high frequency signal down the line? Could an HF signal be detected without a common ground (i.e. detect that there is an oscillating signal, even though you don't know the absolute values of the peak/low points?)

 

[sophiecentaur]

Yes, it would be much easier to detect the presence of the capacity of your opponent when attached to your sabre if you were to use an AC signal. The fact is that, with an AC /HF signal, there IS a return signal path via the ground because of the capacity between your feet and the Earth.
I am surprised that no one has developed such a system already. It would make very good sense.

 

[cfgrs]

Anybody with a reference to something like this? i.e. using an HF signal to detect capacitance so a newbie like me can start learning the basics? most google searches just lead me to touch-screen stuff...

 

[NascentOxygen]

In a typical system, the sabre has +5V or somesuch, and makes contact with the opponent's grounded conductive vest to register a hit.

And if your sabre contacts his sabre no hit is registered because both sabres are at +5v here? Or maybe there is a metal tip that's insulated from the rest of the blade?

Is the sabre blade insulated from the handle?

But if you went with wireless HF, then contact with an opponent's sabre might produce similar effect to contact with his vest.

 

[cfgrs]

Actually, in the system that I'm building now (which is not wireless), I simply alternate between having one sabre "active" or the other (we're talking a couple hundred per milisecond), and when the sabre is not active, it is grounded to an input so that I can distinguish between a hit to my opponent's sabre or a hit to my opponent's jacket.
But I think the older analog systems used something along those lines, yes.

 

[cfgrs]

The sabre blade is indeed insulated from the handle (rubber grips on all of them)

 

[Wetmelon]

Yes, it would be much easier to detect the presence of the capacity of your opponent when attached to your sabre if you were to use an AC signal. The fact is that, with an AC /HF signal, there IS a return signal path via the ground because of the capacity between your feet and the Earth.
I am surprised that no one has developed such a system already. It would make very good sense.

The olympic fencers use this method for their wireless technology, if i understand what you're saying correctly. Essentially it uses the body's natural capacitance to detect closed circuits. There are also RF designs that have been explored and work quite well, though I'm not sure what the status of their commercial availability is.

And if your sabre contacts his sabre no hit is registered because both sabres are at +5v here? Or maybe there is a metal tip that's insulated from the rest of the blade?

Is the sabre blade insulated from the handle?

But if you went with wireless HF, then contact with an opponent's sabre might produce similar effect to contact with his vest.

2 effective lines in fencing:

A - Ground (Lamé aka vest)
B - +5V - Blade, insulated from the hand.
C - (Tied to B in sabre)

In epee and foil the C line is used to detect invalid hits. In foil, the C line is used reverse of foil. Not entirely sure on the specifics, but basically when the pushbutton is pressed, it has to be grounded for foil to get a "hit" (against the vest). In foil if an invalid touch lands, the action is halted, which is why line #3 is required. In epee, no vests are worn and the floor is grounded, which allows one to recognize a valid hit by the fact that the tip is depressed but is NOT grounded. When the epee tip contacts the ground, no lights activate (I think? Idk I fence sabre :P)Note that for any NEW wireless fencing technology to be adopted, it has to be proven to be

1) Reliable
2) Low maintenance
3) Un-"Hackable". This means that there should be no way for the fencers or someone in the audience on behalf of the fencers to remotely trick the box into registering a hit. Encrypted LOS wireless LAN network radios like those used for squad-level communication by the military are probably going to be the most secure.

 

[cfgrs]

Yeppers, a similar system is used at the higher levels, but they actually wear a special (read: expensive) conductive t-shirt under their fencing equipment which allows them to ground to their bodies, and it requires constant calibration and monitoring by on-site engineers and is still prone to frequent issues.

What I would like to do is remove the need for the conductive t-shirt (as most fencers do not have one, on account of wireless systems only being available at the highest level tournaments, not normal fencing clubs) and have the system still work. Not too worried about hacking, since it'd just be to make a simple/cheaper system for fencing clubs.

 

[NascentOxygen]

So there's a switch in the tip of the blade which serves some function?

 

[cfgrs]

Well, there are three weapons. Electronically, sabre is the simplest, and the blade simply needs to make contact with my opponent's conductive jacket to register a hit.

For foil and epee, a hit only counts if it is delivered with the tip of the blade. For these weapons, there is a button on the tip that must be depressed for a hit to be registered. If the tip is pressed down on my opponent's conductive jacket (in foil), then the hit is valid, but depressed on any other ungrounded surface, the hit is not valid.

 

[Wetmelon]

Yeppers, a similar system is used at the higher levels, but they actually wear a special (read: expensive) conductive t-shirt under their fencing equipment which allows them to ground to their bodies, and it requires constant calibration and monitoring by on-site engineers and is still prone to frequent issues.

Oooo. I thought it was an insulative t-shirt (because they have to keep them dry). I've never actually looked into how the current system functions all that much, do you have any resources you were able to find?

 

[cfgrs]

Alas, nothing much, except what little I've heard from speaking with other fencers that had some experience with the systems. What I'd like to do is essentially eliminate the need for the fencer, so that the unit is stand-alone that can be used by any fencer equipped to work on a normal scoring system.