What is the least expensive, most transportable thing that comes to mind? Do you think that I could get away with a version of your idea if I used photosensors instead of a camera?

 

I depends on the number of sensors. Off hand I can not think on anyway to matrix phototransistors. Maybe the EE types know how.

With a camera you only need enough video resolution to map the game board. The real work would be the software. You only need to detect the presence and location of light, no image recognition.

Lets extend the idea a bit. Supose that we want two way communication. Place a photo transistor (or IR reciever) inside each game peice. The game pieces recieve info from the master controller via a few IR leds under the board. When requested to ID themselves, the pieces provide on bit of info at a time as clocked by the under board IR LEDs. That way we can take a image of each bit process it and the ask for the next one. This makes the hardware simpler because we can setup the shot and use a still camera.

It could take a while.

__________________
Please post questions to the forums. PM's are for personal communication.

BCHS/3v0's Tutorials
Junebug USB PIC programmer kit., USB Bit Whacker,
The 15 Minute Printed Circuit Board! (+drill time)

 

I’m not sure what it means to matrix phototransistors. I was thinking of using three phototransistors inside of a light resistant chamber underneath the game board. The light from each game piece would have to be encoded with light on-off patterns. Maybe this is what you meant when you said using a matrix, but I was thinking that when pieces were moved further away from one of at least three photo transistors, distance could be measured as a change in detectable luminosity. Then, the location of each piece could be identified as a function of distance from each phototransistor. If nobody knows about the electrically engineered types of phototransistors that could do this – I’m assuming that is what you meant by EE, then I would have to research this myself. Even I don’t wind up using light to measure distance, I liked your idea about using light to communicate, for lack of a better word, changes in status information. Though, a disadvantage of using a light resistant chamber is that this chamber would make the game less compact and portable. However, this model might be more portable than using a camera above or below the playing surface of the game board.

I can’t afford to hire anyone and my programming skills won’t be near the level that they need to be to complete this project in time. However, I do like the idea. Using this model, I might have to copy the game board design onto a transparency. Or, I might have to paint the game board design on a thin piece of pegboard or holy paper. Both of these options may detract from the aesthetical appearance of the game board and decrease the amount of information that could be communicated on the game board. How might this effect detecting the location of the pieces? Could you explain how you planned on the pieces communicating initial status information with bits? It sounds almost as though I would not only need to make a program that would detect the location of light but also the presence of light in time.

In addition to availability, what are some other disadvantages of a resistive sheet? I’ve made a new diagram of how a resistive sheet could be used as a game board, so I’m going to go ahead and post it in this reply.

Attached Thumbnails

 

 

I modeled the schematic that I posted in the 9th December 2008, 11:31 PM thumbnail off of a schematic that must have been using an “x” to represent ground. I thought that the “x” meant no connection. So, if you are looking for information about how to wire chips, please don’t rely too much on that post. I would appreciate it if anyone would correct that schematic.

 

A while back someone mentioned anyone who mapped a distribution of electron flow would be using electromagnetics simulation software. Does anyone know who uses this type of software, and where I could find a simpler example of the type of map that is created by this software?

 

My electromagnetics professor uses this software in his research. Because it's so specialized I doubt it's very attaintable by anyone without a research grant.

 

It's not too hard to model this, (nice graphics aside), what you have to do is solve Laplace's equation on your surface. You will find plenty of references (and code) on the web. I don't think you need to go this far. You can see the sorts of problems you are going to get from a much simpler problem that can be solved exactly. Consider a sheet that is infinite in all directions, and the two contacts are made with conducting discs of radius 'a', separated by a distance 'D'. In this case the resistance between the discs is proportional to

ln(D/(2a) + sqrt((D/(2a))^2-1))

which goes to ln(D/a) for D >> a

What this tells you is that errors in 'a' are just as important as errors in 'D'. So if your pieces are 5mm dia, spaced at 100mm, if the surface is not perfectly flat and they make contact only over the middle 2.5mm, then you will be out by 100% in the distance estimate. Note that this also says that the worst thing to do is to use a point contact!

Also, other pieces in close proximity will significantly affect the result. So unless you can solve the connection reliability problem, I'd look at other techniques, e.g. the capacitively coupled grid option, or optical methods.

 

Do you know where I could see some graphical models from experiments? Also, I’m using a resistor that is about 100 ohms. The length of the resistor is about a quarter of an inch, definitely less than half an inch. What is the closest material that comes in a thin, flat, sheet-like, and longer form in both dimensions that offers this amount of resistance per this distance? What about a material of a similar shape that offers this amount of resistance per foot. I’d like to learn more about this. So, any tables – in addition to the map, that you could refer me to would be helpful.

 

You didn't get my point.

If you must pursue the resistance method, why not make a PCB with connection points on say a 1cm grid, and on the rear have a grid of say 100ohm resistors between adjacent points. You can then work out the resistance between any two points exactly and you can make the contact resistance have minimal effect.

 

You’re right. I didn’t get your point. I didn’t realize that you were suggesting that I design my own experiment. I’d like to find the right material – which I described in my last post, before I design my own experiment. Does electrical resistance of a sheet figure into any of the sheet equations? Or, am I misunderstanding the concept of resistance? Can a material have a per distance measurement of resistance? If so, I’m looking for materials that have a resistance that is closest to 1 x 10^10 ohms per foot.

Separately, I’m also interested in seeing geometric models from actual experiments - which I don’t plan on constructing from my project.

Would you recommend a grid design, in case this is the best design that I can figure out how to make?

 

You are misunderstanding the concept of resistance in a 2D sheet. This sort of material is characterised by resistance per square not per length. What I was trying to point out was that it is not meaningful to say what is the resistance between two points 100mm apart, you also need to say how big the contact areas are (and actually what the shapes are). It is producing a reliable contact that I think will probably kill your design (even if you can find some suitable material). What those equations tell you is that if you have two contacts 100mm apart and pass a current I between them, if the contacts are 5mm dia then you measure a voltage V, but if the probes made poorer contact and were effectively 2.5mm dia then the voltage measured is 2V, giving you a 100% error in distance estimate. Because the current density is so much higher in the vicinity of the probe, you can get the same voltage drop over the last 1.25mm as over the whole 100mm.

If you want to pursue models, if you are good with programming you can write/modify a Laplace solver (e.g. see http://www.physics.utah.edu/~rprice/PMT/ohmic.pdf for some theory), to play with some simple models there are free packages like QuickField software license types that could help you learn (I've never used this, but it does solve these problems).

But to me, the killer issue you have to solve is making consistent contacts.

 

I plan on using contacts that are about the area that the rounded part of some of the tacks that are sold would make with a flat sheet - or the area that a very small clip would make with a thin pin protruding from a sheet. So, the contacts could be weighed down or mechanically attached in this way. I may be wrong, but I don’t foresee either of these methods resulting in a poor contact.

What is the force that is causing the electrons to flow through the sheet? I thought that the force would be, in my case, a chemical reaction in the battery. So, I was thinking that the electricity would flow through the path of least resistance to the other terminal on the sheet that completes the circuit with the battery. But, what you are saying, if I understand you correctly, is that the electrons will be traveling with the same characteristic in all directions outward from the negative terminal. I want to be really sure that I understand this concept because, if the electrons travel in a manner characteristic of what you are saying, then I might not need a sheet with special resistive properties. The resistance would seem to me to decrease both as a function of area and distance according to this theory.

 

I went ahead and emailed QuickField about their products, but I think that what they are offering – a type of mesh analysis software, is different from what I am looking for. Though, I only looked at their free stuff. Doesn’t mesh analysis only have to do with discrete – not continuous, circuits? If so, I’m still looking for two types of geometric models of electron flow. One that can show the resistance of a continuous sheet as a function of area or distance. And another that can show characteristics of electron flow through a two or three dimensional material having an unevenly distributed composition.

 

I found a suitable picture of the current flow between two contacts



This is the solution to a different problem (which is why the lines are labelled E and H), but the equations being solved are the same and so the picture is relevant. In the picture the E lines correspond to current flow (and the closer the lines are together the higher the current density) and the H lines for your problem correspond to lines of equal voltage (with equal voltage difference between adjacent lines). Note that the current density gets very high around the conductors, and the voltage gradient increases. This is what makes the problem very sensitive to the actual area of contact, and what I think is the achilles heel of the design.

If you want to explore further, you can get Maxwell SV from Ansoft, a serious commercial solver that is free, and has no limitations for solving this sort of problem. Maxwell SV

 

Would it be accurate to say that as the resistance of the material increases, the current density off to either side of the wires and the current density between the wires becomes closer? Is this going to effect the resistance in the wire apart from the sheet? Do electrons take a longer time to travel along the longer red curves from one wire to the other, and is this travelling time measurable just using the two wires? What is this travelling time called? What in the diagram would cause a measureable change of electricity in the wire apart from the sheet as the distance between the two wires contacting the sheet changes?