I have an idea - sort of a reverse of the timex datawatch transmission sytem.
It's to have a sort of cradle into which a stopwatch is placed, and a mode where the watch transmits its stored info by rapidly switching some of the LCD segments, which are read by the cradle.
How fast can you switch an LCD segment, and would power used versus speed be an important trade-off (watch power is usually some smallish lithium cell; the cradle power isn't a factor).
Which kind of hits on a project I'm currently developing at work.
Using a video recognition system to look through an object of rotating wires being cabled with a silicone grease dispenser high pressure pump that files the cavity in the center.
Using optical targeting systems work 75%, but we're trying for a
100% target acquisition to determine, if, the high pressure pump system needs to replenish the silicone as it's being consumed by the water blocked cabled product that is being grouped, with out waste.
You can imagine the cost of materials and profit margin which drives to development of these devices.
It's been 25 years since I've gotten into a project like this, the last time I used camera's in this fashion was with a Z80 computer based system using a B&W camera with high res to perform a line inspection of those large squeeze bottles of the "Hershey Chocolate" bottles to determine if the label was correctly applied on the bottle and reject it, if not. Because the bottle and Label are both the same color and the only variations were the image of data on the label itself, in Silver, made for a tricky out come. But it was successful. And this was done in the computer language "FORTH".
Of course, with today's technology, this project is being done in Windows using delphi as the front end. All is coming along just fine. We're planing on using a mini Windows XP panel screen to operate this when completed.
So, if you are a computer jock, I guess you could use a LCD web cam to look at the watch and then acquire your information! :) OCR
The kinds of LCD used in battery powered watches are slow, gonna take a tenth of a second or so. Other types of modulator are faster (even moving-mirrors, but especially quantum dots...) but aren't ready for low-cost consumer items.
It's possible to parallel a few dozen segments, of course; a tenth second for a 32-bit parallel port, readout by a low-resolution camera...
Didn't the Timex gizmos use a photovoltaic cell? You can power an LED with a cradle that beams light into a PV cell... and maybe use IRDA protocols.
Can the watch draw power from the cradle? Maybe via a pair of contacts, or perhaps some inductive transfer scheme. If so, it doesn't have to rely in the little lithium cell, and watch power isn't a factor either.
There is a second limiting issue, if you want to use some CMOS camera to read the data, and if it does say 30 fps, then Niquist tells you you cannot flip segments faster then 15 x per second else you get aliasing.
Say you have 4 digits, seven segments each, that makes 28 segments, at 15 changes per segment per second, makes 420 bits / second max, you need to be a bit lower, so say 300 Bd is the maximum you can ever do that way.. You could use photo transistors with lenses looking at each segment to up speed, but that would be really complicated mechanicaly. But I think it is a fun idea. Maybe a faster CMOS sensor? Those do exist.
Let's Say you could use a Bandwidth of 10Hz per Segment (mainly limited by the LC Material). Each Segment should give a S/N of 100Db at this low rate. This means roughly 150 Bits/s/segment =3D 150*28 =3D 4200 Bits/s
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.