How do I monitor an angle?

the fastest way to get this going would be to fix a 50k to 250k linear pot to the turntable and connect it to the resistance input of a data logger. i did a quick search of farnells and found a pico technology USB DrDA VERSATILE DATA ACQUISITION unit. goto au.element14.com/data-acquisition_data-logging

the pot wont quite do 360 but start messing with optical position sensors and you might get bogged down

hows that

You can buy 360 degree turning potentiometers, or you could use encoders, accelerometers or a simple led/detector and place a grey scale strip of paper around the turntable base.  You set up the scale so that it gives separate values for each degree.  Just about any graphics program will print you a strip to the correct change.  You could also use tic marks on a paper, but you need to have a zero angle point and keep active track of going right or left to keep track of your position.

Basically, you just need to define how much accuracy you need and repeatability you need and then find a solution that works.

Engineering 101.

DAB

the grey scale strip is a good idea. i have developed a simple circuit to drive a led / opto-transistor reflective detector. it is reasonably immune to ambient light and has high sensitivity. ill give details if anyone is interested. (although there are plenty of chips out there that will do the job nicely, and can be hooked directly to a cpu / mpu)

I would use a liner rotary potensiometer or some encode. Potensiometer is much cheaper. It would be good to use multiturn one. ten You'll ne able to mesure all 360 deg. As for fizical limitation can use something like spring with limiting sides. and a slider or a stick sliding just 360 deg.  do not know how to explane it
I have drawned a ruf skech... As you see I'm not so good with paint

By the way for mesuring the angle I'd use arduino.

Hi DAB,

I think I liked your idea about having a grayscale ring (I'm attaching a file I created based on how I interpreted your reply).  I do have a follow-up question: If I at some point put a night-vision camera on the turntable, would the grayscale ring approach still work?  If not, maybe taking the linear potentiometer approach would be the better route.  The biggest problem I'm seeing right now with the linear pot (as described by Peter and Raimondas) is that I'm still unclear how to attach it.  My current design has a vertical "drive" shaft coming through the center of the turntable and it's getting in the way of how I thought the linear pots would attach.  Maybe I should just order a couple different kinds and play around with it.

Thanks.

You mean you have connected a turntable directly to the motor? What type of mottor is it? You do not use any dears? you could use some gears to make the potensiometer turn when the table turns. And then you can ujust gera ratio (size ot table gear / size of potensiometer gear) to get the maximum acuracy of your potensiometer. For example if you have potensiometer that is turning less then 360 deg, just 270 deg for axample. then you can chose the gear ratio to be about 0.7 then 360 def of table rotation will turn 252 deg for potensiometer. And you'll have spare 18 deg not to breack it. and still have a good resolution.

As for connectin potensiometer and mesuring its resistens, you can use arduino analog input. And if you controll talbes rotation motor with arduino, then it will be perfect.

So what  will control the turning motor and what wil send comands were to turn? Is it pc over usb or what?

By the way, do you know how to connect potensiometer for mesuting it?

If you set up a LED and light detector over the ring, it would work in any lighting condition as it is only focused on the ring.

You determine position by measuring the amount of reflectance from the ring based upon the level of grey that the detector sees.

You can use gears to turn a potentiometer or an encoder to measure movement.  It all depends upon how much you intend to move and how accurate you need to know the position.

Attaching the encoding device depends on access and resolution.  Most application would use the encoder on the primary turning access with a reset when you go 360 degrees.  However, you can use a series of gears or other connections and derrive your position from the movement of the table.  Again, accuraccy and repeatability will drive you to the best solution.

Have fun,

DAB

Yes it depends on how acurat you want it to mesure. As for axample for this robot:

For hand position we ussed 2 regular potensiometers. It is not even a liner potensiometer, but still does the trick.

Yes DAB, define your resolution and develop from there (the three D's - define, define, define). Back to the point of ambient light. Let's assume a high resolution. Using a grey scale ring (presumebly going from black to white along its circumference) and a calibrated reflective led/opto-transistor unit with an optimal focal distance of 5 mm, we will need an interface that will produce an analog voltage that represents the amount of light reflected from the ring (not necessarily linear, it can be calibrated using a lookup table) that will be run into an A/D converter (say 16 bit). Now lets assume our opto unit is sufficiently focused to detect changes in light down to 64,000 parts over the rings circumference (how many pixels does the focal area cover and how many pixels are on the ring). The problem is: changes in ambient light will add an unknown constant error to our voltage, and our reading, and may saturate our opto-transistor under extreme lighting conditions.  If we use a simple resistor / common emmiter setup, we can increase our gain and sensitivity by increasing our resistance, but the transistor will saturate under lower ambient. Or, we can minimise this effect by using a lower resistance, but we sacrifice gain / sensitivity. Saturation has been my biggest problem. In fact, manufacturers go to a great deal of trouble to account for / remove these effects in their interface circuits.

Of course, there are several ways around this:

1: Use a pot. Simple, but they are prone to wear and tear and dirt. Also, start messing with gears and it can get messy.

2: Enclose the unit in a light proof enclosure. Good option, but this adds to the mechanical complexity.

3: Buy a matched detector / interface from farnells. The simplest, but where's the fun factor.

4: Develope your own interface. I have done this and it is a fantastic learning experience if you see it through (it took me years, and the deeper I looked into it, the more satasfying and rewarding it became. I got a real buzz every time I found an improvement or a more elegent solution).

By the way, feedback is always good and my ratings go up (LOL), keep it coming! (cool robot Raimondus)