Controlling a moving stage via Labview

[pannepanda]

Hello everybody,

I have some basic Labview knowledge and I would like to apply it to something that I need to build. it is basically a stage that should move in 1D. On top of the stage are solar cells (very small ones) that get illuminated from the top. I want the motor to move (about 5cm) then stop measure the cell (output current), save and go on. I thought that lego might be sufficient and allow a relatively cheap accomplishment of this job. I had a look online but found mostly robot sets, while I would only need a fraction of what is included there. Could you maybe tell me what I would need to do the job. I am especially unsure about the interfaces between Labview and motor, also compatibility of Labview with Lego (I read the new models can not be programmed in Labview anymore). I have the measuring devices for the cells all set up, I am just unsure about the engineering part (which motor is suitable, do I need an interface, etc.)

I am thankful for any advice!

Ralf

[h-g-t]

If the NXT is carrying out the recording, think you would only need the NXT, one NXT motor, some wheels and axles, a couple of leads, a few bits of LEGO and some scrap timber to make a track.

What measurements you are recording would determine whether a Lego sensor (bought or home-built) would do or whether you need an I2C interface. Can't really say without further details.

The programming language provided with the NXT should be sufficient for for your purposes but NXC would probably be better.

If you search for Lego camera dolly you will find lots of examples, although your rig will be simpler since you don't need to make it pan or tilt.

Basically just a powered cart programmed to pause and take measurements.

If the measuring gear is self-contained and also carries out the recording, an old RXC with one motor and a rotation sensor would probably be sufficient (and cheaper).

[h-g-t]

Info on datalogging using the Lego NXT language here -

http://www.teamhassenplug.org/NXT/DataL ... ogging.htm

And DIY sensors here -

http://www.extremenxt.com/lego.htm#sensors

[pannepanda]

First of all thank you both for your answers! The IV-measurements (potential-current) are done with a separate device that we already posses, fortunately the manufacturer provides the Labview software script for free. Therefore, I would like to program my LEGO bits also with Labview instead of the LEGO provided package. Ideally, I would only have to add the command to move the stage before the IV-measurement command (provided). Would this now mean that I will need a I2C interface to send commands to the LEGO bits with Labview?

The other thing is I would need to use a multiplexer, connected between my cells and the IV-measurement device. That is critical because I don't want to switch cabling after each measurement. Does any of you know a cheap way to switch between the signals? Standard Labview equipment

I hope I gave a proper explanation of the problem. It is basically a stage that should move every few minutes, take a measurement (instrument + Labview code for a single measurement already existing, not the multiplexer though), save and repeat the whole thing. The company that installed the initial setup (1 device measurements only) wants about 6000 GBP for the setup. I hope that I can do it for a few hundred GBP with a bit of LEGO and creativity.

[h-g-t]

Perhaps you could post a photo of the measuring device, preferably with a ruler in it to give an idea of the size?

[pannepanda]

I haven't build the stage yet, because I need to get an idea of the total costs and time that I would need to put in. But it will be around 25 cm long and about 2.5 cm wide with about 10 devices on in (each about 2 x 2 cm^2). Building the stage should be fairly easy with a bit of LEGO and a laser cutter. I am unsure at the moment if I could use the normal Labview to program a LEGO engine. Also If there is a multiplexer (maybe even from Lego) that can do a 30 to 1 signal transduction would be critical for me at the moment. I can however try to sketch my ideas and upload it.

[pannepanda]

here is a sketch of the setup, as I want it to be. Please excuse my poor drawing skills, I never learned proper drawing, therefore I tried to keep it simple and sketched only a 2 cell setup. In the end I want to have 10 cells, but the basic setup is identical, only with a larger multiplexer.

It should illustrate the substrate stage at the bottom, which can be moved by the motor. Above one cell is the light source, this is the cell I want to measure. That is why the multiplexer picks anode and cathode from the illuminated cell. The signal gets processed to the IV measurement device and after the measurement, the data will be saved.

I left Labview out of the sketch, but it needs to control the motor, the multiplexer and the IV-measurement device.

I hope this made it more clear, I am helpful for any advice!

[pannepanda]

can anybody help me with this? I am really thankful for any advice!

[h-g-t]

Sorry, but I have never used Labview and know little about multiplexors, although 4 seems to be the limit 'off the shelf'.

Perhaps you could fit one to each of 3 ports or 'daisy-chain' them - you would have to check with the manufacturers.

Please note that the makers do not supply info on using their devices with Labview, you would have to search the internet or contact them to see if it is possible.

http://www.hitechnic.com/cgi-bin/commer ... ey=NSX2020

http://www.mindsensors.com/index.php?mo ... AGE_id=179

Only Mindsensors seem to have a compatible voltage sensor though

http://www.mindsensors.com/index.php?mo ... AGE_id=162

It is possible to have more than one NXT in a robot though, I believe that you can have one 'master' NXT controlling up to four others by Bluetooth so maybe you could get round it that way. One controlling the movement then telling the others to read the sensors?

Again, I have never done that but if you search the forums here you will probably get some info, or read LEGO Mindstorms NXT Power Programming: Robotics in C by John C. Hansen (Author). This refers to programming in NXC rather than Labview but contains lots of information about the NXT.

If you Google 'labview lego master slave' that comes up with a few websites which might help.

[h-g-t]

Don't know if you have seen this http://www.tufts.edu/~rwhite07/PRESENTA ... TS/NXT.pdf

Should also have said that the Lego Education version of the software is designed for datalogging, but you have to pay for that.