Sciencemadness Discussion Board

Digital data logging

Leander - 15-10-2010 at 13:12

I´m familiar with analog recorders that can process signals from pH, conductivity or temperature meters and convert the mV signal into a mechanical pencil motion producing a chart. They look like this;



I would love to be able to to log data in the form of mV signals over longer periods of time (read multiple days). Using a mechanical system and rolls of (expensive and nowadays rare) paper to do so wouldn't be my first choice. In addition to that I'd like to measure multiple signals at once, something a traditional recorder isn't capable of.

Is anyone of you familiar with software that can do exactly that? In addition to the logging software, it would require hardware to convent the mV signal into 'bits and bites' a computer can read. I am aware of the fact that systems like this are in fact pretty standard in chemical engineering and probably build in most chemical plants. Getting hold of something simple and affordable as a hobbyist is an entirely different matter. :(



not_important - 15-10-2010 at 13:37

There are scads of analogue to digital boards and external converters out there, these days mostly using USB to talk to the computer. Depending on the pH sensor, an additional hi-Z buffer amplifier may be needed, pH and thermocouples are low level and generally need some additional amplification before the ADC, amps for thermocouples often include the cold junction functionality.

Note that PC sound cards are not satisfactory for DC or near DC work, they're designed for audio work.


Software - might look here : http://www.comedi.org/


Some potential sources are not in the lab area - a 4 channel ADC for automotive diagnostics for example http://www.picoauto.com/automotive-oscilloscope.html Another http://www.bitscope.com/product/


[Edited on 15-10-2010 by not_important]

Xenoid - 15-10-2010 at 13:39

Hmmm... have you tried Googling "data aquisition".

There are heaps of options out there!

Here are a few from my bookmarks folder, though it may be a bit out of date and Mac orientated;

http://www.audon.co.uk/data_acq.html
http://www.labjack.com/results.php?category=2
http://sine.ni.com/nips/cds/view/p/lang/en/nid/14604
http://www.perfsci.com/hardware.htm
http://www.dataq.com/194.htm
http://www.dataq.com/products/startkit/di148.htm
http://www.audon.co.uk/daqplot.html
http://www.qubitsystems.com/Merchant2/merchant.mvc?Screen=CT...
http://www.vvi.com/products/index.html
http://www.mccart.us/ThrustLibrarian/#Software
http://www.perfsci.com/hardware.htm
http://www.lennard.net.nz/electronics/ad.html
http://www.edaq.com/index.html
http://www.macresearch.org/usb_or_ethernet_data_acquisition_...
http://www.phidgets.com/index.php
http://www.gnuplot.info/
http://www.vernier.com/
http://www.vvi.com/products/daq/hardware.html
http://www.hytekautomation.com/Products/IUSBDAQ.html?gclid=C...

aliced25 - 16-12-2010 at 07:37

Make it yourself, texas instruments have a microcontroller board for fuck all, the Launch Pad for $4.30US.

watson.fawkes - 16-12-2010 at 07:52

Quote: Originally posted by Leander  
Is anyone of you familiar with software that can do exactly that? In addition to the logging software, it would require hardware to convent the mV signal into 'bits and bites' a computer can read.
Any real question about this has to deal with two important questions.
  1. What accuracy requirement do you have?
  2. What's the signal source and how are you conditioning it?
These two questions have two related answers: the bit resolution of your measurement and the signal-to-noise ratio (S/N) of your source. It you don't have answers to these, you'll all-but-certainly end up with an unsatisfactory purchase, either having acquired something unsuitable or having wasted money.

bquirky - 16-12-2010 at 08:17

I use one of these

http://www.hytekautomation.ca/U120816.aspx?productId=1

Under a hundred bucks

I use it with Labview and it seems to work fine

Hades_Foundation - 26-4-2011 at 15:38

If you have all the instruments and you only need to log the data on rare occasions, or just to check if nothing unexpected happened, the quick and easy solution (and cheap) would be a properly positioned webcam taking snapshots at regular intervals.

As a bonus, you get a timelapse video of the reaction for no extra charge! :cool:

[Edited on 26-4-2011 by Hades_Foundation]

Ozone - 26-4-2011 at 18:09

Not to fan the flames, but in this day in age, analog data acquisition will be more likely to survive EMP and the like. As far as this goes, the good old strip-chart is hard to beat. For long times, just turn the chart speed to the lowest setting.

Otherwise, I've made quite a few suggestions for El-cheapo-DAQ here.

Cheers,

O3