I am looking into the feasibility and any recommendations for building an arduino prototype to record and log data collected from strain guages which will be measuring aerodynamic down force of specific elements in a wind tunnel. Eventually my goal is to produce a purpose built unit.
What constraints am i looking at for sample frequency?
Should I be amplifying the signal for the arduino to increase accuracy?
Should I be amplifying the signal for the arduino to increase accuracy?
Amplification is absolutely required. Instrumentation amplifiers are used, google "strain gauge amplifier" for more info.
Hi,
Look at this.
It uses an amp and AtoD.
google arduino loadcell
You will be surprised at the amount of articles around.
Tom.... 
Strain gage amplifier circuits need to be approached carefully. The actual voltage change over the typical usable range of a gage attached to a solid is actually quite small, on the order of a few millivolts. So it is not uncommon to find amplifier gains in the range of X1000 or more. It really depends on the particular gage being used, so you should choose your gage before you nail down the amplifier gain.
It is necessary to match the amplifier circuit to the gage in order to maximize the sensitivity to changes in strain, especially if you are going to use the 10-bit A/D converters on the Arduino. Keep in mind that you are only going to have 1024 possible strain values, although realistically, you will probably only see 600 or 700 in actual use. If that is not enough resolution, then you would need an external A/D converter and multiplexers, or multiple A/D converters in an SPI chain.
Since the signal is so small, it is highly recommended that you post-filter the signal just before the A/D converter. Signal noise is a big issue, made worse by long gage lines, unshielded cables, etc. I've seen cutoff values as low as 1 HZ in strain gage DAQ signal conditioning modules.
If you are trying to measure highly dynamic signals, the circuit design gets very complex trying to isolate the good signal from the noise, and is not the sort of thing that I would recommend as project to take on. You would be better off purchasing a purpose-built DAQ system that supports strain gage measurement out of the box.
Thank you for the help.
I think it may be very difficult to use an arduino for this then. The signals will be highly dynamic and i think we will produce a lot of noise through vibrations.
I will be using an NI DAQ and signal conditioning for this application.