Brent Brown wrote:

> Hi,
>
> I'm working on a new project that will be using 4 load cells to
> measure the weight of a piece of machinery during operation.
>
> I've seen weighing platforms that use a load cell on each mounting
> foot - presumably they just average the four readings to get the
> total weight eg. (A+B+C+D)/4 ?
>
> Can I minimise the electronics by putting the outputs of the four
> strain gauge bridges in parallel (maybe with some series resistors)
> and then just use one bridge amplifier and one ADC input on my
> micro? (Rather than 4 x bridge amps, 4 x ADC's)
>
> Would this affect linearity much?  My dynamic range of weight is
> very small, I'm looking to measure 0-20kg change (product) on top
> of a constant 500kg or so (the machinery itself), and 0.1 or 0.2kg
> resolution would be fine enough.
>
> Also, what are your favourite bridge amplifier's for a job like this?
>
> Hope these are simple questions to answer for people with
> experience here, and interesting for the rest of us!
>
> Thanks in advance, Brent.
>
> Brent Brown
> Electronic Design Solutions
> 16 English Street
> Hamilton, New Zealand
> Ph/fax: +64 7 849 0069
> Mobile: 025 334 069
> eMail:  brent.brown@clear.net.nz

  If you use one signal conditioning instrumentation amplifier and mux
the inputs you will minimize your electronics error.  You may also reduce
your error by only applying the bridge excitation voltage for the
sampling time.  For example, enable the excitation voltage (better yet,
constant current) select the channel, start the A/D conversion --hold the
excitation on until the conversion ends, disable the excitation, massage
the data, select the next channel and repeat the process, etc.  This will
allow you to use a higher excitation voltage for better S/N and avoid the
I-squared-R error that would normally accompany the High Ex-Volt.

Regards, Richard