> It seems that a "naked" strain gauge or pressure transducer will run $50 -
> $100 when purchased as such, out of an electronics catalog.  Meanwhile,
I'm
> looking at my digital bathroom scale that I paid $25 for, and thinking it
probably
> has a strain gauge in it (and hopefully accessible leads or traces where I
> can siphon the raw voltage signal).  Also, possessing a usable "midsize
motor"
> load range right out of the box.
>
> Any of you folks ever played with this el cheapo
> consumer-technology-crossover possibility?

Modern digital kitchen and bathroom scales often by not always have load
cells in them. The method of implementation varies widely. The kitchen
scales are liable to have superior cells both because the required
resolution is higher and the small cell tends to be cheaper.

I have been considering the possibility of "mechanically amplifying"  a
small cheap load cell. At its simplest this could be a lever arrangement so
that the load cell sees only a small fraction of the total force. Maybe also
a low deflection beam of suitable rating that provided suitable deflection
for the target load cell. Issues such as (but not limited to) stray
resonances, hysteresis and bandwidth would need to be considered. Large
cells to me seem to be unnecessarily expensive compared to small ones.

Bathroom scales that I am aware of include

- Single strain gauge on each of 4 points, each  on a short stubby beam.
Between them they may make a full bridge of sorts

- Single strain gauge on two long metal arms running front-back on each
side. An inferior solution that may make a half bridge

- Some scales still use a mechanical mechanism with a rotating optical
encoder disk. Essentially useless.
_______

The two kitchen scales that I have dismantled both use a true load cell and
are much closer to the sort of starting point one would want. Nicely enough
built single point parallelogram unit in each case. Both out of China with
no manufacturers name on them. The cheaper of the two (a 2 kg unit) had a
5th strain gauge at right angles on a thicker part of the web to allow
temperature compensation. Very impressive for a unit that I bought complete
for about $US15 on special (usually about $US30 ish). Very nasty electronics
construction and assembly though.

The other is a 5 kg unit ($US40 usual price) with a full 4 strain-gauge load
cell and a single PCB for all electronics. Much better designed and
assembled. Temperature compensation probably provided by processor doing its
own local temperature measurements.

Both scales are able to resolve steps of 2g in a 2kg or 5 kg range. I have
plotted accuracy and linearity but haven't got them immediately to hand.
Accuracies of say 0.1% should be easily achieved.

Both scales use a single LM324 quad opamp to make a strain gauge amplifier
and voltage to frequency converter for the processor to read. Very cheap and
nasty approach that is evidently good enough for the target market.

I'd say that for most rocketry applications that one of these load cells
with some form of force divider would provide a very acceptable solution.



        Russell McMahon

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu