RFXCom and BWR101 / 102 Body Weight

[Richard Naninck]

Recently I bought a BWR102 body weight scale from Oregon Scientific. The RFXCom receiver is able to receive the weight data from the scale so I created a theme panel and script which which keeps track of family weights and stores the data in an sqlite2 dbase. Combined with (childrens) age and height the BMI (Body Mass Index) is calculated and displayed as well. A BMI calculator is also embedded in the theme and script. So if a family member measures his or hers weight and a match is found in the dbase within 4% of the previous reading of that person, the new weight is stored in the dbase and used as ref for the next measurement.

If anybody is interested, let me know.

[Osler]

Very interesting. How reproducible are the measurements? Can you plot measurement 1 vs. measurement 2 for a few family members (measurements separated by a moment in time, say seconds or minutes) and let me know what the r^2 value is? If not, can you get me the data points and I'll run the stats?

Osler

[jacco van der Ven]

If somebody will buy an BWR102 body weight scale from Oregon Scientific, I do have them in stock.

I place the measurments in an txt file and GNUplot will make an graph of it.

Regards,
Jacco

[Richard Naninck]

I bought mine from Jacco. New in a box

@Osler:

My dbase holds somekind of an epoch date per stored value and the date resolution is to the minute. My script can run various queries producing Day, Month, Year - average, highs and lows. The childrens age and length / height is also stored in the dbase to be able to recalculate the BMI in the future. BMI is age and height dependant so these values are important over time.

Right now I don't have very many measurement stats so I don't bother about graphs. That becomes more interesting over time I guess. Since everything is stored in a dbase, I can do whatever I like with it in the future.

Not sure what you mean by the r^2 value.

[Osler]

I am a scientist that studies obesity and its related complications. I run a variety of intervention studies and it would be nice to have something like this to give to study subjects to take home to follow their progress. The r^2 (r-squared) is the correlation coefficient and would give me an idea of the accuracy of the device for repeated measures over time (i.e., how similar are 2 measurements seperated by an hour among a group of individuals of different weight?).

Osler

[Richard Naninck]

Interesting..

The BWR102 is a simple scale like any other but it has a 433MHz LCD unit in which you can set the length of 5 persons and therefore is able to calculate BMI. Since RFXCom receives the 433MHz message coming from the main unit, HouseBot can do whatever the LCD unit does and better. It measures in 100 grams (so 88.3 kg) and the BWR measures the exact same weights as my old digital scale which I actually correlated using 20KG weight lifting weights.
I programmed a +/- 4% difference from ref but that will become problamatic when one person (my son in this case) overtakes the weight of the other person (my slightly older daughter). However I do see benifits from this setup for your sience program combined with your hobby;)

I can't really answer your r^2 question, but maybe online docs about the BWR @ Oregon Scientific can.

[terrychen]

The brand for this weight scale i did not hear before, i would like to check it as i need one weight scale