At 01:41 AM 26/05/2013, you wrote:
>I was weighing pills and tried this setup to improve the scale sensitivity=
..
>See photos here =3D=3D>=20
>https://plus.google.com/u/0/photos/102396176247280453443/albums/5882154251=
101853345?hl=3Den
>Max is 5000g says the scale legend.
>I put a hinge at one end of a bar and clamped it to a book.
>I put a styrofoam cup at the other end of the bar.
>I used a marker as a pivot point.   The bar does not touch anything else.
>
>#1  I used the tare function to zero the scale while the cup was at the en=
d.
>#2  I measured in 9 pills and weighed them. ( 97 grams )
>#3  I added all the pills ( 90 ) and weighed them.  ( 969 grams )
>
>969 /97 * 10 =3D .999   That was the best result I had.  My worst was=20
>1.28 when I was measuring smaller pills.
>
>What would be sources of error ?
>#1       the scale     I did not use the calibrate function since I=20
>was doing ratio-metric comparisons.  Maybe cal would have improved things.
>#2      pressing on the hinge where it was clamped to the book=20
>changed the amount measured.  I do not understand this.  I waited=20
>for the scale to settle down and it was still showing a reading as=20
>much as 5% off.
>
>As Napoleon Dynamite would say...... * You idiot*
>
>Gus                         keywords  99centech  99scale   99pills


You can buy a 30g full scale digital scale for under $25 (0.001g resolution=
).

Cheap scales are going to have friction/hysteresis and nonlinearity. Expens=
ive
ones will too, but it will tend to be less. ;-)
As others have said, it's hard to make a scale that's accurate over a wide
range, and about impossible to do it for cheap.

0.1% of full scale is not bad for a Harbor Fright product.

--sp

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.