Log in

View Full Version : Accurate scales in an hour


J
April 11th, 2002, 09:23 AM
The APAN thread has prompted me to write this. I made a set of scales accurate to +/- 0.1g in around an hour for virtually nothing.

It consists of an electric motor from an old cassette deck. The two halfs of the beam (just some equal lengths of wooden dowel) are stuck onto the front (oposite end from the spindle) at a downwards angle. The angle determines the accuracy of the scale; the smaller it is from horizontal, the more accurate it is.

A hole is drilled in a length of square wooden dowel around a cm from the top, just smaller than the motor spindle diameter. The motor is pushed in, no need for gluing if you've got a tight fit. The other end of the wooden dowel is screwed onto a base of some kind.

Holes are drilled at each end of the beam, and containers are suspended with thread to act as containers for the weights and stuff to be weighed. I used a couple of plastic scoops that come with tubs of protein powder, with the handles cut off.

A cocktail stick is stuck on to the back of the motor, just below the spindle, and pointed downwards along the wooden stand. With a ruler, a straight line is drawn on the dowel underneath the stick.

When the stick is not directly parallel with the line, you know that the scales are not balanced and to add more weights/chems until it is.

For weights, there have been some good suggestions already. A penny (British) weighs exactly 1/8th of an ounce (3.5g), according to a friend of mine who has a reason to know this <img border="0" title="" alt="[Wink]" src="wink.gif" /> Another idea would be to use .177 airgun pellets, although the weight varies according to the brand.

To make accurate gram weights, you'll need a weighing machine already, but it doesn't have to have the accuracy of this homemade device. I use a set of 'diet scales' with 5g resolution. These are great for measuring out larger quantities of chems, and cost around �5.

Take a length of fairly thick wire (gardening wire will do fine), and keep adding off cuts to the weighing scales until you have exactly 10g (or 100g if you want better accuracy and have plenty of wire). Find out the total length, and calculate the lengths required to make up 1g and 0.1g weights (and even 0.01g if your scales are accurate enough, and you need this accuracy). There are your weights.

The advantage of using an electric motor is that it's a ready made low friction spindle. The tape deck ones are good, because they are large enough to stick the beam to, and they have very low friction because they're designed to be quiet. This is longer lasting than a spring based device (all springs stretch over time), and it's easy to see if it's not calibrated.

If I haven't explained something clearly enough, please let me know.

<small>[ April 11, 2002, 08:25 AM: Message edited by: J ]</small>

nbk2000
April 11th, 2002, 09:36 AM
In an amatuer scientist article, the describe how to use an electric indicator (the needle that swings back and forth on a stereo, etc) to make a microgram capable scale.

The needle acts as a lever, onto which is hung the pan. The amount of current needed to balance the beam (digital multimeter used) is linear, thus allows for determination of weight once calibrated.

Though I know you're just using the motor as a pivot in this case.

A US nickel weighs precisely 5 grams.

A-BOMB
April 11th, 2002, 10:44 AM
I have plans like that too NBK except mine plans use a bar connected to a varible rezistor a inch from the end of the left side of the beam and a spring hooked up to a eye bolt so that when you tighten the eye bolt it increases the resistance to weight so you can weigh increasing ammounts, then you a formula to calculate the weight from a variation of input voltage.

And NBK you remember that pump action zip gun animation you made a while back, what did you use to animate it? Imageready, Photoshop?
Because I'm want to see if I can make a animation on how the grenade laucher I'm building actions going to work because it is too hard to explain.

nbk2000
April 11th, 2002, 01:53 PM
I created all the elements in PS6, then used Imageready to animate it.

<img src="http://www.sas.org/E-Bulletin/2002-04-05/labNotes2/art/scale.gif" alt="" />

<a href="http://www.sas.org/E-Bulletin/2002-04-05/labNotes2/labNotes.html" target="_blank">Cheap $3 Scale</a>

<small>[ April 11, 2002, 01:05 PM: Message edited by: nbk2000 ]</small>

BleedingLips
April 11th, 2002, 06:05 PM
Here is a site you might like to see:

<a href="http://www.rhodium.ws/chemistry/equipment/scale.html" target="_blank">Microgram Scale</a>

<a href="http://www.sciam.com/2000/1000issue/1000amsci.html" target="_blank">Updated version with PC interface</a>

<small>[ April 11, 2002, 11:54 PM: Message edited by: nbk2000 ]</small>

vonK
April 13th, 2002, 05:49 AM
Pure water weighs 1g per ml so couldn't you simply weigh your chems against water.
That's how I weigh everything, using tap water and a balance scale built from a wooden ruler. To tare it I move a match stick up and down the beam untill it's even.
This leads me to another question I have (a bit off topic).
How do you guys measure you acids and move them from one container to another?
If I had access to proper equipment I'd use a pipet but I dont.
I have been using syringes, H2SO4 doesn't seem to attack them but they do stick after a while causing the acid to burst out, splashing everywhere. I haven't been hit yet but it's a little nerve racking. Also I will be getting some HNO3 soon and know that it attacks rubber so I'm going to need a new method.
Got any suggestions?

Thanks
Kurt.

<small>[ April 13, 2002, 04:51 AM: Message edited by: vonK ]</small>

Mr Cool
April 13th, 2002, 06:03 AM
Good info J.
Vonk: just get a measuring cylinder! They cost a few $, and are accurate enough. That's what I use to measure acid out, and to move it from one container to another I just pour...

Celtick
May 24th, 2002, 07:05 PM
In the Netherlands you can order a similar scale as the one from NBK from <a href="http://www.large.nl" target="_blank">Large</a> (it weights accurate up to 1 gram). It is sold for weighing marijuana :D

<img src="http://217.27.2.76/productimg/582017.jpg" alt="" />

TheBear
October 14th, 2002, 03:27 PM
Bleeding lips
The link to the updated version is broken! Anyone who knows an alternative adress?

BleedingLips
October 15th, 2002, 05:44 AM
No problem, TheBear, here it is:

<a href="http://www.sciam.com/article.cfm?articleID=000A7A1E-1FB2-1C73-9B81809EC588EF21&pageNumber=1&catID=2" target="_blank">http://www.sciam.com/article.cfm?articleID=000A7A1E-1FB2-1C73-9B81809EC588EF21&pageNumber=1&catID=2</a>

TheBear
October 15th, 2002, 03:50 PM
Thanks BleedingLips!
Wow, thats a great article, nice!

frogfot
November 14th, 2003, 04:01 PM
Anyone thoat on "upgrading" usual +/-1g digital scale for better precision?
As I heard somewhere, sensor from such scale gives a nonlinear voltage curve which is red as if it was linear, this gives crap precision of 1g. So, if we could sample output from sensor with several known weights, find the nonlinear equation (with Exel :) ) and then use this equation to convert output to grams..

I've opened up my kitchen scales but sensor had a bit more wires than I expected :eek:
Anyway, scale had also surprisingly small ammount of electronics. Apart from LCD driver, there was only one OP amp and some usual components.. so, they most likely function as I described above..

Ropik
April 23rd, 2004, 08:08 AM
Because I work mainly with pyro compositions ("real" explosives are only small percentage of my hobby time :) ) I just use scale which i bought for 5 dollars. It is mechanical scale, 1g accuracy. For this budget not bad...

Thermiteisfun
September 6th, 2007, 02:08 PM
http://www.sas.org/tcs/weeklyIssues/2004-11-12/Classics/

the microgram scale that nbk2000 was talking about, as well as the schemmy and mechanical setup. You can find all the parts in secondhand electronics parts stores for cheap cheap :) . Also, it's very simple provided that you can actually read a schematic. Enjoy the fruits of someone else mind!!


the other links had incomplete information