[dba-Tech] Has anyone kickstarted?

[John W. Colby]

Visible spectrum analysis of lights (LEDs, florescent etc) is widely 
talked about (and done) in the aquarium community. Basically shine a 
light of interest on a spectrum analyzer and measure the strength of the 
light at various frequencies.  A kickstarter project built an 
inexpensive analyzer using a diffraction grid plastic, a shaped (dark) 
box and a cell phone camera.  Use the cell phone to run software to 
display (and graph) the resulting spectrum.

http://publiclaboratory.org/wiki/spectrometer

Measuring various properties of aquarium water (inexpensively) uses 
measured vials of water (5ml), adding various "number of drops" of one 
or more titrating liquids. You then compare the resulting liquid color 
to a color card to determine the amount of the substance being 
measured.  You can buy kits which measure PH (High and low), Amonia, 
Nitrite and Nitrate.  The process involves taking 4 water samples. then 
dropping the correct number of drops of various things into each sample, 
comparing the resulting colors to an included card.

For anyone who has actually done this... it is an inexact science to say 
the least.  The color cards contain shades which are difficult 
(impossible) to really accurately gauge when holding the tube against 
the card.  As an example, the Amonia test is really yellow turning to 
green, but it is basically impossible to tell which of three shades your 
tube matches.  In this example, any three adjacent shades are close 
enough together (to the eye) that "who knows".  And yet for example, one 
is 1 PPM (borderline bad), the next is 2 PPM (bad) the next is 4 PPM (do 
a water change NOW, do not stop for dinner).

So all you really get is a range.  Unfortunately one end of that range 
is non-toxic, the other is toxic as hell.

So...

Take the tubes of water, beam light through them and measure the 
resulting color intensity with a spectrum analyzer using a camera and 
software.  If the light intensity is known and stable (a white LED), the 
sample is placed in a dark chamber so only light from the LED can pass 
through the sample (and not leak around the sample), the distance from 
the sample to the diffraction box is stable, and the distance from the 
diffraction grid to the camera is stable (and doesn't leak light) then 
you should be able to accurately and reliably measure intensity and 
color of light through the sample.

Reliable analysis.

If you can automate the process, you get automated reliable analysis.  
Twice a day (week, month) measurement of all measurable parameters.  
Throw in water temp.  Throw in logging.  Throw in Wifi and a browser 
interface.  Heck, throw in automated dosing of chemicals to correct the 
issue.

The first part (measuring the color) is relatively simple. Automation is 
not.

I am ordering the USB Desktop analyzer and will then build a little box 
to hold the sample vials such that light is forced through the vials 
(not around) and a powerful white LED as a light source.  I will then 
test the measuring concept with this widget.  If it works then I can 
reliably measure my own aquarium water.

If it works, that is where kickstarter would come in, commercializing 
this.  There is a pretty large community out there of folks doing 
aquariums, even high end aquariums.  Lots of people, lots of money.  
 From what I can tell, and I have done a lot of reading, everyone just 
uses the liquid tests.  So selling into a high end market where people 
already use and are comfortable with these tests, but providing really 
accurate measuring, logging etc seems like a viable business.

John W. Colby

On 3/12/2015 12:28 PM, Jim Lawrence wrote:
> Hi John:
>
> Sounds exciting...so maybe you can give us a hint of what type of project you are thinking of?
>
> Sorry, no experience other than contributing occasionally.
>
> Jim
>
> ----- Original Message -----
> From: "John W. Colby" <jwcolby at gmail.com>
> To: "DBA Tech" <dba-tech at databaseadvisors.com>
> Sent: Thursday, March 12, 2015 7:41:23 AM
> Subject: [dba-Tech] Has anyone kickstarted?
>
> Has anyone ever done a kickstarter project?  If so email me off line. Or
> discuss general details here of how it works.
>