A Salinometer Built For The Science Olympiad

This is a Digital Salinometer which [Daniel Kramnik] built as a Science Olympiad entry. He’s a Junior in High School and when looking for a project to enter into the Water Quality event he was interested in achieving greater accuracy than a mechanical hydrometer provides.

We think the circuit design is very impressive for anyone who hasn’t complete formal training as an engineer, and outstanding for someone as young as [Daniel]. Measurements depend on two main parts, a temperature control and a salinity sensor. These are both necessary because fluctuation in sample temperature will affect the salinity reading.

A Peltier element is used to heat the water sample if it doesn’t fall within a set range of temperatures. From there, an Op-Amp circuit conditions a signal running through the sample, passing an output to the ADC converter chip which drives the three-digit readout. [Daniel] calculates an accuracy within 0.0014%. He must be on the mark because he’s won his regional competition and will soon compete at the state level.

13 thoughts on “A Salinometer Built For The Science Olympiad

  1. Science Olympiad (S.O.) is an academic competition which is coaching intensive. While this student may have done this themselves, it is very likely that they had “assistance” along the way.

    1. The judges were pretty strict about that 20 years ago, Nick, I was certainly run thru the ringer WRT my little homemade current meter, which was much less elaborate than this version. Theory of operation, circuit diagram, explain the ohms law implications, etc. I was even required/suggested/demanded to provide receipts for the electronic parts I used, to prove I was under the $50 limit. I’m thinking they’ve removed that limit, after looking at this kids project.

  2. LOL I did the same thing over 20 years ago and WON at regionals. I just used a 9 volt battery, a 1 mA analog meter, a variable resistor, and homemade probe made out of a plastic pen and a cheap plastic radio shack box. We had a constraint of $50 or less or something like that so I think this elaborate design would not have been possible. I had our teacher/advisor verify they would use distilled deionized water for their solutions so I assumed zero current flow is 0% and they provided a saturated solution which we adjusted to 100% = 1 mA on the meter. Frankly I was shocked everyone else did some mechanical floaty thing when I got there.

    They provided calibration solutions, graph current vs salinity, all done. Was pretty linear as I recall.

    As a chemist I find his claim of ppm level accuracy HIGHLY unlikely. Its probably a sig figs mistake. It would be interesting to see his calibration data at the ppm level. At the ppm level I’m pretty much certain dissolved gasses would be a serious issue, surface contaminants on his probe would be a problem, etc.

    1. Your concerns about contaminants and dissolved gasses are valid, but assuming an ideal case, this can measure to +/- 5ppm, which, at 359 g. NaCl/L H20 saturation, translates to roughly the percentage I calculated.

      I guess I’ll go add something about that in the description, there’s no way it’ll ever actually measure that accurately.

      1. There are three mostly independent quantities to keep in mind that might be getting confused with each other (or at least, I’m confused)

        1) The number of digits in your readout hardware, I don’t think anyone is arguing you’ve got 3 digits.
        2) precision is how wide the standard deviation of your results is, which hasn’t been discussed other than tangentially I mentioned surface contaminants are going to limit your repeatability at the ppm level unless you’ve got good lab technique, which you may have, but you’d need to calculate your stats on multiple runs to prove.
        3) accuracy is how close you can measure to the true value which is what I was talking bout WRT ppm accuracy being a bit .. optimistic.

        Colloquially people randomly interchange the three concepts, but they’re really quite scientifically distinct.

        Don’t get me wrong, high five dude, its a good 9.99 on a scale of 1 to 10, if not better, I’m just pointing out the next 0.01 requires a bit of time with a statistics book and some NIST level calibration runs if you want to claim ppm. A pretty good and short intro to error analysis is a book by Taylor, “intro to error analysis”.

  3. Wow.

    Best I did at regional levels was take first in the new ‘astronomy’ category. We had only to name the planets (Pluto was still a planet back then…) and answer a 10 question T/F.

    Also took second in the Experimental Design category. Which was about making an experiment from crap in a box. Kids these days get to do the cool shit.

  4. I second what Nick had to say in the first comment but only in general. While this project didn’t have “assistance”, many many others go way beyond just assistance. And for the record, I do have personal experience with this. Funny how the schools with the most money always win, especially when some of them hire professionals to design projects for them. Schools can also pass previous years projects on to other students/teams to present as their own. It is not a fair contest in the least. I have corresponded with Dr. Putz (founder, and yes, that is his real name) regarding the rules. My suggested revisions were rejected and I was told that it was too strict. I suggested that each student sign a document the day of the event saying they completed the work themselves *rejected*. I also suggested that contest were made simpler and constructed/built that very day on site, *rejected*.

    Bottom line, is that it is nothing more than a money making machine designed to rule out the less fortunate.

  5. is there a simpler way to make a salinometer? me and my partner are doing the same project for science olympiad amd i was just wondering what the main parts of salinometer are and how it works? please answer

Leave a Reply to Vince MulhollonCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.