- Joined
- Sep 13, 2018
- Messages
- 360
- Reaction score
- 333
Neptune Spear Auto alkalinity tester
youtube.com
I've been wanting to build my own apex based auto alkalinity tester for quite some time but the complexity of programming always left some weak points. I've owned an Alkatronic for a year now and while it's a great, well put together machine- I feel it leaves a lot to be desired. Tests can't be scheduled at specific times, manual tests require opening an app and connecting to blutooth, sample size can't be adjusted, it's noisy and a few other software based issues I never liked. This is in "Beta testing" and there's still a few scenarios I need to tackle- loss of power mid test, running a manual test that overlaps the normally scheduled test and overflow mitigation/detection. I'd love to work with someone who'se experienced in creating 3d print files for an all in one base to hold the vial, mount a magnetic stirrer and catch overflow water with an area for a float switch. Also potentially the upper end to hold the fluid lines and probe.
Alkalinity is defined as the amount(millequivelents) of standard 0.1N acid required to lower the pH of a 1 liter sample to 4.50. I used Randy's article as a starting point for this project. I extrapolated his chart using a 1 liter sample. Through trial and error I found the best combination of sample size and acid dilution for something that could be done with dosers and minimal water. I settled on 1 part 0.1N acid(HCL or Sulphuric(0.05m) diluted with 4 parts RODI water along with a 60ml sample size. This allows for roughly 0.09DKH between 0.1ml titrations. For simplicity's sake I rounded up to 0.1DKH. If you want higher resolution- 0.05DKH you can double the sample size to 120ml. provided you aren't using muriatic acid, waste water CAN be added back to the system though I typically don't. Amazon carries 4L bottles of 0.1N sulphuric acid for $39 for an easy straightforward 1:4 dilution that works out to ~ $0.02 a test. If you would like to have cheaper or larger tests(auto water change) you can use 1.0N HCL diluted with 9 parts RODI to create a 0.1N "concentrated" solution. works out to 1/3 cent per 60ml test.
Early attempts were too complex when attempting to "count" the amount of reagent used to reach a pH of 4.50. I almost took a different direction with the project- adding a set amount of reagent that correlated with my desired alkalinity. if pH was above 4.50 that means my alkalinity is higher than desired and dosing needs to be slowed down. if at or below 4.50 that means my alkalinity is at or below my desired setpoint and I can begin ramping up dosing. I gave the titration method and specific DKH one last stab and I feel I've nailed it. I've tested it alongside my alkatronic- 24 times a day for over two days and it operates flawlessly giving identical readings. I've done about 10 tests where I throw it a curveball either adding an extra .1-.2 ml of reagent or 1.5-3.0ml sample water and it's reacted as expected lowering the reading 0.1-0.2DKH. As it sits, resolution is 0.1DKH and accuracy(compared to the tronic via app) is +/- 0.05DKH. Since the Alkatronic sends the signal to the apex in 0.1DKH increments; the testers are equally precise.
The testing process is triggered by a virtual output named [TEST]. As it sits, tests take 8-9minutes so I opted for a simple line of OSC code turning it ON for 10 minutes every hour. this can be adjusted for fewer tests or to test at a certain time of the day. A VO [MANUAL_TEST] sits on my dash for testing purposes. running a manual test minutes before a normally scheduled test can leave the probe and vial dry. Work still needs done there. Upon triggering a test the alkalinity detector VO's reset to the lowest level(9.0). I haven't tackeled automated tank dosing yet but it'll be important to have dosing off during the test. I have two BRS 50ml pumps immediately kick on to return the storage water to the sump. The dos head drawing sample water runs a +30ml profile to clear the "old" water from the sample line. I've played around with some much cheaper 70ml/min dosers from amazon but went with the BRS pumps because I already had them, they look nice, durable and easy to maintain. The vial I ended up chosing was found at Sherwin-Williams. It comes in a junky $10 "add your own paint" spraypaint setup. I like it because the bottom is convex allowing the last drop to be removed prior to testing. I used a combination of silicon tubing for my fluid lines. The "return to tank" and waste lines go into the vial with a very thin, rigid yet slightly flexible straw found on those silly "fill 60 water balloons at once" devices. They do a great job at getting every last bit out.
Once the vial is empty there is a brief pause before the sample dos runs a +60ml profile. Another brief pause and the main portion of the reagent is added. The mag stirrer turns on with the reagent addition and off when the pH is triggered. A pH_TRIGGER VO turns on once pH dips below 4.50 and stays on until the test is over. If the pH is triggered with just the main reagent profile the test continues on with the drain/refill process. Otherwise the test continues on with the titration step. I'm using a profile that does 6 0.1ml doses at 30 second intervals. My 9.1_DKH detector VO turns ON immediately with the first titration. this causes the 90_DKH to turn OFF. Progressively higher DKH detectors follow each with an additional 30 second defer atop the previous VO with code to turn it OFF. The highest DKH reached before the pH VO is triggered stays on until the next test starts. From there you can direct your tank's dosing a number of ways but dos profiles would be a simple, seamless solution.
all of my coding will be posted tomorrow.