Controlling the pH level of a solution is usually a tedious task. Adding an acid or base to the solution will change the pH, but manually monitoring the levels and adding the correct amount isn’t fun. [Reza] rigged up an automated pH controller to keep a solution’s pH steady.
The build uses an Arduino with a LCD shield, screw terminal shields, and [Reza]’s own pH shield attached. A peristaltic pump is used to pump the pH down acid into the solution. This type of pump isolates the fluid from the pump parts, preventing contamination of the solution. The pump is controlled using a PowerSwitch Tail, allowing the Arduino to control the flow of fluid.
An Omega pH probe is used to read the pH level. [Reza]’s open source firmware has support for calibrating the probe to ensure accurate readings. Once it’s set up, the screen displays the pH level and the current state of the system. The pump is enabled when the pH rises out of the desired range.
After the break, check out a video walk through of the device.
25 thoughts on “Automated PH Control”
One of the most boring Youtube video ever
all i can think of is that would be perfect for keeping your ph levels correct in a fish tank
That is exactly what i was thinking. To allow fish owners with tanks containing thousands of euros/dollars worth of exotic fish to take holidays.
Although the phDown (and phUp) solution(s) that they are using would be more associated with a hydroponic grow system for *cough* tomatoes.
You can take a week holiday if you have exotic fish. It’s called owning a properly sized tank. 90 gallons is the SMALLEST you should own. I have an additional 55 gallon wet sump to have enough buffer to go a month before the tank needs any work on it. Blows my mind that people try and have any exotic saltwater fish in a 30 to 70 gallon torture cell.
It is more out of fear of some part failing – warm salt water is not particularly well knows for helping parts have a long lifetime.
Ugh. I’m building a hydroponic system, and it definitely isn’t for weed. That joke is getting so old. Actually, it was never funny to begin with.
That was my first thought, too.
All I can think: I do not want a computer dispensing acids and alkalies into my aquarium.
Generally speaking, the pH of my aquarium doesn’t need an ounce of help. And if it did, I’d be taking care of it myself in a more proactive way than dumping chemicals in.
I did have pH problems back when I was a kid who knew everything and nothing, but in recent decades it’s not even been a factor: the pH of a stable aquarium is, itstelf, stable.
It’s not even useful for a week-or-two vacation: I dare say that more fish die from automated maintenance attempts than are saved from them.
You want to keep your fish alive while you’re not around? Lower the temperature a bit (and thus metabolism), and have folks stop by to make sure that the filter is running properly. If it’s a very long vacation, give them food to feed the fish every day or several (preferably in individual bags).
Fish can be perfectly healthy for a few weeks of non-feeding as long as the rest of their environment is stable. And in the absence of food, it generally -is- stable.
TFA’s method is useful for scientific experiments, perhaps, but not so much for keeping living things alive.
Great idea but I can’t help thinking it would be easier just to buffer whatever solution you might be using…
Sounds like a good spot for a PhID controller.
I’ve done pH control before using a PLC and I find you often overdose using PID because of the lag of the dose chemical mixing in the solution. I ended up pulsing the dose pump every five seconds but I still had to set the cutoff below the setpoint because it will still change for about 30 sec after you finish dosing. Of course it depends on the size of your tank.
Not the fault of PID in general – it will prevent overshoot if tuned correctly. (and if it isn’t suffering from integrator wind-up, perhaps the controller you had wan’t intended for such time constants?)
Automatic pH control is quite tricky. First, pH is NOT linear, it is logarithmic, second, resolution of final control element (valve or pump) is not fine enough. Generally it takes two or three tanks in sequence to achieve even passably decent pH control. And dissolution time can be problematic; nothing like dead time to screw up a control strategy.
Hmmmm wondering if one could misappropriate this for an electrically active environment, such as a lead acid battery, an electrolysis cell, a plating or anodising bath etc..
Why (have you checked on the price of a Ph sensor google “Omega pH probe” the one mentioned above) ? And to make your own looks like some complex glass work.
To check on the charge in a lead acid battery there are basically 2 ways that I know of:
Check the density of the acid (http://en.wikipedia.org/wiki/Hydrometer#Battery_hydrometer)
Or a temporary (1-2 seconds) short circuit of the battery terminals through a high wattage (1-2KW) low ohm resistor and record the current this value is related to the charge contained in the lead acid battery.
I do not know if the ph changes in a lead acid battery as it charges/discharges so I can’t comment if this would be useful.
Thinking about it a bit
Maybe the lead acid cell was a bad example since voltage/discharge is fairly well characterised for it. The application I actually have in mind, is way up the pH scale on the cusp of the alkaline error range of the pH probes, I see, now that I look into them a bit. Will probably resort to a specific gravity method. It’s a process that would normally work fine at a lower, self sustaining pH, but for efficiency reasons, I am thinking about automating and attempting to run on the verge of thermal runaway. Which will likely cause water loss from the solution as steam, so wanna know when and how much to replace.
Wouldn’t a level sensor give you all the information you need and be more reliable?
Yah, it’s probably all I need, I guess I was thinking, “More sensors = betterer” before you know it, I’d have had a G-M tube pointed at the thing :-D
pH probes are made with specialty glass. Not something a home hacker can readily do. Better off buying in this case.
Hmmmm… if a software daemon was written for this; then it’d make for a quick phd project. :)
its so open source, you can’t find it anywhere.
I’m using a similar setup, not sure why most people think something like this would be necessary in a fish tank for dispensing ph up/down when a small amount of buffer and a balanced tank should be able to maintain itself without constant human intervention. I’m using this sort of setup to activate a solenoid valve that regulates CO2 into my planted aquarium. With high levels of CO2 in the water if the system was to shut of I could see a .6 PH change overnight, same applies with then the bottle runs out. Then again I only have a few prized cichlids which are pretty hardy anyway. It’s more nice to know how much CO2 is in the tank because too little and the plants slow down, too much and thats when the fish start showing stress.
This will really help me auto-manage my hydroponic grow box.. Thanks..
Needs a PID yo.
The idea behind this is nice but for large systems it would be inadequate. Think pH’ing a 60+ gallon system from, say, 6.2 to 5.2. Assuming the system isn’t filled with plain H2O (has something to buffer the pH) and assuming you’re using the GH product shown in the video, you would need to add ~100ml or more to make the adjustment. The system doesn’t allow time for circulation/integration of the acid into the solution before another accurate reading can be taken. This is an aside from the fact that it would take a LONG time to add 100ml one drop at a time. One could also use a stronger pH adjuster like Botanicare which is 2x the GH shown, or AN which is 2x the Botanicare – however, this would require tuning the system to dispense certain amounts at a time, and the medical peristaltic pump used doesn’t allow for such things.
So, great idea, nice design, but depending on your specific use (fish tanks, hydroponics, etc) your milage may vary, greatly.
Is there any reason a PID can’t be tuned for something that takes a while to change? AFAIK they’re pretty flexible, but I’ve never made one.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)