Jump to content

KH Alkalinity Monitor with AIM


dpassar12

Recommended Posts

Yes, I ordered Chemsavers ACS reagent grade HCl. I came to the same conclusion regarding muriatic acid, no idea what the tolerance is on the concentration, or what nasty impurities are in it that I could be dumping back into the reef tank. With the ACS cert I know it's between 36.5-38% HCl w/w and it has 12 impurities that are tested for to make sure they don't exceed the max limit. I presume they select these specific impurities because they occur as part of the manufacturing process, and the presence of other impurities aren't a concern. That's about as much as I know regarding chemical purity standards.

 

 

Great news on the search for a second-source reagent. I'm assuming that you'll use a laboratory grade acid and not hardware-store grade muriatic acid. Is that a fair assumption? I've never really tested the hardware store stuff for consistency or purity. I suspect that it's reasonably well-controlled and may suffice as a source, but that's just an opinion.

Link to comment
Share on other sites

Yes, I ordered Chemsavers ACS reagent grade HCl. I came to the same conclusion regarding muriatic acid, no idea what the tolerance is on the concentration, or what nasty impurities are in it that I could be dumping back into the reef tank. With the ACS cert I know it's between 36.5-38% HCl w/w and it has 12 impurities that are tested for to make sure they don't exceed the max limit. I presume they select these specific impurities because they occur as part of the manufacturing process, and the presence of other impurities aren't a concern. That's about as much as I know regarding chemical purity standards.

I suspect that you're right. HCl is a byproduct of a lot of industrial processes and it's not clear to me if it's bought up from multiple sources, refined and tested for lab use when you get better grades of it. The muriatic acid from the hardware store is strong, but we don't know if it's refined or not, nor do we have any guarantee of the level of impurities it might contain.  

Link to comment
Share on other sites

Graham and I have been talking for some time.  It started when I was trying to create a DIY version of the KH monitor.  I was doing a lot of research trying to figure out the components and to see if they use PH to calculate alkalinity or color change.  Then figuring out the stepper motors and programming them.   Then I got to trying to figure out what the reagent was after discovering how easy it is to test for Alkalinity.  Chemistry is definitely not my cup of tea, but calculating Alkalinity is not rocket science.  It is more complicated than this, but you need an strong acid (HCL or H2SO4(sulfuric acid), a water sample, and a lab grade PH probe.  Add the acid slowly to the sample until you reach a PH of 4.2. Then calculate the amount of acid you used to bring the sample down to that point, run another calculation and you have your alkalinity. 

 

Salifert uses the color change method which adds a

 

I have read that some people have used generic muriatic, but as Graham pointed out you don't want to dump unknown stuff into your tank. 

Link to comment
Share on other sites

Graham and I have been talking for some time.  It started when I was trying to create a DIY version of the KH monitor.  I was doing a lot of research trying to figure out the components and to see if they use PH to calculate alkalinity or color change.  Then figuring out the stepper motors and programming them.   Then I got to trying to figure out what the reagent was after discovering how easy it is to test for Alkalinity.  Chemistry is definitely not my cup of tea, but calculating Alkalinity is not rocket science.  It is more complicated than this, but you need an strong acid (HCL or H2SO4(sulfuric acid), a water sample, and a lab grade PH probe.  Add the acid slowly to the sample until you reach a PH of 4.2. Then calculate the amount of acid you used to bring the sample down to that point, run another calculation and you have your alkalinity. 

 

Salifert uses the color change method which adds a

 

I have read that some people have used generic muriatic, but as Graham pointed out you don't want to dump unknown stuff into your tank. 

 

Yep. RHF wrote about it many years ago on RC and did an article on it. Here's a 2015 copy of the article for those interested in testing alkalinity manually by titration.

Link to comment
Share on other sites

Prepare yourself for a long post, I wanted to document my thought process from beginning to end on this little project.

 

Currently I own three KHG Monitors. While these monitors have greatly aided me in keeping stable alkalinity there are a couple things I would like to improve about them.  

 

First, the cost and availability of reagent for the monitors. A $35 packet is 35mL and makes 1535mL of reagent when mixed with 1500mL of RODI water. The manufacturer advertises this will last for about 270 tests, equating to approximately $0.13 per test. The interval between tests is user adjustable and ranges from 60 to 240 minutes. So, for three monitors, my monthly cost for reagent can range from $70 to $280. I would like to have the monitors test every hour for a smoother daily plot of my alkalinity levels, but I consider $280 a month far too expensive. I am also concerned that after investing in three of these monitors the reagent may one day be unavailable rendering my investment useless.

 

The second thing I don’t like about the monitors is the accuracy of the test. The precision and repeatability is excellent, but all three of my monitors report a value different than my other test kits. My other test kits are a Hanna Checker, Salifert and Lamotte. These other test kits consistently report very similar values. The manufacturer has acknowledged the monitors lack accuracy and have included an offset feature you can use to make the monitor more closely match your other test kits. My three monitors use offsets of [-1.3], [-1.6], and [-1.0] with the manufacturers reagent to achieve values that closely match my other test kits.

 

In an effort to reduce reagent cost and improve accuracy I have been looking into making my own reagent using hydrochloric acid. Randy Holmes Farley published an article for making your own alkalinity test that illustrates the chemistry to accomplish this is quite simple. For a 0.1N (normality) acid the equation to calculate alkalinity is as follows:

 

Alkalinity (dKH) = [volume of acid added added (mL) / volume of tank sample (mL)] x 280

 

When the monitor runs its test it adds reagent drop by drop and reports the total number of drops added and calculates the corresponding alkalinity. When the pH reaches a target value the titration is over and the final alkalinity value is reported. It may also perform some linear interpolation for increased resolution. It’s not clear what the endpoint pH value is, but Randy indicates it is approximately 4.5 for this type of test. This gives enough information to calculate the normality of the acid based on Randy’s formula and a few volumetric measurements.

 

The test chamber holds approximately 17 mL of sample with a pH probe installed. Using the reagent line purge feature I measured the number of drops and resulting volume for several samples. These measurements were 1.02, .85, .85, and .85 mL for samples of drop sizes 30, 25, 25, and 25, respectively. This equates quite accurately to .034 mL per drop. Based on this and the drops / alkalinity values reported by the monitor I believe it is programmed to expect a 0.02455 N acid.

 

I first tested my tank water using my Hanna checker. Reported values were 142, 143, and 141 meq/L. Averaged, these values equate to 7.95 dKH. This number was backed by a Salifert result of 8.0 dKH. So I decided to proceed assuming my tank water alkalinity is 7.95 dKH.

 

I then ran repeated tests on the sample water using the monitor and the manufacturer’s reagent. The KHG Monitor reported alkalinity readings of 8.81, 8.77, 8.82, 8.82, and 8.82. These average to 8.81 dKH.

 

To characterize the manufacturer’s reagent I took a 17 mL sample of tank water and added reagent in 0.1 mL increments while measuring the pH with a new 7/4 calibrated pH probe. To reach a pH of 4.5 it took 2.11 mL of the manufacturer’s reagent. I used linear interpolation in increase resolution similar to what I suspect the monitor does.

 

This provided me a starting point to create and test some solutions of varying concentration. To create an equivalent reagent I acquired 37% w/w ACS reagent grade HCl (36.5-38.0). I first created a .0168 N acid assuming the drop size would be .05 mL, which proved to be an acid that was weaker than the manufacturer’s. It took 2.72 mL of this acid to reach a pH of 4.5. These values suggest my reagent had a normality of .0177. I found this to be reasonably close to my design target of .0168. This small difference is easily accounted for by human error and the tolerances of my tools to measure the weights and volumes. This gave me confidence going forward I could create solutions with normality close to my intended value.

 

From here I had two design targets. First, create a solution that would perform very closely to the manufacturer’s reagent. Second, create a slightly stronger solution that would yield results closer to the alkalinity measured by my other test kits so that I do not have to utilize a large offset value to compensate for the monitors lack of accuracy.

 

I ended up creating additional solutions with normality of .0227 N, .0244 N, and .0258 N. I arrived at these values by doing some basic math in an effort to meet my two design targets.  The .0227 N solution yielded results very close to the manufacturer’s reagent. The .0244 N solution, according to my calculations, would enable my monitors to yield alkalinity values that would closely match my other test kits. Below is a plot comparing measured pH values against mL of acid added to the 17 mL sample for the different solutions. “Baseline 1” and “Baseline 2” represent two tests using the manufacturer’s reagent.

 

VM4gkpI.jpg

 

2hVabT1.jpg

 

Finally, I replaced the manufacturer’s reagent on the monitor and ran a series of tests to see if my .0244 N solution would yield results close to the 7.95 dKH value my other test kits indicated. The results were 7.91, 7.95, 7.90 with an average of 7.92 dKH. This value is close enough to the target of 7.95 dKH that I am confident using this .0244 N reagent for continued use in my monitors.

 

To create the .0244 N acid I used 3.696 grams of 37% HCl in 1,531 mL of RODI water. The 500 mL bottle of ACS reagent grade HCl was purchased for $44.95. This means it now costs me $0.28 to make a batch of reagent as opposed to $35 from the manufacturer. To run hourly tests on all three of my monitors my monthly costs drops from $280 to $2.23.

 

Overall I am very pleased with the results of this exercise. I can now use my monitors to test alkalinity on my three systems hourly at a very low cost, I am able to make my own reagent on demand using a readily sourced acid, and I have been able to improve the accuracy of the monitors in that they now closely match my other tests kits without a large offset value.

Link to comment
Share on other sites

Nice!!! I wanted to run hourly as well but the $ would have been to high even with a single monitor.  Where did you get the HCI? Are you still going to put the waste water back into your tanks?

Link to comment
Share on other sites

I got it off Amazon, search "500 ml chemsavers hcl".

 

I was putting waste water into the skimmers originally, but now that I know what the reagent is I am going to put it back into the tank. With HCl you're basically just adding back a very small amount of chloride, which exists in great abundance already in saltwater. And since it is reagent grade HCl I am not too worried about impurities as they are in extremely low concentration and the volume of acid used is very small anyway.

 

Since the stock reagent is working well for you I would suggest the .0227 N acid solution. To measure the acid you will need an accurate low range scale. I purchased an "American Weigh Scales GEMINI-20" off amazon for about $25, which comes with a calibration weight and is impressively precise / repeatable considering the low cost.

 

Nice!!! I wanted to run hourly as well but the $ would have been to high even with a single monitor.  Where did you get the HCI? Are you still going to put the waste water back into your tanks?

Edited by gws3
Link to comment
Share on other sites

Excellent work, Graham! Really excellent.

 

A couple of quick questions: 

 

1) How does the reagent dropper work? Do you think the measurement error is from inconsistency in the drop size or quality control on the manufacturer's reagent? Alternatively, it could be calibration of the integrated pH probe.

 

2) How is probe calibration maintained in this thing?

Link to comment
Share on other sites

Thanks Tom, it was good to dust off my basic chemistry skills. Although, chemistry is very tedious. I don’t have much in the way of lab equipment, no magnetic agitators, so there was a lot of swirling test samples with the pH probe.

The reagent dropper has a small nozzle and is fed by a very low flow rate peristaltic pump. There is a LED and optical sensor behind the nozzle that is able to recognize when a drop falls. It must be utilizing the refractive index of water, similar to the optical float sensors you guys were discussing in another thread.

 

I don’t think the inconsistency is due to drop size. The way the drops slowly form on the nozzle gives me the impression the drop size is very consistent, but I have not measured across the different units I have or anything like that. The 17 ml sample is measured using a little white float and an optical sensor, I’d suspect more opportunity for variation there.

 

The manufacturer’s reagent is definitely off from what the monitor is expecting given the drop size and sample size. It is programmed this way regardless of what the pH probe reads. So it’s down to the drop size, sample size and reagent. I’m not sure I could conclude which of these is off from the manufacturer’s original design.

 

Probe calibration is the same as any other pH monitor/controller. You pull out the probe and use a two point calibration. Naturally, since pH in the low 4s is the critical reading for the monitor it is calibrated with a reference solution of 7 followed by 4.

 

What are your thoughts on double junction pH probes? I am thinking about transitioning to them as my probes go bad. With three systems, three monitors, and a few calcium reactor set ups I am now maintaining 8 pH probes. Calibrating them is boring and time consuming.

Link to comment
Share on other sites

I think that you'll get better, longer-lasting, and more stable results from a double junction probe as the reference section is better protected from degradation from the outside. They're more expensive and are more commonly seen on higher-end lab equipment. For as much as you'll be using your AIM and because you'll be using the measurements to adjust your dosing automatically (this is key), my feeling is that it's a good expense to incur. For those who monitor pH and use other means to measure and dose alkalinity, cleaning and trying to regenerate single junction probes (in a strong acid bath) is an acceptable way of extending (or trying to extend) their life. 

Link to comment
Share on other sites

  • 1 month later...

Any advice for handling HCL. Getting ready to make a new batch with HCL. Unscrewed the cap on the HCL with safety glasses and rubber gloves and the vapors were very strong! Put the cap back on and decided to ask before using. Been awhile since chemistry in college.. 

Link to comment
Share on other sites

Any advice for handling HCL. Getting ready to make a new batch with HCL. Unscrewed the cap on the HCL with safety glasses and rubber gloves and the vapors were very strong! Put the cap back on and decided to ask before using. Been awhile since chemistry in college.. 

1. Add acid to water, not the other way around. Otherwise you can get spattering. 

 

2. You've already learned that fuming hydrochloric acid can be irritating (and damaging). Work with it outside and, if possible, downwind. Fumes can damage your eyes as well as your lungs. The stress on your lungs can cause circulatory issues (heart), too.

 

3. Wear personal protective equipment: Safety glasses (for spills). Rubber gloves. Plastic apron or raincoat. Good (closed-toe) shoes. There are some respirator cartridges that are effective with acid gas if you want to be extra cautious, but working outside will probably help enough.. 

 

4. Have a spigot or sink with running water nearby. Keep some baking soda handy in case you need to neutralize an acid spill.

 

5. Keep kids and pets away from your work area. 

 

6. When working with muriatic acid (HCl), I'll often uncap and recap the bottle quickly and at arms length. This reduces fumes and my exposure.

Link to comment
Share on other sites

I'm probably less careful than Tom. I keep the bottle at an arms length and avoid breathing in, and recap it as often as possible. I always have saline solution on hand to rinse my eyes.

 

One addition to the points Tom made above, use glass/plastic with HCl not stainless steel. It will eat the surface of stainless. I used one of those little glass vials that came with API test kits. It was sufficient for enough HCl to mix two batches of solution at a time.

Link to comment
Share on other sites

I'm probably less careful than Tom. I keep the bottle at an arms length and avoid breathing in, and recap it as often as possible. I always have saline solution on hand to rinse my eyes.

 

One addition to the points Tom made above, use glass/plastic with HCl not stainless steel. It will eat the surface of stainless. I used one of those little glass vials that came with API test kits. It was sufficient for enough HCl to mix two batches of solution at a time.

Ha. Do as I say, not as I do, right? I've done it that way - probably more than I would publicly admit. I just want to give as many safety precautions for the member to consider. I'd hate for somebody to get hurt because they didn't have a good handle on safety procedures. The fumes can surprise you. That's for sure. If indoors and without protection (normally over the deep sink), I'll hold my breath if I'm handling it for a short time, then back away just to avoid the chlorine-cleansing of my nasal passages.

 

If measuring out quantities of HCl, I'll use a beaker/flask if not precise, or a graduated cylinder or pipette if more precision is needed.

 

What did you use to transfer to the API vials, Graham?

Link to comment
Share on other sites

Ha. Do as I say, not as I do, right? I've done it that way - probably more than I would publicly admit. I just want to give as many safety precautions for the member to consider. I'd hate for somebody to get hurt because they didn't have a good handle on safety procedures. The fumes can surprise you. That's for sure. If indoors and without protection (normally over the deep sink), I'll hold my breath if I'm handling it for a short time, then back away just to avoid the chlorine-cleansing of my nasal passages.

 

If measuring out quantities of HCl, I'll use a beaker/flask if not precise, or a graduated cylinder or pipette if more precision is needed.

 

What did you use to transfer to the API vials, Graham?

 

I took a controlled whiff of it when I first got it at a distance just to see what I was dealing with. Strong stuff. Breathing in the fumes that are directly over the open bottle would definitely not be enjoyable.

 

I just use a 1ml syringe with a tip. But eventually the bottle will get low enough I'll need to pour it into something else. Will be a bit before I need to figure that out.

Link to comment
Share on other sites

  • 1 year later...

Hi, found this thread since I just got my KHG. Awesome info! How has the diy reagent been working and did you ever make adjustments to the formula? Thanks :)

Link to comment
Share on other sites

  • 1 year later...
On 2/11/2018 at 6:32 AM, gws3 said:

Prepare yourself for a long post, I wanted to document my thought process from beginning to end on this little project.

 

Currently I own three KHG Monitors. While these monitors have greatly aided me in keeping stable alkalinity there are a couple things I would like to improve about them.  

 

First, the cost and availability of reagent for the monitors. A $35 packet is 35mL and makes 1535mL of reagent when mixed with 1500mL of RODI water. The manufacturer advertises this will last for about 270 tests, equating to approximately $0.13 per test. The interval between tests is user adjustable and ranges from 60 to 240 minutes. So, for three monitors, my monthly cost for reagent can range from $70 to $280. I would like to have the monitors test every hour for a smoother daily plot of my alkalinity levels, but I consider $280 a month far too expensive. I am also concerned that after investing in three of these monitors the reagent may one day be unavailable rendering my investment useless.

 

The second thing I don’t like about the monitors is the accuracy of the test. The precision and repeatability is excellent, but all three of my monitors report a value different than my other test kits. My other test kits are a Hanna Checker, Salifert and Lamotte. These other test kits consistently report very similar values. The manufacturer has acknowledged the monitors lack accuracy and have included an offset feature you can use to make the monitor more closely match your other test kits. My three monitors use offsets of [-1.3], [-1.6], and [-1.0] with the manufacturers reagent to achieve values that closely match my other test kits.

 

In an effort to reduce reagent cost and improve accuracy I have been looking into making my own reagent using hydrochloric acid. Randy Holmes Farley published an article for making your own alkalinity test that illustrates the chemistry to accomplish this is quite simple. For a 0.1N (normality) acid the equation to calculate alkalinity is as follows:

 

Alkalinity (dKH) = [volume of acid added added (mL) / volume of tank sample (mL)] x 280

 

When the monitor runs its test it adds reagent drop by drop and reports the total number of drops added and calculates the corresponding alkalinity. When the pH reaches a target value the titration is over and the final alkalinity value is reported. It may also perform some linear interpolation for increased resolution. It’s not clear what the endpoint pH value is, but Randy indicates it is approximately 4.5 for this type of test. This gives enough information to calculate the normality of the acid based on Randy’s formula and a few volumetric measurements.

 

The test chamber holds approximately 17 mL of sample with a pH probe installed. Using the reagent line purge feature I measured the number of drops and resulting volume for several samples. These measurements were 1.02, .85, .85, and .85 mL for samples of drop sizes 30, 25, 25, and 25, respectively. This equates quite accurately to .034 mL per drop. Based on this and the drops / alkalinity values reported by the monitor I believe it is programmed to expect a 0.02455 N acid.

 

I first tested my tank water using my Hanna checker. Reported values were 142, 143, and 141 meq/L. Averaged, these values equate to 7.95 dKH. This number was backed by a Salifert result of 8.0 dKH. So I decided to proceed assuming my tank water alkalinity is 7.95 dKH.

 

I then ran repeated tests on the sample water using the monitor and the manufacturer’s reagent. The KHG Monitor reported alkalinity readings of 8.81, 8.77, 8.82, 8.82, and 8.82. These average to 8.81 dKH.

 

To characterize the manufacturer’s reagent I took a 17 mL sample of tank water and added reagent in 0.1 mL increments while measuring the pH with a new 7/4 calibrated pH probe. To reach a pH of 4.5 it took 2.11 mL of the manufacturer’s reagent. I used linear interpolation in increase resolution similar to what I suspect the monitor does.

 

This provided me a starting point to create and test some solutions of varying concentration. To create an equivalent reagent I acquired 37% w/w ACS reagent grade HCl (36.5-38.0). I first created a .0168 N acid assuming the drop size would be .05 mL, which proved to be an acid that was weaker than the manufacturer’s. It took 2.72 mL of this acid to reach a pH of 4.5. These values suggest my reagent had a normality of .0177. I found this to be reasonably close to my design target of .0168. This small difference is easily accounted for by human error and the tolerances of my tools to measure the weights and volumes. This gave me confidence going forward I could create solutions with normality close to my intended value.

 

From here I had two design targets. First, create a solution that would perform very closely to the manufacturer’s reagent. Second, create a slightly stronger solution that would yield results closer to the alkalinity measured by my other test kits so that I do not have to utilize a large offset value to compensate for the monitors lack of accuracy.

 

I ended up creating additional solutions with normality of .0227 N, .0244 N, and .0258 N. I arrived at these values by doing some basic math in an effort to meet my two design targets.  The .0227 N solution yielded results very close to the manufacturer’s reagent. The .0244 N solution, according to my calculations, would enable my monitors to yield alkalinity values that would closely match my other test kits. Below is a plot comparing measured pH values against mL of acid added to the 17 mL sample for the different solutions. “Baseline 1” and “Baseline 2” represent two tests using the manufacturer’s reagent.

 

VM4gkpI.jpg

 

2hVabT1.jpg

 

Finally, I replaced the manufacturer’s reagent on the monitor and ran a series of tests to see if my .0244 N solution would yield results close to the 7.95 dKH value my other test kits indicated. The results were 7.91, 7.95, 7.90 with an average of 7.92 dKH. This value is close enough to the target of 7.95 dKH that I am confident using this .0244 N reagent for continued use in my monitors.

 

To create the .0244 N acid I used 3.696 grams of 37% HCl in 1,531 mL of RODI water. The 500 mL bottle of ACS reagent grade HCl was purchased for $44.95. This means it now costs me $0.28 to make a batch of reagent as opposed to $35 from the manufacturer. To run hourly tests on all three of my monitors my monthly costs drops from $280 to $2.23.

 

Overall I am very pleased with the results of this exercise. I can now use my monitors to test alkalinity on my three systems hourly at a very low cost, I am able to make my own reagent on demand using a readily sourced acid, and I have been able to improve the accuracy of the monitors in that they now closely match my other tests kits without a large offset value.

Did the ph of your .0244 N solution match that of the OEM solution?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...