DIY Alkatronic reagent

[kenchilada]

I have been testing twice a day for 10 months using the Focustronic reagent. So roughly 300 days. At $110 dollars (currently) for 4 liters. A rough estimate is that there is still roughly 40% (conservative estimate) left so another 120 days worth. At roughly 13 cents per test It is definitely not worth it for me to worry whether a home made formula is done correctly unless one is testing several times a day. Not sure the value of that since two tests, IMO, is enough to give one all the info they need.

Oh… so you’re like the Chevy guy that posts in a Ford thread.

 

[Sisterlimonpot]

Oh… so you’re like the Chevy guy that posts in a Ford thread.

Haha.

I thought it was proven that 0.1N sulfuric acid was exactly what Focustronic used anyway. Cutting that $0.13 per test down even further by purchasing it outside hobby sales channels is a no brainer.

But definitely good to hear from various points of view.

 

[Dennis Cartier]

Sigh. I was planning on switching to the muriatic acid DIY version using the dilution that I calculated, but decided to grab some of the OEM reagent since it is now being offered from my LFS at the same price point as the lab grade standard I have been using for the last few years. The reason for my hesitation is I did not wan to mess up my Alkatronic readings as I key all my dosing from my alkalinity readings. I figured the risk was just too great.

Today I noticed I was almost out of diluted reagent and mixed up another 1.25L using the Coralvue OEM reagent. I expected it to come out the same as the lab standard I have been using. Doh, appears to be stronger than the 0.1N lab standard I was using. I even ran a few extra tests to see if it would climb back to the 8.6 range that I hold my tank to. Nope.

Since I am going to have to deal with a jump anyway, I might as well make the leap to the muriatic version as I have enough of that to last my lifetime.

 

[kenchilada]

Sigh. I was planning on switching to the muriatic acid DIY version using the dilution that I calculated, but decided to grab some of the OEM reagent since it is now being offered from my LFS at the same price point as the lab grade standard I have been using for the last few years. The reason for my hesitation is I did not wan to mess up my Alkatronic readings as I key all my dosing from my alkalinity readings. I figured the risk was just too great.

Today I noticed I was almost out of diluted reagent and mixed up another 1.25L using the Coralvue OEM reagent. I expected it to come out the same as the lab standard I have been using. Doh, appears to be stronger than the 0.1N lab standard I was using. I even ran a few extra tests to see if it would climb back to the 8.6 range that I hold my tank to. Nope.

Since I am going to have to deal with a jump anyway, I might as well make the leap to the muriatic version as I have enough of that to last my lifetime.

Couldn’t you adjust baseline calibration?

 

[Dennis Cartier]

Couldn’t you adjust baseline calibration?

I have never had a need to use that before, so I am not sure how well it would work. I only went for the 1L of the Coralvue version as the LFS was out of the 4L, so I will still move to the muriatic version.

That being said, after I swap in the muriatic version, I may find a need to play with that setting to account for any mismatch. My expected dilution is 10 ml muriatic acid to 4990 ml of RODI, so a little finagling might be in order.

 

[thatmanMIKEson]

I just ordered this, it's the best price I've found.

I haven't gotten around to trying a DIY mix yet, I have 2 alkatronics that test every 6 hours and 2L bottles on each last a while if they aren't retesting a bunch. I can't get a flat line like @Dennis Cartier though! wow

59.99 4L

Alkalinity Testing Reagent - 4L Concentrated

Find premium aquarium supplies at Aquacave. we offer top brands like Tunze, Ecotech, and more. Free Shipping available!

www.aquacave.com

 

[kenchilada]

I just ordered this, it's the best price I've found.

I haven't gotten around to trying a DIY mix yet, I have 2 alkatronics that test every 6 hours and 2L bottles on each last a while if they aren't retesting a bunch. I can't get a flat line like @Dennis Cartier though! wow

59.99 4L

Alkalinity Testing Reagent - 4L Concentrated

Find premium aquarium supplies at Aquacave. we offer top brands like Tunze, Ecotech, and more. Free Shipping available!

www.aquacave.com

Since the instructions say to mix 1:4, this must be 0.1N sulfuric acid and you are basically using the same “DIY mix” as me. That is not a terrible price though… The one I use is $65 on Amazon now although shipping is free. It was $39 when I bought it!

 

[thatmanMIKEson]

Since the instructions say to mix 1:4, this must be 0.1N sulfuric acid and you are basically using the same “DIY mix” as me. That is not a terrible price though… The one I use is $65 on Amazon now although shipping is free. It was $39 when I bought it!

shipping was free!

 

[thatmanMIKEson]

Sigh. I was planning on switching to the muriatic acid DIY version using the dilution that I calculated, but decided to grab some of the OEM reagent since it is now being offered from my LFS at the same price point as the lab grade standard I have been using for the last few years. The reason for my hesitation is I did not wan to mess up my Alkatronic readings as I key all my dosing from my alkalinity readings. I figured the risk was just too great.

Today I noticed I was almost out of diluted reagent and mixed up another 1.25L using the Coralvue OEM reagent. I expected it to come out the same as the lab standard I have been using. Doh, appears to be stronger than the 0.1N lab standard I was using. I even ran a few extra tests to see if it would climb back to the 8.6 range that I hold my tank to. Nope.

Since I am going to have to deal with a jump anyway, I might as well make the leap to the muriatic version as I have enough of that to last my lifetime.

how do you make it so smooth!

 

[Dennis Cartier]

how do you make it so smooth!

I use a Raspberry Pi to control my CaRx dosing to target a preset alkalinity level.

To achieve this, I have my Alkatronic test every 3 hours and publish the alk reading over MQTT. The RPI listens for the alk reading and adjusts the flow rate of the CaRx based on it. I run a static flow rate of 8.5 ml/min for the CaRx, and every 15 minute block of time, the RPI toggles the CaRx effluent pump on for a portion of the time and off for the rest of the time. The ratio of on and off is adjusted with a logarithmic back off if the target is exceeded and a balance factor starts to accumulate. The more the overshoot, the faster the balance factor accumulates. The balance factor is used to slow down dosing and persists even if the next reading is no longer exceeding the target. When the reading is not reaching the target alk, if there is a accumulated balance factor still being used, the balance factor starts to be decremented to slowly unwind the adjustment until it reaches zero.

My lights are on from 9 AM to 9 PM, so I run a varying flow rate. From 12 AM until 6 AM, the flow rate is 30% (of 8.5 ml/min), from 6 AM to 3 PM 65%, from 3 PM to 9 PM 75%, and from 9 PM until 12 AM 65%.

I keep my effluent at a fairly strong 90 dKH by auto venting my CaRx everyday.

The varying flowrate gets me pretty close to a flat line assuming there is no increase or decrease of alk uptake.

My 30 day view shows how effective this can be when there is a perturbation in the alk uptake. I recently used Chemiclean after getting frustrated with some cyano that was getting out of hand. You can see the disruptions starting around June 12th, from the addition of the Chemiclean. When I started a water change on the 14th, the alk shot up, and the CaRx went into automatic shutdown. Once the water change ended on the 15th, the RPI brought the CaRx back online and started wrestling the alk back down towards the target. During this process, the balance ratio reached 22%. As the corals started to return to normal alk usage, you can see the sag in alk levels at two points where the increased uptake was exceeding the rate that the balance ratio was being unwound. Around July 2nd, the balance ratio and alk uptake were in sync (about 8% balance ratio), and the alk readings returned to their flat line. Well until the change to the Coralvue reagent messed it up. Interestingly the existing balance ratio is being unwound further, currently 6.38%, and the last 3 alk readings are showing the increase as the flow is sped up to try to get closer to 8.6 dKH.

During the alk uptake disruption from the Chemiclean, I made no changes to the dosing in any way and just let the RPI handle it. I think it did pretty well dealing with it on it's own.

 

[thatmanMIKEson]

I use a Raspberry Pi to control my CaRx dosing to target a preset alkalinity level.

To achieve this, I have my Alkatronic test every 3 hours and publish the alk reading over MQTT. The RPI listens for the alk reading and adjusts the flow rate of the CaRx based on it. I run a static flow rate of 8.5 ml/min for the CaRx, and every 15 minute block of time, the RPI toggles the CaRx effluent pump on for a portion of the time and off for the rest of the time. The ratio of on and off is adjusted with a logarithmic back off if the target is exceeded and a balance factor starts to accumulate. The more the overshoot, the faster the balance factor accumulates. The balance factor is used to slow down dosing and persists even if the next reading is no longer exceeding the target. When the reading is not reaching the target alk, if there is a accumulated balance factor still being used, the balance factor starts to be decremented to slowly unwind the adjustment until it reaches zero.

My lights are on from 9 AM to 9 PM, so I run a varying flow rate. From 12 AM until 6 AM, the flow rate is 30% (of 8.5 ml/min), from 6 AM to 3 PM 65%, from 3 PM to 9 PM 75%, and from 9 PM until 12 AM 65%.

I keep my effluent at a fairly strong 90 dKH by auto venting my CaRx everyday.

The varying flowrate gets me pretty close to a flat line assuming there is no increase or decrease of alk uptake.

My 30 day view shows how effective this can be when there is a perturbation in the alk uptake. I recently used Chemiclean after getting frustrated with some cyano that was getting out of hand. You can see the disruptions starting around June 12th, from the addition of the Chemiclean. When I started a water change on the 14th, the alk shot up, and the CaRx went into automatic shutdown. Once the water change ended on the 15th, the RPI brought the CaRx back online and started wrestling the alk back down towards the target. During this process, the balance ratio reached 22%. As the corals started to return to normal alk usage, you can see the sag in alk levels at two points where the increased uptake was exceeding the rate that the balance ratio was being unwound. Around July 2nd, the balance ratio and alk uptake were in sync (about 8% balance ratio), and the alk readings returned to their flat line. Well until the change to the Coralvue reagent messed it up. Interestingly the existing balance ratio is being unwound further, currently 6.38%, and the last 3 alk readings are showing the increase as the flow is sped up to try to get closer to 8.6 dKH.

During the alk uptake disruption from the Chemiclean, I made no changes to the dosing in any way and just let the RPI handle it. I think it did pretty well dealing with it on it's own.

omg! I'm going to have to read that a few times but I'll never be able to do it, and that's fine I'm ok with that.
thats amazing and definitely on another level of expertise! unobtainium for most, well done!

I'll have to peruse your build thread I'd love to see the tank that correlates with this kind of work!

 

[Sisterlimonpot]

I have my Alkatronic test every 3 hours

Look like your chart suggests every 1 hour... am I reading that wrong?

 

[Dennis Cartier]

Look like your chart suggests every 1 hour... am I reading that wrong?

No, it's every 3 hours, but I did a couple of manual Extra Measurements to try and see if it would settle down and test close to previous measurements and they just happen to be spaced out on the hour interval.

The latest reading from noon today has the tank at 8.56 dKH, so I expect that it will oscillate a bit around 8.6 as the balance ratio finds the appropriate back off to achieve a stable 8.6 dKH.

 

[Redemptioner]

I use a Raspberry Pi to control my CaRx dosing to target a preset alkalinity level.

To achieve this, I have my Alkatronic test every 3 hours and publish the alk reading over MQTT. The RPI listens for the alk reading and adjusts the flow rate of the CaRx based on it. I run a static flow rate of 8.5 ml/min for the CaRx, and every 15 minute block of time, the RPI toggles the CaRx effluent pump on for a portion of the time and off for the rest of the time. The ratio of on and off is adjusted with a logarithmic back off if the target is exceeded and a balance factor starts to accumulate. The more the overshoot, the faster the balance factor accumulates. The balance factor is used to slow down dosing and persists even if the next reading is no longer exceeding the target. When the reading is not reaching the target alk, if there is a accumulated balance factor still being used, the balance factor starts to be decremented to slowly unwind the adjustment until it reaches zero.

My lights are on from 9 AM to 9 PM, so I run a varying flow rate. From 12 AM until 6 AM, the flow rate is 30% (of 8.5 ml/min), from 6 AM to 3 PM 65%, from 3 PM to 9 PM 75%, and from 9 PM until 12 AM 65%.

I keep my effluent at a fairly strong 90 dKH by auto venting my CaRx everyday.

The varying flowrate gets me pretty close to a flat line assuming there is no increase or decrease of alk uptake.

My 30 day view shows how effective this can be when there is a perturbation in the alk uptake. I recently used Chemiclean after getting frustrated with some cyano that was getting out of hand. You can see the disruptions starting around June 12th, from the addition of the Chemiclean. When I started a water change on the 14th, the alk shot up, and the CaRx went into automatic shutdown. Once the water change ended on the 15th, the RPI brought the CaRx back online and started wrestling the alk back down towards the target. During this process, the balance ratio reached 22%. As the corals started to return to normal alk usage, you can see the sag in alk levels at two points where the increased uptake was exceeding the rate that the balance ratio was being unwound. Around July 2nd, the balance ratio and alk uptake were in sync (about 8% balance ratio), and the alk readings returned to their flat line. Well until the change to the Coralvue reagent messed it up. Interestingly the existing balance ratio is being unwound further, currently 6.38%, and the last 3 alk readings are showing the increase as the flow is sped up to try to get closer to 8.6 dKH.

During the alk uptake disruption from the Chemiclean, I made no changes to the dosing in any way and just let the RPI handle it. I think it did pretty well dealing with it on it's own.

Hopefully Focustronic will release the CaRx variable speed continuous dosing control for the Dosetronic sooner rather than later and we can all easily vary the flow through our calcium reactors based on Alkatronic readings with auto adjustment.

 

[Dennis Cartier]

Hopefully Focustronic will release the CaRx variable speed continuous dosing control for the Dosetronic sooner rather than later and we can all easily vary the flow through our calcium reactors based on Alkatronic readings with auto adjustment.

I am not familiar with how they are planning to do it with the Dosetronic. However I have some ideas.

When I approached the problem, I explored the means of control available through my Masterflex's remote port. It offered a whole range of methods. Direct speed control using 0-10v or 4-20ma, along with a simple Start/Stop signal. Though the speed control was tempting, being able to fine tune the speed of the pump directly, I decided that the risk of things going haywire from a bug or a spurious reading from the Alkatronic was too great, and went with just a fixed dosing rate and using the duty cycle of the pumps ON time to set the actual flow rate. This gives the RPI enough control to affect the outcome positively, but not so much control as to cause rapid issues should something go off the rails.

I would think that they could approach the Dosetronic CaRx in a similar fashion. Power cycling the effluent pump is the baseline of what most CaRx setups should support. Hopefully they use something more robust than a bluetooth outlet like they offer with the Alkatronic. I found the bluetooth outlet to be unreliable and quickly migrated away from it.

My method of control is basically the same as the Alkatronic uses for CaRx control, but on steroids. Where the Alkatronic decides that the CaRx will be on or off for the whole duration of the following interval as each test is completed, my method decides if the ratio of on / off is correctly hitting the alkalinity preset, and adjusts the ratio if it is not. Where the Alkatronic method is very granular, with a 3 hour test interval, either on or off for the entire 3 hour period. My method switches the pump on and off 12 times over the same 3 hour period, which gives a much more refined level of control.

I plan to purchased a 4L of the Coralvue reagent and make the next batch from the 4L bottle to see if all the Coralvue reagent offers a steady acid content. I am not too concerned about accuracy in my application, but I do need reliable precision, and knowing that starting a new reagent is going to give readings matching the last bottle is kind of important in that regard.

 

[Dennis Cartier]

I was curious, so I went and read up on the Dosetronic. It appears they have 2 of the 5 pumps setup to offer continuous dosing and act as the effluent pump for a CaRx, or kalk doser. It is a bit weird that they did not build in support for varying the flow through these pumps based on alk readings from that start. It's possible that they are worried about effluent strengths varying over time, which is a real concern, but adding methods to deal with that should not be too hard. Definitely a missed opportunity on their part.

 

[Sisterlimonpot]

I approached this in a different manner than you Dennis. Although I do have an arduino based prototype that was made to vary the speed of a peristaltic pump of my calcium reactor effluent based on alkatronic results.

The goal was to tighten the sine wave of the daily swing.

That project was put on the back burner because of the wispers of a companies patent pending tech that is going to monitor alkalinity in real time.

While waiting for a prototype of that tech, I decided to shift gears and dabble in Kalkwasser controllers.

A few coral farmers convinced me that pH was king over alkalinity and the importance of tighteningthat sine wave. So I made a kalk controller that does pretty much the same thing with pH as I would've done with alk and calcium reactor effluent.

It's very basic, some algorithmic code to speed up/slow down the rotation of a stepper motor that drives a peristaltic pump.

Of course I'm simplifying it because there are quite a few failsafes to prevent certain events but in a nutshell the goal is to tighten that daily sine wave.

For those that missed the thread of this:

DIY Kalk Controller

For about the last 4 months, I've been tinkering with a controller that will adjust the effluent through a kalk reactor based on input from a pH probe inside the tank. It all started as a conversation about pH being the parameter that we need to be focusing on instead of DKH. At the time I...

www.reef2reef.com

There's no surprise that cycling the effluent on and off achieved the same objective.

 

[thatmanMIKEson]

I am not familiar with how they are planning to do it with the Dosetronic. However I have some ideas.

When I approached the problem, I explored the means of control available through my Masterflex's remote port. It offered a whole range of methods. Direct speed control using 0-10v or 4-20ma, along with a simple Start/Stop signal. Though the speed control was tempting, being able to fine tune the speed of the pump directly, I decided that the risk of things going haywire from a bug or a spurious reading from the Alkatronic was too great, and went with just a fixed dosing rate and using the duty cycle of the pumps ON time to set the actual flow rate. This gives the RPI enough control to affect the outcome positively, but not so much control as to cause rapid issues should something go off the rails.

I would think that they could approach the Dosetronic CaRx in a similar fashion. Power cycling the effluent pump is the baseline of what most CaRx setups should support. Hopefully they use something more robust than a bluetooth outlet like they offer with the Alkatronic. I found the bluetooth outlet to be unreliable and quickly migrated away from it.

My method of control is basically the same as the Alkatronic uses for CaRx control, but on steroids. Where the Alkatronic decides that the CaRx will be on or off for the whole duration of the following interval as each test is completed, my method decides if the ratio of on / off is correctly hitting the alkalinity preset, and adjusts the ratio if it is not. Where the Alkatronic method is very granular, with a 3 hour test interval, either on or off for the entire 3 hour period. My method switches the pump on and off 12 times over the same 3 hour period, which gives a much more refined level of control.

I plan to purchased a 4L of the Coralvue reagent and make the next batch from the 4L bottle to see if all the Coralvue reagent offers a steady acid content. I am not too concerned about accuracy in my application, but I do need reliable precision, and knowing that starting a new reagent is going to give readings matching the last bottle is kind of important in that regard.

you should totally reach out to Eric and see if you could collaborate!! I bet just showing him your results could speak for itself, and the progress would be great for everyone!

 

[Dennis Cartier]

I approached this in a different manner than you Dennis. Although I do have an arduino based prototype that was made to vary the speed of a peristaltic pump of my calcium reactor effluent based on alkatronic results.

The goal was to tighten the sine wave of the daily swing.

That project was put on the back burner because of the wispers of a companies patent pending tech that is going to monitor alkalinity in real time.

While waiting for a prototype of that tech, I decided to shift gears and dabble in Kalkwasser controllers.

A few coral farmers convinced me that pH was king over alkalinity and the importance of tighteningthat sine wave. So I made a kalk controller that does pretty much the same thing with pH as I would've done with alk and calcium reactor effluent.

It's very basic, some algorithmic code to speed up/slow down the rotation of a stepper motor that drives a peristaltic pump.

Of course I'm simplifying it because there are quite a few failsafes to prevent certain events but in a nutshell the goal is to tighten that daily sine wave.

For those that missed the thread of this:

DIY Kalk Controller

For about the last 4 months, I've been tinkering with a controller that will adjust the effluent through a kalk reactor based on input from a pH probe inside the tank. It all started as a conversation about pH being the parameter that we need to be focusing on instead of DKH. At the time I...

www.reef2reef.com

There's no surprise that cycling the effluent on and off achieved the same objective.

Nice work on the kalk controller.

I have not done a whole lot with pH yet. I go through great efforts to try and prevent my CaRx from pulling the tank pH down though. I have a powered secondary chamber (old Geo 618) and put the effluent through an aeration chamber on the way back to the tank.

About a week ago, I added 3 part dosing to my system and integrated it so that the controller can decide if it wants to dose each 15 minute interval from the CaRx or 3 part. The 3 part I am using uses NaOH, so it gives the same pH boost as kalk.

I was curious to see if it would have much of an impact on my pH, so I set it up to default to using 3 part from 12 AM to 6 PM. After about a week of running it, the effect appears to be minor, about 0.01higher on average. However the probe in my sump is possibly suspect. The probe was new, but it seem to always read much lower than the Alkatronic. I have it hooked up through Atlas Scientific Tentacle T3 boards along with 3 EC probes and 2 RTD probes. The probe itself is a Milwaukee double junction. I might try a Neptune and see if it reads closer.

Anyway, where I was going with this is that once I have the probe providing readings that give me confidence, I can alter the dosing to decide on what alk source to use based on the pH instead of just time of day. That should flat line the alk and pH at the same time.

Though I will have to think about that. Before I added the 3 part, I calculated that each 2L jug would last me 7 months at my current rate (only dosing between 12 AM - 6 AM), which was fine, but allowing it to get used more frequently means more 3 part mixing, which is not my favourite thing to have to do.

 

[Redemptioner]

Nice work on the kalk controller.

I have not done a whole lot with pH yet. I go through great efforts to try and prevent my CaRx from pulling the tank pH down though. I have a powered secondary chamber (old Geo 618) and put the effluent through an aeration chamber on the way back to the tank.

About a week ago, I added 3 part dosing to my system and integrated it so that the controller can decide if it wants to dose each 15 minute interval from the CaRx or 3 part. The 3 part I am using uses NaOH, so it gives the same pH boost as kalk.

I was curious to see if it would have much of an impact on my pH, so I set it up to default to using 3 part from 12 AM to 6 PM. After about a week of running it, the effect appears to be minor, about 0.01higher on average. However the probe in my sump is possibly suspect. The probe was new, but it seem to always read much lower than the Alkatronic. I have it hooked up through Atlas Scientific Tentacle T3 boards along with 3 EC probes and 2 RTD probes. The probe itself is a Milwaukee double junction. I might try a Neptune and see if it reads closer.

Anyway, where I was going with this is that once I have the probe providing readings that give me confidence, I can alter the dosing to decide on what alk source to use based on the pH instead of just time of day. That should flat line the alk and pH at the same time.

Though I will have to think about that. Before I added the 3 part, I calculated that each 2L jug would last me 7 months at my current rate (only dosing between 12 AM - 6 AM), which was fine, but allowing it to get used more frequently means more 3 part mixing, which is not my favourite thing to have to do.

Remember with your probes in the tank that are staying in the water all the time you need to take much longer time to calibrate. This can be a pain with the limited calibration in the Alkatronic, I find it strange they did not give us the native controls of the Atlas Scientific pH board and let us do 3 point calibration and/or not be restricted by the Alkatronic time out during calibration. It takes from 15mins to 30mins in each buffer solution to calibrate a probe that is continuously immersed, at least you can leave the probe in the ph7 buffer for a while before starting calibration on the Alkatronic but alas not for pH4 buffer and no pH10 buffer despite samples starting well above pH7.

Unfortunately all these companies for some reason persist with using pinpoint calibration designed for a probe only being used for a few minutes at a time, but we are using the probes constantly so they need another form of calibration. You can tell if you probe is incorrectly calibrated, just put it back in some pH 7 buffer and come back in a hour and see how much it has drifted from 7 after that hour. If it drifts less than 0.01 then you good to go, if it drifts more than this then you need to leave it in each buffer solution longer before calibration. The older the probe the longer the buffer soak time.

My hydro setups are almost completely automated and manage the pH control precisely (along with other chemistry and environmental controls). There is one thing you have to have when continuously controlling pH, and that is more than 1 pH probe being used to take the pH readings for the control loop. A probe can throw a wild reading easily and this needs to be managed through multiple probes and a bunch of additional safety controls. I often laugh at how poor the brand name controllers are for reef aquariums, their poor choice of components, the terrible programming and silly limitations baked in especially when you consider how many are using Atlas Scientific for the bulk of the hardware and open source libraries. Just amazes me we don't see more lift and shift from the hydro world considering how much better most of the products are. A prime example is the Alkatronic with a Pi Zero hooked via I2C to a custom daughter board populated with Atlas Scientific components and motor drivers, when for far less you could have just run a standard Atlas Scientific board that stacks on a pi 3 or 4 (and fix the poor responsiveness of a Pi Zero) with Atlas Scientific peristaltic pumps with the motor drivers onboard the motors..... like in hydro. Then bloody formal coat the electronic boards seeing as we are dealing with a hostile environment for electronics inside the case where the electronics are housed....

As for the reagent, I buy 2N Sulfuric acid (about $25USD per litre locally) and use a precision pipette dispenser to make up a couple of 5ltrs containers at a time. It is the cheapest option locally (I have 4x Alkatronics on test every 3 hours) and I already have the precision dispensing needed to keep things accurate (about another $20USD if you had to buy the pipette and tips if you are playing along at home). I dump the waste back into the sump via a couple of chambers (the inline ones used for refillable DI resin) full of reborn and coarse oolite aragonite to reduce any impact on alk or pH. I measure the RO/DI water by weight and the acid by volume when preparing the final reagent solution. Although this is the cheapest and easiest option for me, it by no means is the safest and you should definitely be in appropriate PPE when playing with this concentration of sulfuric acid.