My $25 auto chlorinator DIY project

Jun 13, 2013
224
Hey all...thought I'd share this. It's probably a long post (and is still an experiment in progress, so I *will* follow up with details as it progresses) but I thought it could be useful to those looking to go with automatic liquid chlorine injection, but don't want to spend hundreds on a professional system.

Like many here I'm not big on pucks. I'll use them when we're going away for a weekend because there's few other options, and I hate bugging friends to come do my pool maintenance. And of course, one slip up or forgotten visit to add chlorine and I end up with a frog pond. I keep my pool quite warm with solar (90-91) so chlorine demand can get up there, and with direct sun for many hours a day, it can bleed off quickly. My CYA is 30, and in my comfort zone, leaving me some room for occasional puck use. ;)

But, I wanted to add a basic level of automation to things. Yes, it's not an excuse for still maintaining and checking manually, but it's a backup - I've almost forgot a few times to add the morning and evening liquid chlorine doses and it's only a matter of time before I slip up and end up fighting an algae bloom.

So....I was on the hunt for a peristaltic pump...but being in Canada the prices were high at retail, and anything used I was finding was on eBay after shipping costs would have ended up being unreasonable for me...or worse yet, taking a chance on an "as is" type deal (as many peristaltic pumps seem to be on eBay) could have backfired and I end up out a bunch of money.

So I started Googling for cheap liquid/chemical pumps, and came across a bunch of references on aquarium forums for the "Aqua Lifter AW-20" by TOM equipment. I was hesitant initially figuring the chlorine may eat the internals, but it was listed in many places as designed for "chemical dosing" usage for aquariums, and with a price of only $21 at my local aquarium shop I thought I'd give it a shot.

So, a project was hatched. PIctures of the current "quick and dirty" setup - if it works out I'll tidy up the install and make it more permanent, but I didn't want to spend an inordinate amount of time with a fancy/tidy setup for something that may need to be changed or tweaked in the coming weeks.

The pump:

9329062865_6c65b1354f_c.jpg


2012-02-27_14-52-57_489.jpg


As you can see...it's small. Best as I can figure (based on parts availability) it's very similar internally to an aquarium pump except that it's plumbed for both inlet and outlet. The pump itself does appear to be rubber so longevity may be the big question, but a few days of initial testing have proven it resistant to liquid chlorine so far. I have plumbed it with chlorine tolerant tubing.

So, I hooked it up today and tested it's flow rate which conveniently came to exactly 100ml/minute. That makes dosing calculations easy. Your mileage may vary, it's rated for twice the flow rate versus my testing, but I attribute it to the fairly long hose run I have (probably 20 or so feet), so testing is necessary to confirm and time injection cycles accordingly.

9329063607_354b671793_c.jpg


One thing I noted is that the flow WILL SIPHON through the pump if the tank is higher than the pump. Don't ask me how I found this out initially, but thankfully it didn't end in disaster. Keeping the pump ABOVE the chlorine supply is absolutely necessary for this pump in particular!

So, continuing, I drilled a small hole in my skimmer above the water line and ran the feed line from the pump, securing it with hot glue. (Yes, my skimmer needs a good scrubbing. Out of sight, out of mind.. :wink: ) Thankfully the underside of my skimmer is exposed under decking so this was a viable spot for me to basically "drip inject" into. Your mileage may vary, but timed carefully (to ensure my pump is running whenever injection happens) this was a viable spot for me. More on that later.

9329066669_7ff8487069_c.jpg


Again, as mentioned earlier, quick and dirty for now, I ran the lines and situated my chlorine container in an out of the way area (with little/no sun) under the back edge of my deck. I drilled a small hole in the lid (obviously I'll need a new one now, or a plug, for when I transport the jug to/from the pool store to refill) to run the feed line. I weighted the feed line to ensure it sat on the bottom as initially it wanted to float and sucked air.

9329064261_bc29aca7e9_c.jpg


9331860068_7f4a19124c_c.jpg


Again, I temporarily setup the pump out in the open. I will tuck it carefully away in the cabinet in the picture before weather arrives, although in all honestly, it looks pretty waterproof - it is designed for aquarium use so although it's obviously not submersible, it's probably reasonably waterproof. However, if this project works out I will certainly tidy up the install - run the plumbing neatly under the deck, make a bracket or something to hold the chlorine jug (although this section of my yard is basically a dead zone, no activity goes on there), and put the pump inside the cabinet in this pic..

9329065041_b372c1dfdd_c.jpg


Last but not least, and most importantly, a good high end electronic timer with multiple on/off program cycles. I've used this timer for a year or so now to do a nightly reset on my DVR security system and it's worked perfectly, so I trust it. Best of all it had 20 programmable on/off cycles (at as little as 1 minute each) per day so it's very versatile and suitable for turning the pump on/off for small doses of chlorine through the day to maintain levels. It also has a built in battery backup to maintain programming. Yes, I know it's NOT waterproof, but again..quick and dirty at this point - it will be tucked away in the waterproof cabinet before weather is expected next.

9331860964_4908afe811_c.jpg


So, I programmed it initially with 4 dispensing cycles. Each cycle corresponds to when my pump (which is also on a timer) is running to ensure that the chlorine is promptly picked up and mixed. The pool filter pump runs for at least 15 minutes before chlorine injection starts, and runs for at least 45 minutes after each injection cycle ends...so there's NO worry about any chlorine damaging the liner near the skimmer from building up too strong. Yes, I understand how absolutely important it is that the pump runs when injection is occurring, and I take no responsibility for anyone destroying their liner trying this project and screwing up the timing. For those who run their pumps 24/7 this obviously isn't a worry, but if you use a filter pump timer like me for whatever purpose, careful timing is an absolute necessity obviously!

So far, so great - instead of my usual peaks and valleys of chlorine levels the system (which is injecting metered doses about 4 times per day now) is maintaining levels much more consistent, and best of all, I can go away for a day or two now without fear of coming home to a green disaster. Adjusting the duration of the injection cycles will require some homework still - as the weather changes my current settings may end up with too much or too little chlorine being injected, so obvioully it's not a "set it and forget it" solution, but at least I can go to work in the morning now and not have to freak out if I forgot to glance at the pool before leaving.

Worst case scenario, I'm out about $25 if the pump fails in short issue (because of with Chlorine intolerance) and I abandon the project. Rebuild kits for this pump are fairly inexpensive, so even if I need to rebuild it once or twice a season I'll still be happy.

Thoughts/Comments?
 
PrivatePilot said:
I was hesitant initially figuring the chlorine may eat the internals, but it was listed in many places as designed for "chemical dosing" usage for aquariums, and with a price of only $21 at my local aquarium shop I thought I'd give it a shot.
Keep in mind that aquarium chemicals never involve chlorine, so I won't be surprised if it does go sour. That said it is a neat experiment and I will be watching to see how it goes. I am hoping it works, may be something I would consider playing with if the gaskets don't disintegrate. :p
 
That would work great, as long as your chlorine use is fairly stable. I like it!!

I think I can see you going all wiz-bang on us and designing a microcontroller project that can handle variable dosing cycles based on observed pool use, day and time.
 
Donldson said:
Keep in mind that aquarium chemicals never involve chlorine, so I won't be surprised if it does go sour. That said it is a neat experiment and I will be watching to see how it goes. I am hoping it works, may be something I would consider playing with if the gaskets don't disintegrate. :p

Yes, I agree on all fronts. The possible saving grace is that the pump is simplistic at best - it's a vibrating diaphragm style pump - there's not much to go wrong unless the rubber itself disintegrates (possible) or the flappers inside the diaphragh do the same - also possible. There's not a birds-nest of moving parts or seals/connections/etc inside - just one of these plumbed to the inlet/outlet:

41eTMy6AMjL._SY450_.jpg


In the end, it was (and continues to be) a $20 gamble. I can handle a $20 loss if it all falls apart. :)

Charlie_R said:
That would work great, as long as your chlorine use is fairly stable. I like it!!

I think I can see you going all wiz-bang on us and designing a microcontroller project that can handle variable dosing cycles based on observed pool use, day and time.

Don't count on the whizbang. I'm fairly into home automation and tech (my solar pump is remotely controllable from my iPhone as well as on a UV-based automated schedule via a Belkin Wemo), so I'm into that sort of thing..but changing dosing schedules based on observed use is way more than I'll accomplish. The closest I may come would be to reprogram Saturdays and Sundays separately to introduce more chlorine (based on anticipated heavy use on weekends) but even that's unlikely since sometimes we don't use it on weekends if we're away camping or jetskiing. As mentioned, there will still be a good amount of monitoring going on and I'm sure there will be times I will have to supplement the pumps dosing (I'd rather have it under vs over though), but it reduces the risk of my levels ever falling into the danger zone at least..something that I've been constantly worrying about given that I work long shifts at work sometimes and have come home to <1PPM levels after long sunny hot day.
 
bugsy714 said:
I sold marine aquarium goods for many years those are usually used for kalwasser aka pickling lime dosing among other chemicals but I think the full strength chlorine will eat it aliv
Good luck

Yep, entirely possible. I looked far and wide for technical details on what the diaphragm and flappers were actually made of (heck, it could actually be a polymer or some compound that *is* chlorine tolerant for all I know), but I couldn't find a thing. The only way to find out for sure was to basically, well...sacrifice one. Someone has to be the guinea pig, right? :)

This is all very much an experiment. There will be a lot of water tests in the next few weeks to make sure that it is actually dosing reliably. I can override the timer momentarily as well and flip off the skimmer cover to physically see the flow, so that's an easy method of confirming that the pump hasn't crapped out as well.

I should also mention for anyone following along, I did a measurement test on the flow again this evening (after routing things a little more hidden vs the earlier out in the open setup I had) and the increased length of hose and routing reduced the flow to 20ml/minute - WAY down from the "sitting on the deck pumping water to/from the pool" test that yielded 100ml/min initially. This obviously changed the calculations a LOT on how long the pump needs to run to dose sufficiently. Potentially better depending on how you look at it - a slow gradual release of chlorine is never a bad thing.

Moral of the story is that anyone trying this out is going to need to do some careful measurements with the system in place and set the timer accordingly. My initial settings on the timer based on the earlier flow tests would have yielded a woefully inadequate amount of chlorine in the end.
 
Hehe..for the record its still going strong today - I watched it inject the noon-hour dose troublefree.

Given the significantly reduced flow I am now experiencing I think that I will program it to do quite a few small doses throughout the day versus larger doses twice or three times a day – the flow rate is just too slow to do significant volumes in short periods of time – it would take about 50 minutes to pump 1000ml (1L) of chlorine - what I was sometimes adding to bring levels back up quickly when I had dropped dangerously low a few times.

3 30 minute cycles (~600ml/each time) seems to be doing the trick, but I might change that to six 15 min cycles @ 300ml/each) instead to add to the reduction and peaks and valleys in my CL levels.

We've had two pretty chilly nights in a row and I lost quite a few degrees on the pool temperature, so my chlorine demand has dropped. I will definitely be keeping an eye on things to make sure I'm not overdosing now.
 

Enjoying this content?

Support TFP with a donation.

Give Support
I really dig this kind of stuff...as you mentioned I'd even do a rebuild or two each season if needed, well worth it. Keep us informed as time goes by!

Still not able to source more than 5 gal. bulk chlorine in this area, and the supplier wants too much for the bulk 5 gallons (more than the .21/oz I'm paying for 1 gals. of 12.5% at the pool store). Would like to find 13-15 gal carboys, even @ .21/oz. Keeping an eye on the liquid chlorine threads for ideas.
 
It's still working great, for the record. I'm having to greatly extend the pumping times though as I got the pool back up to the 90's and chlorine demand increased accordingly.

Chlorine is cheap and plentiful around here, my 20L jug (~5 gallons) costs about $10 to fill at most pool stores. 21c an OUNCE? My metric to imperial conversions show that a 5gallon jug is costing you an arm and a leg?
 
Forgot to mention, that's factoring in the concentration per the chlorine cost calculator. 128oz of 12.5% @ $3.40/gal....I can only find it in 1 gallon jugs, can't find anyone local to do refills on carboys or 5 gal buckets. Always looking though.

Thanks for the update on the pump, glad it's still working for you.
 
WhoaTed said:
I can only find it in 1 gallon jugs, can't find anyone local to do refills on carboys or 5 gal buckets.

That's odd, around here every pool place has huge bulk chlorine fill station. I could go buy 1000 liters tomorrow morning if I wanted, so long as I had a proper container.
 
PrivatePilot said:
It's still working great, for the record. I'm having to greatly extend the pumping times though as I got the pool back up to the 90's and chlorine demand increased accordingly.

Chlorine is cheap and plentiful around here, my 20L jug (~5 gallons) costs about $10 to fill at most pool stores. 21c an OUNCE? My metric to imperial conversions show that a 5gallon jug is costing you an arm and a leg?

If/when it dies on you, you might want to take a look at this Peristaltic pump, it's only $25. It's what I'm thinking of using when/if I build an auto chlorinator.
 
thrystan said:
If/when it dies on you, you might want to take a look at this Peristaltic pump, it's only $25. It's what I'm thinking of using when/if I build an auto chlorinator.

I saw those, they're available on eBay for even less than that. ;)

The problem is several fold - first, they're 12 volt, and they probably require a few amps at that...so you need a reasonably hefty power source. Second, their flow rate is not adjustable, and rather fast actually at 100ml/minute - good for big pools, but unless you have a timer that is able to split doses down to less than 1 minute (most won't) then at minimum you're dosing 100ml. Those with smaller pools in cool weather would have to be careful to not overdose. The flow rate may be somewhat slowable by voltage reduction, but that again adds complexity.

I have those pumps firmly in the back of my head as plan B, though. ;)
 

Enjoying this content?

Support TFP with a donation.

Give Support
Thread Status
Hello , This thread has been inactive for over 60 days. New postings here are unlikely to be seen or responded to by other members. For better visibility, consider Starting A New Thread.