The Planted Tank Forum banner

Wasserpests new airpump microdoser

10K views 31 replies 8 participants last post by  vidiots 
#1 ·
As you might know, I am slightly obsessed with autodosing. I have tried more designs than I care to remember. Sometimes an idea gets stuck in my head, and a few days later you find me going through scrap parts trying to assemble yet another monstrosity.

Some designs have worked out really well. Dosing N and K via the waterpump method, and P via the test tube method is extremely simple and reliable. There have been a lot of attempts for micro dosing that for different reasons did not work out long term. Dosing diluted solutions lead to undesired growth in the containers. Timer driven syringe dosers did not last. Peristaltic pumps are expensive, and one of the two I have tends to strip the plastic rotor and stops working.

There was another design which seemed to be promising. I spent many, many hours trying to make it work and never could. What would happen was that either the "refill hole" was too small and the tubing would not refill, or it was too large, which lead to little droplets entering when all the solution was pushed out, and when the pump stopped, the droplets would again prevent the tubing from refilling.

Still, this seemed like a real smart idea, so I kept playing around with it, and I think finally found a bit different design which seems to work better. The trick to making this work is to have a restriction ("small hole") while the pump is pushing the fertilizer out (which keeps the tubing free of more solution), and having open flow when the pump turns off, so the tubing refills readily.

After many hours of thinking I came up with a somewhat *******, but simple and cheap solenoid valve. I used a piece of latex glove, wrapped around the end of some air tubing with a rubber band. That went into a plastic syringe. As the air pump pumps, some air is diverted to blow up the latex bubble, which blocks the flow through the syringe, therefore restricting the flow and keeping the tubing clean.

Okay -- here is the design:

 
See less See more
1
#2 ·
So here is how it works:

1) An air pump (~$10), controlled by an electronic timer ($8), powers the dosing. The airflow is split into two via a T.

2) A little airvalve adjusts the airflow through the dosing line. This is not extremely critical, but I find that if the flow is too high, droplets splash through the tubing and can collect after the pump has turned off, preventing an easy refill.

3) This check valve allows nutrient solution to flow into the tubing, and prevents it from exiting when the air pump is running.

4) This check valve prevents solution going up the airline, and therefore skewing the dosed amount depending on the level of the nutrient solution.

5) This tubing contains the volume that will be dosed, which depends on the length and diameter. In my prototype, I am using a 3in piece which results in just a little under a milliliter dosing volume. Extremely easy to adjust, of course.

6) I use mini tubing for this part, to minimize the amount of liquid that fills it to the level of the solution. Also, mini tubing should help to push the liquid out completely.

7) As described in the first post, this serves as a solenoid that is closes while the pump is working, therefore creating back pressure so no more liquid can enter through check valve 3. This syringe shows clearly how many ml are being dosed. First, the liquid is pushed into the syringe. As the "solenoid" closes, pressure builds up and finally pushes the liquid through a tiny needle hole out of the syringe into the dosing line 9 and into the tank.

8) This air valve determines how much air goes into the "solenoid", which is a very small amount, unless you want the little pop. :smile:

That's pretty much it, dosing a consistent amount each time the timer kicks of the cycle. I am sure this can be beautified and maybe simplified, but the basic concept seems to work.
 
#3 ·
Hmm, I was thinking a rubber test tube stopper with a hole thru it for the airline tubing might be used in place of your latex glove as a cork for the end of the syringe.

The potential weakness I can see for this design would be the use of cheap aquarium check valves which tend to eventually fail and get stuck especially when exposed to liquids with stuff dissolved in them.

I'd be curious to know how long it runs for before it is in need of repair.
 
#5 ·
Hmm, I was thinking a rubber test tube stopper with a hole thru it for the airline tubing might be used in place of your latex glove as a cork for the end of the syringe.
It might work. The difference in my design is that the latex glove collapses, which opens up the syringe part completely. If you use a rubber stopper, there is still the back pressure from the air pump membrane which might prevent the refill.

The potential weakness I can see for this design would be the use of cheap aquarium check valves which tend to eventually fail and get stuck especially when exposed to liquids with stuff dissolved in them.
You are right... that might or might not be an issue over time. Here they sell check valves with a different design, might be worth to check out.

I'd be curious to know how long it runs for before it is in need of repair.
Me too. It's still in the trial phase. Once I have it running for a few weeks or months I will let you know.
 
#10 ·
Trying to simplify the design a bit... a smaller nutrient container will be beneficial with small dosages.



Currently searching for a different design check valve. The ones we commonly use have two rubber lips and need a little bit of cracking pressure to get the flow started. There is a different design with a movable flap or cylinder that covers up a hole in one direction, and allows flow in the other direction. That design has almost zero cracking pressure, that is, allows fairly free flow as soon as the pressure is zero. That would be optimal and might even alleviate the need for my super latex glove solenoid. :)
 
#13 ·
... The trick to making this work is to have a restriction ("small hole") while the pump is pushing the fertilizer out (which keeps the tubing free of more solution), and having open flow when the pump turns off, so the tubing refills readily.

After many hours of thinking I came up with a somewhat *******, but simple and cheap solenoid valve. I used a piece of latex glove, wrapped around the end of some air tubing with a rubber band. That went into a plastic syringe. As the air pump pumps, some air is diverted to blow up the latex bubble, which blocks the flow through the syringe, therefore restricting the flow and keeping the tubing clean.
I've read this a number of times and I still dont understand exactly what you mean by "which keeps the tubing free of more solution" or "therefore restricting the flow and keeping the tubing clean"

I've played with these airlift dosers myself and by "keep the tubing clean" I assume you mean preventing the retention of solution in the uplift tube so that air can escape as the submerged dose tubing refils?? ...if so... I dont understand why restricting the outflow prevents solution being retained in the narrower uplift tubing. Can you perhaps take another crack at explaining exactly how the solenoid improves the dosing and exactly which bit of tube you are trying to keep "clean"?

.. or have I missed something else?

oh and btw I think those disk based checkvalves sound perfect for this application!
 
#14 ·
Yeah, it is a bit hard to explain.

With the original, simpler design you have only one tube going straight up from the container. It fills up with solution on the bottom, and when the pump kicks in, that solution is pushed into the tank. Now as soon as all the solution has left the tubing, the back pressure against the lower checkvalve becomes zero, and slowly, more solution trickles into the tubing and is pushed out to the tank. When that happens, little droplets remain in the tubing and prevent it from refilling.

So if you restrict the outflow a little bit, all solution is pushed out, but then no more solution enters the tubing since the air pump creates some back pressure against the check valve. Once the pump stops, the back pressure is gone, and the tubing fill again.

The issue was that I couldn't get that to work. Either I restricted the outflow too much, which prevented the refill, or it was too open, which lead to droplets making it through the check valve while the pump was running, and then prevented the refill.

The work around with the DIY rubber glove solenoid makes that work. Back pressure while the pump is working, no back pressure while pump is off.
 
#15 ·
A picture is worth a bunch of words, so here is one of my currently working prototype.



It has been dosing my 100gal tank for the last 3 or 4 weeks. Weak points are the check valves. The one on top dries up a little and it takes quite some pump strength to unglue it.

I really want to try one of the other check valves, but I can't get them without incurring a $10 shipping charge, and that goes against my frugal nature. If I knew they would work I would buy a bunch to distribute the shipping cost, but I don't know that.
 
#17 ·
Simpler than this? Come on... :) Not sure how. I mean the simplest solution is to take a bottle of micro something, measure some and squirt it into the tank. Obviously I want to automate this, while keeping it inexpensive.

Gravity usually means that you have to place or hang something above your tank. Me no like that.

Not sure about spirits in pubs... sound a bit like manual dosing?
 
#20 ·
#23 ·
Yay. After much searching I finally got my free samples!



And, as expected, it works awe-some-est. I tried to do it the simple way first, without my ******* Rubber Glove Solenoid, but no go. With the new design, it works like a charm. Comparing the various check valves (by blowing and sucking on them, if you must ask), it seems like the 402 flavor works best, along with the tiny 400 (if you have very small tubing). The other two felt like they had an ever so tiny cracking pressure point. Might still work, but I think the 402 was the best of the samples.



Tubing refills in a few seconds after the rubber glove bubble has collapsed.

I might set up a real "Microdoser" with micro dimensions for my 10gal tank, using the 400 checkvalve. But first I will enjoy this one for a few months.
 
#25 ·
Thanks... and I was so proud of my DIY solenoid. :icon_cool

I think this design has some potential for semi/commercial marketing. Some good points:

Cheap
Easy to adjust dosing volume
Pretty precise dosing

Peristaltic pumps are somewhat more compact and simpler though. And if you add a real solenoid you will also add some cost, and the potential of another thing gumming up with micro sludge.

This one is something for the geek in me... there is so much to adjust... how fast the little balloon blows up, how much air pressure lifts the fertilizer solution into the tank... One could probably add a third airline that mixes up the nutrients gently. And maybe add a flush mode to clean any settled down stuff when you add new solution. Oh, the possibilities! :)

Thanks for asking, the canister filter is going semi-well. Got the filtration sponge, and ran the thing actually for a few minutes. Worked very well, only the pump is one loud beast. Need to check if it is missing a grommet or two. Anyway, I'll post things in the homebuilt canister thread soon.
 
#26 ·
Thanks... and I was so proud of my DIY solenoid. :icon_cool

I think this design has some potential for semi/commercial marketing. Some good points:

Cheap
Easy to adjust dosing volume
Pretty precise dosing
All it really needs is to be cheap (cheaper than peristaltic) and reproducible. You can handle the adjustments with the solution concentrations if need be. Your DIY solenoid is awesome indeed, and the price can't be beat, but I think you'd have a hard time charging money for it. Not that money was your motivator; clearly you just like to build things.
 
#27 ·
In life, things don't always work out. What seems like a great idea one day soon catches up with the harsh reality and quickly breaks into pieces.

While the new check valves work very well for this little machine, some other problems with the "solenoid" and balancing has made me decide to abandon this for the moment. I needed something to dose micros in my 100gal tank, and this was taking too long to get perfect.

I might work on this design again sometimes next year, for now I have set up my proven waterpump doser, with a very small outlet hole to reduce the dosed solution and keep it pretty concentrated to prevent mold. I will need to check regularly to see if the outlet clogs, apart from that it is very reliable.

I ordered some timers that can be set from 15 seconds to a couple of minutes, which turn on when they are powered on, then turn off after the preset time. I am going to play around with them, reducing the pump time further will allow to concentrate the solution even more, maybe then it will be near perfect.

A nice side-effect of the waterpump dosing is that the solution is thoroughly mixed every day. That was one concern here with this method.
 
#29 ·
Okay, now I'm offended, ******** do not use them big words like "Solenoid".

Actually, this tiny dosing pump that cost 11$ from Tom's might get you thinking:

http://www.marinedepot.com/ps_ViewItem~idProduct~OE1137.html

Now since it has a max head of 30 inches, placing it up at 24 inches and on a timer.......will allow you to dose continuously by VARYING THE HEAD LEVEL, and/or with a timer.

Simply measure the rate of fill in a flask for 1 hour.
Adjust to suit with the head level height and /or a timer.

Very easy also.
And 11.99$+ your time and timer.
Add a needle valve to it if you wish to control the flow more than with just the height increase(which might not be practical in some cases).

Regards,
Tom Barr
 
#30 ·
Considered that a few times, but I am not sure if the membrane withstands prolonged exposure to the micro solution. I don't want to dose weaker solutions, they tend to grow fungus and "stuff".

So the dosing volume might be an issue too. Needle valves clog fast when dosing Flourish...

I have some little timer circuits on order which let me dial down the dosed amounts (15 sec minimum). That, in combination with the waterpump doser, will be my weapon of choice. Cheap, reliable, fairly precise, and self-mixing. Sweet.
 
#32 ·
Been using mine continuously for over a year and have had no problems with things gunking up on me yet. Granted my reservoir and tubing have become stained over time and don't look as nice as they did at first but they still function just like they are supposed to. I am using 2 of those Tom aqualifters one for micros, and one for macros. I'm using 1gal milk jugs as my reservoirs.
 
#31 ·
If you add a little HCL in the micro mix, then you do not get that funk in the line........

Also, more dilute solutions make it easier to use timers and dose more volume and more precision on the amounts that you dose.

It's easier for most any system to dose a larger volume more precisely.
3.5 gallons per hour is pretty low, so a simple timer with 1-15 min increments (intermatic digital etc) can be programmed to dose 220 mls at zero head.

Now adding a valve or better yet, head pressure(less clogging issues with small orifices) can tweak that down to much less.

Whatever amount you really want.
You can also use a larger adapter for the tubing, but you still have the small pump part that's the constriction/clog point.
However, some simple things get around the issues of funk in the lines or rot of the traces.

This is really old news about dealing with fungus etc:
# Put your traces in a dark brown or opaque bottle (I use well rinsed glass cough syrup bottles). Light can destablize chelates.
Refrigerate stock (mark it well so someone doesn't take a swig) and leave a small 25ml bottle by the aquarium for convenience.
Since dry ingredients are more stable, only mix up a month's worth of PMDD at a time. Kevin Conlin adds a little HCL to the brew; (about 0.8ml 9M HCl/500ml). SeaChem acid buffer also or H3PO4 can be added to stop fungus(it hates really acid conditions....as well as bacteria, think peat acid bogs). Excel also will work pretty well, but dose not last that long, it will not oxidize anything at least.

Should help out a lot no matter what auto dosing routine you want to use.


Regards,
Tom Barr
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top