by Tony
Alternate air pump performance data
7:24 pm in Materials and Resources by Tony
Alternate pump performance data
In the spirit of R&D, here is the performance data to my alternate pump design that I had shown in my first post. http://our.windowfarms.org/2010/03/10/just-getting-started-with-slightly-different-pump-design/ The design is easy to assemble and has been very reliable.
The data shows the relationship between resevoir height, pumping height and flow rates. I do not know how the air pump pressure effects these since I can not change that. I think the general conclusions will hold true for any pump design including the standard 3 plant windowfarm design.
As mentioned in the troubleshooting guide, http://our.windowfarms.org/2010/03/05/airlift-troubleshooting/ the resevoir level is very important. Small changes in the level really effect your pumping rate and how high you can pump. My one gallon resevoir level naturally drops about an inch a week. As it drops the flow rate goes down, but it keeps working. If I let it get near 5 inches, then the pump would stop working completely.
Excellent, (if a bit nerdy!) I have been experimenting with a narrower pipe (6mm.) which has the advantage of being black (it’s irrigation tubing) but it changes the way the air lift works, and if fact makes things harder, I’ve tried various air lift designs, angles, depths and all of them seem to suffer from occasional stalling (just pumps out air into the water)
The length of the pipe between the air supply join and the water intake seems quite critical (didn’t realise that ’till today – mine were too short)
It’s a fact that a shallow reservoir (I’ve been using 12 inch deep 2 litre bottles) makes things difficult – people should use the deepest reservoir they can.
My 6mm. tubes though (when I tease them into working) are good in that I can run them through the 3/8 inch drain/return pipe and actually up through the bottles – so I don’t need any external piping or support for piping – so I think it’s worth trying to find a reliable formula for air lift performance.
Nerdy? Thanks I’ll take that as a complement. Geeky works for me too. I’m an engineer… If you want to mess around with some formulas take a look at this. http://www.plantservices.com/articles/2002/12.html
This weekend I want to plug my data through the equation and see if things match up. The one thing I am missing is the specifications on the air pump. I need the psi and cfm. I can not find this data on my aqua culture pump nor the petco pumps online. Maybe it comes with the literature with the pums, but I have an old pump. Maybe someone has a petco 9901 or 9903 and can send me the data. My pump is rated 2watts which might be similar to the petco 9901 (3watt). They recommend the 9903 which is 5 watt. People could be using a smaller less expensive pump with a better pump design.
As far as the equations go, I think the key is the amount of air flow. More air flow means more lift and more liquid flow. I think an issue with the pump design is the air needles. They restrict the air flow. I would guess it does not work with one, but when you add the second it works. Well, you are doubling the air flow (cfm). I have neither the air needles or petco pump so I can not test this theory.
Well, I’ll play with the data this weekend and also try to hit the library and find the other article that it references. How’s that for nerdy…
Hey – the planet needs us nerds!
Yes, she sure does. Just a quick update. I plugged my normal operating conditions data through the calculation in this article and the air flow it calculated looked to be in the general area of the small Petco pump, but not exactly. I went to my local Petco and wrote down all the data on the pumps off the boxes so I have that now. I still have nothng on my pump though, but I am still digging around there.
Note knowing anything about the assumptions that went into this formula, I am not sure if it scales down to the small distances and flow we have. I went to the local community college this last weekend and they were able to do an interlibrary loan on the referenced journal article. It came in right before my trip this week. I have been reading that the last few days. As it turns out, this article from AICHE actually is not where the formula can from. This article does have it’s own set of formulas that I’ll look into as well. However, it does reference the original equation and included a reference to it. It’s from a Chemical Engineers Handbook. When I get back home, its back to the library…
You could get a relatively accurate guess of the volume of air you pump puts out by:
Measure the ID of your lift hose
Place a yard stick next to your lift
take a pic of it running so you can determine how much air to water is in it
time how long it takes for a bubble to rise a specific distance
Then [(pi r^2)*(height - water)]/time
Well DONE as far as it takes us. I am a retired Chemical Engineer from the R&D side of the house, so I will look into this a bit more and come up with some improvements in the future. However, I am off on a 2 month vacation next week.
Great work.
I got fired up a few months ago just after WF got their Kickstarter funding achieved, and did a one window top reservoir system after troubles getting an airlift system to work for me (winter in Oklahoma, couldn’t get my lettuce to do as well as my basil, yadda-yadda & etc) but now I’m looking to give it another go after reading Tony’s “Alternate Air Pump Performance Data” post.
I was wondering if anybody could shoot down an air-lift idea I had before I go and try it: using a mini-soaker hose (conected to the air pump) inside a slightly larger lift tube to achieve greater lift heights?
Anyone have an opinion/data on whether distributed bubbler-lengths can be beneficial?
It might make things a little difficult to re-enter your window farm bottles but I suggest piping the water down in a U from the bottom of your reservoir and having a t joint letting in the air at the end of the U. You have much more control over the “submergence” then. Submergence has a very big effect on pumping rates. Another thing that has a big effect on pumping rates is air speed through the “up” pipe. Has anyone considered a little variable bleeder valve to let out air as it leaves the pump and actually have less air flow?
Sometimes lower air speed means more! liquid holdup and therefore more liquid pumped! I tested my little aquarium air pump in a blue rain barrel. It can pump air at least 25 inches deep. That means any one of you could have 25 inches of submergence too.
I put a video on youtube to show how submergence makes a difference.
It is at http://www.youtube.com/watch?v=TpfHRViress I have no windowfarm. This is purely a demo to help people along.
I did lots of experiments with tromps and low pressure airlift about 20 years ago for my coffee jar vacuum pump and pulser pump experiments. So I have more experience than most.
Another thing I noticed is that in the diagrams on window farms, the aquarium pump is always lower than the reservoir. In my view, it should always be higher than the reservoir just in case there is a power outage or a pipe comes loose. If either thing happens, you could get water inside your pump or all over the floor.
Hope this helps
Brian
Just a note that the t-joint and u shaped tube to prevent backflow of air works really well to pump water. Using 2 ft of submergence, (and a pump that can pump air maximum 3 ft under water) I easily pumped water to 13 ft high! I used 1/4 inch tube or 3/16 inch. (Internal diameter). People should note that the right air speed is important. If you pump too much air into the tube, you get “churn flow” and pretty much no pumping. If you ease it off a bit on the air, you get plug flow and significant pumping of water. My video to 13 ft is at http://www.youtube.com/watch?v=lKtB1YKoMxk If anyone has a suitable space, why not try find out the limits of pumping with the t-joint system?