[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Pushing a thousand litres of water

From: "Greger Lindstrand" <greglind at algonet_se>

I use a Ehiem 2252 which (at least when it was new 18 years ago) is able to push 1250 l/h of water.
With the CO2 system engaged on the Reaktor S, the Eheim manages to keep 192 l/h output of CO2 enriched water.

Ok, I figured that it wasn't strong enough so I got myself an AquaClear 5000 powerhead to do the job. It's good for 3550 l/h and can lift 200 cm of H2O.
I was astonished when measuring the output of CO2 enriched water from the Reaktor S. It was exactly the same 192 l/h!!!

The problem lies in the way that pump specs are reported to hobbyists and the fact that you don't know the pressure drop over the reactor. The LPM ratings on pumps are frequently reported at zero discharge head or at best with a specified discharge head that varies from pump to pump making practical comparison impossible.

I think a lot of mystery of practical fluid mechanics vanishes when you think of an electrical system as an analogy... the pump is analogous to the battery, the lines (size, shape of inlets, sudden constrictions, valving and relative heights) and equipment (pressure drop over reactor) supply the resistance, you are interesting in the current (flow in this case). Saying a pump drives xxx LPM is like saying xyz battery provides xx current without specifying the resistance. The zero discharge head velocities are like reporting the amperage of a battery when you short it; pretty worthless.

What you really are looking for is a pump curve (P vs. Q), which gives discharge rate as a function of total discharge pressure drop. Expect to loose 0.43psi per foot of vertical climb, pressure drop due to line resistance in aquarium system is basically 0 (and would require the friction factor of the tubing which is unknown anyway), tees and elbows and frequently estimated in terms of equivalent pipe diameters. The real problem you have is that pressure drop over the reactor is generally a complicated function of flow... To really get this right, you would need to measure this, the best you can do right now is a simple proportional fit as follows.

If you have a pump curve for your current powerhead/pump, measure the flow through the reactor, calculate the total pressure drop by backing that off the pump curve. Then subtract the effect of gravity on the system so you have DeltaP_reactor (pressure drop just due to the reactor) and Q (flow), divide the two to get a proportionality constant. Use this to estimate the pressure drop at the desired flow rate, add back in gravity then look at pumps curves and figure out which pump would give you Q at that pressure drop. Hope this makes sense, I know a lot of this info isn't available for aquarium pumps but it something a little better (and cheaper) than just guessing if the info is available. Check with the hardcore marine guys over at www.reefcentral.com, I think they might have some pump curves for standard pumps. Let me know if you need more info.

Jeff Ludwig