Continuous vs intermittent water change

[alemone]

As I am setting up my automated water change system I was wondering- how much "fresh" water do we waste by doing the water exchange continuously? We start removing the new water right away, right?

The answer wasn't obvious to me so I did some calculations. Skip to the end if you don't care how the results were derived.
Let's say we do 10% water change in 1 month (1 week, 1 year- the interval is not important).
1. If we do 1 water change, we are left with 90% old water, 10% fresh water.
2. If we do 2 water changes 5% each, we are left with 0.95*0.95- 90.25% old water, 9.75% fresh water.
3. If we do 3 water changes 3.33% each, we are left with 0.9667^3- 90.32% old water, 9.68% fresh water
x. If we do x changes totaling 10%, we will be left with (1-0.1/x)^x old water

Continuous water change will result in the following amount of old water after 1 month (or week, year, etc): see image

The final result is 90.484% old water, 9.516% fresh salt water vs 90%-10% with a single change.
Continuous water change does not waste much replacement water. Once you get to 50% water changes, the wasted amount of fresh water becomes more meaningful (you are left with 60% old water and 40% fresh water if done continuously).

 

[lolgranny]

Good thing I'm on a well and don't have city water!

I believe randy did a article about continuous vs one time changes. I'll see if I can find it, but if I'm correct the one time remove more of the unwanted then continuous did. Hopefully someone can chime in


Sent from my iPhone using Tapatalk

 

[alemone]

Good thing I'm on a well and don't have city water!

I believe randy did a article about continuous vs one time changes. I'll see if I can find it, but if I'm correct the one time remove more of the unwanted then continuous did. Hopefully someone can chime in


Sent from my iPhone using Tapatalk

Only marginally so according to my calculations.

 

[reefgeezer]

There is also a stability benefit to continuous water changes. Slight variations in temp, salinity, ph, & alk have less impact.

 

[bearpeidog]

Here is Randy's article:http://reefkeeping.com/issues/2005-10/rhf/index.php

 

[d0ughb0y]

your example (and formula) is to remove old then add new.
does it make any difference if you add new then remove?
or simultaneous add and remove?

 

[reefgeezer]

your example (and formula) is to remove old then add new.
does it make any difference if you add new then remove?
or simultaneous add and remove?

It matters only slightly. The math says the the most effective way is to remove then add, but we're talking about a difference that is so small it isn't enough to sweat. Not that I think it matters, but "add then remove" is the least efficient.

 

[sde1500]

Easy to add and remove at the same time, add by return pump, remove by the overflow outlet, unlikely to remove much new water that day. And as far as daily small vs weekly larger, the difference is so minuscule as the math shows above that I prefer the added stability of a daily change.

 

[d0ughb0y]

since we are this thread is about quantitative analysis, I was looking for more than just "isn't enough to sweat".



also, how about the consumption and replenishment of elements? Doing the change instantaneously vs over a longer period of time would not have the same consumption/replenishment. looking for quantitative analysis as well. I think the article referred to by second post may have covered this. I vaguely recall reading that article.

 

[alemone]

your example (and formula) is to remove old then add new.
does it make any difference if you add new then remove?
or simultaneous add and remove?

It does but if you do it sequentially there is really no reason to first add, then remove. The example shows how the effectiveness of the water change decreases (minimally) as you go from a single change of 10% to 2 changes of 5% to 3 changes of 3.33% to 4 changes of 2.5%...to infinitesimal intervals and tiny amounts i.e. continuous simultaneous removal and replacement. The amount of new water is always the same- 10% of the total.

 

[alemone]

If you do 100% change, continuous replacement/removal will decrease the amount of new water added over the given interval of time from 100% to 63%.

 

[ReefsandGeeks]

For normal every day water changes for the sake of depleating naturally rising elements, or raising naturally depleated elements, It would be more beneficial to perform small changes, as close to continuous as you like. Yes, you do lose a small amount of new water that you essentially add then quickly remove, but this will give great stability to your tank chemistry. Doing a 50% change all at once is more efficient, but also risks a quick swing in water chemistry or temperature that the inhabitants need to acclimate to.

If there was something wrong with the tank, and I wanted to do a water change as a reactive measure to remove some accidentally added unwanted chemical (for instance, if a kid dumped something in) than I'd do a one time 50% or so change all in one shot to remove as much as possible, while sacrificing the stability of the chemistry in eschange for faster removal of some detramental thing. And of course some GAC would be changed out also as a reactive measure.

The math is nice to show that, while many small changes is less efficient, it's not that much less efficient unless performing a very large % water change. Great info to have so you can personally decide at what water change % it would be worth an all at once change vs smaller changes. Just keep in mind too, that there's more to the equation than old water vs new water. The bigger the difference in water change, the bigger the effect of any differences in water perimeters, for better or worse.

As far as adding first vs removing first, you'd also be limited on how much you could add before you'd flood your sump. This could have a benefit of being able to add first, then drain and still keep your return pump running during the change. Otherwise, this would still be marginally less efficient than removing first. Again, would depend on the % change, with higher % changes being increasingly less efficient than removing first.