Skimming Principles

[BeanAnimal]

What do you feel is making the adjustment so hard to increment? Do you see a particular design flaw or weak point aht is causing the all or nothing?

[tinygiants]

I am thinking that my riser is to tall. I may need to raise my water level higher than my hartford loop is plumbed for in order to compensate. I can remove a pipe and add a longer one, but it will have to wait for the weekend. If I use the air to compensate, I get skimmate. I just get more than I am thinking I should. I get 3 cups of gingerale looking stuff. I am after coffee.

I may try a quick test install of my cup right off the 4" fernco. That will lower my cup by about 8". Then I can see the difference without total commitment.

Dale

[goby1]

What's the problem with using the recirculation pump to just deliver more water from the sump? Same flow into the skimmer top... you're just delivering more water from the rest of the tank as opposed to from the bottom of the skimmer. Maybe I'm just zoked right now, but I can't figure out/remember why you did that....

Do we have any more solid data on the bubble size or bubble size distribution from high-end water pump driven skimmers? Apart from the theoretical questions so far, it would be good to compare your bubble sizes/rise rates to more examples of other skimmers.

G1

[tinygiants]

I am calculating Spazz's bubbles right now.

I fed from the bottom to improve current while keeping a good dwell time.
Dale

[BeanAnimal]

goby the difference is simple.

Extra water from the sump is "new" water, or the same as increasing the turnover rate of the skimmer.

Pump water out of the skimmer bottom back into the skimmer top (or vice versa), allows the same water to continue to be skimmed and make contact with the air.

It is the same as using a pump to mix your coolaid, or using the pump to pump new water into your container (displacing the coolaid that is already there).

Bean

[samtheman]

Originally posted by BeanAnimal
goby the difference is simple.

Extra water from the sump is "new" water, or the same as increasing the turnover rate of the skimmer.

Pump water out of the skimmer bottom back into the skimmer top (or vice versa), allows the same water to continue to be skimmed and make contact with the air.

It is the same as using a pump to mix your coolaid, or using the pump to pump new water into your container (displacing the coolaid that is already there).

Bean

If the contaminates you are trying to remove are generally spread in all of the systems water, where the return water comes from makes little difference. In fact "new" water would be higher in concentration than recirculated water, so it would skim more efficently. Your analogy about coolaid is incorrect in that all of the water in the system is already coolaid.
You can skim the tank a few gallons at a time or move the whole tank through it and it won't effect the skimmer efficiency negatively.

[BeanAnimal]

Sam that could not be further from the truth.

1.) Recirculating water that is already in the skimmer and in contact with air will presumably help remove some of the more stubborn protiens.

2.) If your logic were to hold true, then the flow rate through the skimmer is irrelevant over a long period of time (say 2 days).

3.) How do you figure that "the more protien rich water" would skim more efficiently? More efficiently as compared to what? The same skimmer with a higher flow? A bigger skimmer with a smaller flow? This statement directly contradicts your premise that it does not matter where the water comes from.

4.) All of the water in the system is NOT cooliad. The water that has been agitated with air is more apt to release it's protien. (see point 1). This is why surface skimmed water has more protien that that of water that comes from even a 1/2 of water. (read Shimek's, Escobal's, Calfo's, or anybody elses studies and observations regarding this fact).

5.) Cleaning the water until it is 50% clean and returning it to the tank twice is more efficient than cleaning the water 25% and returning it to the tank 4 times. The same itterations can be used to show the dilution of 5 gallon water changes vs 10 gallon water changes. The terms in use of electricity will follow the same pattern.

You can skim the tank a few gallons at a time or move the whole tank through it and it won't effect the skimmer efficiency negatively

This makes utterly no sense and is actually 100% wrong. You need to look up the word efficiency and also have a look at the principle behind skimming (foam fractition). I highly recomend Esobal's book as a starting point. I also recomend a hard look at linear overflow lengths and how they relate to the ammount of protiens that end up in the water going to the sump. You will also find a lot of information in the wastewater industry that uses these same principles (this is us reefers stole the technology from).

Bean

[tinygiants]

I am a water processor by trade. I use filters, carbon, resin, peroxide with UV, and evaporators. Most of what I do does not apply to our skimmers, but processing by dilution does.

Look at the original post about efficiency. You have to process water 9.2 times in order to get 99.9% totaly processed. If you only get 1/2 of the processable contaminants, you have to process the volume anther 9.2 times just to get back to par. I prefer to get more out in the first pass, cause whatever I let get by will take another 9.2 cycles inorder to get a shot at it again.

The single most efficient way to process is called single pass. In this method, all the dirty water starts in tank A, goes through a filter (skimmer, filter.....), and exits to tank B. The efficiency of that process is 100%. If the volume to process is 100 g, then you only had to pump 100 g to get all the processable items out.

If you use the dilution method (return processed water to the same source tank), you have to pump 920 gallons through in order to achieve 99.9% of the water processed.

Dale

[samtheman]

Nonsense!

[jnarowe]

LOL...Dale, maybe you should re-vamp your "occupation" so people understand that there is probably no one on RC that knows more about flow than you! I won't give you up though because I know that the odds are high I shouldn't know either. Just don't go to work ****ed OK?

[tinygiants]

Originally posted by samtheman
Nonsense!

Where do you get your rebuttal basis from?
What part do you feel is nonesense? The efficiency factor?
That is a fact that you could easily verify on google. It is a wonderful thing not to blindly believe things just because it was reported on some bulletin board. Even in this thread I used info that was provided here to later find out there was 10X error rate in the math. So please do your own research. If you can find a source that disputes the efficiency factor, please share it. But the source I used is Escobals book Aquatic Systems Engineering. I also have sources at work, but they are not allowed to be used off site for security reasons

Scenario to show what I mean:

Imagine you have 100 gallons of koolaid. You want to process out the koolaid and have pure water. For easy math sake lets say you have 1 ounce of koolaid per gallon of water. You need to process out 100 ounces of the koolaid from the water. Now lets look at the 2 ways to process. For the sake of this comparison we will assume that the processing method is 100% efficient (never is, but for the demonstration it will prove the point).

Single Pass at 2 gpm
You will process 100 g in 50 minutes with a net of 100 ounces of koolaid removed.

Dilution process at 2 gpm (120 gph)
(total tank and sump volume / feed rate gph) X 9.2 (this is purity coefficent that will yield 99.9% processed water) = hours till processed
So (100 g/120 gph) X 9.2 = 7.66 hours to remove 99.9 ounces of koolaid.

For our tanks, there is no way to single pass process. We will always have to process by dilution. However, the best efficiency is to get as much out while the water is in the processing chamber. Anything that gets through the chamber will take an average of 7.6 hours to get back into the chamber again. The reason is that gallon of water we clean does not stay clean when we dump it back into the dirty tank. It just dilutes the dirty water.

Please post any sources that disprove the above efficiency factors. I love to learn new things.
Dale

[tinygiants]

Originally posted by jnarowe
LOL...Dale, maybe you should re-vamp your "occupation" so people understand that there is probably no one on RC that knows more about flow than you! I won't give you up though because I know that the odds are high I shouldn't know either. Just don't go to work ****ed OK?

I wish it were true. I am just a manager now, but I was the processor for the last 15 years. Last year my crew processed 100,000 gallons (drop in the bucket for waste water treatment, my processing was a little more regulated due to thhe constituants involved) of water back to clean after it was used for its original purpose. I do not write the processes (my engineers do that), but I utilize them in practical application. My system was multiple banks of demineralizers and filters. I used multiple dirty tanks and clean tanks. The reason was the efficiency factor. When you are dealing with 10,000 gallons of dirty batch, you can not play around with dumping clean into dirty.

Dale

[jnarowe]

yeah well the word you use, "constituants", tells the story anyway. (unless the reader is a total knuckle-head)

[tinygiants]

http://www.hawkfish.org/snailman/skimmer101.htm
http://aquariumadvice.com/showquesti...Auto=39&page=2

Also see page 6 & 7 of the attachment from University of Virginia

Since my experience in the trade (industrial water processing) does not count, here are some links to consider before dismissal.
Dale

[tinygiants]

Here is the file. It was found here

http://www.healthsystem.virginia.edu/internet/ iacuc/policies/policy_aquatic_amphibians.doc

Dale

[tinygiants]

I should have used the word purity coefficient. This will yield better results on google.
Dale

[BeanAnimal]

tiny, that was the point of my rebutal in the first place (to sam). He is also missing the point that the longer a particular water molecule is in contact with air, the more apt it is to be stripped of it's protiens. He is missing many other principles as well. In any case to continue down this path is pointless... Waste dilution and efficiency of multi-pass systems are methematicly proveable and accepted science. This point alone invalidates sams entire case, without having to take into consideration the benefits of recirculation and dwell time increasing the skimmer efficiency. Sam simply stuck his foot in his mouth. I guess we all do it once in a while.

Bean

[jnarowe]

Or he's a Troll?

[goby1]

Lot of writing since I last visited.... Seems my question sparked some discussion. I'll have to read what was written in more depth (pun!) and think some more about this.

G1

[tinygiants]

I can bring more water straight from the tank overflow itself. I decided on the feed rate based on goal turnover rate of 2 times per day. Looking into it further, I wanted more current. That is why I have the recirc loop. I get current without sacrificing dwell time.

The skimmer is starting to pull out some really nasty stuff. I just cleaned the cup, and it was foul.

Dale

[goby1]

Well I'm not convinced yet. Of course I understand the dilution theory. But we're not comparing dilution to not diluting, we're comparing two different dilution scenarios (if we're still talking about my question regarding the recirculation pump, that is). Let's say you have 100gph from the sump to skimmer, and you have another 100gph pump and you need to decide what to do with it. Pulling partially filtered water from the bottom of the skimmer or more water from the sump. I just don't see that pulling it from the bottom of the skimmer is better. Hope we can continue to discuss this in more detail. Good job on the skimmer upgrades. Is the new collection cup performing as you thought?

Cheers,
G1

[jnarowe]

I would say without a doubt that pulling water from the bottom of the skimmer and reintroducing it to the top of the skimmer would be more effective than pulling water from the sump. That essentially gives you miore passes on already partially cleaned water.

Think of it like water changes. Some people do water changes by removing 20g and replacing it with 20g of new water. That is a 20g water change. Other people have a 20g resevoir and circulate the sump water through the resevoir to mix and dilute the tank water. Is that really a 20g water change? I don't think so. However, it may be more practical to do it that way.

[BeanAnimal]

goby, Myself and Tinygiants already answered your question, sam begged to differ. Pulling water from the sump and a recirc loop are not even remotley the same thing. It's a matter of how the skimmer works (contact time) and dilution vs the number of passes. The results can viewed from an efficiency of time or electricity. In both cases pulling water from the sump for "recirculation" is a loser.

Bean

[jnarowe]

uh, I didn't ask a question.

[tinygiants]

Originally posted by goby1
Well I'm not convinced yet. Of course I understand the dilution theory. But we're not comparing dilution to not diluting, we're comparing two different dilution scenarios (if we're still talking about my question regarding the recirculation pump, that is). Let's say you have 100gph from the sump to skimmer, and you have another 100gph pump and you need to decide what to do with it. Pulling partially filtered water from the bottom of the skimmer or more water from the sump. I just don't see that pulling it from the bottom of the skimmer is better. Hope we can continue to discuss this in more detail. Good job on the skimmer upgrades. Is the new collection cup performing as you thought?

Cheers,
G1

Well it comes down to dwell time. If I feed my skimmer water at 4 gpm (2 gpm from overflow and 2 gpm from sump) i will only get a little over a minute dwell time. It has been proven that longer contact time yields better protien stripping. Just look at all the becket and NW designs that were fed by the main pump then converted to recirc with seperate feed. The reason for improved performance was the slower feed rate. Every other aspect of those designs was the same (air, water volume, skimmer shape).

I am not certain if it is better to underskim and get 2 turnovers a day or overskim and only get 1 turnover in a day. That is a different debate in itself. I actually think that if that is the case, then the skimmer may be improperly sized for the tank volume. So I did the math and found I need 2 gpm in order to turn my tank over 2 times a day. So I picked a 6" body. At my feed rates I get 3 minutes of water dwell time. Longer is better. Now I could speed up the flow and still get the reccomended 2 minutes of dwell.

This reccomendation comes from the thought that some protiens need 2 minutes to bind to the air water interface. So I refuse to do anything that drops me below the 2 minute dwell. It has been discussed that the true dwell time may be better applied to the air bubble instead of the water. So I set out to try and get both. I get it with the water, but I am far off on the air. (Spazz's NW tested very nice - I posted some of the bubble characteristics in his thread. His average bubble was 2 inches / second. Mine was 5 inches / second.)

In addition, the source of the water is important. Feeding a skimmer directly from surface skimmed water is more efficent than letting that water dilute into the sump and try to get it back out by skimming. The air to water interface is where the protiens are attracted to. There is a higher concentration of skimmable protiens at the surface of your tanks water than any other place in the water column or sump.

Dale