Skimming Principles

[barryhc]

Has anyone ever used a Tunze to recirc an airstone skimmer? This would only add 11 watts to the equation but could make for some incredible dwell times. Let me hear your thoughts...

Well, Mr. mad "scientist-engineer", welcome to the "fray"!

I don't know why you would particularly pick a "Tunze", but then I don't know that much about them, yet.

Please "expound" on this idea, and we will chew it up, and spit it in whatever direction that seems appropriate, including "straight ahead"!

Thanks for "coming in", > barryhc :beachbum:

 

[ChemE]

I specifically picked a Tunze because one can get 1600 gph with 11 watts at a low linear velocity with little to no turbulence (think low speed underwater fan). This is ideal given a large diameter recirculation loop (I'm thinking the same diameter as the skimmer body so as not to increase the flow velocity too much). As we all know, turbulence is our enemy when it comes to removing those weakly adherent proteins yet recirculating skimmers always use insane mondo pumps which beat the froth to death; little chance that those proteins remain captured by the bubble wall in the face of so much turbulence.

So what I am proposing is a wide, low-turbulence, slow recirculation loop driven by a very very energy efficient pump (airstone skimmers are low electricity and that is why I like them, want to keep it that way). The flow velocity down the recirculation loop needs to be just greater than the incipient velocity of the air bubbles (the velocity at which they become entrained by the fluid are are thus dragged downward through the recirculation loop). Sorry for all the technical lingo, you know us engineers just love to drop the science.

Chew away...

 

[ChemE]

Barryhc,

I actually joined the "fray" on the first page on a similar point; contact time.

I'm still unconvinced based on my knowledge of fluid flow and chemistry that what we are shooting for is 120 seconds of water dwell time. A fairly straight forward surface chemistry arguement would indicate that what we really need to strive for is 120 seconds of contact between an air bubble and the bulk fluid.

 

[barryhc]

Barryhc,

I actually joined the "fray" on the first page on a similar point; contact time.

I'm still unconvinced based on my knowledge of fluid flow and chemistry that what we are shooting for is 120 seconds of water dwell time. A fairly straight forward surface chemistry arguement would indicate that what we really need to strive for is 120 seconds of contact between an air bubble and the bulk fluid.

I may be wrong here, and I may very well need to review, but I thought that the original "princiapls" adressed averages of "bubble-dwell, factored to "water dwell", and then "bombardment rate".

Now, without having done this review, yet, "Bombardment rate", may be the point that is being currently covered to some degree by "looking toward" air driven skimmers.

In any case, I follow your point to some degree regarding contact time of an air bubble to the bulk fluid. However, this contact time with the bulk water, will not be adequate if we only pass one ounce of bulk water past the bubbles, per day. Now I'm using the "infinity argument" here, but you get my point.

Actually, the 120 second time, I think is "subjective", and some "dissolved compounds" need more of a "gentle path" in order to attach, than others, and this is getting "quite heavy" to say so, but nonetheless, skimming technology has progressed now , to this point of "finness".

I certianly appreciate your response, and I apologize for not having noticed that you have been involved since "the beginning'.

Keep us "straight" here with your input, It's all good, when we find an answer.

> barryhc :beachbum:

 

[ChemE]

Agreed, we need to pass the required water flow through the skimmer in order to skim all the water twice a day but at the same time we need to ensure that some bubbles exist for at least 120 seconds in the bulk fluid; this can only be reasonably done with a gentle recirculation loop. It would seem that if both these objectives can be met we would have a design that conforms to all of Escobal's rules.

 

[tinygiants]

The reason for not recircing an air skimmer is the independent air source. If you need longer dwell time for the water, just slow down the feed rate.

ChemE proposes a recirc not for water dwell, but air dwell. That is a different idea altogether. I am no chemist, but my logic does not follow the difference in 120 seconds of air dwell being different than 120 seconds of water dwell. I followed the thought on it being the airbubble attracting the protien, but that would assume the same molecule of water stays with the same molecule of air for the 120 seconds. If I managed to get a standard design to meet the principals and have 120 seconds of air dwell in counter current design, how would the water molecule stay with the airbubble? Doesn't the bubble time in the foam tower count in that binding time?

Dale

Well Dale, that dwell time might lead us "back" to a larger dia. to accomodate "dwell time requirements. Various "wild a$$ options" come to mind here, for dwell time, in conjunction with "large dia.-short" skimmers.

There is no reason why we can't incorporate recirculation, in an air driven skimmer, but the design might be very far away from what is considered "normal".

Give me a "say what?" on that one. > barryhc :beachbum:

 

[tinygiants]

Interesting points there ChemE. I had not considered the info in that way. If the 120 seconds is a valid number, and the bombardment rate is a valid ratio, the water dwell would have to be 20 minutes.

Water dwell / air dwell = bombardment ratio (goal of 10 according to Aquatic Systems Engineering by Peter Escobal)

water dwell / 120 seconds = 10
Water dwell = 1200 seconds = 20 minutes

With a 2 gallon feed rate my skimmer would have to be 40 gallons big to have that water dwell. In a 12 inch body, that is over 80 inches tall. This of course does not guarantee the air bubble dwell.

There are so many factors when trying to balance everything.

Dale

Getting back into ChemE's point, meeting all of Escobals principals. If height was not a concern, you could could make the body skinnier and taller till the bubble took the 120 seconds to rise through the 40 gallon volume. (Based on the numbers above) A 8" skimmer would be about 15' tall to get the bombardment ratio right.

Dale

 

[tinygiants]

Thinking about it more using both barryhc and ChemE ideas, the recirc does seem to be feasible. The water dwell ill take care of itself based on feed rate. However at 2gpm feed in my tank, I am not going to create much counter current to slow down the bubbles. Using a standard recirc of the water to create an increased counter current would lengthen the dwell time. It would also decrease the amount of air needed to maintain the 13% volume if the bubbles stayed in the water for 2 minutes.

How fast would the water have to flow to in a 6' tall body to get 120 seconds of dwell?

 

[ChemE]

Dale,

To explain the difference between water dwell time and air dwell time reread my post on the first page; it was long winded but I thought thorough. I'll summarize here.

The protein will remain captured by the bulk fluid until it embeds in the bubble wall. This process can take up to 120 seconds. Allow me the use of an analogy that might make things clearer...

I have terribly slow reactions, so much so that it takes me 120 seconds to catch a basketball thrown at me. Throwing millions of basketballs at me very quickly will not help me catch even one. What you need to do is throw the basketballs extremely slowly, so that they take 120 seconds to pass near to me thus giving my slow reflexes time to kick in and catch it. Now, it is impractical/impossible to throw things this slowly since gravity makes them move faster than that so we'll throw me too. If you throw me at the same speed as the ball, then I get my 120 seconds next to one ball and am able to catch it.

It is the same way with bubbles and proteins in the water. It doesn't do us any good to pelt a protein with a cloud of fine bubbles because it doesn't spend enough time next to ONE of them. Sure, it might get pelted for 120 seconds before it exits the skimmer but that is not how embedding works, it needs 120 seconds next to the SAME bubble to embed. If it is in contact with one bubble for 2 seconds and then it moves away and is in contact with a new bubble the embedding process must by definition start all over. 60 such restarts before exiting the skimmer will produce no result (at least not the one I want).

I think bombardment rate is a load of crap unless I completely misunderstand it (possible).

The only thing that makes any sense whatsoever is one bubble being in contact with one protein for long enough that that hydrophobic portion of the protein moves inside the air/water interface and becomes captured by the air bubble. This increases the surface tension of the air bubble and if this surface tension is increased enough, it will not pop until it is in the skimmer collection cup. Then and only then is a protein removed from the system.

 

[tinygiants]

Agreed, we need to pass the required water flow through the skimmer in order to skim all the water twice a day but at the same time we need to ensure that some bubbles exist for at least 120 seconds in the bulk fluid; this can only be reasonably done with a gentle recirculation loop. It would seem that if both these objectives can be met we would have a design that conforms to all of Escobal's rules.

Would a really long loop with the air entrained in the fluid be different than the recirc loop using water flow rate to slow down the air bubble? If the water and air travel down the loop at the same speed (trying to keep the bubble with the same water molecule) wouldn't the airbubble just collect at the top of the pipe?

This is good points being discussed here. I am enjoying this information exchange. Thanks to everyone who is sharing.

Dale

 

[barryhc]

Well, here is "the thing". Many of Escobals reccomendations are based as much on emperical observations as any "scientific" calculation, and some of them particularly so.

I have seen "wet neck" applications that run as high as 2000 gph of water flow, for a 750 gal. tank. That is about 70 turnovers a day VS two. Escobal was not observing "wetneck" systems when he made his observations.

I am not necessarily even promoting "wet-neck" technology, when I say this. Still, you are charting new ground, and further observation will tell "the tale".

> barryhc

 

[barryhc]

Beautiful job there ChemE, were "passing replies" again, oh well.

I really love the analogy, and I'm going to sleep on it.

Thanks, "tommorow". > barryhc

 

[tinygiants]

ChemE,
Thanks for explaining the reason for the air bubble dwell being important. Your analagies make sense to the pipefitter in me. I am just having trouble with the concept of the same water molecule with protien staying in contact with the same air bubble. As the bubble moves through the water column isn't the surface of that bubble constantly changing which molecules are on it?

 

[tinygiants]

barryhc,

I am still in the design phase, but I am trying to do my homework. I am about ready to order my air pump, stones, and uniseals. I have just been trying to make the skimmer as "efficient" as I can. I have a downdraft skimmer that is converted to beckett, but it is weak at best. There is no dwell time for air or water since it is not a recirc design.

 

[ChemE]

Tinygiants,

In a word: no. The no slip boundary condition used throughout fluid dynamics has been observed time and again (see my talculm powder experiment in my first post to this thread to prove it to yourself). This no slip boundary condition assures us that at the moment an air bubble forms, the water molecules at its surface will remain there until it pops. Now this still allows for proteins to diffuse to and away from this boundary but hopefully they will diffuse to it and then stay there long enough to become captured.

In response to your earlier question about the closed loop, I think that if done properly we could keep a gentle recirculation loop going without getting a build up of air bubbles at the top. But like barryhc says, we need to experiment to know for sure.

By the way, I too am very much enjoying this thread. My skills with DIY are very good but I haven't yet begun practicing on acrylic so I'm going to rely on others to prove or disprove my theories. I am very sure that we can come up with a air driven recirculating skimmer that uses 20 watts yet will overskim a 300 gallon tank. Most of the skimmer designs out there are terrible and have little or no thought behind them. They basically make up for their lack of design finesse through 300-400 watts of power draw.

 

[Puffers]

I think bombardment rate does come into play. If we only have one bubble in our skimmer that had a contact time with the bulk water for 120 seconds it won't take very much out of the system (unless it's one big bubble). So how many bubbles are too much or too little?

Is there a point where many fine bubbles in a skimmer body decrease the amount of dwell/contact time needed? The thinking here is that because there are so many small fine bubbles there is not much room left for the proteins to go....?

I also think air driven skimmers have come a long way and think there is huge potential in this area of skimming.
:bum:

 

[barryhc]

I think bombardment rate does come into play. If we only have one bubble in our skimmer that had a contact time with the bulk water for 120 seconds it won't take very much out of the system (unless it's one big bubble). So how many bubbles are too much or too little?

I think that the bombardment rate is the means to an end, and that end is maximum saturation. Maximum saturation is supposed to be 13%. I'll have to check back, to see if this represents water or air. It seems terribly low, for air, when you look at a "milky froth".

Is there a point where many fine bubbles in a skimmer body decrease the amount of dwell/contact time needed? The thinking here is that because there are so many small fine bubbles there is not much room left for the proteins to go....?

I think this may actually be a valid point. Let's think about something here. Depending on the various designs, there are often times, two different "areas" of oxgenated water. Even three, if you like. There is the "foam" in the neck, the "foam" near the top of the skimmer body, and the "bubble water" mixture that is lower in the skimmer body.

Now this second area of "foam", is "leaking" water concentration, and therefore concentrating the skimmate to water ratio.

The third ( or top ) "foam" area in the collection "neck" is obviously 95 to 98% air. The second "foam" area in some skimmers could be as high as 80 to 90% air. This is sitting of course, on top of the first area, which is where I think that most of this "maximum saturation", bombardment rate, dwell times, etc. is going on.

I think that it is this first area, where all of these considerations are taking place, and I wonder if "we" are failing to consider the importance of the "second area", and the "dwell time" that is occuring there.

I don't think it is at all the same as what is occuring in the "lower-first-area". The "foam" in the "collection neck" is obviously getting a lot of dwell time, in fact it self regulates to some degree, and "waits" until it has had enough dwell time ( and water leakage ) to finally "push out the top".

The same kind of thing can be occuring in the "second area", and we may be able to control this to some degree with the design.

It's something to think about. What do you think?

> barryhc :beachbum:

 

[gbtower]

I don't think you can count any other "area" other than the first when figuring the "dwell time", because that time is what is supposed to be required to have the protein stick to the bubble. At that point, the water should be separated from the air -> which is what is happening in the other 2 "areas" you delineated.

I wonder about the whole 120 sec contact time figure. How was that determined? I would think that a protein would latch to a bubble once it becomes oriented in a favorable way. I don't think there is some other chemical bonding/binding interaction going on. So, is the 120s the average time that these "difficult" proteins take to orient? If so, would a little localized turbulence speed up the time it takes to orient by spinning it around a little quicker?

I'm also not so sure if the no-slip boundary condition would "help" in pushing for more counter current flow, since some proteins may be too large to be contained within the boundary area. Sure, the talc may stick to the windshield, but a significantly smaller percentage of cat hair would.

Moving on - how do we go about trying to remove larger particles with a skimmer? The way I understand it is that we push so much air, that the froth becomes a mech filter, like a simple screen, since there is not enough space between bubbles for the particles to pass through. To do so, we would/should most likely be pushing more than the magic 13% air.

Amd just to throw in something from BB principles, if we want to try to get stuff out before it is broken down, perhaps we should concentrate more on detritus skimmers than protein skimmers (protein skimming would be a secondary effect of the detritus skimmer)? Or is there a better way to get the larger stuff out than oversaturated wet skimming?

Just dumping some mental garbage... You guys put too many ideas in my head.

 

[barryhc]

To do so, we would/should most likely be pushing more than the magic 13% air.

The 13% limit is supposed to be a physical limit that we can't avoid ( or pass ), not a "target".

Amd just to throw in something from BB principles, if we want to try to get stuff out before it is broken down, perhaps we should concentrate more on detritus skimmers than protein skimmers (protein skimming would be a secondary effect of the detritus skimmer)? Or is there a better way to get the larger stuff out than oversaturated wet skimming?

Seriously, some filter floss, that we throw away once a day, shortly after feeding, would perform miracles.

> barryhc :beachbum:

 

[tinygiants]

The 13% is supposed to be the maximum saturation of the small bubbles we are after before the volume of air causes the bubbles to join. The reason joining bubbles is bad is based on surface area. Multiple small bubbles have more surface area than fewer larger bubbles. The surface area is where the binding takes place.