Dave W's 3000 gal plankton/reef tank

[Chad]

It does look nice...

 

I love the get er done method!!

 

If you want to try to steal my way of getting around google's inconvienence... I set the image in sketchup like I want it to look, then press the ctrl and prtsc buttons at the same time. Then open MS Paint and press ctrl and v... the image (and everything else that was on your screen) should appear on the screen. You can then modify it in paint as you want to and save it as a jpeg for easy upload.

 

edit: exactly like David pointed out already.

Edited November 7, 2010 by Chad

[ctenophore]

 

Recently, I read a study somewhere that said calanoid copepods are mostly absent from captive systems because their antenae are damaged by impellors, which mortally wounds them. I think there is merit to what you are doing. Hopefully, it will be successful.

 

Maybe Jon will figure out his Archimedes screw out of acrylic. I would think it could be done on a capable CNC machine.

Calanoids are mostly absent from captive systems because they get eaten faster than they can reproduce. Way too many mouths in far too small a volume of water. Forget pumps, if a plankter has to dodge 20,000 polyps on a trip through the tank to even think about breeding (once every 10 days?) it doesn't stand a chance.

 

It would be easy to make an Archimedes screw on a 4-axis mill and some high end CAM software. Maybe one day we can upgrade our router to a 4th axis. You might be able to find pre-molded plastic augurs that would work though. I saw some small ones for chocolate fondue fountains but nothing big enough for a 3000 gallon tank.

[dave w]

It looks very nice. You probably mentioned it somewhere in this thread and missed it but what type and thickness of material are you using for the viewing panes?

 

You could hit the "print screen" button and paste that into Windows Paint, then save and post it.

 

Actually I don't think I ever mentioned pane thickness and a lot of other pertinent details. I had an old 600 gallon tank made from 1" acrylic, and I am using these 3' by 9' panes for five of the seven viewing panes. much of the tank. I only need to buy one piece of 1" acrylic that is 36" by 96" to cut in half for the two 45 degree panes. After accounting for the plywood and fiberglass flange holding the acrylic, the tank will be about 40" high.

[dave w]

I forgot to mention the piston idea when you were over at my place Dave. You won't need a 24" diameter piston. A 6" wide one will set your entire water volume moving. The trick is finding the resonant frequency to set up a standing wave. I can demonstrate on my 750 gallon tubs. I used a capped 2" PVC pipe inside a larger diameter pipe as a test, and within a few minutes of pumping it up and down by hand I was able to create a standing wave that sloshed water out of the tanks. It's even easier with e.g., a 6" dia. 1-gallon bucket. Building a piston drive exactly like Garratt's will cost about $200 in parts, assuming using 6" duct PVC for the cylinder and a custom cut inner piston. It will move ALL of the water in your tank. I would still use a few of the large props to circulate that water- the wave just moves the water back and forth over a small area, you still need some (but not a lot) of force to cause true circulation. Plan on keeping at least 6" of clearance to the rim of the tank if you go this route so as not to slosh water out with the piston.

 

Justin

 

Thanks again Justin for your generously sharing your knowledge. Everything else being equal, I wouldn't want to lose 6" of water, but If the benefits outweigh the cost that is what I will do. I can see great gas exchange by this method, and it probably looks nice aesthetically.

[dave w]

Your fiberglass idea is good, what about using a sonotube for the mold to wrap fiberglass around? It might by difficult to get it out after the glass cures, but it is an idea. The disc's will be easier to size than the diameter of the tube.

 

As far as the flange, I would think that seating the center "tube" in a groove and holding the whole thing together with a strongback that doubles as a mounting point for the piston rod, like this. Stainlss steel is an acceptable material.

 

 

and the back

 

 

That looks like a great modification.

[dave w]

Calanoids are mostly absent from captive systems because they get eaten faster than they can reproduce. Way too many mouths in far too small a volume of water. Forget pumps, if a plankter has to dodge 20,000 polyps on a trip through the tank to even think about breeding (once every 10 days?) it doesn't stand a chance.

 

Like other things Justin says, this makes good sense. But with a continual phytoplankton drip and 25 plankters per drop, 3,000 gallons would have about 270 million plankters, so each of the 20,000 polyps would have to eat faster than 145,000 plankters every ten days to stay ahead, assuming a 10 day plankton reproductive cycle. Many plankters reproduce much faster than a 10 day doubling, in fact most have entirely female populations and throw off several to many larvae daily. For example, rotifers reach sexual maturity in 18 hours and produce about one offspring per day. Ciliates reproduce by splitting in half very fast. Benthic amphipod adults stick to the substrate, but produce about 30 planktonic larvae prior to each molting. And as I said earlier, 2 million oyster veligers a day (even though they are only planktonic for about 8 hours) can keep a few corals fed.

 

And this is just talking about the free floating adults. The sponges, fanworms and sand inhabitants produce many offspring also. One thing I like about refugiums with filamentous algae is the many millions of plankton they produce. But as they say, the proof is in the pudding. Looking at a microscope slide a year after setup will tell us whether Justin is correct or if he owes me. In reality, some types of plankton that resist polyps will become dominant, and other types easily grazed will be very scarce. All in all, I think a plankton tank can keep ahead of the predators if my phytoplankton drip stays continuous. I could be wrong.

 

Oh, and one other thing after going "all technical" on Justin, for which I apologize. Justin would say that with adequate food 20,000 polyps this month becomes 40,000 next month and so on until the food supply is consumed, and he'd be exactly right. But there are other strategies which can augment plankton. For species which take a long time to reach sexual maturity like artemia, I can drip phytoplankton in to a large tank to keep the adults fed, then screen the overflow so that the nauplii constantly flow into the tank. So even helpless plankton like artemia can be set up to live long enough to reproduce. It all stems from the ability to culture adequate amounts of phytoplankton in semi-continuous cultures. I can say that after a long learning curve, it became much easier for me.

Edited November 7, 2010 by dave w

[ctenophore]

ame='dave w' date='07 November 2010 - 01:12 PM' timestamp='1289153552' post='340490']

Looking at a microscope slide a year after setup will tell us whether Justin is correct or if he owes me.

 

Yep, pretty much I hope I am happily surprised that the system is teeming with "quality" planktonic life once corals & other predators mature into stable colonies. Time will tell.

 

Maybe you could get away with 4" of clearance for the piston wavemaker, but I would do it for the aesthetic value. That big of a surface should create a standing wave pattern several peaks across @3" tall, it would look like a slice of an actual ocean, especially with sunlight glimmer lines playing across the bottom. You'd probably get sand ripples too, not to mention nice movement of corals.

[Coral Hind]

I think the shape this tank will make it harder to reach a good standing wave as the angles may cause some cancelling out of the wave. What do you all think?

[Chad]

David, I think it depends on the individual lengths... I was thinking that too. If the long section ends up such that its resonant frequency wavelength is a multiple of the short length, it would be ok... I think it would probably need at least two wave devices that are driven by the same prime mover (or electronic signal), one in each back corner.

 

It would certainly be a tough thing to tune in with the horseshoe shape.

[dave w]

David, I think it depends on the individual lengths... I was thinking that too. If the long section ends up such that its resonant frequency wavelength is a multiple of the short length, it would be ok... I think it would probably need at least two wave devices that are driven by the same prime mover (or electronic signal), one in each back corner.

 

It would certainly be a tough thing to tune in with the horseshoe shape.

 

Hopefully a little tuning will determine where to place piston devices to achieve resonance. I don't mind three versus one device if that is what it takes. Won't pipes connecting the two legs of the tank lessen a resonant effect?

 

Regarding overflow with the piston setup, I will have a 4" plywood interior coping (like a swimming pool) inside the top frame of 2" x 2" stainless angle, so peak waves may not overflow much. This also makes a nice shelf that invariably collects stuff. A design principle I follow is to never have tank penetrations below the water line, so I won't be able to drill holes until a wave test sets a waterline.

[davelin315]

I think the shape this tank will make it harder to reach a good standing wave as the angles may cause some cancelling out of the wave. What do you all think?

 

What about having 2 pistons on opposite ends of the tank that go much slower than the other. Timing would be hard, but if you could have the waves meet at the junction of the two parts they would create turbulence there and then hopefully rebound a bit.

 

Oh, and surge devices are great as well. Your key is getting it up high enough where it will actually siphon all of the water down into the tank. You honestly don't need to have that big of a pump driving the water up, either. You could do this slowly and have it only surge every few minutes or you could even rig up some sort of water wheel to carry it up to surge if you wanted to get really fancy with it (or unfancy as the technology may be!).

[dave w]

What about having 2 pistons on opposite ends of the tank that go much slower than the other. Timing would be hard, but if you could have the waves meet at the junction of the two parts they would create turbulence there and then hopefully rebound a bit.

 

Oh, and surge devices are great as well. Your key is getting it up high enough where it will actually siphon all of the water down into the tank. You honestly don't need to have that big of a pump driving the water up, either. You could do this slowly and have it only surge every few minutes or you could even rig up some sort of water wheel to carry it up to surge if you wanted to get really fancy with it (or unfancy as the technology may be!).

 

My shortcomings in mechanics ask what I am missing. I see how pistons create waves, but would this energy penetrate deep enough to create turbulence? No matter how resonant the timing, it seems that moving a few gallons every few seconds is too little energy to create more than surface waves in a big tank. If it really caused bottom sediment to sway back and forth with the wave action, sea fans and gorgonians would look just like a real reef.

 

I think a surge device could dovetail nicely with water coming out of the refugium. Could surges be timed closely enough to create resonance? Right now I plan 4 refugiums dumping into three places. I can change the layout so the fuges dump into a surge device on each of the two corners about 20' apart with a much larger volume of water (and longer intervals) than the pistons. If dump times became too inconsistent to cause resonance, could a timer be used to keep them in tune? I guess my question is: what is better, the more perfect timing of a low energy motor driven system to create resonance, or the higher energy of a surge system that may not have consistent timing?

 

Thanks again for everyone's creativity.

[dave w]

I think the shape this tank will make it harder to reach a good standing wave as the angles may cause some cancelling out of the wave. What do you all think?

 

You're right that the shape will make resonance much more problematic than a rectangular tank. But each of the inside corners are at least 6 or 7 feet deep (front to back measurement), so waves might turn this corner better than we think. Another option would be to forget the side legs of the horseshoe and just try to time waves in the 20' long dimension of the tank.

[dave w]

David, I think it depends on the individual lengths... I was thinking that too. If the long section ends up such that its resonant frequency wavelength is a multiple of the short length, it would be ok... I think it would probably need at least two wave devices that are driven by the same prime mover (or electronic signal), one in each back corner.

 

It would certainly be a tough thing to tune in with the horseshoe shape.

 

The long section is 20' and each of the two short sections are about 11.5'. But isn't it the case that a natural resonant frequency exists for this shape, it's just a matter of moving the wave devices until they are in the right positions? I don't mind using several wave devices if that's what it takes. With 43 feet of back wall to use, I could even space the refugiums over a dozen dump buckets if that would help.

 

Up to now I've spent my time thinking about total flow. The concept of frequency in a tank is new to me. Thanks to everyone for exploring this.

[davelin315]

The difference between a surge and a piston or a fin is that the surge will simply flow water from one side to the other. It could cause you to have the effect you're looking for, but it's more of a violent, if you will, surge of water from one place to another. A piston or a fin is more like picking up the tank and rocking it back and forth. They have similar results, but accomplish them in slightly different ways.

 

Best way is to demonstrate it in a container of water (go fill up your bath tub!). The surge would be like taking a container of water out and then pouring it back in through a smaller diameter tube. It'll rush out with a great deal of pressure if you have enough water in it. It's actually the same concept as a toilet when you hit the flush valve. A fin would be like putting your hand in the water and swishing it back and forth, only you'd be using a better object to move water in your tank. A piston would be like taking a bucket of water and lifting it up and down in the tank, causing displacement to go back and forth from a large area that is both pulling water in when it's going up and pushing water back when it's going down. If you've ever bobbed up and down and seen the effects on the water around you as you do so, that's what a piston would do.

[dave w]

The difference between a surge and a piston or a fin is that the surge will simply flow water from one side to the other. It could cause you to have the effect you're looking for, but it's more of a violent, if you will, surge of water from one place to another. A piston or a fin is more like picking up the tank and rocking it back and forth. They have similar results, but accomplish them in slightly different ways.

 

 

 

So it seems best to have both. If surge devices can be timed to create resonance that would be a big bonus. If pistons generate enough energy througout the tank to create back and forth motion on the bottom, the sand ripples would look very reeflike. Waving action from sea fans would complete the picture, with anthias swimming through the surge for fuge plankton. It sounds pretty nice.

[Chad]

The long section is 20' and each of the two short sections are about 11.5'. But isn't it the case that a natural resonant frequency exists for this shape, it's just a matter of moving the wave devices until they are in the right positions? I don't mind using several wave devices if that's what it takes. With 43 feet of back wall to use, I could even space the refugiums over a dozen dump buckets if that would help.

 

Up to now I've spent my time thinking about total flow. The concept of frequency in a tank is new to me. Thanks to everyone for exploring this.

 

Yes, a natural frequency will exist for most given systems regardless of shape... HOWEVER, it is a matter of whether that frequency is an achievable one... it may be too low meaning wavelengths much longer than the physical dimensions of your actual tank or not achievable for your system - most likely. Or too high to be resonably achieved with pistons or pumps because of material concerns - less likely. In addition, the horseshoe shape of your tank adds an added complexity... three separate paths where waves can travel in different directions which means that in order to achieve a stable and standing wave in your tank, each of the frequencies must be a multiple of one another. Here is a visual example... if you take a rope and wave it up and down such that there are two stable waves (looks like this ~), that is the natural frequency of the rope, now if you attached two additional ropes along the original rope and tried to maintain a wave in all three... it is possible, but more difficult.

 

If possible, I would seriously consider modifying your tank size such that the short lengths are a multiple of the long lengths (i.e., 10' and 20' or 11.5' and 23'). I think it would be MUCH easier to accomplish a standing wave effect that way.

 

Also, since maintaining a standing wave requires pretty precise timing, I think that attempting to do so with a surge device would not work. Fill rates change over time as the inner walls of pipes gets normal fouled, as the pump impellors get fouled, as the exact inititating event changes slightly. If you are going to do a standing wave, I think it would best be done with either two pistons or four pumps (like a large wavebox) in the back left and back right of the tank each directing flow toward the front (short leg) and center (long section). In theory, you could have a standing wave that would meet in the center of the long section and terminate at the end of the short section (since they would be the same length.

 

Finally, more stuff to think about concerning surge devices: they are not quiet and they produce microbubbles (some find them visually distracting, I do not, but they do make a mess with salt spray).

 

I can support my position with math and pictures when I get a chance later if desired.

Edited November 8, 2010 by Chad

[dave w]

Yes, a natural frequency will exist for most given systems regardless of shape... HOWEVER, it is a matter of whether that frequency is an achievable one... it may be too low meaning wavelengths much longer than the physical dimensions of your actual tank or not achievable for your system - most likely. Or too high to be resonably achieved with pistons or pumps because of material concerns - less likely. In addition, the horseshoe shape of your tank adds an added complexity... three separate paths where waves can travel in different directions which means that in order to achieve a stable and standing wave in your tank, each of the frequencies must be a multiple of one another. Here is a visual example... if you take a rope and wave it up and down such that there are two stable waves (looks like this ~), that is the natural frequency of the rope, now if you attached two additional ropes along the original rope and tried to maintain a wave in all three... it is possible, but more difficult.

 

If possible, I would seriously consider modifying your tank size such that the short lengths are a multiple of the long lengths (i.e., 10' and 20' or 11.5' and 23'). I think it would be MUCH easier to accomplish a standing wave effect that way.

 

Also, since maintaining a standing wave requires pretty precise timing, I think that attempting to do so with a surge device would not work. Fill rates change over time as the inner walls of pipes gets normal fouled, as the pump impellors get fouled, as the exact inititating event changes slightly. If you are going to do a standing wave, I think it would best be done with either two pistons or four pumps (like a large wavebox) in the back left and back right of the tank each directing flow toward the front (short leg) and center (long section). In theory, you could have a standing wave that would meet in the center of the long section and terminate at the end of the short section (since they would be the same length.

 

 

Chad, what you suggest is entirely possible. I could cut 1.5' off each of the shorter legs for refugia. I was planning on a pairs of Navarchus and Regal angels anyway so I need to protect the sponges. As long as the top foot of the divider is solid acrylic, I assume the lower part can be open.

 

Is the wave frequency in addition to or in place of other effects? Am I trying to do too much, juggling four goals: 20x hourly tank turnover, 6 hr. tide changes, surges and waves. If this is too much, please give opinions on priorities. None of us have perfect answers but my absence from the hobby makes me uninformed of the past 8 years of discussion on water flow.

 

Some effects could cancel each other. Surge buckets can dump clockwise, CCW or perpindicular to the tank flow, but would lessen the tidal effect every 6 hours. It seems that tide is not as important as a nice wave effect. I plan on using connecting pipes to turn the horseshoe into a circle for efficient flow. I like circular flow because energy is turned into water speed instead of flows that cancel each other. Or is this an outdated concept when thinking in terms of waves and turbulence?

 

Hope this isn't too many questions.

[zygote2k]

I think you should stick with the original plan and see how it works. You've already got the motors and the big prop blades- go with those until you decide that they don't work, then go from there. The ATS's and the geyser pumps will be enough of a challenge.

Edited November 8, 2010 by zygote2k

[Chad]

I have never seen a system with a standing wave and a surge device, but the standing waves are not usually affected by other circulation pumps (lots of water moving vs. a little water moving = local randomized effects = good). I would think that a random surge (unless it was huge) thrown on top of a standing wave wouldnt affect the standing wave much, just disrupt it locally for a few moments. Remember, standing waves are moving ALL of the water in your tank back and forth, so anything that is moving significantly less water than that wont detract from it much. Or if you connected it all together and had a general circular water flow going, the standing wave still not be affected... The water would move fast, then slow, then fast, etc. but the average speed around the tank would be set by whatever is making the flow go roundy roundy. If you changed the direction tidally, that may be all the flow you need. IME, having a single circular flow setup (i.e., gyre), detritus tends to get caught behind rocks as the water flows by, changing direction of these things is a good idea if you are going to do that.

(my initial impression is that it may cause lots of additional complexity to the system, though)

 

Here is a picture explaining my recommendation on modifying the dimensions slightly:

 

[dave w]

I think you should stick with the original plan and see how it works. You've already got the motors and the big prop blades- go with those until you decide that they don't work, then go from there. The ATS's and the geyser pumps will be enough of a challenge.

 

Rob, I agree that the geyser pumps may prove problematic. It could be that the bubbles that rise up the wet tube will be too small to displace all the water and a DIY system will be ineffective. I don't think the refugia will be so difficult.

[dave w]

I have never seen a system with a standing wave and a surge device, but the standing waves are not usually affected by other circulation pumps (lots of water moving vs. a little water moving = local randomized effects = good). I would think that a random surge (unless it was huge) thrown on top of a standing wave wouldnt affect the standing wave much, just disrupt it locally for a few moments. Remember, standing waves are moving ALL of the water in your tank back and forth, so anything that is moving significantly less water than that wont detract from it much. Or if you connected it all together and had a general circular water flow going, the standing wave still not be affected... The water would move fast, then slow, then fast, etc. but the average speed around the tank would be set by whatever is making the flow go roundy roundy. If you changed the direction tidally, that may be all the flow you need. IME, having a single circular flow setup (i.e., gyre), detritus tends to get caught behind rocks as the water flows by, changing direction of these things is a good idea if you are going to do that.

(my initial impression is that it may cause lots of additional complexity to the system, though)

 

Here is a picture explaining my recommendation on modifying the dimensions slightly:

 

 

Chad, thanks for a really well done diagram. I am amazed that such small devices would impact 24,000 pounds of water but I trust your experience over my lack of it. It is also nice to know the wave system will not be affected by other water flows in the tank.

 

Even though refugia on top of an open tank may not be the prettiest view, if the peak wave occurs exactly in the middle of the tank, refugia or a brace over this area can help prevent wave peaks from overflowing.

 

I'm not sure that tides will add much complexity. Each individual motor turns on for 6 hours, then rests for 5.5. It's just a matter of a higher number of motors pointing different directions. So the motors are stationary. I like the idea of a motor moving along a light rail idea through the 20' leg of the tank because I think it does well to stir things up. No matter how many different directions the propellers hit, dead spots will still develop.

 

So what size pistons should I build? Up to 6" or 8" I can use PVC pipe, above that I can wrap fiberglass. I don't mind using a sonotube as a mold, a layer of wax paper on the outside usually removes male molds pretty well from fiberglass.

[ctenophore]

You could build a scale model from plywood & pond liner to test the piston/wave device. I think 6" is all you need. I can slosh water out of my tubs with just a few dozen pushes of a 1-gallon bucket. It really gets water moving well, and it doesn't seem to matter where I position the piston. Maybe that's because it's a nearly square tank but I have a feeling that it will create a wave even in a horseshoe shaped tank.

[dave w]

You could build a scale model from plywood & pond liner to test the piston/wave device. I think 6" is all you need. I can slosh water out of my tubs with just a few dozen pushes of a 1-gallon bucket. It really gets water moving well, and it doesn't seem to matter where I position the piston. Maybe that's because it's a nearly square tank but I have a feeling that it will create a wave even in a horseshoe shaped tank.

 

In that case I can use 6" pvc pipe and get a 6" hole saw for plywood or sheet pvc pistons. I don't have a large body of water but the blockwork on the sumps will be done soon and I can drape them with pond liner or a roll of EPDM rubber I bought to waterproof the walls. Can anyone recommend a good motor? This will also be a good place to put a 2" or 3" pvc pipe through an upside down 2 gallon bucket and see if the DIY geyser pump works.

[dave w]

OK folks, I followed directions and loaded the water flow sketch with better resolution.

 

 

The two round tubes shooting toward you will meet each other and turn the horseshoe tank into a circular one to facilitate water flow. The green trays are refugiums. The rectangular blocky horizontal things are meant to signify water flows. I have about 14' between the two side legs of the tank, enough to put several people on a sofa and be surrounded by tank.