Large volume laminar flow machine

[thejuggernaut]

Is there a home use laser cutter?

Not really. But you could commission someone to do it. I've done some engraving on mine for fun. The problem with most of the smaller ones that are made in china is that the controllers aren't very well suited for engraving. Mostly cutting.

 

[thejuggernaut]

What are you wanting the etching for? Maybe I can help. This project looks interesting.

 

[thejuggernaut]

Ok I just read the entire thing. Pretty cool. I could probably help you etch the acrylic if you'd like. Shoot me a PM if you want. If you get the acrylic, I'll volunteer my time and CNC for the project. You live about 4 hours from me. We could even meet half way to swap the acrylic panels.

 

[theatrus]

There are the $800 special Chinese laser cutters on eBay but they are generally total rubbish and would be frustrating to work with at best (and downright dangerous at worst - their safety interlocks basically don't work).

 

[karimwassef]

Wow... So laser cutting isn't in my wheel house.
Juggernaut, I really need cutting, not etching. Can you help with the cutting?

I'm not getting any feedback on the design - gorgok, herring, salty, dkeller, erica, sculpin, james ....?

You guys usually find what's wrong with my designs... Is silence an indication that you think this will work?

 

[theatrus]

I'm hardly an expert on flow, but my main concern is increased head pressure due to all of the sharp 90 degree corners preventing a nice laminar flow. The rough idea works from a 10k foot view.

 

[karimwassef]

I have the Legos. I'll build my first prototype.

 

[salty joe]

I don't understand all the calculations-all I can do is believe that you brought the head pressure way down. I will say it seems completely counter intuitive that cutting the tall baffle into a series of short baffles reduces drag. I'd have expected that to increase drag. Anyway, I am fascinated with this and if I see something where I maybe might could help, I'll be on it like white on rice.

 

[Floyd R Turbo]

This may have been mentioned before but, no time to read it all:

Sharp edges = drag
pockets = drag

I see a lot of these

As far as a laminar flow goes, this just popped into my head. A friend of mine has a pool and wanted to make one of those dancing waters type things. You know, the thing that kids play in where a 'water snake' shoots out of a jet and it's a perfect tube of water, then it hits the ground and another one shoots out, etc. Or like those fountains in a hot tub. The point is, like a clear water 'tube' or 'jet' but not spraying like a hose sprayer.

So he took a PVC pipe and crammed it full of drinking straws and had a little template CNCd to fit over the end, and voila! Took him a few iterations but it worked. The point is, couldn't you achieve a very similar laminar flow with the sam type of concept, splitting up the flow via a simple in-pipe divider with a long draw, then branching that outward to cover a larger area?

FWIW I love this idea, one of my reefkeeping buddies totally wanted to make something like this a few years ago after he when diving in Mexico and the whole concept of total tank side-to-side flow struck him.

 

[karimwassef]

Straws = massive back pressure. In fact, that's why they work.

They'll also clog up in a living reef. I have 1/4" slits and I'm concerned. To get full laminar flow would require 1/8" straws and pressure pumps. We looked at it a few pages ago.

 

[herring_fish]

I didn't re-comment because I don't want to be a nay saying stick in the mud. You are going to have to do a scaled down test to see if it will work. It may work great.
I personally think that the flow will be lack luster compared to the work that is produced by the motors. Besides the bottlenecking, I don't think that the calcs capture the turbulence that the corners produce"¦.but I could be all wrong and I hope that I am. I personally hope that you also can test the louver idea.

Not because it's my idea. You are the one that found it but I think that it is worth a try and could be easily made (particularly for a test) with slices of PVC pipe and some glue.

Yes, there would be a difference between the front flow rate and the back but the display tank flow rate might me fairly laminar anyway and flow should be comparatively high, over all. Flow that approaches a smooth obstruction slows down close in but stays laminar, until or unless the flow burbles in its wake.

You have a few other ideas that should be tried before you invest in a project as big and expensive as this.

 

[Floyd R Turbo]

Straws = massive back pressure. In fact, that's why they work.

They'll also clog up in a living reef. I have 1/4" slits and I'm concerned. To get full laminar flow would require 1/8" straws and pressure pumps. We looked at it a few pages ago.

I wasn't suggesting straws per se, just the concept of them. You could sectionalize a pipe many ways and condition the flow stage by stage, like one 50/50 divider, then a few inches downstream, you expand the orifice and split again, continue. In the end think it would look like a miniature pipe organ (church kind, not coral kind)

 

[OllieNZ]

Could you get the assembly 3d printed?

 

[karimwassef]

I didn't re-comment because I don't want to be a nay saying stick in the mud. You are going to have to do a scaled down test to see if it will work. It may work great.
I personally think that the flow will be lack luster compared to the work that is produced by the motors. Besides the bottlenecking, I don't think that the calcs capture the turbulence that the corners produce"¦.but I could be all wrong and I hope that I am. I personally hope that you also can test the louver idea.

Not because it's my idea. You are the one that found it but I think that it is worth a try and could be easily made (particularly for a test) with slices of PVC pipe and some glue.

Yes, there would be a difference between the front flow rate and the back but the display tank flow rate might me fairly laminar anyway and flow should be comparatively high, over all. Flow that approaches a smooth obstruction slows down close in but stays laminar, until or unless the flow burbles in its wake.

You have a few other ideas that should be tried before you invest in a project as big and expensive as this.

When I ran the louvre idea through the math, the answer was this:

If the inlet and outlet pressures for two pipes/ducts is the same, the flow between them is only a function of the travel distance.

My first design was essentially a sideways louvre and the results were terrible on paper. One path was 70% longer and had 70% less flow. It was worthless.

That's where I reset with this fractal concept... Split in 2... Then again in 2... And again in 2. Etc... It forces the paths to be equal length. That's the whole point.

 

[karimwassef]

I wasn't suggesting straws per se, just the concept of them. You could sectionalize a pipe many ways and condition the flow stage by stage, like one 50/50 divider, then a few inches downstream, you expand the orifice and split again, continue. In the end think it would look like a miniature pipe organ (church kind, not coral kind)

Yes. That's exactly what this design is. Split in 2, then 2, then 2...

That's why the number of outlets can only be 8, 16, 32, 64, 128 etc... There are no versions that end with 98 outlets... The paths won't be equal length.

 

[karimwassef]

Could you get the assembly 3d printed?

Yes. And I can even round the corners and smooth the outlets. I could also skinny the walls to 1/8" or maybe 1/16".

That would be amazingly better...

But this is a big piece. Even split into 5 subsection stacks 3" high, rack is 10" x 32" ... Big 3D part...

Who can print an ABS model that's 10" x 32" x 5" ?

If anyone can make it, that would be great and I would run all the CAD drawings for it.

 

[karimwassef]

This may have been mentioned before but, no time to read it all:

Sharp edges = drag
pockets = drag

I see a lot of these

As far as a laminar flow goes, this just popped into my head. A friend of mine has a pool and wanted to make one of those dancing waters type things. You know, the thing that kids play in where a 'water snake' shoots out of a jet and it's a perfect tube of water, then it hits the ground and another one shoots out, etc. Or like those fountains in a hot tub. The point is, like a clear water 'tube' or 'jet' but not spraying like a hose sprayer.

So he took a PVC pipe and crammed it full of drinking straws and had a little template CNCd to fit over the end, and voila! Took him a few iterations but it worked. The point is, couldn't you achieve a very similar laminar flow with the sam type of concept, splitting up the flow via a simple in-pipe divider with a long draw, then branching that outward to cover a larger area?

FWIW I love this idea, one of my reefkeeping buddies totally wanted to make something like this a few years ago after he when diving in Mexico and the whole concept of total tank side-to-side flow struck him.

So, when I created the title of this thread, I used one word: LAMINAR. As we've iterated, I think that's misleading. There are several key performance requirements here and they're not all equal in importance. Here's what I see this as now:

1. BULK
2. CURRENT FLOW
3. UNIFORM & DIRECTIONAL
4. LAMINAR
5. HIGH FLOW
6. ALTERNATING
7. PROGRAMMABLE
8. COMPACT
9. EFFICIENT
10. COST EFFECTIVE

Laminar is important but not the most important. I'll take some time to clarify these next.

 

[karimwassef]

1. BULK

This really means that a large volume of water in the tank, preferably all or most of it, moves at once in the same direction.

This isn't achievable with a powerhead or return pump or even a surge (unless it was one the order of magnitude of the tank).

Think of all the water - 500 gals for this example - moving together to the left or right.

 

[karimwassef]

2. CURRENT FLOW

This means that the water motion isn't a surface effect like a wave. It's a real push/pull of a current. Think of a rip current that can pull a swimmer hard to one side. It's a current, not a wave.

Currents flow in circles. To go somewhere, it needs to eventually come back from there too in a big loop.

 

[karimwassef]

3. UNIFORM & DIRECTIONAL

So this is what the diverter/baffle box does.

Uniform means that the water at the top front of the tank, middle of the tank, and bottom back of the tank moves at the same speed in the same direction.

Any time you try to go around a corner, or from a small inlet to a large outlet, the flow will not naturally be uniform. This is because the distance travelled for each current of water flow is different. Short paths go faster. Long paths go slower.

Also, if the output goes the same speed but in different directions, that won't work either.

Think of it like a parade... Straight together in the same direction.

So uniformity is key here. A bulk non-uniform non-directional flow is a mess.