Large volume laminar flow machine

Forgive me in advance as I have not read all 15 pages of this thread if this has been mentioned before, but this is my idea. If you were to have a chamber on the back of the tank that ran the length of the tank minus 2 or 3 inches on each side fitted with slots on the sides for water to run though but not tank your companions, you could then have a flat "wedge" that would travel back and forth within this chamber on a track. You could power it with a simple DC motor and belt drive system. It wouldnt have to move all that fast to move a massive amount of water and should still get you a very laminar flow. I cant see this costing more than $200 to build at most if you have decent DIY skills. Also please excuse my poor Scetchup skills as this was done very quick. Red cylinder is the DC motor, and I did not render the track system as I have no idea how to make it, but hopefully you get the idea.

a>
 
Last edited:
Looking at that... I think I realize my mistake in laminarizing flow...

I've been trying to push water through a grate or a baffle or diffuser to get a uniform directional flow...

So I put a propeller pump pointing in the right direction, and then I follow it with a diffuser. Sounds logical.

But that's not what these students are doing. They're doing the OPPOSITE.

They're pulling water out of the system and letting the return water enter through the straws... There's no propeller there. No cone of flow. The water is being pulled out, and the incoming water is uniform and directional with only a couple of inches of straws.

Is that the answer?
 
That was one of the things i thought of way back in the 'channel behind the tank with props instead of the paddle' idea. I thought i did post about putting the props basically in the far corner pointing down the channel of the side you want flow to be at, but the other half of the props would have some negative effects sitting idle in the path of the water from the set that is on.

I checked back, its post #249.
Gorgok said:
With pump outlets/powerheads you can gain some flow just placing them in ideal locations i think. Putting typical powerheads in the middle of the back loop will be worst case i think, as they end up fighting their own wake to generate more flow. But powerheads or closed loop outlets them right by the entrance of the return side of their respective loops the water they work on may get pulled further along like in a venturi. Closed loop outlets could also be angled in along the straight back for the same effect i think, just like fuel injection on an intake manifold.

Powerheads may be less ideal than a monster of a closed loop as well, as its harder to make flow past a typical powerhead laminar...

Typical powerhead design is actually lacking for this application as you are really after a propeller more so than a fan. A trolling motor is better suited to this idea... You could have those in the middle back and the little disruption to the flow it creates won't matter one bit the the time it curves back around the front. If you sized it right you could even turn it around when flow changes like an azipod... Or cobble up a Z drive unit out of something even slimmer and stick a motor out of the tank to drive it. Blended fish will always be on the menu though.
Click to expand...

Those examples above take that last bit to heart. Instead of a motor block and prop they have a skinny shaft to get it far from the motor block.
 
Ok, i though that's more about not expecting the output to be laminar because of the fan force. If you're pushing in the corner before the outlet, that introduces the problem of using baffles to make the flow on the output uniform.

I'm wondering about the other side though, is pulling (not pushing) through the straws on the opposite side a more effective way of getting uniform, directional, laminar flow?

This is where a picture is worth a thousand words but I'm on my phone so here goes...

I thought you meant:

|------------------------|
| <=....................=>|
|....|----------------|....|
|:=:....................:=:|
|:=:....................:=:|
|:=:....................:=:|
--------------------------

And I'm talking about this:

|------------------------|
| >=....................=<|
|....|----------------|....|
|:=:....................:=:|
|:=:....................:=:|
|:=:....................:=:|
--------------------------
 
http://www.advancedaquarist.com/2015/2002/1/aafeature

This is an article that I have read many times over the years and was referring to it in an earlier post in that he noted many different designs like over the top and cylindrical.

As I think that I mentioned earlier, my quest has been to create laminar flow like you are but also to create it in a way that does not kill plankton. That is why I have more input on paddles than on propellers.

Anyway, my observations on power heads probably agree with most everyone else. Chaotic flow from the pump turns laminar pretty quickly. While I waited for a dream flow setup, I had a 6 foot tank with the big prop pumps at either end of the tank. On the right side one was pointing left along the top and on the left side, I had one pointing lower and to the right. In 1 to 3 feet, the water was very laminar. The stuff in the middle of the tank got good laminar flow going left at the top and to the right lower down in the tank.

Of course I wasn’t trying to reverse the flow and that might be a big factor but chaos dies pretty quickly. Any attempts to laminarise the flow via straws or baffle, also create localized chaos just like an airplane wing causes burbles and vortices after it so those efforts need to be as early in the stream as possible.
 
Last edited:
autocad.png


The light blue showing suction, and the squigglies at the far side pumps showing the interferance from the second set of pumps.

Shielding them to cut down interferance can't be too hard though.
 
So how do you ensure that the entire surface is uniform flow?

Why would water from the lowest and furthest reach have the same flow as that right at the end of the duct?
 
The idea that I was more or less thinking is that when you push water you will create a wave, but with the design that I had in mind you would also create a vacuum effect and pull water in as well. Which I dont think should cause a standing wave in the tank, but rather move the entire colum of water left and right. I could be wrong but think of when you fill a bathtub up with water and move it with your hand. No surface waves just flow.
 
I get that, but the direction of the flow based on the location of the outlets is into a wall...

The flow has to turn a corner and it would be a narrow push based on the cross-section of the outlets/inlets.

I'm looking for motion that's almost the complete width and depth of the tank... And travels down the full length of the tank.
 
karimwassef said:
So how do you ensure that the entire surface is uniform flow?

Why would water from the lowest and furthest reach have the same flow as that right at the end of the duct?
Click to expand...

The answer to those issues can be seen in the experiment examples. The channel size is the tank size... And the flow is so high the 'pipe' is pressurized against the restriction of the straws. Not practical, so it will be less than ideal.
 
That's what I thought too. The small cross-section they use compared to the inlet ducting forces a uniform distribution. In my case, the duct is 1/6th the width of the tank. Unless my "straws" structure creates a tremendous back pressure, the flow won't be uniform... It's that or I go back to baffles and diverters.

Do you agree?
 
I ran through the numbers on the distribution baffles, etc... And it's pretty poor.

Basically, the only way to have equal distribution is to ensure that each path has the exact same length of path. That's very difficult to do in a short span~10in.

Well, it's difficult unless I can mold by own manifolds or acrylic shape casts. :)
 
That was on the props direct to diffuser? I would expect that to be pretty hard to rectify just because of the nature of the props. They are called screws for a reason...

Back to the squirrel cage fans i guess. I actually forgot what their major malfunction was other than being entirely un-reversible, which on further thought is the case for all examples so far.

With a row of them vertically sitting there, intakes on the bottom into the channel box (extensions on the bottom making it U shaped) along with a bearing in the bottom of that, a shroud of PVC with one hole for the outlet slightly offset... Then the motor up top.

Its very common practice with these fans to enlarge the outlet to full fan diameter really quickly too (examples of this probably found in every actively cooled laptop), so the little fins in an earlier design would help that.

Other than the power needed to turn the long fans, i am still missing a real problem but this is only first pass thinking again.
 
ok. I went underground for a while to rethink a lot of this.

Thanks Gorgok for your feedback. It really helped me.

Reversibility is key, so using the propellers at the outlet doesn't work. The same goes for the cross-blade blowers.

I used a mechanism that I have on my current tank to create a reversible structure with a rotating motor.

<a href="http://s1062.photobucket.com/user/karimwassef/media/1A09593D-4091-427F-ABB2-2896FD239F10_zpswovymxvh.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/1A09593D-4091-427F-ABB2-2896FD239F10_zpswovymxvh.jpg" border="0" alt=" photo 1A09593D-4091-427F-ABB2-2896FD239F10_zpswovymxvh.jpg"/></a>

It's not very elegant, but it works. The current use is to rotate a PVC fixture inside a cross fixture to divert flow to different parts of the tank.

But if I use a rotating fixture for reversible flow injection, my old nemesis "uniform laminar directional" flow rears its ugly head...

To convert the injected water into a uniform flow, I have to use outlet diverters. The problem is that I don't have the ability to make these things... I'd need to reshape plastic or use PVC pipe... or make it out of LEGOs!

So this is a crazy idea I came up with one night at 3am... but... go with me...

Instead of gluing LEGOs (I still that might work), I use sheets of 0.5" acrylic cut to increments of 0.5". It's good that I've got a local supplier who'll cut me a deal on quantity and cut for free...

So.. here it is.... I have one major uncompleted gap - but I'll get to that after the unveiling...
 
<a href="http://s1062.photobucket.com/user/karimwassef/media/0_zpsvpokik4r.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/0_zpsvpokik4r.jpg" border="0" alt=" photo 0_zpsvpokik4r.jpg"/></a>

That's the fully assembled view

<a href="http://s1062.photobucket.com/user/karimwassef/media/1_zpsd2txjatc.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/1_zpsd2txjatc.jpg" border="0" alt=" photo 1_zpsd2txjatc.jpg"/></a>

and that's the exploded view with each of the components.

The glass tank is on the bottom, along with glass overflow and internal side members.

Left and right are the diverter boxes (no power here). These are acrylic assemblies with an eggcrate face and can be removed from the tank for maintenance. The one on the right is separated into it's constituents: The eggcrate, the stacked acrylic baffle structure (green), and the box it goes into.

center top is the back duct cover with ramps to direct the flow into the diverter boxes on either side.

In the middle, and attached by a completely waterproof PVC assembly, is a 10" PVC cylinder that has two large rectangular openings, one in the front and the other in the back. This cylinder has an acrylic face on the outlet (yellow). On this face are 10 power propeller pumps pushing a combined peak 52K gph and consuming about 600W (peak). The cutouts for the pumps outlets separates the water being pulled from the injection region in front of the assembly.

<a href="http://s1062.photobucket.com/user/karimwassef/media/4_zpsqbw7zvrv.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/4_zpsqbw7zvrv.jpg" border="0" alt=" photo 4_zpsqbw7zvrv.jpg"/></a>

The top and bottom are also sealed with acrylic sheets cut into circles. The rotating assembly has a PVC axis that seats into a larger PVC holder on the floor of the glass tank's duct space. It's centered there.

<a href="http://s1062.photobucket.com/user/karimwassef/media/3_zps7vfypteq.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/3_zps7vfypteq.jpg" border="0" alt=" photo 3_zps7vfypteq.jpg"/></a>

One either side, four glass strips (green) are mounted to establish another source of alignment.

The glass duct is constructed so that the clearance on either side is 1/16".

<a href="http://s1062.photobucket.com/user/karimwassef/media/2_zpszms81o2o.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/2_zpszms81o2o.jpg" border="0" alt=" photo 2_zpszms81o2o.jpg"/></a>
 
The baffles... this took some time.

So the old duct diverter concept is worthless. When I ran the numbers, the closest exit was going to push 3x the furthest exit...

Back to the drawing board: Basically, here's the simple logic that would have spared me a bunch of formula work... to share flow, the paths must be equal in length and resistance... DOY! Ok - shouldn't have taken so long to get there, but there it is.

<a href="http://s1062.photobucket.com/user/karimwassef/media/5_zpsuonlfion.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/5_zpsuonlfion.jpg" border="0" alt=" photo 5_zpsuonlfion.jpg"/></a>

So, all the flow must enter at the same location perpendicular to the final exit... that took some space, then it must split evenly, over and over... until it's at the highest possible resolution I have with my 0.5" acrylic... 0.5" openings/0.5" closed...

This is basically the structure geometry

<a href="http://s1062.photobucket.com/user/karimwassef/media/6_zpsqrozmll0.jpg.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/6_zpsqrozmll0.jpg" border="0" alt=" photo 6_zpsqrozmll0.jpg"/></a>

It takes about 11" of total side distance, but it should work.

The LEGOs are still an option, but I think I can build this thing.

The hard corners are actually a good thing in this case. Smooth corners tend to keep flow going in the direction it comes with. The hard 90 degree corners force a pivot and shares the flow...

Will it cost some flow in the end? absolutely. But I have margin. I need 40K and the assembly gives me 50K.

This is also so much better than a bunch of 1/8" straws.

It's got 0.5" open eggcrate on the front too, but it's basically never any tighter than 0.5" and that only lasts for the first 2" or so.
 
You must log in or register to reply here.

Similar threads

P
New installation / No a new Hobbits
Replies
14
Views
265
Pump_Guy
P
E
Wall Mounting Light Bar
Replies
3
Views
159
wvned
wvned
S
Re: bean animal help!
Replies
2
Views
173
wvned
wvned
BigEZ77
DIY Plumbing Repair Options?
Replies
3
Views
1K
Uncle99
Uncle99
J
My dream of a fleece roller
Replies
2
Views
491
SantaMonica
S
Back
Top