Large volume laminar flow machine

[JamesHolt]

<a href="http://s1062.photobucket.com/user/karimwassef/media/2_zpsecrtmzr3.png.html" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/2_zpsecrtmzr3.png" border="0" alt=" photo 2_zpsecrtmzr3.png"/></a>

I am going to call these pieces blue,yellow,white,green (I am color blind so...)
the blue piece (bottom section) you make out of acrylic..
The other three, figure out where to put blind holes for a few plastic pins (top and bottom) and have a local community colleges CNC class make them out of Delrin or UHMW whichever is easier/cheaper to get..
Might be a couple hundred to get it made but it would be an interesting lesson for them and cheaper for you...
I have had stuff like this made like that when I didnt have access to a CNC mill, normally the are happy to do it..
All the inside edges will be rounded as well...
If you do the blind holes for pins dont glue it together, it will be easier to clean that way. You could throw it in the dishwasher for a quick rinsing...

These would be very impressive milled from a 3" piece of acrylic, with the edges flame polished

 

[JamesHolt]

Just looked at the cost of some Delrin,,, ouch
Have them mill it out of a chunk of hardwood, and use that to make a sand mold, and pour it out of resin..

Or you could scale this up http://www.instructables.com/id/Home-Plastic-Injection-Molding-with-an-Epoxy-Mold/

 

[karimwassef]

How do I make a sand mould?

If I go 3D, I can make tweaks. With a hard mould, I would need to be a lot more confident first.

I can have a single plate printed for $100. The problem is that I need 20 plates. I may go 3D until I get it right, then try to cast it in acrylic.

 

[JamesHolt]

Basically you are pushing the shape in the sand and it leaves the pattern to be filled in with whatever you choose to cast it in

 

[uglybetty]

I haven't read all 21 pages of the thread post for post but have you thought about a reverse carlson surge device instead of spending a couple grand on something like the renderings? If you want a massive amount of laminar flow it could definitely do it.

 

[karimwassef]

I have a massive surge now. Mine is actuated with valves with the ends fully submerged constantly so it has no bubbles. It's available on demand from my Apex.

It has two outlets with the max release at 20gal each, of 40gal both on. That's 10% of my 380g DT.

It's not laminar- since it comes out of a PVC end. It's turbulent.
It's not uniform- the flow is highly directional, but it's only a cone of flow, not the whole tank's cross sections (depth x width).
It's not a current of flow- since the water entering and the water leaving are not equal and continuous in direction and rate.
It's not bulk volume flow because it's only 10%, of the water. I need 80-100%bod the tank water moving in the same direction.

My fish have to literally swim at full speed (or hide) and power to the counteract the insane flow my dual surge generates. It's only limited by my tank's overflow capability and I have that maxed out too. The DT pushes to near overflow with each double surge. And I have a new secondary siphon design to potentially double that again.

The problem isn't flow. It's the flow pattern, distribution and uniformity.

 

[karimwassef]

Posts 457-483 are the ones that define the desired specification.

 

[uglybetty]

gotcha.......I'll take the time this evening and re-read the whole thread and see if I can help in this brainstorming process

 

[karimwassef]

Thanks.

I was hoping the whole thing would cost < $1000 with most of that being equipment, not plastic.

I can probably reset to ~ $1500?

Right now we're at $2500 for plastic + $1000 for pumps and that doesn't include the motor and the acrylic covers.

 

[Rosenberg1984]

If you were to use DC powered power heads I believe you could do this very easily. A DC motor with neutral timing can run in both directions. You could build a controller that would ramp up and down multiply heads from a push to a pull rotation and achieve a truly unique and diverse movement in the water.

 

[Rosenberg1984]

And the power head pulling water in would then surge it into the tank wall creating even more diverse water currents.

 

[karimwassef]

Do you know of any existing powerheads that can be reversed electronically?

 

[Rosenberg1984]

Any power head that is DC powered you would be able to run in reverse.

You would run 2 seats of leads to the motor. You could use relay switches on the positive feed closing and opening the switches oppositely to reverse the power.

 

[karimwassef]

I thought that but Gorgok back in post 377 said it can't be done. Take a look.

 

[OllieNZ]

I thought that but Gorgok back in post 377 said it can't be done. Take a look.

It's not the reversibility of the motor that's the issue it's that the impellers are optimised to push the water in a single direction and running it in reverse will give you sub optimal performance (most pumps will be the same). I've got a cheap AC power head that decides to run backwards when switched on every now and then and the flow is pretty pathetic when it runs in reverse.

 

[karimwassef]

The cylinder assembly made me wonder if I couldn't make one massive 10" paddle wheel type pump powered from above and reversible.

I just don't know enough about pumps to presume that I can make an efficient reversible 40k gph pump

 

[JamesHolt]

Any power head that is DC powered you would be able to run in reverse.

You would run 2 seats of leads to the motor. You could use relay switches on the positive feed closing and opening the switches oppositely to reverse the power.

You would only need one set of wires to the motors, use a DC controller to handle the reversing..

Most of them have the actual controller inside the power head (epoxy filled),
the pendant just passes dc through at full voltage (2 wires) and the third wire is a signal (0- 10v or whatever they use)..
I dont have an ecotech to check it but I imagine it is the same..
To me this is a dumb way to build a powerhead, if a $.02 cap goes bad on the actual controller you have to replace everything...

Video for a Jebao Teardown
<iframe width="560" height="315" src="https://www.youtube.com/embed/Nv2TA-c9XVE" frameborder="0" allowfullscreen></iframe>

 

[salty joe]

I got to post 526 and thought, cool he got it! But darn it, not quite yet.

I believe the Jebao powerheads are built for high volume and have sacrificed pressure to get the most gal/watt.

The paddle wheel idea might be the way to go-just like the inside of a garden variety powerhead.

 

[karimwassef]

I think the back pressure on the pumps will be low enough to use propeller pumps. I'm not creating full laminar flow, just bulk uniform current.

 

[Gorgok]

Any power head that is DC powered you would be able to run in reverse.

You would run 2 seats of leads to the motor. You could use relay switches on the positive feed closing and opening the switches oppositely to reverse the power.

The thing is these pumps aren't true DC motors (with brushes, running on direct current). These are BLDC (brushless DC motors), which are more like 3 phase AC motors run off DC supplies with inverters and sensors/timing control to turn on and off the correct windings both positive and negative (hence the inverter) built into the controller. Now again, that wouldn't really be a big deal if the controller was exposed, or replaceable, as they are in the RC world, and the motor just had the 3 wire leads coming out. But as the video JamesHolt posted shows the controller is embedded in epoxy with the motor and to get at it requires destructive methods.

It would be interesting to see what one of these pumps would do if you feed it non-standard control signals though. Assuming one is expecting 0-10V and you feed it -5V what would it do? I don't recall what signals RC controllers expect, but those are generally reversible, or at least provide motors slowing functions on some signals so maybe they have something similar and just don't provide the signal to reverse it? It wouldn't surprise me if these controllers were 'standard' BLDC controllers capable of running the motor in each direction with either safeguards against reverse signals or just hope that they never are provided reverse signals.

One thing to keep in mind as well, the propeller, run in reverse, will want to run away from the motor (screw out) and will more than likely succeed if not positively held in place (not done on Tunzes, Neptunes, or Koralias, but appears to be done on Vortechs and Jebaos).