Sinusoidal Valve

[weatherson]

Well, I finally finished the sinusoidal valve I've been working on for some time now. I'll share some details but I'll be cutting and pasting from my thread on Reef Central. It's just too much to retype.

I guess it's about time I add some information here as it's been a while. I did manage to complete my latest tank project today so this makes a good subject to share. But first, I need to go back in time and explain my initial plumbing plans for the tank regarding flow. I knew I would be doing a closed loop system to simulate tidal currents that flow from left to right for 6 hours and then the opposite direction for another 6 hours and so on. I did work in a 15-minute overlap with both on to create some turbulence and prevent dead spots within the tank. This has worked very well and I am pleased with the resulting currents.
As for the tank return pump, the tank manufacture, Oceanic, supplied 3/4" returns at each of the two prefilter boxes. Adequate, but not what I considered ideal. Therefore, I ran three lines from this pump with the third to the back-center of the tank. My original plan was to do a Sea Swirl type unit that I had planned to fabricate myself. Have I mentioned that I greatly enjoy DIY type projects? After hearing of the reliability issues of this design or rotating returns and the fact that with my light box in place, space was now limited where it was to be located, I decided to rethink my possibilities. I still had the parts needed, 2 RPM motor and such, so I turned to a device I had seen many years ago and was intrigued by its design. This device would be a Sinusoidal Valve that I first learned about at Dr. Robert C. Michelson's website here.
OK... before you start typing the address of OceansMotions website, let me reiterate that I enjoy doing DIY projects. I have recommended to many their product as I feel it's one of the best to hit this hobby in many years. Paul is to be commended for his ingenuity and creative ideas.
So, back to the sinusoidal valve. One of the main issues with this design was the shaft that attached to the rotating drum within. This had to exit through the casing in a watertight fashion. No simple task and this proved to cause issues for many that tried to recreate their own versions. This is where I had one of those early morning epiphany moments and came up with a design that would alleviate this issue. More on this shortly. But first, let's take a look at some photos of the pieces I made to create my sinusoidal valve.
I started with the acrylic housing and decided on a 3" OD cat acrylic tube. Cast acrylic is more smooth and even than extruded acrylic tubing. I also decided to stick with 1/2" flat acrylic for most all the other parts for strength and sheer beauty once polished. Just because it's DIY doesn't mean it has to be ugly. Here is a shot showing the main housings parts.

[weatherson]

Here's a photo after having glued the end piece in place. The round end at the top is the side opposite the drive motor and has a rubber gasket to seal.



Next is a photo of my idea to eliminate any possible shaft water leaks as mentioned above. What I did was to use an old impeller and housing from an Iwaki magnetic drive pump. Are you getting it now? Basically, the motor turns a open ended drum made of magnet. This surround the plastic, watertight housing that attaches to the valve housing and seals with an O-ring. Inside the housing, the magnetic impeller rides on a ceramic bearing (part of the Iwaki design) and turns when the outer magnet is turned. So you have the ability to turn the internal diverter without any direct physical connection. Here's a shot of the scavenged, old Iwaki pump impeller parts. The magnetic drive unit is at the top left, impeller in the foreground and housing at the right.

[weatherson]

Next up is a photo of the internal diverter drum. I went with UHMW (ultra high molecular weight) plastic for its "slippery" characteristics. This material is the same often used on pads to move furniture more easily. What I needed at this point was to make an interface between the impeller and the drum. Rather than create something new to do this, I simply used a Dremel tool to cut grooves in the drum that accept the stock impeller vanes. After carefully centering the impeller on the drum, I traced the shape of the vanes and used this to cut the grooves. Then, the impeller mates to the drum without the need to attach it in any means.
The UHMW drum I special ordered from my local plastics store and unfortunately, it wasn't available in a diameter that would fit well into the standard acrylic tube. It was larger by about one 16th of an inch and required me to turn it down on a lathe to the proper size. I turned it to about a 16th smaller that the ID of the tube. Here's a shot of the impeller end where I created the grooves.



On the opposite end, I drilled out and glued in a ceramic bearing at the center. I also glued on the mate to this bearing on the removable acrylic end piece or the valve's housing. Here's a shot of the drum and bearing.

[weatherson]

This next photo shows the unit assembled with the motor attached. I did this with a piece of L shaped aluminum to create the mounting bracket. The motor is a 2-RPM version that when I ordered it many months ago, was supposed to be a two speed unit that included 1 RPM too. Unfortunately, upon recent inspection, it's actually a reversible 2-RPM model. Reversibility is useless in this application but I could definitely use the 1-RPM feature. I may have to purchase another at a later date but it will do for now. I'm not sure if I ordered the wrong model or they shipped the wrong one but it's been so long ago that it's too late to do anything about it. Anyway, in this shot you will see the motor and magnetic drive assembly installed as well as the end cap at the opposite end.



Here's a slightly closer view of the drive assembly.

[weatherson]

By the way, all the nuts, bolts and washers are stainless steel where not plastic. This next shot shows the holes already drilled for the three outputs. I went with three so one would face centered to the front of the tank and the other two aim right and left. Keep in mind that these enter the tank at the center back. As luck turned out (I should have factored this in my design), the diameter of the drum and the hole sizes allowed me to have the first output on full and then transfer the flow to the other two in series and then all off before repeating. I desired this to create more erratic flow caused by the backpressure during this cycle being diverted to the two main returns. So essentially, when the sinusoidal valve is allowing flow through it, pressure from these main outlets is slightly decreased, then increased when the flow is restricted at the sinusoidal valve. Make sense? Here's the shot.

[weatherson]

Note in the above photo the staggered placement of the holes in the drum.

Next was time to create the threaded plumbing pipe attachment points. Otherwise known as "where the pipe attaches". Of course acrylic would be the material of choice. I purchased two, 2" thick (!!!!) bars that I cut down to six 2" by 2" by 3" blocks.



Next, I machined these so that they would have an inside radius on one side where they were to be glued to the housing cylinder. I did this by using a 3" hole-saw in my drill press and ran it down one of the long sides. Then I drilled through a hole in the center that I tapped to 1" pipe threads. This is where the pipe screws in and makes the watertight connection with the help of some Teflon tape. Once these holes were drilled and tapped, I radiused (beveled) all the sharp corners and polished them up nice and pretty. Here's the result.

[weatherson]

Here's a shot after all the blocks were glued in place. I call this my "Porsche design" stage. Those familiar will get that. Here's the photo.



And a shot at a slightly different angle and the UHMW drum in place.

[weatherson]

After having completed all the wiring, it was simply a matter of creating and assembling the needed plumbing. I placed unions at each connection for ease of dissasembly when cleaning. I had intended to use dual wyes but space required I use a 4-way plus connector instead. This isn't ideal with both outer lines passing through 90-degree elbows but had to be. There's also a single union beneath this to enable the removal of the complete unit. I used dual 45-degree connectors at the top where LokLine attaches for flexible aiming of the outlet nozzles. Fortunately, the black acrylic back is still removable with the valve in place. Here's a shot of the valve at its location at the back of the tank.



And a shot a bit closer.



And finally, a shot from slightly above and the side.



So there you have it, the completed project (finally) working as intended. As mentioned above, I will need to eventually get a 1-RPM motor but the flow action within the tank is very nice none-the-less. There's definitely a lot more current in more areas of the tank than before. All-in-all, this hase proven to be a great addition and an incredibly fun project to build. While I wouldn't recommend anyone considering this type of device not take a look at Paul's OceanMotions devices, it is possible to create your own with somewhat basic acrylic and DIY skills.

Joseph

[blubfish]

thanks for sharing your dyi projects with us they sure help and they show us that with a little thought and planing that just about anything can be done. thanks mike

[whodah]

incredible...

[weatherson]

Mike,
You are very welcome and it's my pleasure.

Jeremy,
Thank you very much. Bet you thought this was never going to get done. I know I had my doubts.

Joseph

[Smaug]

Beautiful

Richard

[weatherson]

Thanks Richard.

Joseph

[Bobsreef]

VERY NICE JOSEPH!

Bob

[weatherson]

Thanks Bob.

Joseph

[Stoker]

So, If Your Skill's Are "somewhat basic" I Guess I Would Call Mine,
Neanderthal ....... Either Way, Very Incredible ..... How Long Did It Take To Build In Man Hours?

[wonrib00]

Good Lord Stoker, this is a very old thread. When I was looking at it, I didnt even look at the date, the content seemed like deja vu, then I saw some pics and started to think "this is looking very familiar."

[weatherson]

*cough*... *cough*... Sorry. It's just so dusty in here. May of 2004, huh? Seems like just yesterday.

That was so long ago that I can't even recall exactly how long it took to make. I would guess that it was roughly a couple of full-on days worth of just machining and assembly. Of course, this doesn't include design, parts gathering and plumbing.

[krusty the clown]

Mad DIY skills!! How's it working three years later?

[weatherson]

Thanks and it's still going strong.