DIY- Ultra Beckett

Zephrant

Zephrant

Premium Member
DIY- Ultra Skimmer

I had been looking at my Monster Beckett that I am running on my 200g system, and thinking that I'd like something bigger, so off to the shop I went.

This time I went with a 8" dia. cast main tube, and dual 3.5"x30" cast injector tubes. The mixing box size is about 14"x10"x12", and made out of 1/2" cast acrylic.

As before, start out with making the box. Cut on the table saw, then run it across the jointer (or setup a router table to joint) to clean up the edge. Build the 4 sides of the box first (see the Monster Beckett thread for details.)

Make the top of the box next- The 3.5" injector tubes penetrate the top of the box for the strongest joint. The downside is that they have to be carefully aligned with each other before you glue them, or they may be crooked. Get a tight fit here, and use Weldon #4. The 8" main tube sits in a 1/8"x1/8" groove in the top, which precisely locates it and allows more surface area for the glue to adhere too. If you have a snug fit, use Weldon #4, if it is a little loose, try some Weldon #16 in the groove, then push the tube down in to it, and put a weight on it.

Thinking inside the box:
sk8000-box.jpg



Since I had two injector tubes, I couldn't do the mitered tube construction I used in the Monster Beckett, so I came up with a simple baffle setup in the mixing box. I have a false bottom installed that is walled off from the main tube, but not from the injector tubes. The foam is forced down below that baffle, and is only allowed to escape up through the 4" coaxial tube in the middle of the 8" main tube.

Thinking outside the box: (and gluing on the main tube and 10.5" flange)
sk8000-maintube.jpg


The flanges are all made as done previously- MDF patterns are made up by printing out a CAD drawing 1:1, and using 3M77 to stick it down to a square piece of 1/2" MDF. Then it is center-drilled, and a 1/4" shank is installed in it. I then put it on the mill, and spin the MDF against a bit to get it to the correct round shape. Leaving it setup there I move it over and drill the holes for the key-holes and all, using the printed pattern as a guide. (I'll do a flange DIY with more detail soon. ) This design used 13 flanges made from 5 patterns.

The injectors are similar to last time too- I used 3.5" cast acrylic again, with 6" dia flanges, but I went to 4 holes instead of 6 to make removal easier. The O-ring grooves are 1/8" wide x1/16" deep, and I use Buna-N O-rings available from All O-rings in packets of 10-50, depending on the size.

sk8000-injector1.jpg


The finishing touches seam to take the longest time. I hand sanded all the edges with 220, 400 and 600 Wet/Dry sand paper, then hit it with a $20 car buffer and Novus #3 buffing compound. The result is a mirror-smooth finish, but is extremely time intensive. Look for tips on flame polishing in the future. :) (I still don't like flame polishing for tanks, but I'm thinking it's a great idea for low-stress areas like flanges. )

I found some "Banjo" 1" quick disconnects at the local farm supply store, and used them ($12/pair) on the inputs to make it easier to setup/teardown. The output is a 2" gate valve from Aquatic Ecosystems . I also used some 1/2" rubber grommets to mount the 3/8" air valves. This required drilling a 1" hole in the cast acrylic, which I don't like to do, but I think the rubber will help protect the joint as it will flex some when the valve is bumped. Hard gluing that joint would be asking for a leak IMHO. Reports from the test running are that 1/4" air valves would be adequate too.

The acrylic was purchased from the local Laird Plastics in 6' lengths.
Tube cuts:
8" tube: 2' section for the main tube, and a 6" section for the overflow cup
4" tube: 18" section for the coaxial injection, 8" section for the neck
3.5" tube: two 30" injector tubes, two 3.5" sections for the Beckett injectors

Look for a DIY tube-cutting jig for a table saw thread coming soon.

I used SCH80 elbows and nipples mostly for looks- Grey looks better than the white PVC, but also for strength. if someone over-torques these, they won't break nearly as easy as SCH40 threaded elbows do.

Injectors, ready to run, with the 3/8" speed-fit air valves, Monster Beckett still running in the background.
sk8000-injectors.jpg


Finished, and wet-testing it:
sk8000-wet-test.jpg


Over all I really like the design. The baffling in the bottom box is far easier to assemble than the mitered design on the Monster Beckett, and the "Banjo" connectors give it a "high-tech" look. :)

More (and better) pictures after it gets run some.


Zeph
 
Very nice!

Can you please explain the location and purpose of the 18" length of four inch diameter tubing used for coaxial injection? Thanks, and please more photos!

Brian
 
DANG man...that is AMAZING. what next or do i dare ask. please do update us w/ some pics of it running and the NASTY crap i bet it pulls out.

umm, do you just dream this stuff up or do you just smoke a lot of drugs and the visions tell you how to do it ;)

keep on kicking DIY butt
Lunchbucket
 
Zeph,
Lookin great as usual! That thing is gonna rock. What pump are you going to run on it and what size is the outlet? That skimmer must be HUGE! But it looks great.

Jeff
 
Cheapreef- I didn't get any good shots of that part- Hard to get pictures of black baffles inside a black box.

Brian- The 18" tube forces the air/water mixture from the injectors deep in to the main column. That means that no water can bypass the main tube and exit the skimmer, it is all forced at least 18" up the skimmer, and another 18" down to the exit, guaranteeing another 3' of contact length.

Other than that, it looks cool. :)

LB- I spend most of the day in a cube farm staring at the computer screen, by night time I'm dying to make something that I can actually see. I do use AutoCAD to rough out the design, but often make modifications once I start.

Jeff- I am hoping an Ampmaster 3000 would be adequate, but some research indicates that it might not be strong enough for a very large system. I think a Sequence 6000 would be more than enough, maybe a 3600 would work too. For now, we are running dual Iwaki 100's, but that is overkill. A single Iwaki 100 might be ok for smaller systems too, but would require some replumbing on the tank side to achieve.

Zeph
 
Zeph,

Your previous becket has the main tube with the upward flow, and smaller holes surrounding it for the exiting water to go down.
I dont see this in your "ultra" model. Can you explain a little about the main tube here, I think I am missing alot.
Nice job and looking forward to more pics.

Thanks.
 
I may be very off here, but the 2 iwaki 100 would probably be your best bet to make it the most efficient possible. The thing about beckets for what I ahve read is that it limits outflow regardless of inflow to about 1000 gal/hr. It works best by using pressure rated pumps, neither of which the sequence or the ampmaster is. I would suggest, and again I may be wrong here, to use larger valves for the air intake to the becket chamber. Those RO valves are ratehr small and may cut down on your air intake significantly. I would also try to cut down on the amount of 90 degree elbows coming to the top of the becket since as I am sure you know they cut down the flow by adding resistance.
If this was my skimmer design, I would certainly make the main mixing tube a bit taller. Can't really tell from your picture how tall it is, but I would make mine at least 4 feet tall if not taller even. Also the water coming into the mixing chamber, I would re-direct using some PVC to make it swil inside the chamber (again, more contact time).
It looks awesome. Keep up the good work. By the way, how much would one of these monsters cost?? Ballpark figure would be more than enough.:D
 
Another fine machine. I agree that you should upsize the air intakes a bit, but thats it. Nice Ride!
 
Hey all-
Cheapreef- Here is a bad picture- imagine the piece with the hole in it is flat, not angled. In this picture it fell down seconds before I snapped the shot, and I did not get another shot.
sk8000-baffles.jpg


Note that if any bubbles form in this lowest chamber, they tend to bubble back up in to the injector tubes, where they get broken up in to smaller bubbles. This helps keep the unit from "burping" bubbles up the main column, which adds turbulence.

Pantinor- This design has the same idea, but is easier to manufacture. It still forces the main water up the middle tube, then allows it to go down between that tube and the main tube to exit.

sk8000-1.jpg


sk8000-base.jpg



dgasmd- Some good points- I've read the same about the Becketts, but it is not quite true. They are not some kind of a pressure-compensated valve that only allows 1000gph though them. They are simply a 3/4" inlet, with a restriction in them. Throwing a bigger pump at them will force more water though them, but it does taper off. i.e. a 1000gph pump might get 800gph though it, a 2000gph might get 1100gph though it, while a 4000gph might get 1500gph though it. (All made up numbers just to illustrate the point, I don't have those kinds of pumps on hand to get real numbers. :) )

The valves on this one are really 3/8", and not the 1/4" RO style valves. They have not needed to be run wide open, so I think they are big enough. You are right though, 1/4" valves might not be big enough. To get a good foam pattern, you need to restrict the air a little.

sk8000-inject1.jpg


The 90 degree elbows do restrict the flow some, but it cleans up the plumbing greatly. The user can always change to 45 els or plumb direct if needed. It is all 1" pluming, which gets restricted to 3/4" at the Beckett. In this case, I knew the pump would be on the floor near the unit, so it had to go up and over the edge, so to speak.

I could certainly make the main tube taller- I had an arbitrary design goal of staying at 48" high, so that is where I set it. Considering the long input path though, I'm not sure additional height would really add much.

I'm not sold on the swirl idea yet- You have a set volume of water going in to a fixed volume, and the same amount going out. Swirling it cannot add to the "contact time" or the unit would be blowing water out the top. It does accelerate the water in a spiral, making it move faster in relation to the tube, but I'm not convinced that this is a good thing.

That said I certainly could install some baffles in the 4" tube to give the water a spin though. Since that tube is removable, it might be worth a shot. Good idea!

I don't like to quote prices in DIY threads, this is the area to help people make their own. :)
After it tests out successfully, I'll post prices on my web site. For now, it is still in R&D.

Keep the suggestions coming please-

Thanks-

Zeph
 
"I don't like to quote prices in DIY threads, this is the area to help people make their own.
After it tests out successfully, I'll post prices on my web site. For now, it is still in R&D."

Man, you are to be commended for that! Very well said, by a true DIYer.
Chris
 
im with zephrant on teh swirl design, i thought of it that way for a while, so much goes in and so much goes out, if some is staying in for 5 minutes then some is ony staying in for 5 seconds.

Has anyone built identical skimmers with and without the swirl design and seen differences?

Zeph since you mentioned this in another thread maybe you should take on the job of building another one with a swirl design and see which peforms the best. :D
 
Does it matter if you put the air valve on the acrylic tube or could you also put it on that nipple section you have between the 2 pvc elbows? or else drilled into the second elbow like in your "monster" model?
 
Thanks Chris- This hobby is way too expensive as is, besides, it's a good excuse to by more tools for the shop. :)

Wedfr- That is certainly possible, but it would take a lot of effort to determine if it made a difference, and even then, it would only prove that it did/didn't work in one particular instance, in one design. A real test would have multiple designs, all with/without the swirl. Like I said, I'm not convinced either way.

Pantinor- The air intake must be in the chamber with the Beckett, as the air is being drawn though the holes in the side of the Beckett and mixed with the water. My original design of the Monster Skimmer had the air being drawn in though the lid of the column, and the Beckett was not "sealed". It turns out that it is not nearly as efficient that way, so I modified the design to have a sealed Beckett like most other designs out there. Much better performance IMHO.

I just updated my Monster Beckett thread with a new picture.

Zeph
 
You mentioned that most air bubbles that get trapped below the false bottom would go back up the injector tubes, but I was wondering if there would still be considerable burping from the bubbles that catch the rim of the false bottom. Of course it wouldn't make much of a difference on such a sweet skimmer, but then again, I wonder if trading a little head pressure from 90 deg ells would be worth it to avoid any burps at all. Oh well, toss that question on the queue behind the swirl question.

Nice work,
Gerald
 
Good point- The 4" center tube penetrates about 1" below the baffle though, so any air that does collect in that area will travel back to the input tubes. The Ells are a possibility, but I could not think of a good way to combine both input tubes to the single 4" tube with them.

Zeph
 
You must log in or register to reply here.

Similar threads

adova
300g DIY Acrylic Tank Recommendations
Replies
5
Views
178
griss
griss
T
Beckett Skimmers
Replies
1
Views
450
griss
griss
T
replace joint adhesive or just reseal?
Replies
4
Views
378
griss
griss
DjPhato
Fluctuating water level in overflow compartment
Replies
6
Views
182
kharmaguru
kharmaguru
R
DIY AI-coded ESP32 based aquarium controller
Replies
0
Views
393
ronnyking
R
Back
Top