230 gallon in-wall, room divider, reef build

[Higher Thinking]

The one down side is this entire idea of accessibility to the bulkheads without removing the tower is pretty much out the window. Best case, i'd have to unscrew the bulkhead nut from under the tank, pull the standpipe up an inch or so, slide out the old gasket from underneath the tank and slide in a new one. Would be a huge PITA, but possible in the event of a double point failure (overflow tower leaks AND the bulkheads fail). In general though, the bulkheads shouldn't be seeing water.

Can you elaborate on this? Maybe I'm not following. The gasket goes inside the tower, yes? Not under the tank. Or are you just referring to the nut?

Also, are your bulkheads screwed in under just the tank itself or screwed under the stand as well?

 

[MarsRover]

Can you elaborate on this? Maybe I'm not following. The gasket goes inside the tower, yes? Not under the tank. Or are you just referring to the nut?
Also, are your bulkheads screwed in under just the tank itself or screwed under the stand as well?

Yes the gasket goes inside the tower (on the inside of the tank). Yes the bulkheads are screwed in under JUST the tank. the stand is not being clamped by the bulkhead nut.

Let me take the time to fully explain all that has happened and why i landed on the overflow design i am going with incase someone in the future wishes to make something like this for themselves.

When approaching finding a solution to a problem, I always revert back to what i know as an engineer. Define the functional requirements and what has to happen to the system to be considered a "mission ending failure," define the desirements of the system (nice to haves), then weigh possible solutions that meet the requirements and some/all of the desirements.

Mission Ending Failure: I decided to define this as "having to drain the tank" be it to remedy a bulkhead failure/leak or whatever else. Right now, I am trying to come up with a solution to the overflow box / bulkheads and so having to drain the tank means the bulkheads leaked/failed and if nothing is done, i'll have water leaving the tank at some uncontrolled rate.

Functional Requirements: Provide a means for transporting surface skimmed liquid from the main display tank, to the sump located below the display, through a centrally located means (not attached to a tank wall) while demonstrating two fault tolerance to a Mission Ending Failure.

Desirements: Minimize the volume of the overflow box-structure so to maximize the swimming/coral area within the display tank and not intrude on the aesthetic, visible from all 4 sides. Limit the amount of noise in the overflow system. Limit the amount of stagnation volume in the overflow box.

Simple Definition of "Two Fault Tolerance": Mission ending failure will occur after TWO SUCCESSIVE failures occur in the same fault tree branch. This means, you have to have double failure in series to result in mission end.

Simple Definition of "Stagnation Volume": Water that does not circulate or move around in the system because, for one reason or another, it is stuck in something (a corner, a tube, an area of the tank without flow). To prevent this, we all stuff our tanks with powerheads. I want to make sure my overflow box doesn't have any dead water.



To meet these demands, this is the thought process i went through:

I don't want my overflow attached to a wall of the tank. Thus it is in the center of the tank. To reduce the impact to aesthetic, i'll make it cylindrical. Because of the spacing of the bulkheads, this would mean i need a 10" cylinder in my tank. This was way to big for my liking, so i made a tube that goes from 10" to 4", then back from 4" to 10" at the top.

How do i design the standpipes for my overflow. There are a few options. I went with bean animal. Without any thought put into it, my bean animal solution would look something like this:

But what does this solution result in? Stagnation of the volume around the bottom of the tower.

So, lets then remove the bottom of the tower and do something that resembles the concept of a coast-to-coast overflow box in the sense that the box doesn't extend from the bottom of the tank to the top, and modify it to work in the center of the tank:

Well now, this solves the stagnant water issue...so our overflow tower now looks like this:

Where we would have the overflow box be suspended by the standpipes themselves (glued/sealed to them) so that we could have this centrally located and NOT attached to a wall of the tank. Okay sure why not!

Okay so now we have met all the desirements (minimize the volume of the overflow box-structure so to maximize the swimming/coral area within the display tank and not intrude on the aesthetic, visible from all 4 sides. Limit the amount of noise in the overflow system. Limit the amount of stagnation volume in the overflow box.) But what about the requirements? We have not met the "two fault tolerance" requirement. Right now, with the design we have above, if the bulkhead leaks (1 fault) then the tank volume is vulnerable to draining uncontrolled. Furthermore, if i wanted to fix the leaky bulkhead before the whole tank is drained (say it is a slow leak), i can't replace the bulkhead's gasket without draining the tank! Not acceptable to me...so lets rewind a little and see what we can do...

In my mind, the best way to keep the bulkheads from having access to the bulk of the water volume in the display tank is to go back to the tower-cylinder idea. But to combat the stagnant water issue, i'd have to also include the coast-to-coast "floating box" idea too. Now we having what looks like a workable solution:

What we see here now is that we satisfy all of our needs. What we loosened up on to meet our requirement is a desirement: minimize the tower size. We had to increase the tower size to meet our requirement of two fault tolerance while still maintaining access to the bulkheads and we were still able to meet all other requirements fully. So lets talk about this solution and make sure it truely meets all of our needs:

1. Is it two fault tolerant? Yes. If the seals, sealing the overflow tower to the tank bottom, or sealing the partition at the top of the tower, leak, then the tower fills but the bulkhead seals will support the water. It would take failures of one of those seals AND failure of one of the bulk head gaskets in order to have an uncontrolled transport of the display tank volume. CHECK!
2. Does it have volume stagnation points? They are minimized or zero. By virtue of the top box, if the siphon is strong enough, all water inside the box eventually makes it way into the sump and re-circulated. CHECK!
3. Is it designed so as to minimize the overflow's volume and intrusiveness to the visual aesthetic of the tank while still providing sufficient surface skimming and access to the bulkheads? It is definitely bigger than the "floating" box design previously considered, but we had to expand the tower diameter because of the other requirements. We also need access to the bulkheads, and so the diameter of the cylindrical tower needs to be big enough to support that. So how do we change a bulkhead seal? More on that shortly....

We have met all our design requirements and, generally, met all of our desirements while remaining within the confines of our requirements!

Now, one thing i didn't go into too much detail on is how i sized the 4" cylinder that makes the main body of the tower. This goes hand in hand with how we will be changing the bulkheads should we have a leak.

Let us talk about contingency situations and plans for dealing with them with this design:

Failure Type 1: The seal for the top partition in the overflow tower leaks, filling the dry section of the tower with water.

Solution 1: We can turn off our return pump to stop overflow into the tower. We will loosen one bulkhead nut (under the tank) and drain the tower of water via a controlled leak. We will then let the overflow box dry out. Replace the bulkhead nut, tightening to reseal the bulkeah. Then, we will re-seal the partition with sealant. Let it dry. Turn back on the return pump and resume normal tank operations. No big deal and we resume having our nominal dry tower overflow.​

Failure Type 2: The seal between the overflow tower and the bottom of the tank fails, filling the dry section of the tower with water.

Solution 2: One thing we won't ever be able to do without mission failure (drain and dry the display tank) is reseal the tower to the bottom of the tank. If that seal fails, and leaks, the overflow tower will fill with water, exposing the bulkheads to the main tank volume. Well, in this case we will have to live with it. And we can! As long as our bulk head gaskets stay strong. We could attempt to characterize the rate of the leak into the overflow tower, and regardless, it is likely to be a slow leak into the tower. Thus if a bulk head then subsequently fails, The leaky seal of the tower will be a bottle neck, aiding to slow the uncontrolled release of display tank water out of the display tank.
​

Failure Type 3: Bulk head must be replaced. For this to happen, we have to first have a Failure Type 2 occur and then have a bulk head seal fail too. This is our two faults. But, because of the way this system is designed, we don't quite yet have mission failure...we just have a PITA situation to deal with.

Solution 3: We have just had the worst case occur, two faults in the same fault tree branch have occurred subsequent to one another. Yikes. We have a bulkhead leaking, and uncontrollably releasing the display tank water.

The leaky tower seal is slowing the flow available to the leaky bulk head, hopefully, to some degree.... but we can't survive forever like this. Here is how we will attempt to get out of this situation.

If the bulk head seal HAS to be replaced, we will go under the tank and loosen the leaky bulkhead's nut. We will then push the bulkhead, and attached standpipe, upward.

This will drain the tower while also supporting the slow drain due to the leaky tower seal. The standpipe/bulkhead is able to move upward enough to be able to go under the tank, and fish out that bulkhead's gasket, which is inside the tower/tank, through the hole in the bottom of the tank. The bulkhead/standpipe can only lift an inch or so. This is just enough for me to be able to fish the bulkhead gasket out of the hole from under the tank.

I'd then have to take a new gasket, squeeze it up through that same hole back into the tank/tower, and put it around the bulkhead flange. Then, push the standpipe/bulkhead back down, over the new gasket, and reapply the bulkhead nut. All from under the tank. Now tighten the nut and pray it holds. This is the last-ditch effort to recover from a two fault contingency situation. It may, or may not work. But it is possible in this event.
​

Hopefully this more fully explains what happened over the last few weeks and explains why i made the decisions i did and how i plan to operate this system.

 

[MarsRover]

So got quite a bit done yesterday. Pretty much all the overhead plumbing is done. Left a few things in the overflow unglued since I need to get back in there tomorrow to leak test the dry overflow tower and then after that silicone the wet-dry partition.

My Marco sand came on Saturday and the rocks come tomorrow!

Going to be a fun week!

 

[MarsRover]

Welp, I did it. I just bought the ecotech vectra L1. Should arrive later this week. I decided to go with the L1 over the M1. The reason being that I can always turn down the flow, but it's very hard to convince a pump to give you more than it is capable of.

Man those pumps are expensive....hopefully it's good construction so I don't ever have to buy another one....

 

[justingraham]

Welp, I did it. I just bought the ecotech vectra L1. Should arrive later this week. I decided to go with the L1 over the M1. The reason being that I can always turn down the flow, but it's very hard to convince a pump to give you more than it is capable of.

Man those pumps are expensive....hopefully it's good construction so I don't ever have to buy another one....

Love both my vectra m1s you will to

 

[MarsRover]

Love both my vectra m1s you will to

I hear some folk purchase some other kind of fittings for inlet or outlet to facilitate maximum flow. Do you know anything about this?

 

[Higher Thinking]

I must say, you have hyper analyzed this. I don't think I've ever read just a detailed write up about over flow assembly..With pictures to boot! I hope it works out for you. Looks as though it should. Great effort!

 

[justingraham]

I hear some folk purchase some other kind of fittings for inlet or outlet to facilitate maximum flow. Do you know anything about this?

Yes I know all about it.

Cost ten dollars extra on an l1 it makes the intake two inches and the output 1 1/2 inches. Intake wouldn't matter if you are planning on running the pump in the sump.


I don't quite remember what size pipe u ran thru ur overflow box but it it's one inch then u mine as well not even worry about it.

Also if you keep the factory pipeing u will need ABS to pvc glue. It's green in color and when you attach the vectras collar to the pvc pipe with the glue the collar is going to push away from the pvc so I recomend having a vice or c clamp to hold it together.

Here is the link to the adapter set

http://www.bulkreefsupply.com/vectra-l1-sch-40-adapter-ecotech-marine.html

 

[805reef]

Welp, I did it. I just bought the ecotech vectra L1. Should arrive later this week. I decided to go with the L1 over the M1. The reason being that I can always turn down the flow, but it's very hard to convince a pump to give you more than it is capable of.

Man those pumps are expensive....hopefully it's good construction so I don't ever have to buy another one....

I was wondering which return pump you would be going with. How high is that return going? I've read nothing but good things about the Vectra's

 

[MarsRover]

I must say, you have hyper analyzed this. I don't think I've ever read just a detailed write up about over flow assembly..With pictures to boot! I hope it works out for you. Looks as though it should. Great effort!

Thank you I do this level of analysis, and much deeper....., at work regularly for a mars rover development and so it's kind of hard for me to turn that off when I'm working on my own projects at home. Glad you appreciate it!

Yes I know all about it.

Cost ten dollars extra on an l1 it makes the intake two inches and the output 1 1/2 inches. Intake wouldn't matter if you are planning on running the pump in the sump.

I don't quite remember what size pipe u ran thru ur overflow box but it it's one inch then u mine as well not even worry about it.

Also if you keep the factory pipeing u will need ABS to pvc glue. It's green in color and when you attach the vectras collar to the pvc pipe with the glue the collar is going to push away from the pvc so I recomend having a vice or c clamp to hold it together.

Here is the link to the adapter set

http://www.bulkreefsupply.com/vectra-l1-sch-40-adapter-ecotech-marine.html

Yes, it will be floating in my sump and all my lines are 1" so I suppose there's no point. Thanks for the info!

I was wondering which return pump you would be going with. How high is that return going? I've read nothing but good things about the Vectra's

It's going up a good 9'-10' and there's quite a few bends going to cause pressure drops from pump to outlet so that's why I figured the L1 is a good bet. At work, I rarely get the opportunity to design overkill in, especially at such a low price point, so i figured I'd take the liberty where I can

 

[MarsRover]

Welp, I got rocks and I got about 190lbs of sand. I think that's a fair start. I'm figuring I'll use these 50lbs of Marco rocks (box was open on delivery.....I don't have a bathroom scale to check if it's all there....didn't feel like 50lbs to me though....I'll see what Marco has to say....) and then probably make some more to fill in and really shape my aqua scape.

I'm expecting to have to buy more sand. Luckily, I can get 40lbs crushed coral at $23 so I'll be using that to make and cast any rocks with my white Portland.


Time to fill the overflow tower and hope it holds water!

Update:

Filled the tower with water and...... it leaks.....just not where I was anticipating! Kinda awesome actually, it wasn't until the tower was almost full that it began to leak....through the plastic! I feared this as a possibility, likely with how 3D prints are made, but luckily it is isolated to a single feature and I have a few possible solutions:

So, solutions:

RTV
Acetone
ABS glue


I have heard at work that folks will do an acetone vapor bath to melt the surface of prints making them less permiable to fluids and strengthening them so they don't delaminate.

ABS glue serves a similar purpose. Like PVC glue with PVC, ABS glue will chemically fuse the ABS. It actually melts it away some when you apply the glue so I could apply and mold.

RTV-speaks for itself. Slather it on and hope it holds.

The issue with RTV is it would likely hold fine in the true operations pressure direction (water pushing on the RTV, into the plastic, sealing the cracks. But it might not be able to hold the water pressure coming from under the plastic. I suspect it would though....

Hrmm....will need to think on this.

 

[KingJason]

Just getting caught up on our build. Going to be cool seeing it all working once you get the sump in.

 

[kswan]

I mess around with 3D printing ABS and you can get get a really smooth surface with acetone vapor or by simply dipping or brushing in acetone. I would probably just brush it onto the areas of concern. This will melt and fuse the surface without the need to remove your tower. The other option is to do a vapor bath. This could be done in the tank however I would be concerned having that much flammable vapor in my house...

 

[MarsRover]

Just getting caught up on our build. Going to be cool seeing it all working once you get the sump in.

ugh yeah, I still need to get acrylic. Maybe this friday when i have my day off i'll be able to go grab some from a nearby place that doesn't sell it for the exorbitant amount that hardware stores sell it for.

I mess around with 3D printing ABS and you can get get a really smooth surface with acetone vapor or by simply dipping or brushing in acetone. I would probably just brush it onto the areas of concern. This will melt and fuse the surface without the need to remove your tower. The other option is to do a vapor bath. This could be done in the tank however I would be concerned having that much flammable vapor in my house...

Doing a vapor bath in the tank is a bad idea. The base of the tank is PVC and the RTV silicone probably wouldn't fair well either.

That said, I do think i am going to go with painting it on really light and have a metric s**t ton of air flow pulling air out of the tank while i do it.

 

[MarsRover]

I was wondering which return pump you would be going with. How high is that return going? I've read nothing but good things about the Vectra's

so the actual measurement of head height for the pump is 80".

I also realized i needed to order a LOT more sand. So i ordered another 140lbs from marco rocks and another 120lbs from fosters and smith of crushed coral.

Today i am going to try and get over to plastic's depot to pick up some acrylic sheets and then MAYBE go use a laser cutter.

I resolved the overflow tower issue by coating it with RTV. I didn't want to risk ruining the tank, even though it's an extremely low risk, or ruining my warranty by using acetone on the overflow tower in the tank. RTV was a safe bet and so i'm biting the bullet and waiting the additional couple of days for that to dry to test the tower.

Hoping today to put together some of the under tank plumbing and at the very least pick up the acrylic i need for the sump.

 

[MarsRover]

whoo....

Okay so i placed an order for another 50lbs more of Marco rock. Today the last of my marco sand came (total ~280 lbs). I put all the sand in the tank today after finally leak testing the tower and successfully showing it won't leak.

I'm waiting on the remainder of my crushed coral to come in. But that will be added last to top the sand.

We started crafting the 50lbs or Marco rock i already have and I think we have settled on an aquascape.

Looking from above it will look like a claw that makes two arches, a main arch that bifurcates off an secondary arch. Heres a drawing. Side views (mine is on the left and GF's on the right) and her top view:

Then we went ahead and drew it on the glass. In the foreground glass, you can see the main arch (black red) which comes towards the fore-glass and drawn on the back glass (blue green) is the secondary arch that will angle towards the back glass from this perspective.

So, i'll have to wait for more rock to show up this week!

What do you all think?!?

Yesterday I ordered the acrylic to make my sump. should be picking that up monday which should keep me preoccupied this week.

 

[sundog101]

Wow! This has got to be one of the best designed builds I've seen!

 

[KingJason]

Looks pretty sweet!

 

[MarsRover]

Wow! This has got to be one of the best designed builds I've seen!

Looks pretty sweet!

Thanks folks!


Finishing up laser cutting the sump partitions right now. Took a few hours, but haven't wasted any acrylic! So it's going well.

I'll post a video when I can figure out how....just babysitting a laser....

 

[MarsRover]

put together the filtersock holder and the very first piece of the first partition for the sump last night.

Also put together the ATO box