Acrylic fabrication questions? I can help!

[adova]

Would you recommend using a round over bit on the outside edges of the tank and then flame polishing?

 

[Floyd R Turbo]

Roundover is fine but never flame polish

 

[adova]

Ok - and one more. I need to cut out a toothless overflow on one of the short ends for a BeanAnimal overflow. I am aiming for was much width as possible. How much, if any, acrylic should I leave from the edge and top seams?

 

[adova]

BTW - I finished all of the glue up and, with all of your guidance, I am extremely pleased. All of the seams look super clear. Without the help, I am sure I would have been headed for a big $$ disaster!!

 

[dcforester1]

Ad you don't need to stop talking I like the different views. LOL. I am just saying I like to tinker, so I thought it would be fun to work on it. I am having a tank built by Miracles so this needs to last until that came in. I am taking my current tank down so a few months is all I need. I love this tank so I was hoping it could be reused by someone. I hate to toss it out if I can fix it. The panel I got was extruded. I called him today and he spent so much time taking and showing me stuff he got sidetracked and got the wrong type. The part that stinks is its a long weekend and we are getting another 18" of snow so it would have been a good project to work on. I am not going to use it but what is the difference between the two and why is the extruded so bad.? I know for a windshield you want the clarity that cell cast gives, and retains but why the bottom of a tank? Also the rest of the tank is 3/8th, just the bottom was 1/4. Thanks for all the good info.

Derek

 

[adova]

No problem. Here is a quote from James ("Acrylics"):

Major difference between the two is the amount of stress they can handle.

In virtually all physical properties, they are both either equal or close enough to equal to not make a difference. The major difference is in molecular chain lengths which (for acrylic) determines the amount of stress they can handle. No other physical quality, optical quality, or any other quality matters for our purposes. The two are nearly identical in all qualities other than the previously mentioned chain lengths. Due to these high molecular chain lengths (aka molecular weight,) cell cast simply can handle stresses that extruded can't handle. Since aquaria are pressure vessels, it is advised to use a high quality cell cast acrylic for aquaria.

HTH,
James

 

[dcforester1]

Thanks. I am waiting to get the new stuff. Can you mix the two? If I make a sump with cell cast can I use extruded baffles? I don't know I would do that just curious

 

[Floyd R Turbo]

Ok - and one more. I need to cut out a toothless overflow on one of the short ends for a BeanAnimal overflow. I am aiming for was much width as possible. How much, if any, acrylic should I leave from the edge and top seams?

If you are talking about routing out a slot along the back edge to act as a weir, which would dump into an external overflow box, this can get sort of dicey. The issue is that you are essentially disconnecting the structure that the eurobrace provides the back panel. This means that all the bowing stress on the back panel that the euro takes care of is transferred to the slot, specifically, the ends of the slot. What has been recommended if you do this is to bond the overflow box on such that it is flush with the top edge of the back panel, and then extend the euro (one solid piece, not a patch) over the overflow box as well. This, if done right, should effectively transfer the stress through the overflow box and into the euro

I suppose you can do this after the fact also (I'm guessing you have the euro on, or at least cut already) but it might not be quite as strong.

the other option if the tank is built is to make an internal/external overflow box. The internal box can be a very thin profile, just 2" is how I make them, and then held on with bulkheads. The holes you drill for the pass-through bulkheads are nearly zero-stress.

I know that doesn't answer your question directly but maybe the question is now moot....

Thanks. I am waiting to get the new stuff. Can you mix the two? If I make a sump with cell cast can I use extruded baffles? I don't know I would do that just curious

You definitely can mix the two. I don't typically, even for a sump, but what I do for a sump is use 3/8" walls minimum and then any internal non-pressure baffles I use 1/4" and for pressure baffles I use 3/8". But in a pinch I have used extruded for non-pressure baffles and there is no problem with that IMO. You can also build the entire sump out of extruded and for most applications 3/8" will work just fine - 1/4" bows really crazy under pressure. I don't like 1/4" for walls ever, just a rule of thumb.

Just a comment on sumps, I always attach all the baffles to one side (front or back) and then bond them all in one shot to the other side before putting the bottom/top on. Otherwise you will end up with weak joints because you can't get pins into the joint easily, and if you do, they are in there so tight you can't easily pull them out. Bonding them in with capillary action works but leaves a weak joint and the baffle can pop under pressure over time.

The only tricky part is that when you have baffles that also need to be bonded to the bottom to create separate chambers, they (along with the ends) have to match - either they are all dead-on square, or they have to be equally "off" and oriented as such. Even then, with variations in thickness of the front/back panels, you can end up with some joints that just don't want to line up and you have to scrape them down. The bigger (wider) the sump, the easier it is to force baffles into position when bonding the second side joint.

 

[adova]

If you are talking about routing out a slot along the back edge to act as a weir, which would dump into an external overflow box, this can get sort of dicey. The issue is that you are essentially disconnecting the structure that the eurobrace provides the back panel. This means that all the bowing stress on the back panel that the euro takes care of is transferred to the slot, specifically, the ends of the slot. What has been recommended if you do this is to bond the overflow box on such that it is flush with the top edge of the back panel, and then extend the euro (one solid piece, not a patch) over the overflow box as well. This, if done right, should effectively transfer the stress through the overflow box and into the euro

Yes - that is pretty much it. One thing that may make this a bit more acceptable is that the overflow is on the short side of the tank (peninsula build). I would have to extend the euro at this point. I would rather not do the internal / external. Do you think that, considering that this is the short (21.5" side) I can avoid extending the euor, or better of doing it anyways?

Shawn

 

[Floyd R Turbo]

Ah I forgot about that. That certainly makes it less of an issue as the main stress the euro takes up is the long panels.

If you build a rather beefy external box then that might be able to do a good enough job of transferring the stress.

You have the end panel / euro seam and the fillet, after a flush trim it should be pretty smooth but it's not always - there can still be a slight ridge where the euro meets the top of the panel. So you would want your external bonded-on box to be just a hair below that end/euro joint so that you get full contact and a strong bond.

When I say a beefy box - what is your tank wall thickness?

 

[adova]

The tank wall is 3/4", overflow is 1/2" measuring 6" deep x 11" high x 21.5" long, and the euro is 1/2" thick...

 

[Floyd R Turbo]

Should be good

On the slot, I'm not sure on the distance from the top edge since I've never really done that. Do you know what you want the tank water level to be at when it's running? How wide would you make the slot? What is your target GPH through the overflow (actual flow)?

I think the most important aspect is making sure that the ends of the slot are a nice even half-circle, and then make sure to knock down the edge on both sides, at least with a razor blade at a 45

 

[m1k3]

Actually I was having an interesting conversation with Bean about that recently.

Basically what I have been doing is allowing for the footprint dims to allow for 1/2" of clear space from the edge of the drain bulkheads to the inside edge of the box.

Then, the height of the box is driven by how deep the pass-through bulkheads need to be in order to keep them fully submerged during operation. Usually this is not an issue as you generally want the top of the outer box to be roughly lining up with the top of the tank. The factor here to consider is which is worse: the tank overflowing, or the box overflowing, in the event of a total failure (2 pipes clogged and third cannot handle flow, or all 3 pipe clogged).

The one in the pic, I made so that the outer box is actually even with the bottom edge of the trim on the tank. So the outer box is lower than the rim of the tank. What I have realized since is that in a bad scenario where the drains can't keep up, and then let's say you have a sensor that shuts off the return pump when the water level in the external box rises to a certain point, or it overflows. What happens in this situation is that (let's say it's a total block of all 3 drains) the external box will continue to overflow until the water level in the tank drops such that it is below the level of the external box, or reaches the bottom of the weir or notches (in a notched overflow). Because of this, I now make these so that the external box stands off the tank so that it can be installed even with the rim of the tank by adding a laminated sheet of 1/4" material to that side of the external box.

Then, you have plenty of "overhead" for your emergency to kick in full and siphon & flush as the box level oscillates.

Increasing the footprint dim of the external box will give you extra volume to allow for the time factor in which the box fills up to a critical point. This turns out to be a bit of a complex calculation, because you have to figure out

1) how fast the water level rises based on the footprint of the box and in influx of water,

2) how long it takes for the emergency to turn into a full siphon (which depends heavily on your pipe routing, which is why it's important to have no 90s and a straight shot to the sump)

3) what water level you need in order to cause the emergency pipe to turn into a full siphon with no air intrusion from the surface (vortexing)

4) what the full-siphon capacity of the emergency pipe is

Usually, the only time this becomes an issue is when you are pushing 3000 GPH or better. 1.5" emergency pipe with a straight shot to the sump (or maybe just a few 45s) will siphon pretty quick once the water over it reaches about 1" or so, so you just have to allow for the rate of rise of water that gets it to this point and allow for a comfortable overage.

At 1500 GPH, if you have 2" of headroom from the top of the emergency pipe to the top of the box and it's a 1.5" pipe - in reality, no worries, not at all.

Floyd,

I am waiting on material to arrive to do just this. I am doing this on a rimmed tank and you mentioned something I had not taken into consideration. You posted that you add a layer of 1/4" material to essentially make a spacer laminated to the box. Can you elaborate on the technique to do that lamination?

You also mentioned a pic of one that you had posted. Can you give me an idea on which thread to find that in? If it was this one I missed it.

TIA

Mike

 

[thejuggernaut]

I think a good idea would be to make an acrylic spacer that looks like a washer. Use two different size hole saws. Cut one about 1" larger than the hole you'd drill for the bulkhead then drill in the center the size of the hole for the bulk head. That will give you a 1/2" lip all the way around in order for the bulkhead gasket to seal well. It's important you cut the big hole first, cause you need the pilot hole you make with the first hole saw in the middle to make sure you center the smaller hole.

Disclaimer, I've never tried this, but it think it would probably work. You'd just need to make sure the thickness of the "washer" is thicker than the outer lip of the rim.

 

[Floyd R Turbo]

Floyd,

I am waiting on material to arrive to do just this. I am doing this on a rimmed tank and you mentioned something I had not taken into consideration. You posted that you add a layer of 1/4" material to essentially make a spacer laminated to the box. Can you elaborate on the technique to do that lamination?

You also mentioned a pic of one that you had posted. Can you give me an idea on which thread to find that in? If it was this one I missed it.

TIA

Mike

It's post #4013, earlier in this thread

http://www.reefcentral.com/forums/showthread.php?p=23429364#post23429364

That one actually has the top of the external box against the bottom of the trim on the outside. It works just fine but I didn't really think about what would happen in the case where the pipes in the external box are overwhelmed to the point where they can't keep up with the influx of water from the tank. It's a pretty slim chance of that happening though, you would have to have something happen like 2 turbo snails crawl up and over the notched box, into the internal box, they would have to somehow get through the pass-through bulkheads (the openings of which are less than 1" from the inside face of the front panel of the inside box - so literally, no chance) and then one would have to get into the siphon line and the other into the open channel, the latter of which has no valve on it so it would pass through. Chances of a worst case failure and the external box overflowing are really, really low. But they are there.

So to answer your question, what you do is take the panel of the external box which has the pass-through holes on it (with the paper masking in place) and double-stick-tape a piece of 1/4" to it. If you have one straight edge, line that up such that the external box would be at the tank trim height (top of tank) and the 1/4" piece would line up with the lip. Then use a flush trim bit to route off the material off the other 3 edges.

If your holes are already drilled in the box part, then you just drill out the 1/4" piece and flush cut the hole to match.

Now you take them apart, peel off the masking, clean & prep, apply the solvent, and sandwich them together. Warning, it's not going to look pretty. But if you have a black internal box that doesn't matter.

How you do it is you take WO #3 or #4 or even #16, lay the back panel face up on a table (preferrably, on a board with foam strips, you'll see why) and then essentially draw a circle around the 2 pass-through holes with the solvent. This is all you *really* need to bond, the panel is not structural but you need it to be watertight, so you need a full seal around the hole between the panels. Theoretically you could just silicone and clamp also, again, because it's not structural and doesn't have to be. Anyways, once you have the solvent in place (and you can add more all over the area that will be touching together) you will want to start with the 1/4" piece at a slight angle, touching on one edge, and then hold it there with one hand while you lower the piece down and make contact from one side to the other. It's a form of art to do this and make it bubble free. It's a form of art to do this and not make it look like total %$#% really. But generally it doesn't mater how it looks.

Make sure the holes and edges line up and you are good to go. Come to think of it, it's better to just cut & trim the holes, then bond, then after it cures, flush trim the other 3 edges, that way the will all line up for sure.

HTH
Bud

 

[Floyd R Turbo]

I think a good idea would be to make an acrylic spacer that looks like a washer. Use two different size hole saws. Cut one about 1" larger than the hole you'd drill for the bulkhead then drill in the center the size of the hole for the bulk head. That will give you a 1/2" lip all the way around in order for the bulkhead gasket to seal well. It's important you cut the big hole first, cause you need the pilot hole you make with the first hole saw in the middle to make sure you center the smaller hole.

Disclaimer, I've never tried this, but it think it would probably work. You'd just need to make sure the thickness of the "washer" is thicker than the outer lip of the rim.

Yeah that would do the trick too. But it might actually be tougher to make 2 rings than to just laminate the whole sheet.

 

[m1k3]

Think I found the pics you were referring to at post 4013.

Jugg, I get what you are saying, and as I was contemplating this on the drive home from work, I was coming up with the same idea. Except I would create a pattern out of MDF and route the ring. I can see how that would be narrow enough to get a good bond with WO#4. When I read Floyd's post, I read it as a layer over the full back panel. I think I would also add strip along the bottom edge to alleviate some of the stress that might otherwise be induced.

Floyd - What say ye?

 

[Floyd R Turbo]

Yup. Remember that there will be a gasket between the external box and the tank wall, and those are about 1/10" thick. So you will actually need a piece of something that is 0.236 + 0.10 = 0.336" to hold the external box vertical and relieve the stress. Or, you can just add the 1/4" acrylic and take an extra gasket and cut a few chunks off and glue them to the bottom corners

 

[m1k3]

Ah Floyd, I looked at the page, then started typing my short response and in that time you wrote a book. I did have a wife distraction though in my defense. Thanks for the info!

One last question, since I am a bit green with working with acrylic. I know that some manufacturers use different formulations in their acrylic. I believe I only have enough of the black acrylic I ordered to make the boxes. Are there any issues when bonding two acrylics from two different OEMs and/or cast vs extruded? Keeping in mind that this would only be used for the filler we are talking about. The rest is all 3/8 cast.

 

[CatsnClownfish]

This thread is huge. Like, days of reading huge. I'm working my way through it, but I don't suppose there's some place where the main advice from this (some of which gets repeated here) is concentrated into a smaller amount of space?