Cylindrical Acrylic Tank Fabrication .. a couple Q's

Æonic

New member
Hi all,

I'm going to try to form a sheet of 1/2" acrylic into an 18" diameter, 15" tall cylinder.

Originally I was planning on just using a heat gun and a form and drape form it a bit at a time. But after a little testing, a heat gun is too slow and probably too uneven for this.

The new plan is to build a small "oven" that will let me heat a foot or so of this material to 300, evenly, and let that part drape over the form. Once it cools enough to hold its shape, move it a little, with some overlap, heat the new section, etc, until I have the full cylinder.

I had a few questions, if anyone with experience with this would be so kind:

How long do you generally heat a sheet of 1/2" acrylic at around 300 degrees for the acrylic to reach around 300?

Do you think the edges will remain square or will I have to "plane" them down to be flat again?

Do you see any other problems with this plan?

Thanks!

I'll keep this thread updated too as I go.
 
> Yes, it won't look good

Could you elaborate? Specifically, what about it will be bad? I do totally respect and appreciate your experience, but it would be very helpful to know the potential pitfalls I may face.

Several sites that will sell a cylindrical aquarium, do seem to create them by drape forming them in an oven. I think they make a few segments.. other than that, seems to be the same idea.

> spend the money and buy the chunk of tube

What do you think a 16" tall chunk of 18" diameter tube should cost? If I make a big paperweight, I might end up going this route.

Thanks ;)
 
Last edited:
Ok so what I've seen so far as a result of looking at some older threads...

Bending + glueing a cylinder will always show a seam. I'm ok with that. But it's obviously a critical part, getting the two sides to match up cleanly.

Bending will flare out the edges due to compressing the inside and stretching the outside. This does sound like it will be a pain, I'll have to find a good way to cut or route out the flare, assuming it will happen in my case, which is far less bend than the typical bend you see on the corner of a rectangular aquarium.

I'm not too worried if it's not perfect - like, as long as it's reasonably round and fairly uniform. I know if I bought a prefab tube it would be "perfectly" round and "perfectly" uniform. But honestly once it's full of fish and coral, and because it's a small tank, and because of the shape -- it won't have any hardware or any "front" or "back" or edges-- I feel the imperfections will be easy to miss. Maybe with strategic placement of the seam. Heh:)

My main concerns right now are, possibly overheating the sheet, possibly leaving marks on the sheet (It will be resting on felt over a pretty nice, uniform form), and longevity of the seams (not unique to my build) and the material in general after being gently bent like this.
 
the glaring problem is half-inch acrylic isn't cheap. after one failed try you might as well have just bought the acrylic tube.
 
i agree with the above posters that it will probably not work. i have only formed 1/2 inch acrylic into braces for the top of a frag tank. getting a 3 inch wide piece of 1/2 inch acrlyic bent to a 90 degree angle was difficult and not even. i cant imagine bending a large piece evenly esp doing it in multiple bends. i think if you cant heat and form the whole sheet at once you would be better off cutting it into segment and heating and bending those pieces, cleaning up the edges and then joining and sanding the joints. i cant imagine it being cheaper and the it would be very hard to make it look acceptable. good luck if you do pursue this project and if you do please post your progress so we can learn from it.
 
Edited... Hehe.. I got a little excited.

This is the DIY forum. Buying a tube, is not an option in this forum (unless you're going to buy a tube and try to make it flat so you can make a rectangular tank, which knowing me, will be my next project). This acrylic sheet is as good as dead. It's either going to be a cylinder, or a big steaming pile of fail. I'm perfectly willing to accept either, but I want to give the former my best shot.

That's the fun of this forum. Encourage trying new or crazy things. The only time great things are accomplished in life are when someone doesn't have the good sense to give up when they should.

I'll definitely post my progress. I will say that half inch in real life is a lot thicker than you might think. It's like, half a whole inch. That's a lot. If I had to try again, I'd go 3/8. But half inch it is. You'll be able to drive a tank over my tank (or over my big steaming pile of fail) and it'll probably survive.

Also: I tend to be very tongue-in-cheek. Please don't take any of my crazy ranting personally. Originally I had posted a paragraph about how Tony Stark was locked in a cave and made an Iron man suit powered by a "arc reactor" that FLEW.. and all I am trying to do is make a plastic cylinder... made it sound easy, but ... Well look there it is again, guess I'll leave it this time. :)
 
Last edited:
I say build the oven!

But what's stopping you from building it the full length of the acrylic? Seems in my imagination it'd be best to do the whole length at once.
 
I'd make a semicircle form. Just sit it in your shop, sawhorse like. Make sure you include guides on the sides to keep things square.

Lay the sheet on it and use a radiant heater to uniformly heat it until the material lays flat around it. At some point you can rotate it some more and pickup the part that hasn't bent yet. Work the whole thing around and around until you're happy with it.

Doing it that way will let you touch it and work with it in ways you won't be able to in any oven.

For a radiant heater that cranks out ungodly amounts of heat for cheap, use one of those clamp-it-to-a-propane-bottle radiant space heaters. I have one that heats my whole living room to just below flash point in about 1 minute. Scary hot from ten feet away. (Yes I know all about CO, I used it only when the windows were missing.)

Use the heater on a hose so you can wave it around and use it as needed.

The only thing missing from this equation is that it's typically a good idea to heat both sides, i.e. the form face too. Perhaps make the form out of metal and paint the inside flat black and use the smallest radiant heater you can find underneath.

Here's just a picture example:
http://www.onlinesports.com/pages/I,CLM-5017A751.html
 
In the above example I would worry slightly about heating points to hot and getting some "burns" or blisters in the surface, so be carefull... and I like the black undersurface....good call!
 
In the above example I would worry slightly about heating points to hot and getting some "burns" or blisters in the surface, so be carefull... and I like the black undersurface....good call!

Thanks.


BTW The black is because surface emissivity issues are heightened by using radiant heat.
 
The only time great things are accomplished in life are when someone doesn't have the good sense to give up when they should.
Or listen to others who have the experience or knowledge to accurately predict the results beforehand...

Iron man suit powered by a "arc reactor" that FLEW.. and all I am trying to do is make a plastic cylinder... made it sound easy, but ... Well look there it is again, guess I'll leave it this time. :)
When reality and science fiction meet, expectations need to be lowered...



I don't think James is telling you it "can't" be done. he is trying to kindly tell you that given your experience, described method and materials, the end result is going to be less than most of us would consider acceptable.

You are likely going to see every "segment" that you heated, as there will be slight differences in the flexibility and bend radius. You are going to have a hard time matching the seams and truing the bottom edge.
 
What will *probably* happen is the tank will look rather wrinkled(wavy), uneven, and gluing the seam will most likely be quite problematic. Once you get say 1/2 of it formed - forming the other parts quickly become an issue.

As a generality, tanks that are formed do not last *nearly* as long as spun-cast tubes. Crazing is almost always seen at the seam(s), if not immediately, in the near future.

If you want a certain diameter, you're probably going to have to make it several times. Acrylic shrinks as it cools so if you make a form for a 17" inside diameter (18" OD) - the piece will end up closer to 15-15.5" ID. This part is not an exact science; depends on material thickness (varies on cell cast sheet,) latent moisture, cooling rate, etc. For a form, you can do something simple as MDF/partical board wrapped in Formica, the better forms are rolled sheet metal. All of this is a time/money thing vs. just buying the tube.

And as a FWIW, acrylic tubes are hell to scrape, formed tubes are worse, segmented formed tubes would be a nightmare.

Not telling you *not* to do it, go ahead, for culture if nothing else. Just that for the time & money you're probably going to have to put into it, and the expected result - IMO you'd be better served buying the tube :)

James
 
You are likely going to see every "segment" that you heated, as there will be slight differences in the flexibility and bend radius. You are going to have a hard time matching the seams and truing the bottom edge.

That is very useful. Thank you. The bottom edge (and really, the top too, but that's more for asthetics) does worry me. Right now, I feel I'll be spending a lot of time setting up my router to at least get it flat. Hopefully I won't have to remove too much of it. It's kind of a bridge I'm going to have to cross when I get to it. Maybe it'll look like a tree stump when I'm done. Hehe :) I can say I planned it that way.

The segments, I feel are going to be more forgiving, and I can reheat one if it's too obvious (I think).

kcress, thanks for the suggestion. Even with a badass heater like that, I think I'd be risking scorching the surface, or heating it unevenly. I don't trust myself to wave it patiently for the time it would take to get a uniformly softened segment, going by what I saw with the heat gun. It would also be hard to know if I had any cold spots, which I think would cause a lot of trouble.

So with my current plan, the oven will contain the heat and maintain a constant temperature across the entire segment, and will fit around the form with the sheet draped over. If it takes an hour to get the sheet to the right temp, I don't care, it just sits in there. The parts of the sheet that stick out will have a graident of flexibility, which I hope will help blend the segments, which themselves will overlap some. I'll have to support them to prevent a collapse as the piece softens.

Then... "Work the whole thing around and around until you're happy with it." Exactly this. But slowly.

> But what's stopping you from building it the full length of the acrylic
Mostly it's that I want to tackle this in small chunks, rather than a large "hail mary". I've done the all at once method for different things and its a lot more risky. But there's also concerns like gravity wanting to work against me (instead of for me as it will with small segments) and just how unwieldy the full sheet would be.

One question I still had is that, I've seen you can use felt to prevent marking the surface. That's true for a 300 degree slab of acrylic with all its weight pressing into felt? Or will that side take on the texture of the felt? If nobody knows, I'll have to test it. I don't want a frosted texture reef tank ;) Like a "privacy tank" so the fish can be themselves. Lol :)

Thanks all for your time.
 
What will *probably* happen is the tank will look rather wrinkled(wavy), uneven, and gluing the seam will most likely be quite problematic. Once you get say 1/2 of it formed - forming the other parts quickly become an issue.

As a generality, tanks that are formed do not last *nearly* as long as spun-cast tubes. Crazing is almost always seen at the seam(s), if not immediately, in the near future.

If you want a certain diameter, you're probably going to have to make it several times. Acrylic shrinks as it cools so if you make a form for a 17" inside diameter (18" OD) - the piece will end up closer to 15-15.5" ID. This part is not an exact science; depends on material thickness (varies on cell cast sheet,) latent moisture, cooling rate, etc. For a form, you can do something simple as MDF/partical board wrapped in Formica, the better forms are rolled sheet metal. All of this is a time/money thing vs. just buying the tube.

And as a FWIW, acrylic tubes are hell to scrape, formed tubes are worse, segmented formed tubes would be a nightmare.

Not telling you *not* to do it, go ahead, for culture if nothing else. Just that for the time & money you're probably going to have to put into it, and the expected result - IMO you'd be better served buying the tube :)

James

Thanks James!

I had a few questions about your post:

> Crazing is almost always seen at the seam(s), if not immediately, in the near future.

How is this seam more prone to crazing than a seam on a rectangular tank? Is it just because of the bending, or the heating, or because it's hard to match up so probably isn't matched up as nicely as on a rectangular one? Or maybe because they're often polished?

> Acrylic shrinks as it cools so if you make a form for a 17" inside diameter (18" OD) - the piece will end up closer to 15-15.5" ID.
That's extremely good to know, thanks. I thought it might shrink in length but didn't consider the radius shrinking.
 
That's extremely good to know, thanks. I thought it might shrink in length but didn't consider the radius shrinking.

The length, wrapped end to end is what makes up the diameter. If the length shrinks, then so does the diameter of the circle that the length makes up.

The eneven shrinkage (as a result of heating in segments) is also going to contribute to the eneven (wavy) appearance of the finished product when filled with water.
 
How is this seam more prone to crazing than a seam on a rectangular tank? Is it just because of the bending, or the heating, or because it's hard to match up so probably isn't matched up as nicely as on a rectangular one? Or maybe because they're often polished?

Likely because the end result of heat formed acrylic will result in a high surface tension. (same as flame polishing)

For crazing? Just add glue. Of course your seam will be glued so the fumes from the glue will be around that area and could lead to crazing.
 
How is this seam more prone to crazing than a seam on a rectangular tank? Is it just because of the bending, or the heating, or because it's hard to match up so probably isn't matched up as nicely as on a rectangular one? Or maybe because they're often polished?
A few reasons;

On a rectangular tank, unless the joints are mitered, the force of the water is never "trying" to pull the joint apart. On a cylindrical tank, it is.

You have several secondary operations giong on there; the forming itself, the machining/prepping the area for bonding, the bonding agent itself, and then further polishing.

This area also tends to be primary stress point(s) on a cylindrical tank. Cylinders can use thinner material because the stress is distributed evenly around the circumference of the tank. Unless you get an *absolutely perfect" bend and then *absolutely perfect* machining to prep, you will have a slight "out of roundness" at the seam. If this is the case as it typically is, then these areas become areas of concentrated load. Couple this with the secondary operations, and crazing is a real potential issue.

From what I've seen, even annealing the tank as a completed unit will not alleviate the issues above; virtually every formed cylinder I've ever seen over 4' diameter gets crazing in the vertical seams.

> Acrylic shrinks as it cools so if you make a form for a 17" inside diameter (18" OD) - the piece will end up closer to 15-15.5" ID.
That's extremely good to know, thanks. I thought it might shrink in length but didn't consider the radius shrinking.
Even if you make it in semicircular segments and glue them together, the radius still shrinks, so keep it in mind :)

This, along with the time, money, etc., required to actually do it lead me to recommend buying the chunk of spun-cast tube rather than forming it.

HTH,
James
 
Ok, so I'll admit when I'm beat :) I was able to bend about a foot or so, but as everyone predicted, it seemed pretty much not worth it. I ended up with a curve that looked not too bad, actually somewhat promising, but with some flaws. The flaws were a visible boundary between bent and unbent areas (I was hoping it would be more of a gradient), and a curling of the sheet along the would-be height of the cylinder. And I could tell getting it flat again was going to be a nightmare as well as progressing past the very easy first 10 degree angle worth that I did.

So you guys were right, but I had fun giving it my mostly best try.

I think if I were to try again (and I may someday!), I'd definitely go with a thinner sheet. 1/2" is a bear to get heated up and to deal with once it is (and it was complete overkill for this size). I'd spend more time getting something safer and where the heating would be more even (ie a bigger oven with a regulated temperature)... that's quickly getting beyond my weekend quota of free time, heh.

So live and learn. I lose a foot of acrylic and a little pride ;)

With the rest of it, I'm going to just make a 19"x19"x16" tank with a center overflow. It's still going to be awesome... and meanwhile I'll keep my eyes peeled for chunks of large diameter acrylic tube for less than $200 a foot ;]

Just wanted to update, since I said I would.
 
Back
Top