Acrylic fabrication questions? I can help!

[Phixer]

super, because if you bond 2 surfaces directly together with WO40 such that they are a "perfect fit" the joint will push out the WO40 and you will end up with a dry joint.

If you read enough of this thread, you will see references to angles on the pieces to create the void. This I believe is the proper method. It is in this thread, you have to really search for it though. Try using the search function on this thread only for the term "40" or "42" and you might find it, also go back to the splits I think it's buried there.

Phixer, in reference to you question here is what James e-mailed me

Thanks Floyd...

Forgot to mention to Supernemo earlier, the decision to use a CNC to mill these panels was recommended to me by the originator of this thread Steve Frosti aka Acrylicman many years ago when I talked to him on the phone. The other reason was because of the weight of handling these panels (500lbs ea) over a router table and the difficulty in doing this with a hand router was not something I was up to. In addition I wasnt sure if I I could replicate each edge as uniformly as a CNC could.

A CNC with a vacuum hold down table and laser sight was just a better option for me. The edges are good IMO but do have some minor lateral grooves in them. Probably due to feed rate as James mentioned. I hope they will be OK and not trap air. Upon closer inspection, the pitting is on the unmilled sides. Thank you for the reply and thank you James for your advice, it's always appreciated.

I plan on clamping a piece of 1/2 plate glass on the bottom of the V to keep the cement from running out.

 

[smartwater101]

I have a few questions about my design and another member suggested I post here.
I'm wondering how much my cutout in the back will effect the strength of the tank. Would It be smarter to make two slots so the middle would connect?

35.5 x 18 x 18

Also... some thickness calculators suggest 1/2" acrylic and others suggest 3/8". What are your thoughts?

Thanks!

 

[Floyd R Turbo]

The reason I decided to use 5 degree bevels on each 4 ft section was reached after about 3 years of research.

Although two 22.5 degree bevels equaling one 45 degree void

I decided against a 90 degree butt joint because I didnt feel this was enough surface area for the cement,

For me.. two (5) degree bevels making a 10 degree void

I'm not quite understanding something here. Are you talking about making a V joint with 2 pieces with 45 degree cuts on them with both of the angles facing up so that there is a 90 degree V, and then you fill that in with WO40? Or 2x 22.5 for a 45? Not having worked with this I didn't see this as the right way to do it.

My interpretation of the using 45 degree joints would be to mill them and position them such that the ends are parallel to each other like this // and then you have a small void in the middle, but a large surface area.

I've seen cylinder tanks with 2 40 joints and you can see the gap between the sheets (i.e. the difference in clarity of the material vs the WO40 - this was on a tank that was being refinished).

Just not quite understanding why you would position a cast joint with the 2 pieces creating such a large void, that does not seem logical.

During my research and observations... one commonality I found was that all of these aquariums meaning every public acrylic aquarium I inspected throughout the world is made using 2 part polymerizable cement. PS30 or WO 40/42. Solvent bonding has it's place. I've just never seen it used on the big public aquariums.

I have no doubt about this, WO40 can make perfectly invisible seams and is the only thing to work with for casting vertical joints, no surprise there.

Forgot to mention to Supernemo earlier, the decision to use a CNC to mill these panels was recommended to me by the originator of this thread Steve Frosti aka Acrylicman many years ago when I talked to him on the phone.

Is that guy still around??

 

[Floyd R Turbo]

I have a few questions about my design and another member suggested I post here.
I'm wondering how much my cutout in the back will effect the strength of the tank. Would It be smarter to make two slots so the middle would connect?

35.5 x 18 x 18

Also... some thickness calculators suggest 1/2" acrylic and others suggest 3/8". What are your thoughts?

Thanks!

I think this is fine but you have to realize a few things. When you cut away the top edge of the back panel, or even when you cut a slot in it, you disconnect the euro from the back panel. All the stress then transfers to the inside corners of the back panel. To offset this, the external box needs to be very sturdy because it essentially transfers the structural load.

Ideally, you would want to extend the eurobrace over the external box. But then creating access holes is the issue. If you do it right, you would only need a few round holes big enough to get your hand in in order to adjust plumbing.

If you wanted it to be a rimless external box, I would make it really thick. Like 3/4" or even 1".

 

[smartwater101]

I think this is fine but you have to realize a few things. When you cut away the top edge of the back panel, or even when you cut a slot in it, you disconnect the euro from the back panel. All the stress then transfers to the inside corners of the back panel. To offset this, the external box needs to be very sturdy because it essentially transfers the structural load.

Ideally, you would want to extend the eurobrace over the external box. But then creating access holes is the issue. If you do it right, you would only need a few round holes big enough to get your hand in in order to adjust plumbing.

If you wanted it to be a rimless external box, I would make it really thick. Like 3/4" or even 1".

Thanks! I think i''ll go with an overflow divided up like this. I'll def go with 3/4" to be safe.
Also, would adding a couple inner braces or top braces to the box provide the same strength as an extended top?

Also, what thickness do you recommend for the actual tank?

Thanks.

 

[SuperNemo]

Floyd, still around. Been busy doing research.

I think Phixer might have mistake what I mean by 45degree bond. Like you mentioned the pieces are more like "/ /", than "\ /".

Floyd, can you give me your opinion on clamping the two pieces together. I had asked James, but guess he is a little busy. Would you choose option 1, 2, or 3 for the layout of the surface you would clamp the two sheet together? I am more incline to use option 2, which would have one solid piece of plywood with a grove cut into it.

 

[smartwater101]

I'll be using 1/2" but I do I have one more question about the tank top. My plan was to have the top inset instead sitting on top of the four walls. Although I don't know the benefits of each method. Which is more structurally sound? Thanks

 

[Phixer]

I'm not quite understanding something here. Are you talking about making a V joint with 2 pieces with 45 degree cuts on them with both of the angles facing up so that there is a 90 degree V, and then you fill that in with WO40? Or 2x 22.5 for a 45? Not having worked with this I didn't see this as the right way to do it.

My interpretation of the using 45 degree joints would be to mill them and position them such that the ends are parallel to each other like this // and then you have a small void in the middle, but a large surface area.

I've seen cylinder tanks with 2 40 joints and you can see the gap between the sheets (i.e. the difference in clarity of the material vs the WO40 - this was on a tank that was being refinished).

Just not quite understanding why you would position a cast joint with the 2 pieces creating such a large void, that does not seem logical.



I have no doubt about this, WO40 can make perfectly invisible seams and is the only thing to work with for casting vertical joints, no surprise there.



Is that guy still around??

Floyd

Yes. Ok I got you. Sorry for the confusion.

What your describing "//" both at 45 degrees is a kerf joint. Used quite frequently in finish carpentry. I considered this, it was my second choice any yes lots of surface area but a lot of room for bubbles to get trapped. For some reason I thought he was talking about a 45 degree V joint??? Yes, that would be huge void and not practical at all.

I didnt see any of the tanks I visited that used kerf joints, all used a V or straight 90 degree butt joints. I dont know why ? Maybe they were made so well I couldnt tell and they just looked like 90 degree butt joints. In retrospect I may heft the viewing panels back to have them cut this way.

As you decribed I would think the clairty of a kerf joint would be good since your looking thru both acrylic and cement. However your looking thru a wider patch of cement also. With a V your looking thru cement only but a much narrower patch of it. Distortion?

I considered the internal pressure wanting to push the vertical panels outwards and how each joint would react under this pressure? When designing I wanted to use the compressive strength of the cement rather than the shear strength since the cement has great compressive strength.
Both joints provide good resistance to shear strength as well but in different ways. Splitting hairs on this I know, but thats what I do LOL.

For the bottom, sides and top edges I believe a wedge is a better joint but for the viewing panels Im thinking that kerf joint may be a better option now because of that additional surface area.

Acrylicmans repututation has been ... interesting. I saw some unique comments about him on the net but have not heard from him in many years. Hope he is doing ok.

Finally... Some photos of the tanks. These are from Dubai, made by Reynolds. You can see the vertical seam if you look closely. The curved tank is a 90 degree butt joint, the big tank is hard to tell.

 

[Floyd R Turbo]

Acrylic fabrication questions? I can help!

I'll be using 1/2" but I do I have one more question about the tank top. My plan was to have the top inset instead sitting on top of the four walls. Although I don't know the benefits of each method. Which is more structurally sound? Thanks

Much more difficult to build I would think, and not much benefit. Why do you want to do it this way?

 

[Phixer]

Attached is what the machined ends look like. Tried to get a good pic of what a 5 degree bevel looks like. Disregard the piece above, this is an unfinished end and wont get bonded.

 

[smartwater101]

Much more difficult to build I would think, and not much benefit. Why do you want to do it this way?

I just noticed other plans had it built that way. It doesn't really mater to me. If there isn't a benefit I'll just stick to having it sitting on top.

Also,I'm assuming that 2 cuts for the overflow would help with strength? I may even just do slots instead having the cut go all the way to the top. What are your thoughts?

Thanks again for the help. Much appreciated!

 

[SuperNemo]

Attached is what the machined ends look like. Tried to get a good pic of what a 5 degree bevel looks like. Disregard the piece above, this is an unfinished end and wont get bonded.

Those pieces look very well machined.

Phixer, can I get your opinion on how you would go about clamping the two piece to bond them side by side. I've only seen what the two members on here have don't.

Hopefully Flyod would chime in as well.

 

[Phixer]

Thanks for the feedback. They are actually pretty smooth also, Im probably knit picking it too much. I tend to do that.

Would be glad to explain. The sides are bonded using a different method (I make wedge joints). If your talking about butt joints, here's how I do it and, very similar to what the other guys said.

I made four 4x8 ft steel tables (like hospital gurneys) with wheels on the bottom. Each table is hinged on one side along the 8ft side. The table tops are made of two sheets of 3/4" MDF with the top sheet being melamine to allow the acrylic to slide. The slabs are laid horizontally on the tables and wheeled into position with the 4 ft edges to be bonded facing each other like this... end to end: _________ _________

They rest on the tables under their own weight. Just ensure they are lined up parallel to each other, (will explain how to do this). I then clamp a sheet of 1/2" plate glass with wax paper on the bottom of the slabs to bridge the gap and to keep the cement in the void when casting. After the plate is clamped in place I use a small Bosch laser level at one end of the panel to ensure I have a straight plane from one panel to 16ft away at the end of the other panel to be joined. I shim as necessary across the full 4ft width of the panel using cardboard drywall shims along with an 8ft carpenters level as a straight edge to make sure everything is flat and use a dead blow rubber mallet to gently adjust the panels into place where required. Using lasers combined with carpenters squares and standard levels promotes redundancy and ensures everything remains square prior to bonding.

I use aluminum foil tape to create a dam at each edge of the panels to keep the cement in and tape wax paper to the adhesive side of the tape that will be in contact with the cement because I dont want the glue residue from the tape to come into contact with the cement. Probably unnecessary yes, but I dont want to take chances on contaminating the cement in any way. I tape parallel to the void on top of the panels where I pour the cement in and peel it up slightly along the edge of the tape so that it keeps the cement in the void and allows me to overfill the void slightly due to shrinkage. Kind of like a spoiler on the back of a car deflecting the air where you want it.

42 is injected from the gun into a large syringe to allow better control and ensure the cement is bubble free prior to filling the void with the syringe. The gun is Swiss made and is basically a bulky calking gun with a long mixing tip, I like to be very close to the joint when filling. Good ventilation and a mask help also. I find the syrenge allows me more control than the gun when filling the void because I can use my thumb. After filling, suck out any remaining bubbles with smaller syringe and hypodermic needle.

The void is overfilled, after it dries wet sand flush using alternating 90 degree (perpendicular) sanding patterns with progressively finer grits of sand paper until you can barely feel the ridge with your finger nail, ( I learned this repairing aircraft canopies). This method of perpendicular sanding prevents swirl marks and allows you to see how much material is removed during each course of sanding). After sanding you finish flush by hand polishing until all minute scratches are removed and there is minimal distortion. You dont want to sand too much and create a concave surface as this will cause distortion.

Some use heat lamps to cure and others do not, really depends on temp and humidity of the air.

I am really considering the 45 degree kerf joint mentioned by you guys earlier, researching this now and think I may end up using this joint for viewing panels instead. Bonding process would be the same, I'll still use the 5 degree wedge joints for the edges. Sorry for the long winded reply, hope this helps.

 

[SuperNemo]

Long winded is a good thing when it comes to something this technical.

I would guess that making two separate tables with wheels in your case would be a great idea. 2.5" material is pretty heavy to maneuver around.

Love the idea of the little damn for overfilling and shrinkage.

The main question that I would have for your setup is how do you plan on moving the two sheets closer together after the initial fill of 42. Based on James explanation, you fill the joints and then after a couple minutes push the two pieces slightly closer together to squeeze out some of the 42 (I'm guess due to shrinkage). My only guess is that you have compensated that with the little damn you will be making. But this only accounts for the shrinkage on the top, do you think that there will be a problem with shrinkage on the bottom/sides if you do not plan of pushing the two pieces together slightly. Also, having the two sheets on separate movable tables is a great idea and makes it very easy to move. But if you need to push the sheets together, based on the above procedure, would that cause problems if they were shimmed.

 

[Phixer]

Long winded is a good thing when it comes to something this technical.

I would guess that making two separate tables with wheels in your case would be a great idea. 2.5" material is pretty heavy to maneuver around.

Love the idea of the little damn for overfilling and shrinkage.

The main question that I would have for your setup is how do you plan on moving the two sheets closer together after the initial fill of 42. Based on James explanation, you fill the joints and then after a couple minutes push the two pieces slightly closer together to squeeze out some of the 42 (I'm guess due to shrinkage). My only guess is that you have compensated that with the little damn you will be making. But this only accounts for the shrinkage on the top, do you think that there will be a problem with shrinkage on the bottom/sides if you do not plan of pushing the two pieces together slightly. Also, having the two sheets on separate movable tables is a great idea and makes it very easy to move. But if you need to push the sheets together, based on the above procedure, would that cause problems if they were shimmed.

Got ya, I dont. Here's why..., on public tanks constructed with standard slabs 4x8' or larger and material over 2" (due to size and weight) panels are normally assembled and bonded on site, and most are not annealed. As you can probably imagine maneuvering such large slabs of acrylic within a fraction of an inch within the time you have before the cement sets after casting is a very small window. For this reason I prefer not to move the acrylic after casting the joint.

Imagine what would happen if the the void was full of cement and while squeezing/tapping/nudging/adjusting... two 500 lb panels together end to end one was accidentally nudged only 1/16" too much, (the initial gap is only about 1/8 "). The cement would be pushed out more in this area, the seam would be uneven and a dry spot might even be created if the two slabs contacted each other, in addition to a crooked seam on the viewing panel. The longer the seam the more significant slight variations are. You would then have to pull them apart to the proper gap and refill ,(a large chance of introducing air pockets). And during the whole time your racing against the clock because the cement is curing.

For something more manageable I would do exactly as mentioned and adjust them after casting. But for the larger really heavy stuff (over 500lbs per slab) , it's much more difficult (for me) to line up uniformly without introducing air pockets or dry spots. Im not skilled enough to attempt this on the monstrous panels. For this reason I prefer to not squeeze it together after pouring but rather to fine tune before casting and then: fill the void partially, inspect, fill a little more, inspect, then slightly overfill for the last pour.

Because contraction/shrinkage occurs within the joint on all sides it is not uniform (although minimal) because the top is overfilled. Over filling the void allows continuous pressure within the void the same way the inside of a container maintains contact with the liquid as evaporation occurs from the top. The more a container is filled the more hydrostatic pressure is created due to the height of the column. No problem with surface contact on the bottom and sides due to shrinkage (contraction of cement) because each side is dammed, (bottom with plate glass and sides with tape), the top is overfilled and the level of cement is monitored and adjusted during curing to ensure it remains above the surface of the panels being bonded or (convex). This way all sides maintain equal contact/pressure as the cement cures. Overfilling acts as a reservoir. Gravity / density provides the pressure.

Wedge joints on the top, bottom and sides also serve to add structural integrity to aquariums constructed using 2 part cement. But thats another post LOL... Once again sorry for the long winded reply. Hope this helps.

 

[Floyd R Turbo]

Phixer, great info. How long have you been building big tanks like this?

 

[SuperNemo]

Phixer, you make a good point on the issue of pushing two panels together to compensate for shrinkage. Your idea of overfilling and monitoring would probably be better than having to maneuver the sheets after filling with 42.

When do you plan of constructing the tank? Please post lots of pics. Thanks in advance.

 

[Phixer]

Thanks you guys. I normally build a tank this size about every two to three years but dont really document the builds (not really my thing). I figure since most folks deal with more reasonable sized tanks there isnt much interest in these. Most are done by the big boys like ATM, Reynolds or AquaArt for those with deep pockets.

As soon as I get caught up, I'll start back to work on this one. Might be awhile though, but if you guys think there would be an interest in documenting a large all acrylic build using the 2 part stuff I'll do it.

I used to work with a lot of acrylic and composite materials in the Navy, after 30 years you get exposed to everything from deep submergence vessels, to aircraft canopies. Funny though, now that you mention it I've actually constructed more of these monstrosities than the normal sized tanks. Probably why you guys know so much more about the solvent cementing process than me.

Hoping to provide some information in this method so it moves our hobby forward. This thread has been really good for that.

 

[SuperNemo]

Thanks you guys. I normally build a tank this size about every two to three years but dont really document the builds (not really my thing). I figure since most folks deal with more reasonable sized tanks there isnt much interest in these. Most are done by the big boys like ATM, Reynolds or AquaArt for those with deep pockets.

As soon as I get caught up, I'll start back to work on this one. Might be awhile though, but if you guys think there would be an interest in documenting a large all acrylic build using the 2 part stuff I'll do it.

I used to work with a lot of acrylic and composite materials in the Navy, after 30 years you get exposed to everything from deep submergence vessels, to aircraft canopies. Funny though, now that you mention it I've actually constructed more of these monstrosities than the normal sized tanks. Probably why you guys know so much more about the solvent cementing process than me.

Hoping to provide some information in this method so it moves our hobby forward. This thread has been really good for that.

Documenting a large build is always interesting to read.

I'm a long time lurker, but I think the main reason most people do weldon3/4/16 is because it is readily accessible. Solvent welding requires a special gun if you want to do 42 and mixing with 40, so most people shy away from that. Most tanks are about 8ft long, so no need to bond two sheets together sided by side.

As for the corner and bottom/top seams, are you going to be using 42 as well. Don't you need to have the pieces annealed to make the bonds stronger. This part I am not too familiar with, still researching this. Would you be able to shed some light into this process?

 

[Phixer]

Yes, all joints cast with 42.

Annealing acrylic is basically the same as heat treating metal. Worth noting: as the material is machined stress is imparted into it by the cutting tool, (this stress can be insignificant or major depending on many factors). On a molecular level the material gets compressed which forces the molecules of the acrylic closer together creating sections of the machined surface that are more dense.

The material wants to return to it's original state. If the joint is cast prior to relieving, this stress gets encapsulated with no where to go and will manifest itself as a series of small fractures or crazing as the material tries to return to it's original state. This is the reason for annealing. Annealing relaxes the material thru the application of heat and allows the molecules to return to their original "relaxed" position.

IMHE unless the bit is gouging the crap out of the material i.e. scooping tiny chunks out it, (or the material is overheated due to the wrong feed rate or bit speed) very minor stress is imparted into the base material. Most of the panels on public tanks are too large to fit into an oven prior to assembling on site and therefore do not get annealed.

I plan on giving the edges a minor hand scraping to remove a few thousands of material prior to bonding. Some I have spoke with will wipe the machined edge with denatured alcohol, (yes I know) and if any crazing manifests they will hand scrape it down about .020 to remove any stress in the material from the machining. Then wipe again until no crazing occurs. Once that happens, they glue it up. I questioned this but over the years found several fabricators who do this to check a freshly machined surface for internal stress. Normally most of the time less than .050 needs to be removed if any at all.

The strength of the joint can be increased by raising the temp during curing, but this comes with it's own issues. Was thinking about curing under heat lamps. There's a chart out there somewhere for this, a joint cast at room temp is strong enough for most applications.

UV cured cements are also interesting.