Dissecting a 6 year old plywood tank (why it failed)

[rotccapt]

Hello all 6 years ago I designed and built a 130 gallon plywood tank with three windows. If you want to see the build it can be found here.

The tank measured roughly 24x48x24 and I used fiberglass and epoxy to seal it. The tank was designed with a 3-pipe overflow system with an internal overflow draining to a 55-gallon sump. overall, I was very happy with the build and it ran relatively problem free for its whole life. a few years ago, I decided to put some pathos in the tank to help with nitrates and although it did work it eventually caused a problem because I placed the plants inside the overflow box. After a while the roots grew through the silicone that held the box to the tank the main body of the tank started leaking into the overflow box. I then built a new overflow box and screwed it into the back of the tank. This was the beginning of the end for the tank. The original overflow was also screwed in which is the cause of the failure. after removing the old box, I plugged the screw holes with silicone and went on to install the new one.

luckily, I caught the imminent failure before it happened. During a water change the other day I noticed that there was some mold on top of my water pump return line. after further investigation I found that the whole back of the tank was bowed out about 2 to 3 inches and water was seeping from the bottom seam of the plywood. I decided at that point that the tank was beyond repair and promptly drained it and moved the inhabitants to temporary living quarters.

Today I started to disassemble the tank with hopes of saving the 3/8 glass that I used for the windows. I also wanted to see where the failure was so that I can account for it on the rebuild. I started the dissection with the bottom of the tank it did come off but it put up a fight, the plywood actually separated from the fiberglass in the process. I then went on to pull the back of the tank off, the plywood was completely waterlogged and came off without any trouble. Once the back came off, I discover the smoking gun of sorts. It turns out when I did the new overflow a couple of the old screw holes did not get fully sealed and it allowed water to seep past the fiberglass and compromise the wood of the tank. When I build the tank, I completely sealed the back wall inside and out so the water was able to stay inside and destroy the wood.

So long story short I have some take away from the experience.

First off, had it not been for the fiberglass holding back the water, this would have been a giant mess that would have ended with 150 gallons of water on my floor.

Second, I need to find a way to install the overflow without using screws. If I had done that, this problem would not have happened

Last, I need to perform a closer inspection of the tank to try and catch problems before they become catastrophes.

I hope you all can learn something from my failure, I love my plywood tanks and have built three now and currently have a 300-gallon running as well. Every time I build one, I learn from the last build. As they say you learn more from your mistakes that from your success, so here is to the next mistake.



Keep building.

 

[Ulu]

It's a wonder you caught it in time. This is a common problem with old boats. Someone drills a hole, and puts a screw in with sealant, but it eventually seeps water. The reinforcing wood swells and bows.

I have a fiberglass boat which had that problem. The transom was cored with thick western red cedar plywood. It took decades to rot but it did, right at the bottom, which caused the transom to droop. This was 34 year old but when I cut the transom open you could smell the aromatic cedar like a new hope chest.

I rebuilt it with steel tubing epoxied into the transom and a plastic infusion treatment of the redwood. I opened the flotation chambers, and removed the rotten bottom 2" of plywood, then rebuilt it with fiberglass and foam.

 

[celebrist]

Glad you dodged that bullet

 

[phreeflow]

Thanks for the write up. Good info

 

[the_deeb]

Thanks for sharing. Glad you caught it before it became a complete disaster.

If using screws in a plywood tank (from the inside), I would suggest first drilling a deep countersink hole . Dip the screws threads in epoxy before driving them in and then, after they’re in place, completely back fill the remaining space in the countersunk hole with epoxy so the screw head is completely sealed off from the water.

Similarly, if you need to fill old screw holes, don’t use silicone. First, drill out the old hole to a uniform diameter. Then, use a q-tip to cover the sides of the hole with epoxy so it soaks in and saturated the open end grain. Then take a wood dowel the same diameter as the hole and trim it down to slightly shorter than thickness of the plywood. Dip it in epoxy and insert it into the hole, and fill all the remaining spaces with epoxy.

Alternatively, best to avoid internal screws entirely. If attaching an external overflow, you can use pocket hole screws to drive them in from the outside so there’s no possibility of contact with the water.

 

[fishdance]

Thank you for the write up and advice. I decided to leave the outer wall of a concrete aquarium unsealed to make it easier to detect seepage. And I placed bare speaker wire and LED light on a direct current incomplete circuit at the bottom. Any moisture will complete the circuit in theory to provide early warning.

I was thinking you could fiberglass over your screws to stop water seepage? Have you considered using HDPE sheets (on steel or wood frame) next time. You can plastic weld over any screws/penetrations and it will never rot.

 

[Ulu]

The problem with seepage and weepage is its inevitability under some circumstances, like concrete things with construction joints.

For this reason they build seepage control on large concrete structures. Even brick and stucco houses have weepage control.

If I was going to build a large aquarium I would definitely have controlled seepage from construction joints and control joints to a location where you could monitor and drain it.

 

[Ulu]

R rotccapt
Thanks for posting the forensic analysis and photographs. I've been through a few of these for other people over the years.

Not aquariums, but other structures like stadiums and basements and steeples and balconies. How and why are they failing, how are we going to repair it, or remove it, remediate it, or eliminate it in the future.

This is a big business, but being retired I don't see much of it anymore.

 

[rotccapt]

On the next tank I plan on building the entire overflow out of chopped fiberglass. This way I can attach it to the tank and fiberglass it in. This will allow a complete waterproof joint with no screws.

On another note, I discovered that the rubber sealing washers for my bulkhead fittings were also completely dried out. this may have also led to the failure