Large Scale Central

TiteBond III vs. The Elements -- A Tropical Sort-of Study

Cliff (@Cliff_Jennings ) had posted a question about using TiteBond III to rot-proof wood on our railroads. This arose from my own questions about preserving and restoring our first scratchbuilt project as part of our currently ongoing Triple O Salvage Project - Pu’uoma’o Yard Office and Observation Tower. I had promised to hang some pictures, but I thought a separate thread might be more discoverable.

To set conditions, I thought I’d define our climate a bit, our railroad sits about 1.5 miles inland from the ocean. Average temperatures range from 71 F in the winter (brrr… :cold_face:) to 85 F ( :nerd_face:) in the summer. Average humidity is about 74%. Winds are around 15 mph, and they typically come from the east. A 5’ fence surrounds the yard, so this does cut the winds/

Our control are some 1/2" x 1/2" basswood timbers we once installed to hide cinderblock tunnel cores. These timbers faced the prevailing winds. Overhanging lava rock partially protected them.

They were not treated with TiteBond III or anything else. Installed in 2015 or 2016, they finally rotted off this spring.

Plants have grown in, blocking any view of the portals…

…so there are no current plans to install new timbers.

During some MIK or another, someone mentioned coating wood in TiteBond III to protect it. We have since “painted” (smeared?) TiteBond III without cutting it with onto any wood we plan to leave outside exposed to the elements. Wood is, again, basswood, maybe some poplar, and popsicle stick tree.

One example, 2021 Mik Challenge – Hale Ipukukui o Haluku’ilio (The Lighthouse of Dogwallow), had a craftstick upper deck and some basswood supports. Here’s what that deck looks like in July 2025:

Note also the rot at the foot of the support beam. The rest of the beam as well as all the other beams, is sound. Replacing this deck, probably with scribed plastic, is on the 2026 goal’s list!

On the other hand, the M&K Sugar Mill has been exposed to the elements since 2020. The core is foam covered in crimped can and craftsticks. As of July 2025, the craftsticks look like this:

I think they’ve weathered quite well, and, even better, there are no softspots on the planks at all. Minor warping has exposed some of the core, but you can barely tell.

So, what might we conclude?

  • TiteBond III can serve as an effective sealant on vertical surfaces exposed to the elements 24/7 in tropical regions.
  • It has little or no protective properties for horizontal surfaces similarly exposed.
  • Inclined surfaces likely have the same protection as vertical surfaces unless they are in direct physical contact with rotten wood.

Hope this helps!

Eric

4 Likes

You better take that 2nd to last picture down before National Geographic sees what you did to that turtle! Either that or find the pieces and get the Titebond III back out.
It’s quite disturbing!!

That is an interesting study Eric. It is surprising how badly the decking has failed in only 4 years, or maybe the turtle ripped it up? :upside_down_face:

In my experience T3 is the best but it will eventually fail. E6000 is another good all around glue that will also fail.
I live in MA and we have temp swings from 0 to 100 over 4 seasons with rain, snow and ice so the glue that holds structures together really takes a beating.

Recently a 14 year old pony truss bridge needed some regluing. I noticed one side was wiggling, the glue had let go and the pin nails were the only thing holding it on. I squeezed glue into the gap and tapped it closed and hopefully it will stay solid for many years to come.

Moisture is the biggest enemy of the glue.
I have coated small projects with T3 particularly board and batten sheds that use stir sticks. It helps but they still eventually warp and pull away.
If you coat your wood do not try and stain it but paint goes right over it.

Thank you for the wonderful report Eric. Some nice hard facts there, very helpful!

Something’s still nagging at my mind, referring to something I’d posted on your prior thread.

FWIW, I’m reading an older book on making structures for British garden railroads, and the author repeatedly says to adhere and then “paint” the structure with slightly diluted PVA glue such as “carpenter’s glue.”

Oddly, doing a search, PVA is widely said to not be good for outdoor use. For example, Titebond says don’t use their PVA (“Speed Set”) outdoors, but use TBIII instead.

FWIW, I think I’m understanding my disconnect here. The author (the famous Peter Jones) gives this description of his preferred glue:

Polyvinyl Acetate (PVA). The most useful glue for our purposes is white PVA, but it is very important to remember that it must be the waterproof variety… The other observation I would make about waterproof PVA is that, although admirable at surviving adverse conditions, it is not entirely waterproof and can dissolve if in direct contact with the outside world – it is prone to go milky and finally dissolve if used as an outer protective layer. It has proved itself time and again to be reliable if shielded by an outer layer of paint.

Bottom line, it appears that waterproof PVA, or perhaps TBIII as well, will last much longer if painted.

Eric, do you have any examples of TBIII-coated, then painted, surfaces?

Cliff,
Can you just let the man work with his kids ?

Unfortunately, no, Cliff. It would be worth incorporating that into our current restoration project. The core will be painted where I am replacing planking. I plan a second story, where I can certainly try paint-on-TBIII. Do note, I’ve no intention to leave the building outside on the railroad. Between sessions, it will come onto the lanai, where it won’t be exposed to rain or the prevailing winds.

Eric

P.S. With regards to the claim…

…I do not remember observing this at all.

Eric