Cracking of wood due to strong humidity deviations – need help

[Drazen]

Hello,

 

I have a big problem with cracking of wood on a nice model. Hope, somebody can help.

Since few years, I am building a larger ship model of “DE 7 PROVINCIEN” in 1:45. The hull alone has a length of 106cm (0verall, it is 136cm), Width of the hull 28cm. So, it is a bigger model. The main problem is that the room where I am building has strong humidity deviations between summer and winter. Unfortunately, I cannot move anywhere else into better conditions for the time being. I am measuring with a professional humidity measuring device (TESTO) in winter constantly below 30%, sometimes below 25% and in summer higher than 75%, even 80% humidity. It is a cellar, but of a modern house – not the typical old cellar of old houses. I have no mould, and the working atmosphere is very pleasant. Still, it is very difficult to regulate the humidity in this place. In my flat, which is just half floor higher, but above the ground, well isolated, the humidity level is normal - measured from 42% in winter up to 58% in summer.

 

The construction of my model:

The ship model is built of linden/lime wood, the planks are mostly of pear wood and decks of maple.

As you can see on the photos. Here are several phases of the built. More detailed steps, you can see here:

https://modelshipworld.com/index.php?/topic/911-de-zeven-provinci%C3%ABn-by-dra%C5%BEen-cari%C4%87-scale-1-45-1665/&

 

Lime wood as base. Unfortunately, I did not set the base in both directions of grain but was positioning the grain of the base, so they have less deviation in the longitudinal direction. At the beginning, I was not expecting such problems. Today, I would set the base in both directions, and protect (seal) with some layer additionally.

The wood base (lime wood) was initially dry, below 20% humidity.

[Drazen]

I am getting cracks on the decks and this is disturbing on a really nice model. Here are some photos.

 

There were ideas to reinforce the model with screws, but this is nearly not possible to set everywhere and especially not in this stage of work.

 

Does anybody have an idea about how to solve the problem and how to get these cracks away? I have a feeling, this is not a process which is finished, but is going on and on. The problem is getting bigger from time to time.

How to stop the cracking?

How to remove these cracks?

 

Making several planks on decks would be possible, but will not solve the problem on other areas (hull below water, or the like.

 

Looking forward to your hints!

 

Drazen

[druxey]

Wood will move with humidity change, whether you try to stop it with glue, nails or screws. If the variation is large, then you need, as Y.T. suggests, try to lessen this by either humidifying or dehumidifying to even things out. Traditional cabinetmakers allowed for this by using 'floating' panels and joints that move with seasonal change. Even at model sizes wood will move, particularly across the grain, so use wood that had sat long enough to acclimatize in your local environment as well as control conditions.

[Hubac's Historian]

Hello Drazen,

 

I’m sorry that you are having this problem.  I’m going to go back and have a look back at the early stages of the build.  I will come back, after I’ve had a chance to process a few things.

[Keith Black]

15 minutes ago, Hubac's Historian said:

I’m sorry that you are having this problem.

 Second that!

 It's beautiful work that you've done and to have this happen to you is a shame. I hope you find a solution in quick order and can recover without great difficulties. I'm going to follow your build, the very best to you in your efforts.......Keith

[Hubac's Historian]

A few questions for clarification:

 

The bulkhead formers and keel plate are plywood?  

 

Lime is your filling wood, which initially had a moisture content below 20% - what has the average humidity in your cellar been since this problem first surfaced?

 

[Hubac's Historian]

And, also, what is the relative humidity now?

[mtaylor]

My solution is a large tub of water in the shop all year around.   But... my shop has air conditioning for summer and dry heat (electric baseboard) for winter.

 

You might consider a de-humidifier for summer and try the tub of water first before investing in a humidifier for winter..

[Bob Cleek]

I can’t be sure from the photos whether your decks  are made of wide pieces of grown wood (not plywood.) If they are, cracking as you’ve pictured is a common problem I’ve seen when restoring models. Most all wood species expand and contract most across the grain as the ambient humidity fluctuates. When a thin, wide piece of wood is fastened down and then moves, it’s highly prone to checking. It simply doesn’t have the strength to resist cracking when it is thin. A one Inch thick piece wouldn’t crack, but a eighth inch thick one will. 

 

Sometimes it is possible to fill the crack, but this often looks unacceptable and if the filler is hard, it will likely crack even more when the wood swells again. The only real solution is to plank the deck with scale planks which will allow for movement between the planks without cracking. Usually, smaller pieces tend to move less than larger pieces. I also find it a good practice to seal all wood with a thin coat of shellac. Shellac is one of the most effective moisture barrier coatings around. It won’t prevent the effects of humidity changes completely, but it slows them down considerably which reduces the amount of movement. 

 

[vossiewulf]

As others are saying above, cracks appearing in wood objects is always a moisture problem, either the humidity where it is stored is varying over a wide range or pieces of wood used in construction were not fully seasoned and stable. 

 

If it's a humidity problem, as others noted a humidifier is the solution. If it's because not fully seasoned (dried) wood was used, there wouldn't be much you could do other than rip that wood out.

 

And sorry you're having this problem, that's a sad thing to happen to a pretty model.

[Drazen]

Thank you guys !

 

Your advices keep me focussing on the best solution: to buy a house and set the working room in the living area and well isolated. No joke, I anyhow will need new place for this big model (the heigh is also significant - without stand: 127cm.

🙂

 

In the meantime, let' see...

 

OK, but, let's answer some of your questions...

[Drazen]

18 hours ago, Hubac's Historian said:

A few questions for clarification:

 

The bulkhead formers and keel plate are plywood?  

 

Lime is your filling wood, which initially had a moisture content below 20% - what has the average humidity in your cellar been since this problem first surfaced?

 

Yes, you are right: bulkheads of plywood (they started to crack first), lime as filling material with initial moisture approx 20-25% measured.

Lime filling is set that the swelling goes in the width of the ship, not the length. See please photos below.

 

My cellar has:

- about 8m²

- in winter constantly below 30%, sometimes below 25% (yesterday measured 20% - this is dry like a gunpowder 🙂  )

- in summer constantly higher than 75%, even 80% or 85% humidity

- further problem is that this is a part of the cellar-complex in a bigger modern block of flats. I could partly separate this room by closing nearly all openings with foam. Still, If I heat or climatize in this room, it is soon away since not well isolated. Thus it is separated by concrete walls, what is good.

 

So, to answer your question, no constant humidity in my cellar, but strong variations within a year cycle.

 

The temperature is more or less constant: summer 19°C till winter lowest 17°C. The deviation is fine, but the temperature is low and supports the humidity deviations. Having constantly e.g. 24°C would be better.

Drazen

[Drazen]

Edited January 27, 2019 by Drazen
cannot upload files

[Drazen]

I don't know why, but I cannot upload photos any more??

 

What I get is: " There was a problem processing the uploaded file. -200 "

[vossiewulf]

26 minutes ago, Drazen said:

 

- in winter constantly below 30%, sometimes below 25% (yesterday measured 20% - this is dry like a gunpowder 🙂  )

 

- in summer constantly higher than 75%, even 80% or 85% humidity

That's going to put serious stress on just about any wood object, please don't leave any good musical instruments in your cellar, either.

 

You don't need to close off everything, just get a humidifier with an automatic setting so it turns on whenever the humidity drops below a set point. With your summer humidity, I'd set it to 50%, few things have a problem with a 20-25% humidity change over the year. I had one at a previous house that I bought on Amazon for about $150. I had to fill it with water every day which was a bit of a pain, but it was better than damage to my wood stuff.

[Hubac's Historian]

Okay, so it seems as though the degree of tangential shrinkage of your lime wood substrate is the culprit here.  The cracking appears most prominent, along the centerline of the ship, where the lime is pulling away from the central keel former and causing the plywood, there, to split along its laminations.  Your planking strakes show irregular gapping and cracking because they are alternately fixed to the shrinking lime and the stable plywood bulkheads.  So, that’s a description of the problem.

 

It’s a bit surprising that you are seeing this degree of shrinkage, given how dry the lime was when you first glued it in.  You mention that, at one point in your early build, there was some flooding in the basement that damaged the work surface, but not the model.  Did the model get even a little wet?

 

Anyway, as for fixing the problem - I agree with the advice of Vossie and others that you are going to need to regulate the humidity, to the degree that the cracks close up again.

 

Personally, I would remove the cracked planks that will never be sufficiently “repaired,” so that I could get a better read on the substrate to know when the cracks had truly closed.

 

Once you have done so, and stabilized the hull, I would think about injecting some thin cyano (or perhaps there’s a better adhesive for the application?) into the crack faults - particularly along the central bulkhead former.  You might also consider marking the path of the cracks with pencil, in the event that they close so completely, that you can’t reliably find them for your injections.

 

Once you are satisfied that you have stabilized movement, I think it would be safe to go ahead and make repairs.

 

I feel for you, Drazen.  This is one of my favorite models on MSW.  I think it is fixable, just a bit of a bear to set things to right, again.

Edited January 27, 2019 by Hubac's Historian

[druxey]

With that range of humidity, you would need both humidifier and dehumidifier, I'm afraid. It's a difficult situation you have in your cellar.

[Drazen]

I am working on this problem right now. Out of the help received in the MSW, I decided to buy a humidifier for winter and a dehumidifier for summer. I am right now testing the humidifier. It works fine (an ultrasonic device),  but has some problem with measuring humidity and thus steering the humidity on 50%. I will report in the other link about this appliance and how much it helps.

 

This cracking is older, but is getting worse with the time. I just hope, it will go back somehow if I manage to keep the humidity between 40% and 60%. Till now, I did not do much effort on adjusting the humidity in my cellar.

 

Drazen

[vossiewulf]

Sounds like you're on the road to recovery Drazen, congrats. Unfortunately although after a couple weeks of more humidity chances are the cracking will stop, it won't fix the damage that has been done.

[Drazen]

1 hour ago, vossiewulf said:

Sounds like you're on the road to recovery Drazen, congrats. Unfortunately although after a couple weeks of more humidity chances are the cracking will stop, it won't fix the damage that has been done.

What a pity...

 

I am wondering if this effect occurred with expanding or drying the wood... ??

 

Drazen

[vossiewulf]

A piece of wood acts with respect to humidity like a balloon with a skin that doesn't stretch well. As humidity goes higher, internal compressive pressures build because every cell in the wood structure is trying to get bigger, with the vast majority of these stresses oriented in the radial direction of the tree rings. When humidity drops, the process reverses and internal tension stresses build until humidity stops changing and the wood reaches a steady state again. Now wood doesn't handle that well in the long run even if it's totally free to move. However if you add glue someplace that is constraining the wood from moving in the direction it wants to go, the failure process is much faster and usually much worse.

 

With your ship, I see what looks like both compression and tension failures, like it sat in place in the cellar for most of a year and went through both cycles.

[lehmann]

Drazen,

 

By the pictures of the deck planks it looks like you used flat grain boards (maple).  The effect of moisture on wood shrinkage/expansion in the tangential direction is about twice that of the radial direction.  In other words, edge grain boards will move only half as much.   That is why wood floors and wide door sills are usually edge grain (that, and the wear resistance is higher).   

 

A few numbers:

Relative humidity        Equilibrium Moisture content

      20%                                     5%

       70%                                    13%

                         Difference          8%

 

Flat grain maple shrinks 11 % from  green (28% to oven-dry (0%).    So, your flat grain planks, over a width of 250 mm will want to shrink

       (250 mm) x (11%) x(8%)/(28%)  = 7.9 mm   

 

The shrinkage for edge grain maple is only about 4.5%, which results in a width change of 3.2 mm.  Either, way, since the plywood bulkheads/frames will have negligible shrinkage, a lot of stress will be generated.  

 

In looking at the shrinkage data for various species, I can see why pine was often used for decking.  While is is much softer than oak, the edge grain shrinkage is Eastern White Pine (North America)  is only 2.1%.   As is teak, at the other end of the price scale.

 

My second point, is that the cracks in the picture seem to be near transitions in how the deck is supported.  In other words, near openings.  The deck parallel to an opening can shrink with little constraint, but the deck at the ends of the opening is probably attached to a bulkhead and can not move.  The difference in constraint creates the stresses and cracks you are seeing.  I also suspect that since maple is so strong it does not crush like a softer wood, which would have relieved the stresses before a crack formed.

 

So how to fix, or avoid in other large models?   

1. Edge grain decking is an obvious must do.  
2. Use a wood with a low radial shrinkage.
3. Avoid gluing the edges of the boards together so that the shrinkage can occur at each join rather than finding one place to fail.  Either no glue, or a glue that never hardens, so that it acts like tar - fills the gap, but does not restrain movement.
4. Many narrow planks means the gaps that appear will be smaller, assuming the gaps are uniformly distribute across the width of the deck.
5. Install the boards when they are bone dry (dry in an oven, or a small box with a light bulb inside (not LED light!)).  That way, even at 20% humidity, the boards will swell.  Full size ships rely on the planking swelling when the ship is put in the water to achieve water tight seams.   However, I would firmly attach the water channels to ensure that railings are not pushed out when the deck expands.

Last thought:  I wonder if Admiralty models at 1:48 scale are often left half planked to avoid these issues.

 

 

 

[lehmann]

18 minutes ago, Y.T. said:

sorry. I have issues with understanding these terms.

whst is flat grain boards?

what is tangential direction?

what is radial direction?

what is edge grain boards?

thanks

I'll send you to the Wood Handbook.  Wood Handbook CH 4 - Effect of Moisture  See page 4.5

Figure 4.3  shows how wood from various places in the tree shrinks and warps 

 

Tangential - parallel to the grain rings

Radial - 90 degrees to the grain rings or in a line from the center of the tree

Flat grain (or flat sawn) - the grain rings are somewhat parallel the wide side of the board

Edge grain - the grain rings are parallel to the narrow side of the board 

Another term you'll see for edge grain is quarter-sawn.  

 

I hope that helps

 

 

[Drazen]

For me as a guy who studied mechanical engineering, these terms are fine. Still, thank you a lot for sharing this book chapter. I will go through it thoroughly.

I see now, that this topic is educative for many and for me especially.

 

The question is for my case now:

1.  What to do to stop the process on the top of keeping the moisture between e.g. 40% and 60%?
2. After I have solved the first problem: what to do with the damages I have got?

@1.

As you can see on the basic construction of the hull and later on the ship how it looks now (please, try to visit previous stages of building process), the hull is opened on many places (gunports, gratings, but also the planking itself does not seal 100% since there are fine cracks everywhere). My idea was to seal the inner part of the hull. It does not help to seal the decks and planks when the humidity has the chance to "free exchange" through these openings.

Well, an idea is to shoot a big amount of shellac through the gunports by closing at the same time the other gunports in order not to get the spray out where I do not want. I have several very good airbrush guns and a good compressor and this may help sealing the ship from inside. Through gratings, it may not be possible, since the area below is closed and painted dull black.

Maybe this is one of the possibilities to lower the stress and minimise the moving process since the "entrance areas" would be much smaller. Still, I am not sure how good an airbrush mist would reach areas like inner corners, but many areas will be shot and covered.

 

Inserting long screws through the gunports would be nearly not possible and also would not solve the problem. It may hold only in one direction. This would not be enough.

 

@2.

One possibility is to do some planks again. I was hoping the cracks will go beck with increasing of moisture, but, as it had been said here. this is an irreversible process. Previous to this, I may pour some thin epoxy into the gaps to close them and additionally plank on this solid base.

 

 

If you have any ideas, please, be free to post them.

 

Drazen

[vossiewulf]

Drazen, a very good rule of thumb when working with wood is whatever you do to one side of a table/board/wooden ship, you do exactly the same thing to the other side. If you want really stable furniture for example, when machining down from boards it's good practice to constantly flip board faces when you get to the planer stage so that the same amount of wood is removed from each side of each board. If you live in high or low humidity conditions and you don't do that, you'll come in the next day to find all sorts of sweeps and winding and checks in your boards.

 

So yes sealing the inside to the same extent that the outside is sealed is a good idea.

 

Once you establish a controlled humidity you should give the ship at least a few weeks to equilibrate to the new moisture baseline. Guitar makers will frequently leave guitar necks for months in the shop before using them in a guitar to ensure they're fully stable within their environment. Ship models that aren't solid hulls will become stable much faster because of greater relative surface area, but you still need to give it some time. And remember it's been under enough stress to break things.

 

With respect to repairs, you seem to know what you're doing in building ship models, so let your own judgment be your guide. Only thing I will say is if you are in doubt about any piece, rip it off and replace it. That's likely to be the smart thing to do.

[paulsutcliffe]

Expansion can then become just as big a problem though as the shrinkage, my cheerful hull below after a year in the garage the wood is holly and was dry and seasoned but has obviously swollen after fitting and painting with watered down acrylic

Regards

Paul

[druxey]

Unfortunately you probably made things worse by the water content of the acrylic paint that you used!

[paulsutcliffe]

I have liked that but don't if you know what I mean

[Katoom]

2 hours ago, paulsutcliffe said:

my cheerful hull below after a year in the garage the wood is holly and was dry and seasoned but has obviously swollen after fitting and painting with watered down acrylic 

 

  Ah nuts, that is really disappointing to see the damage after all of your efforts to produce a nice hull. I have no doubt that you can sort it when the time comes but still....

[Bob Cleek]

3 hours ago, paulsutcliffe said:

Expansion can then become just as big a problem though as the shrinkage, my cheerful hull below after a year in the garage the wood is holly and was dry and seasoned but has obviously swollen after fitting and painting with watered down acrylic

You betcha! I avoid water-based acrylics as much as possible for exactly that reason, even though I seal my wood with white shellac before painting. Sometimes I cheat and thin them with alcohol, if that works on a particular brand. Airbrushing with alcohol solvent allows the solvent to evaporate quickly. Trying to airbrush paint thinned with water results in a wet mess every time. Oil-based, solvent-thinned "modeling" paints are getting harder to find, although the fairly new acetone-based Truecolor acrylics, claiming to be equivalent to the old Floquil, are getting great reviews. Artists' oils (Grumbacher, Windsor and Newton) are fine if you don't mind conditioning your own with acetone, turpentine, Japan drier, Penetrol, and linseed oil, raw or "boiled," provided the pigment is ground finely enough for modeling scale purposes.

 

If it's any consolation, Paul, the seams on your model look quite like the seams on many newly built carvel planked boats I've seen. In fact, in full-size construction, it is expected that the swelling of the planks after launching will cause the stopping in the seams to bulge outwards exactly how the putty or filler you apparently placed between the seams on your model has done. On full-sized boats, the bulging stopping is sanded fair at the next haul out and paint job looks to be doing. This phenomenon can continue with a new boat for two or three haul out cycles before the planking "settles down," or stabilizes somewhat. Of course, if a boat is on the hard for a long period of time and dries out significantly, the problem will usually arise again upon the following launch.

 

The same problems occur in building models and full-sized wooden boats. Wood is a dynamic medium, always moving, to one degree or another, depending upon temperature and humidity changes. The degree of that movement depends also upon the species of wood and the grain orientation of the wood. In the case of the model in this thread, Lime ("Basswood" in the US) shrinks 6.6% radially and 9.3% tangentially, Pear shrinks 3.9% radially and 11.3% tangentially, and Maple shrinks 4.8% radially and 9.9% tangentially. These percentages are for drying from a live cut. Once fully dried, the "movement" factor due to changes in relative humidity may vary from "live cut" shrinkage, and is referred to as the species' "stability."  Stability will generally be a lower percentage of movement than "drying" from a live cut and, as said, will vary from species to species. The ratio of movement radially and tangentially after initial drying will likely be proportional to the stability of a species.  The Wood Database is an excellent resource. A handy hyperlinked version of their print book can be found at https://www.wood-database.com/basswood/ 

 

It's important to remember that the actual distance of movement multiplies relative to the size of the piece. Thus, a one quarter-inch wide plank might move +/- 4% with wide fluctuations in humidity, which would be nearly negligible over the width of a quarter-inch plank, but if, say, sixteen quarter-inch planks are butted against each other, the total movement of the four inch width of quarter-inch planking will move in total almost an eighth of an inch, which, if it's shrinking, is a lot more than the elasticity of the wood is going to endure without splitting, and if it is expanding, will be enough to induce stresses in the structure to which the planks are attached to cause catastrophic failures. In full-sized vessels, transversely cracked frames are a frequent result of planking expansion, often caused or exacerbated by driving caulking too hard or filling seams with inflexible, hardened stopping or splines of harder wood species. Regardless of species, wooden boats are always moving structures, some more than others for a variety of reasons. Some, the Vikings' longboats as a classic example, are even intentionally engineered to flex in operation. Models are different only in a matter of scale. What doesn't "scale," however, are the laws of hydraulics. The amount of pressure created by the water in the wood doesn't vary according to the size of the wood. The water in a tiny piece of wood expanding isn't going to compress any more than the water in a large piece, even if the amount of expansion is proportional to the size of the piece.

 

Sealing wood by coating or impregnating it with some water-resistant material will slow, but not stop, movement because it retards the absorption and evaporation of moisture in the wood. Much as many have tried, there is nothing found as yet that totally "encapsulates" wood. Indeed, the unintended consequence of limiting wood's ability to "breathe" in the marine environment can often create a more favorable environment for decay fungi. Balsa-cored fiberglass decks were once thought to be an innovative solution to achieving lightweight panel stiffness until it was discovered that "water will always find a way" and moisture entering through poorly sealed fastener holes and such soon caused the balsa cores to decompose to rotten mush. As models won't be living in the water, the problem is less severe. One of the best sealers is also one of the oldest. A coating of white shellac is an excellent sealer for model parts. Coating planks with white shellac ("white" because it's the colorless type versus "orange" shellac, which is unbleached.)  It won't prevent the wood from moving ever, but it will slow down the absorption and evaporation cycles causes by ambient humidity considerably. The fewer movement cycles, the less stress is placed on the structure over time.

 

I'm sure a lot of experienced modelers already know this. Those who have experience in full-size wooden boatbuilding certainly are familiar with these principles. These comments are offered for those who are in their modeling "learning curve" where others of us were decades ago. As for myself, most of what I know about this was learned from experience, A mistake is a lesson that isn't soon forgotten.