Get out your jinashi root-end D flute. It probably weighs somewhere around 300 grams and about 8-10% of that is water. There's nearly an ounce of water in your flute right now. When the bamboo was green it's moisture content was 100% or higher, so 250 grams of water had to evaporate to get it to the state it's in now. Back in the day, your culm was green, plump, happy, and unstressed, but nearly 9 ounces of water evaporated and your flute has been stressed ever since. That's why it has already or may yet crack--the missing water. If the flute were submerged for awhile, it would swell back up, any cracks would disappear and it's weight would tend toward the original 550 grams and all stress would be gone.

Let's get a grip on how much water we're talking about. The volume of the bore is somewhere around 150 cubic centimeters. One cc of water weighs one gram, your flute is missing 250 grams of water, more than enough water to fill the bore--nearly twice. Now you know why flutes crack.

Moisture content (MC) is a little bit like talking about exerting 110% in terms of effort/commitment/etc. To calculate MC, divide the moisture weight by the dry wood weight and multiply by 100 ( (M/W) x 100 ). So green bamboo often has a moisture content well over 100%.

Moisture in bamboo is divided into two parts: free water and bound water. Think of a wet sponge. You can squeeze out the free water but the sponge remains moist with the bound water. Something called the fiber saturation point (FSP) identifies the point between free and bound water, which in bamboo runs around 20%.

As bamboo dries the free water leaves first and the MC drops to 20%, at which point the bound water begins evaporating. An important point to remember is that dimensional changes (shrinking/swelling/cracking/etc.) only occur as bound water is lost or gained. Placing dried bamboo in 100% humidity indefinitely will only raise the MC to 20%. Once the bound water is replenished bamboo stops absorbing moisture from the air. To replenish free water the culm needs to be submerged. Bound water is held in the cellular walls in molecular form by forces of attraction, free water fills the cell cavities. Bound water is directly involved in the structure of bamboo, free water is just along for the ride.

The hydration graph below plots bound water versus humidity. All the moisture action you need be concerned with (in regards to a bamboo flute cracking) is contained in the graph--it's all about the bound water. To be clear, the graph (below 20% MC) only becomes operational after all free water has left the culm. Shrinkage doesn't begin until bound water starts evaporating.

The graph indicates the moisture content of your flute probably never rises much above 10-12%. Note also that varying the relative humidity surrounding your flute from 40% to 60% results is just a few percent change in the moisture content of the flute. 50% humidity (6% MC) is usually considered the minimum to remain crack free. Bamboo fishing rod makers need to keep the MC above a certain level in order for the glue to set and work right. Otherwise, the fishing rods start coming apart latter. For those particularly interested in the MC of their flutes, weighing a flute is probably the simplest and most accurate way to determine water gain and loss.

Here are two links with current US relative humidity: National Humidity and Regional Map.
Check your location and get a number for relative humidity
apply it to the graph above and derive the Bamboo Moisture Content that can be sustained.

---

Throughout the long history of the craft of Japanese flute-making one fact has remained true: bamboo is a poor material choice for flutes. Bamboo can't stand up to the demands placed on it by the rigors of flute playing, the material is too fragile and it cracks. Being a very linear material, only in a single direction is it strong and resilient. Radially (across grain), bamboo is virtually without merit. Which is a long way of saying it cracks.

At what MC a particular piece of bamboo may crack is dependent on many factors, chief among them, age/development of the wood structure and wall thickness--thicker walls being more prone to cracking. But for many flutes there is some point of moisture content where the stress becomes too great and the wall ruptures. If the MC of your flute is under 20% it is experiencing some degree of stress.

Samples are all from the same piece of bamboo.
They've been split to facilitate quick visual indication of stress.

If you were to band-saw a section from your flute right now and split it, it would spring open in the manner of the left sample above--indicating the amount of stress it was under. This desire to crack is perpetual and will reside in your flute as long as it exists. As can be deduced from the center sample, the stress level in your flute is directly related to its degree of hydration. When a hydrated sample (right) is left to dry the stress returns. We've taken bamboo dried to less than 5%, immersed it until the MC is above 20% and dried again, to demonstrate the absorption/desorpsion cycle. It also becomes obvious that the blond color associated with bamboo is largely the result of empty cells cavities filled with air.

There's a difference between the inner and outer portions of the bamboo wall and more of the vascular bundles (the water/sap carriers) are in the outer wall. Thus, as bamboo dries and the bound water evaporates, the outer wall shrinks proportionally more than the inner, creating stress which, if it becomes great enough, will lead to wall rupture and a crack.

In simple terms, the bamboo cell wall is a composite made of a rigid cellulose polymer in a matrix of lignin and the hemicelluloses. The lignin polymer is thermoplastic; that is, it softens upon heating. The glass transition temperature Tg of the lignin in the matrix is approximately 170C (338F). Above the matrix Tg, it is possible to cause the lignin to undergo thermoplastic flow and, upon cooling, reset in the same or modified configuration. This is the principal behind bending bamboo with heat.

The strength of bamboo is virtually unchanged while free water is leaving the culm. But once the free water is gone and bound water starts evaporating the values for strength, hardness, and stiffness begin rising and culminate at zero moisture content after having risen substantially. Oven dried bamboo is at it strongest with relation to moisture. Surprisingly, even greater strengths can be obtained by lowering the temperature. Dropped well below freezing the strength of bamboo continues to rise.

Speaking of temperature in general, the strength of bamboo falls as the temperature rises--about 5% for every 10F rise. So, placing your flute in a hot car such that the temperature goes from 70 to 100 degrees drops the strength about 15%. If the flute is under stress anyway, elevated temperatures could cause it to crack.