what humidity and how do you keep it?

[dcopley]

I still refuse to believe that a wooden flute reacts near-instantaneously to humidity changes, such that its internal moisture content varies immediately to remain in equilibrium with its surroundings. The wood's simply too dense. I go now in search of answers!

Yes, we have been oversimplifying things by talking about equilibrium conditions. If you leave a flute long enough in constant humidity, it will reach equilibrium, but I'm still not sure how long "long enough" is. The oven test speeds things up because of the increased temperature. The picture of what happens before the flute flute reaches equilibrium is a far more complex problem, and I imagine it would need some sort of computer simulation of the diffusion process to quantify it. Intuitively, you can imagine that if the flute is in a drying atmosphere, the exposed ends will dry most rapidly (if un-sealed) and the mid-section furthest from exposed surfaces will retain moisture the longest.

Meanwhile, back at the torture test mentioned earlier. After 32 hours in the oven, the crack has extended slightly from 38mm to 41 mm long. The crack width at the bottom is around 0.5mm. I was not smart enough to get a width measurement at the start, but it appears to have widened. At 25 hours I put it in a plastic bag with a hygrometer as suggested by Terry. The hygrometer reading stabilized at 35%.

Dave Copley
Loveland, Ohio

[GaryKelly]

Yes, we have been oversimplifying things by talking about equilibrium conditions. If you leave a flute long enough in constant humidity, it will reach equilibrium, but I'm still not sure how long "long enough" is.

Well, I found some disturbing stuff on the 'net, published by one Dr. Tim Padfield, something of an authority in the field of conservation. Disturbing because from what he's written, it would seem that artificially humidifying a flute could be one of the causes of cracking rather than a cure, depending on where you live!

It looks like this chap has done a good deal of research in his career in such places as Conservation Department, Victoria and Albert Museum and the Conservation Analytical Laboratory, Smithsonian Institution.

Terry & Dave, it's fascinating stuff, well worth a dekko, especially page 3, which would seem to suggest I should've left my Bleazey headjoint in the bag for a few months, not days... that's never going to happen for as long as I'm able to the play the thing.

I'd be interested in hearing your opinions!

http://www.padfield.org/tim/cfys/wood/wood1.php

[dcopley]

Well, I found some disturbing stuff on the 'net, published by one Dr. Tim Padfield, something of an authority in the field of conservation. Disturbing because from what he's written, it would seem that artificially humidifying a flute could be one of the causes of cracking rather than a cure, depending on where you live!

Thanks for drawing our attention to this paper - really an excellent summary of the issues we are talking about. I'd be interested to hear how you drew your conclusion about humidifying being a cause of cracks, as I did not get that from the information presented.

I'm particularly interested in the last but one graph, the one showing rate of water desorption by a panel. I've been trying to waffle about the same thing but without having any hard numbers. I plan to contact the author and ask for information about the model used to generate this chart. If we could re-calculate it for African blackwood and other flute materials, we would be a lot further ahead in our understanding. Also it would be nice to know more about what's happening near the surface and in the first few hours and days - something which could easily be done if we had the means to do the calculation.

Dave Copley
Loveland, Ohio

[Terry McGee]

Well, I found some disturbing stuff on the 'net, published by one Dr. Tim Padfield, something of an authority in the field of conservation. Disturbing because from what he's written, it would seem that artificially humidifying a flute could be one of the causes of cracking rather than a cure, depending on where you live!

I don't think it's so much rehumidifying per se, but the speed of rehumidifying (or even more the speed of dehumidifying) that can cause damage. EG, throwing a very dry flute in a barrel of rainwater would rehumidify it very quickly, but involve a lot of stress in the wood. One can almost imagine it delaminating!

The opposite happens during sudden drying - the outside shrinks fast, putting intense pressure on the wood inside. An extreme case of that is what the kiln-drying people call "collapse" - the internal cell structure of the wood is actually crushed by the drying forces, and the piece of wood ends up looking like a cake that rose in the oven, but then fell on the inside.

Fortunately for us, we have a few things on our side. Our timber is very dense so even big changes of humidity can only produce moderate rates of exchange. The timber is strong,a nd can resist extreme forces. The timber is thin, so that there isn't scope to build up a really big moisture differential between surface and middle. And our shape (round) is naturally strong, unlike the big flat portrait he mentions.

Our aim of couse should not be to let the flute get too dry and require gross rehumidification, but to try to keep it within reasonable bounds. Irish maker, Hammy Hamilton, suggests flutes shouldn't be allowed to drop below 60%RH, which is consistent with the figures I presented for 19th century London-made flutes. That would probably be too high a figure for flutes made in Central US - again, ask the maker, and buy the little hygrometer to keep in the case.

The portraits mentioned in the web page are an interesting reminder of the benefits of "balance". The protrait that was painted on one side of the piece of wood warped, while the portrait that had the other side also painted didn't. This is commonly known in furniture making - the inside of veneered craftwood doors are veneered also, often in a cheaper wood, so that they are balanced to humidity change and will not warp.

It's interesting to speculate whether "balance" has any significance to the flute. The head and barrel, being lined, are "unbalanced", the rest of the flute is "balanced" (air and moisture can get to both sides equally). So all change in the head has to come in via the outside - it gets worked harder, so to speak. Is this in any way part of the failure mechanism, or does it pale into insignificance compared to the static forces that build up in very dry weather? Hard to know, but worth keeping in the back of the mind.

Terry

[Terry McGee]

A tidbit more information, hot off the presses. Starting to investigate the effects of cycling - I took a flute that had been drying, let it re-equilibrate to ambient, then dried it again. (ie 20%, 45%, 20% RH) It levelled out at exactly the same weight as when it was first dried, however the minimum diameter was very slightly smaller - .04mm or 0.15%. So no dramatic sign of a non-linear, cumulative effect, but maybe a hint of one. I'd probably need to repeat this many, many times to be sure, but I can't with this one - it's destined to be a flute sometime soon!

So my money is still on straightforward shrinkage of wood versus non-shrinkage of metal simply exceeding the rupture strength of the wood. Dave's experiment will probably reveal whether I'm right. If not, we think again.

Terry

[GaryKelly]

Thanks for drawing our attention to this paper - really an excellent summary of the issues we are talking about. I'd be interested to hear how you drew your conclusion about humidifying being a cause of cracks, as I did not get that from the information presented.

Well, this was a deduction I made from Dr. Padfield's article, page three, where he says:

"Rapid RH change can also causes internal stresses in massive wood with no surface coating. As the graphs show, the penetration of water into wood as the surrounding RH changes is a fairly leisurely affair. Large transient stresses can build up when the outer skin tries to change size against an inner core that has not yet noticed that the RH has changed."

It occurred to me that the following scenario might be perfectly likely: A flute is kept humidified in a tupperware box at say, RH 75%. Spends most of its life in there (while the owner's asleep, and while it's not being played). Say, 22 hours out of 24? Then whipped out of the box and exposed to oh, for argument's sake, RH 20% for two hours before being put back in the box. Or, in a region where natural RH is say 60%, and the flute is being kept in a box humidified to 40%...

That's got to be adding to the stress.

I was interested in the differences in expansion, radially and tangentially, according to how the wood was cut from the tree. It'd be interesting to learn the grain orientation of the stock blank used to produce a flute headjoint.

Can either you or Terry let me know the dimensions of a tuning slide barrel please? Mass (without lining!), length, outer diameter, wall thickness? I'd be much obliged!

The time it took for that flat panel to achieve equilibrium in his charts was certainly in the order I was expecting. 50 days to reach the halfway point.

[andrewK]

There seems to have been any amount of work done on what happens as wood seasons, wettens, & dries, and how it's properties change under drying & humidification. An article at http://www.ces.purdue.edu/extmedia/FNR/FNR-155.html
for example is interesting.
It may be fun, but not necessary to go through the basics all over again.

The sort of splitting I come across I don't think is related to wetting. It relates, I expect. to the shrinkage of wood over a long period whether it is wet or dry. Think of the pegs in old furniture sticking out where the wood has shrunk more one way than the other.

[andrewK]

Have you all tried this one ?
The following formula can be used to estimate the dimensional changes that a piece of wood will undergo due to changes in its moisture content. It is applicable in situations where the moisture content is at or below the fibre saturation point.

Di (MCi - MCf)
Dc = ----------------------
FSP(100)/S - FSP + MCi

where:

Dc = dimensional change
Di = initial dimension
MCi = initial moisture content (percent)
MCf = final moisture content (percent)
FSP = fiber saturation point (average = 28%)
S = shrinkage percentage from green to ovendry
(radial, tangential, longitudinal

It is guaranteed suitable for those who have difficulty sleeping.

[GaryKelly]

Or you could simply use "The Shrinkulator" here: http://www.woodbin.com/calcs/shrinkulator.htm

which'll do all the calculations for you on the fly.

It's interesting to select "African Mahogany" and play around with wildly different levels of RH and observe the dimensional changes. But it's also interesting to note that these things refer to a plank or board. But there are many other factors to consider, the flute being round as Terry says. There's also the question of grain orientation in the stock blank to start with. It'd certainly explain the "going out of round" though.

[andrewK]

Which must magnify the problem a critical amount, I imagine.
Whether the effect of freezing temperatures is low humidity or shrinkage this still seems a major danger. Steamhammer in his arctic wilderness certainly seems to find it so.
The cracking in old flutes I have experienced is nothing to do with humidity as far as I can see, but mere shrinkage ascerbated by pins, ring damage and the like.
Most of the flutes I have around are 100 + years old so are stable, but have all shrunk from new. The oldest ones are only seven inches long now. ( No, I am lying ).
It is heat & cold I fear.
Unready wood would also seem a problem. not all makers have their products cracking ( except in extreme circumstances ), but I am not free to discuss those of my personal experience.

[dcopley]

It occurred to me that the following scenario might be perfectly likely: A flute is kept humidified in a tupperware box at say, RH 75%. Spends most of its life in there (while the owner's asleep, and while it's not being played). Say, 22 hours out of 24? Then whipped out of the box and exposed to oh, for argument's sake, RH 20% for two hours before being put back in the box. Or, in a region where natural RH is say 60%, and the flute is being kept in a box humidified to 40%...

That's got to be adding to the stress.

OK, now I see where you are coming from, and yes, those conditions would affect the stress. Then the question is whether the total stresses exceed the failure point.

I was interested in the differences in expansion, radially and tangentially, according to how the wood was cut from the tree. It'd be interesting to learn the grain orientation of the stock blank used to produce a flute headjoint.

The joints should be parallel to the original grain, so the grain is running straight up and down. Usually the embouchure hole and tone holes are drilled in a radial direction (straight toward the center of the trunk if you imagine the flute piece in the original tree).

Can either you or Terry let me know the dimensions of a tuning slide barrel please? Mass (without lining!), length, outer diameter, wall thickness? I'd be much obliged!

The barrel joint is usually curved on the outside and stepped on the inside (for the socket of the joint to the body). A simplified version would be a tube with 20mm ID, 28mm OD, 70mm long, and weight around 23 gm.

Dave Copley
Loveland, Ohio

[ChrisA]

FWIW,
Size change in wood due to temperature is negligible, however in metals it is significant, so
for temperature related stress look to the metal-wood interface.

Conversely, the size change due to humidity for wood is significant, and for metal nonexistent.

When gaining and loosing moisture, the end grain is the most rapid loss of moisture. The cross-grain is a significant loss of moisture. The parallel (concentric) faces of the wood are the least vulnerable to moisture exchange.

I recommend the book 'Understanding Wood', and I can doublecheck the expansion rates this evening, for some wood one might make a flute out of.

Anyway, leaving facts and entering opinion, I think the reason for keeping the flute in relatively high humidity is because, when you play it, you will -drench- the inside of it. If you keep it in a 20% humidity environment, it will be shedding moisture rapidly until equilibrium, and then when you play it, it will be suddenly receiving vast amounts of moisture. If you're playing daily, I can't think it matters much, you'll be keeping it thoroughly humidified anyway. For a flute that you only bring out once a week though, I think the dramatic shifts in moisture would be very stressful as the wood compresses and decompresses repeatedly. At once a month, it'd be almost as bad as it can get. (Yeah, it takes six months to season wood...
but that's wood in thick slabs, it won't take nearly that long for flutes to reach equilibrium
with the environment)

When making things that you don't drench in spittle, of course, it makes sense to season the wood to the destination's average humidity, but flutes are a special case, I think.

[Steampacket]

"Whether the effect of freezing temperatures is low humidity or shrinkage this still seems a major danger. Steamhammer in his arctic wilderness certainly seems to find it so. " Andrew K.

Yes, Andrew's right. I'm very apprehensive of the low humidity here during the Swedish winter. I have just two blackwood flutes so I try to play them on a regular basis in order to keep them played in, and to keep them (the inside bore at least) from drying out. I oil the outside of the head joints occasionally and the outside of the rest of the flutes. The plastic bag to keep the instruments in, together with a small piece of damp sponge and hygrometer, seems to work for me and there have been no mishaps so far this winter.

[andrewK]

Sorry about you name, Steampacket. At least I got the Steam bit right !
Can you remind us how, when your headjoints went, it actually happened ? It could be instructive.

[Steampacket]

I purchased an 2-3 year old English made blackwood flute in late November 2003 from a seller in New Hampshire, USA. The flute was shipped to Sweden and should have arrived within a week, but went missing and both the seller and I were very worried. I finnally located the flute on December 17th and picked it up at the carrier's depot where it had been stored in a very cold dry storeroom for 10 or so days. For some inexplicable reason they hadn't notified me when the flute arrived in Sweden??? I must add the flute was very professionally packed, and undamaged. Anyway early in February 2004 I took out the flute to play and to my great dismay noticed that there was a 3 cm long, thin, crack in the back of the headjoint opposite the embouchure hole. It didn't affect the playing of the flute, but it was unsettling to say the least. Those of you that have had a fine flute crack will know what I mean. I contacted the maker and he suggested placing the flute inside a container together with a damp sponge to close the crack up. We agreed that I send the headjoint either for repairing or/and a new headjoint. So from then on I kept the flute in a plastic bag with a damp bit of sponge and a hygrometer, the crack remained closed up, you couldn't even see it, and the flute played fine. I then became aware that even with the central heating turned off my flat only had a humidity of around 45-50 which was/is too low. In late August 2004 the maker had a new headjoint ready. I sent off the "old" headjoint to England and the maker was also of the opinion that the old headjoint was much too good to scrap so I have it as well. I must also add that I'm very pleased with the service and help I've had from the maker. So since February 2004 I keep my 2 flutes, a Wilkes & a Williams, in their own humid environment when not being played and it seems to be working. I try and play them both on a daily basis so that the bores don't dry out completely, and I oil the headjoints and outside of the flutes occasionally. Seems to be working, touch blackwood.

In the autumn of 1996 I had bought a Martin Doyle blackwood flute, and in February 1998 I went to play the flute and found a dramatic 1 mm wide crack right through the embouchure hole and running the length of the headjoint and even the barrel. The flute was unplayable. I never played that flute much back then, being a piper, so it had dried out, and then the dryness of winter took it's toll. Bummer as they say.