Does wood finish itself cause wood to expand? Shella/varnish

[Tetsuaiga]

I have an old very thin probably 5mm thick piece of walnut which I French polished to learn on. Over time it cupped a lot, I was wondering, is it the finish which is absorbed and expands the wood or the moisture barrier?

My walnut piece cupped so the finished side is crowned. To me that suggests that the finish itself is responsible for the expansion of wood and not water.

I know you should seal the other side at least when finishing but am curious, I often read wood is hydroscopic, but I presume perhaps it's all liquids the wood changes shape with?

What do you think?
Thanks

 

[dzj]

Was this 5mm piece ripped from thicker stock or have you had it in its current shape for a long time?
Finish usually doesn't affect wood in such a way.
Perhaps in your case, because it is so thin, I can't say. I've never had anything like this happen,
but I suspect it has something to do with the wood reaching EMC, rather than the finish being the culprit.
As for "sealing the other side", this isn't really necessary, particularly with traditional finishes.
Most antique furniture isn't sealed on the inside and has done very well for a couple hundred years.

 

[CHJ]

Dependant on how and where it is stored it could be a case of the unfinished surface loosing moisture quicker than the surface that is sealed to the surrounding atmosphere.

Lay it down on a damp cloth or newspaper (finished side up) and see if it moves.

 

[ED65]

...or the moisture barrier?

Apparently shellac is a brilliant barrier to water vapour so that should be a major part of it.

I think the cupping here is largely because the piece is so thin. That and it's unsupported, in a frame or with battens fixed to the underside it might have held flat.

I know you should seal the other side at least when finishing

That's not always necessary. On thicker wood in particular it appears to make little if any difference. Almost all tabletops are not finished on the bottom and they don't cup as a rule.

 

[Jacob]

Finishes can have a dramatic effect on thin stuff. Anybody who ever made balsa/tissue model aircraft can tell you that wrongly applied the finish can destroy the aeroplane by curling it up. Thin veneer similarly - you might expect it to bend out when applied and then curl in on the varnish side as it dries, along the axis of the grain.
If your piece went the other way it could be that it wasn't dry at first but is now?

 

[Tetsuaiga]

Thanks for the replies. I do have a few other pieces cut from the same peice of walnut, they warped a little also. I think I would generally expect very thin pieces to cup a bit over time so maybe perhaps the finish isn't such a factor.

I might have to try some experiments with the other pieces I cut that were left bare. Its well over a year since I cut them so will be fully dried now.

I found the walnut is actually more like 2-3 mm too.

 

[custard]

Apparently shellac is a brilliant barrier to water vapour

I don't know where you got that from but it's wrong. A thick layer of quality epoxy might be a robust barrier to water vapour, but virtually any finish that you're likely to apply to furniture offers very little long term protection against water vapour. Water spillage that'll be quickly mopped up yes, water vapour no.

 

[ED65]

Apparently shellac is a brilliant barrier to water vapour

I don't know where you got that from but it's wrong.

I didn't mean to imply it's a complete seal like epoxy can be, but it is a good barrier.

All film finishes are inherently some barrier to water vapour by their nature, in contrast to oil finishes which are highly porous. As the thickness of the film is directly proportional to the protection given and since this was a practice piece for French polishing it's safe to assume a fully built coat, which would provide significant protection to that face of the board.

 

[custard]

Apparently shellac is a brilliant barrier to water vapour

I don't know where you got that from but it's wrong.

I didn't mean to imply it's a complete seal like epoxy can be, but it is a good barrier.

All film finishes are inherently some barrier to water vapour by their nature, in contrast to oil finishes which are highly porous. As the thickness of the film is directly proportional to the protection given and since this was a practice piece for French polishing it's safe to assume a fully built coat, which would provide significant protection to that face of the board.

No, it's not "a good barrier". At best it slightly and temporarily slows the transmission of moisture vapour. Think about it, a thick layer of fibreglass resin won't halt the transmission of water vapour which is why fibreglass boats can suffer from osmosis, a layer of shellac that's at most a few tenths of a mill thick has very little effect, doubly so if it's a traditional shellac containing the natural waxes because wax will act to wick the vapour through.

 

[ED65]

That sounds reasonable custard but I'm sorry, there is some easily observed proof that shellac is quite a good barrier to water vapour: its ability to retard rust formation on iron and steel. But if you want something in a wood context, Bob Flexner states this in at least one of his books. I can go look up the exact wording if you'd like but I'm pretty sure he uses the words "good" and "barrier"

 

[custard]

That sounds reasonable custard but I'm sorry, there is some easily observed proof that shellac is quite a good barrier to water vapour: its ability to retard rust formation on iron and steel. But if you want something in a wood context, Bob Flexner states this in at least one of his books. I can go look up the exact wording if you'd like but I'm pretty sure he uses the words "good" and "barrier"

Why would anyone put shellac on metal when it's acidic?

Thanks for offering, I would be interested in the exact Flexner reference, because I'm sceptical that he said shellac offers a good moisture vapour barrier. Shellac provides a terrific barrier against all manner of wood finishing nasties, but water vapour isn't one of them.

 

[AJB Temple]

Hmm. Consider the size of water vapour molecules. Often less than one ten thousandth of a micrometer. You need a pretty effective barrier to for it to be regarded as "good". I have used shellac as a musical instrument finish for a good few years (though I am no expert) and it is an accepted fact that it is not water protective. Water vapour condensing or in contact with a surface is the same as water in effect with potentially better penetration of wood than a simple spill. I regard it as effectively porous. Peter Sefton states similarly. I tried to post a link but it is three lines of code long.

However, shellac was used in the past as an electrical insulator as it seals out moisture. For that to be effective though it needs to be in perfect contact with the base material and built up rather thickly. So it would seem likely that shellac could be effective if it was applied thickly enough. Most times when applied in french polish form it is going to be minimally thick and little more effective as a moisture barrier than tissue, especially at the molecular level.

Something 3mm thick will go wavy on you whether you finish it or not if it is just left lying flat as one side is always likely to be more exposed to air flow than the other.

 

[ED65]

Why would anyone put shellac on metal when it's acidic?

Because it works to give long-term rust prevention? If the japanning has begun to flake off the bed of a plane and you don't want to keep on having to oil or wax all the nooks and crannies to keep rust at bay a couple of thin coats of shellac will prevent rust from spreading or forming, apparently for years.

Thanks for offering, I would be interested in the exact Flexner reference, because I'm sceptical that he said shellac offers a good moisture vapour barrier. Shellac provides a terrific barrier against all manner of wood finishing nasties, but water vapour isn't one of them.

Well I was wrong, he didn't use the words "good" and "barrier"...

• Shellac forms an excellent barrier against water-vapor (humidity) exchange, silicone contamination, and
  existing stains caused by water, grease, crayons, or natural wood resins.
  
  From the first edition of "Understanding Wood Finishing", p.145

 

[custard]

I guess Flexner's realised the error of his ways then, because in the updated and revised 2010 edition there's none of that! The only finish that's rated as excellent as a barrier for water vapour are two pack finishes, in the section covering pros and cons of shellac (p 124 Understanding Wood Finishing, 2010 Revised Edition) Flexner describes shellac has offering an "excellent barrier to silicone", which I'd agree with, but he's dropped any claims about water vapour...presumably because he realised he was wrong.

Oh, and one other thing, I see Flexner agrees that shellac is acidic and should be kept away from metal, he comments that when pre mixed shellac is offered in metal containers those containers are specially coated to prevent any contact.

 

[ED65]

I guess Flexner's realised the error of his ways then, because in the updated and revised 2010 edition there's none of that! The only finish that's rated as excellent as a barrier for water vapour are two pack finishes, in the section covering pros and cons of shellac (p 124 Understanding Wood Finishing, 2010 Revised Edition) Flexner describes shellac has offering an "excellent barrier to silicone", which I'd agree with, but he's dropped any claims about water vapour...presumably because he realised he was wrong.

Hmm, that's possible. I'd still hold that the long-term rust prevention it provides is good evidence that it is a barrier, but I'll leave room for it not having quite the same properties when applied to wood.

Oh, and one other thing, I see Flexner agrees that shellac is acidic and should be kept away from metal, he comments that when pre mixed shellac is offered in metal containers those containers are specially coated to prevent any contact.

Surely that's referring to it as a liquid, not the dried film? After all, some of the original japanning is a baked shellac finish. As for just normal unpigmented shellac brushed on, a light coating is used by some owners of old planes and their observations are pretty conclusive that it doesn't promote rust at all, quite the opposite in fact as I refer to above.

 

[custard]

You do what you like with your tool collecting, that doesn't really concern me. The issue I'm focussed on is your claim that shellac provides "a brilliant barrier to water vapour", if a newcomer to woodworking buys into such nonsense they may think there's no need to design their furniture to accommodate wood movement. Anyone in that position needs to understand that a coat of shellac has minimal effect on water vapour transmission and therefore minimal effect on timber movement.

 

[Jacob]

You do what you like with your tool collecting, that doesn't really concern me. The issue I'm focussed on is your claim that shellac provides "a brilliant barrier to water vapour", if a newcomer to woodworking buys into such nonsense they may think there's no need to design their furniture to accommodate wood movement. Anyone in that position needs to understand that a coat of shellac has minimal effect on water vapour transmission and therefore minimal effect on timber movement.

Shellac is brilliant as a primer or knotting under paint. I've stripped a lot of old stuff over the years and the shellac knotting is often as new. It also works well in glazing rebates under putty and keeps them in excellent order for 100s of years.
But how this relates to water vapour transmission I don't know. I expect it does transmit but still manages to stay stuck down. All paints and varnishes are "micro porous" as far as I know.
nb "micro-porous" is a pseudo technical term used by paint salesmen - there's nothing special about "micro porous" paint, in fact many of them are craap.

 

[ED65]

The issue I'm focussed on is your claim that shellac provides "a brilliant barrier to water vapour", if a newcomer to woodworking buys into such nonsense they may think there's no need to design their furniture to accommodate wood movement.

That's a bit of a stretch, I didn't say you could do anything like that and had no intention of implying it. As I already stated, I didn't mean to imply it's a complete seal and if the OP had required clarification I would have given it.

I thought initially you were hung up on the terminology I used in an off-the-cuff comment more than anything, making this somewhat of a semantics debate, but your last sentence shows otherwise.

Anyone in that position needs to understand that a coat of shellac has minimal effect on water vapour transmission and therefore minimal effect on timber movement.

I think if you read on you'll see that Flexner would not agree that this is true.

Film finishes slow water-vapour transfer into and out of the wood and, obviously, how much is related to the film thickness.

I think you don't believe this includes shellac, but Flexner makes reference to finish in the broadest terms slowing water-vapour transfer in an article written in 2006 which he repeats (I think verbatim) in two of his later books, from 2010 and '11:

• No matter how a finish deteriorates, the end result is increased moisture exchange in the wood...
  ...
  Furniture requires a finish in good shape to protect the wood from moisture exchange and to make the wood look nice.
  ...
  The easiest way to keep moisture exchange to a minimum is to keep the finish in good shape.

That's pretty clear, and further I think the accompanying graph plainly shows he believes the effect to be significant:

So I feel comfortable with my later statement that a well-built shellac coat, as you'd expect in French polishing, does provide significant protection to that face of a board

 

[Jacob]

.........
So I feel comfortable with my later statement that a well-built shellac coat, as you'd expect in French polishing, does provide significant protection to that face of a board

It slows it down which will reduce highs and lows, which amounts to a little protection, but in a constant environment it will reach the same water content as bare wood.
BUT the main benefit of a good coating is to shed water which falls on it - protection from running water, rather than high humidity.
So architectural weathering details may give zero protection from humidity, but 100% from running water e.g. roof tiles.

 

[ED65]

So what was that you were claiming again custard?