What the heck just happened?

[hawthorn1213]

Looking to understand what just happened to my sound plates....

I've been slowly and happily moving along through my first build when I had to go on a short hiatus to move into a new house.

After moving and getting settled, I took a look at sound plates which had been left off just after shaping the braces and they have completely gone from convex to concave.

Needless to say I will be starting over, but just trying to get some info as to why this happened and what I can do to prevent it next time...

[Tom Haywood]

That happens. You can try putting them somewhere flat where there is some humidity. with weight on top, for a couple of weeks, and see if they flatten out. I predict they will.

[sunburst]

You've had a change in moisture content of the plates, almost certainly from a change in relative humidity.
Can you post a picture of the bracing? That can make a difference in things like this. The plates will probably go back to the original shape if the relative humidity returns to where it was when the braces were glued. No need to start from scratch, at the worst you might have to remove the bracing and redo it.

[HoGo]

What John said. I see a cross-brace, so if the humidity dropped (more than a few percents) then the top shrunk across width but the bracing did not (wood doesn't shrink much along its length) so the curve appeared. For such sensitive tops you need to know where your humidity is when you glue bracing on or you can get some surprises - this time you are lucky it hapened before assembbly. If the top stays like this in normal humidity (50%) then you need to remove the bracing and glue new braces when top stabilizes at that humidity (or slightly lower, 40-45%)

[hawthorn1213]

What John said. I see a cross-brace, so if the humidity dropped (more than a few percents) then the top shrunk across width but the bracing did not (wood doesn't shrink much along its length) so the curve appeared. For such sensitive tops you need to know where your humidity is when you glue bracing on or you can get some surprises - this time you are lucky it hapened before assembbly. If the top stays like this in normal humidity (50%) then you need to remove the bracing and glue new braces when top stabilizes at that humidity (or slightly lower, 40-45%)

That's good news if I just have to redo the bracing. Does anyone have any good ways for removing the current bracing?

[sunburst]

Carving bracing off of the plate is about the safest and easiest way. Releasing brace glue joints with heat and steam can result is releasing top center glue joints.

[HoGo]

Yes, carve away almost all the brace using gouges and fingerplanes leaving tiny layer of wood. Wet this layer slightly and let sit for few minutes and it will peel off easily, or just use wide flat chisel (or well sharpened scraper) to safely slice away the last bits without cutting into the top.

[hawthorn1213]

Is there anyway to prevent this in the future? I guess I'm having a hard time wrapping my head around how any mandolin can be made if it can only be maintained the same same humidity it was built. (Thanks so much by the way, this is a huge help!)

[Marty Jacobson]

In general, you should try to keep your shop at around the middle of the RH range in which the instrument will be used. I used to be pretty fastidious about keeping my shop at 40% RH. Now I try to make sure my shop is in that range for 24 hours before and after critical glue-ups, but I don't worry about trying to keep it there constantly. Lining onto rims is not a critical glue-up, nor is a headstock overlay, for example.

The extent to which something moves, before glue-up, is not a clear indication of how much the assembly will move after glue-up. It's a clue, but flex it in your hands and see how much force it takes to flatten it out. Not a whole lot, right? The sides can handle that, no problem. Where it gets to be problematic is when you have larger spans, like guitars, or especially weird wood (not quartersawn wood which moves in funky s-shaped ways, etc.)

[sunburst]

Is there anyway to prevent this in the future?...

It is best to glue the plate to the rim as soon as possible after gluing the braces. Don't give it time to move, in other words.
Take this is a lesson in why we should keep instruments in good conditions of relative humidity as much as possible. Wood moves, and when it is restrained from moving (like when it is glued up into an instrument) stresses and strains are involved. If your plate had been glued to the rim it wouldn't have moved like it did and you wouldn't have had much indication of how much it "wanted to" move.

[HoGo]

I read that Martin used to have a special room with elevated temperature where they kept their tops and backs few days before gluing braces. They didn't have RH control, but higher temperature drops the RH of the air somewhat. It is said that pre-war Martins didn't use radiused bracing anywhere near what you see with modern builders (if any at all) and the top and back radius ir result of this procedure when tops are removed from the dry room and bracing glued they will form the dome upon return to original humidity equilibrium. Exactly opposite effect to what OP got.
Also reduces risk of cracks due to low humidity. Some modern violin makers also leave tops and backs to dry in drying cabinet before gluing then to ribs for the same reasons.

[Levi S]

It’s amazing what a little humidity will do.
I’ve made a lot of things from thin wood I’ve resawn and have been surprised at the way it can change shape
In ways I didn’t expect. Sometimes just laying a piece flat will cause one side to draw more moisture more than the other.
And don’t even start about what can happen when you use a water based finish on just one side of a thin piece. (Though usually it will flatten out after the moisture evaporates / equalizes. Gluing together two pieces with equal moisture content can still be a problem, because if they will expand or contract differently depending on whether with the grain, or across the grain.
I’m not sayin it will fix your problem, but as an experiment I would take a dump cloth and uniformly moisten the concave surface, lay the piece flat, come back in an hour and see what happens to it.