Official Luthiers Forum!
http://luthiersforum.com/forum/

Tolerances of to be glued surfaces
http://luthiersforum.com/forum/viewtopic.php?f=10101&t=56521
Page 1 of 1

Author:  Juergen [ Sat Apr 13, 2024 2:47 pm ]
Post subject:  Tolerances of to be glued surfaces

What tolerances do you work with on surfaces which will be glued, and thicknesses of the different guitar parts as braces, tops, backs, sides etc. and rightangeled surfaces?

Author:  meddlingfool [ Sat Apr 13, 2024 3:16 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

I do not understand the question.

Author:  Mike_P [ Sat Apr 13, 2024 3:35 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

I'll bite...

there are no tolerances only perfection...

wood glues aren't meant to bridge any gaps...

perfection being relative as I'm sure there are plenty at the microscopic level and it's these imperfections that give a good glue joint

Author:  Dave m2 [ Sat Apr 13, 2024 4:53 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

In the woodworking world, not just instrument making, it is held that glueing surfaces should match as closely as possible. They should also be smooth. A planed surface is best and the glue up should be done soon after the surface has been prepared. Glues such as titebond, probably the most commonly used glue is not a gap filler.

With all our curved surfaces in guitar building a planed surface is not always going to be possible but going to a fine sanded surface is good, better still a sharp scraper used before glueing is good.

The only exception I make is where I am glueing a cutaway side to the headblock where I find it really hard to get the mating surfaces to really match by in
I use epoxy which will fill any small gaps

Author:  Clay S. [ Sat Apr 13, 2024 6:54 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

When it comes to gluing surfaces, the greater the stress on the glue joint, the closer the fit should be . As to the tolerances of thicknesses of "parts" it has to do more with their mechanical and sonic properties than it does with a set dimension.
Factories often do it otherwise and dimension the parts to a set standard that guaranties structural stability and reasonable sonic performance.

Author:  Juergen [ Sat Apr 13, 2024 8:25 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

I'll bite...

Mike_P wrote: there are no tolerances only perfection...

wood glues aren't meant to bridge any gaps...

perfection being relative as I'm sure there are plenty at the microscopic level and it's these imperfections that give a good glue joint

sorry, Mike_P, but by the meaning of the word perfection can not be relative. Either something is perfect or not.

Which tools do you use to check for perfection? Do you use a knife-edge square and knife-edge straightedge? That would mean tolerances of about 0,0003 inch.

Author:  Hesh [ Sun Apr 14, 2024 5:09 am ]
Post subject:  Re: Tolerances of to be glued surfaces

Hey my friend.

For HHG it's not a gap filler so we want the fitment to be as good as we can get it. That's why I hammer home here on the OLF that for a bridge if you want to use HHG I want to see a complete contact fit of the entire bridge everywhere with only light finger pressure. If clamps have to force something in place it does not fit.

With say epoxy which is not used much in Lutherie but does have a place it will fill gaps and is more tolerant of a lousy fit.

Other glues are in the middle so to speak but we pretty much always want the best wood-to-wood fit we can get for anything in guitar building.

Some folks are keen to shoot for the least glue line possible and that's certainly a worthy endeavor. When guitars age the finish tends to sink into glue lines that are of any thickness and show and although it's not a structural problem it is visually not great....

Jointing a top is a very good exercise in how well something fits. People "candle" joins meaning hold them up to a window or even a "candle" to see if the light reveals a gap. My friend Link Van Cleave who is a Master Woodworker, had his own woodworking TV show on the West Coast and he was Jim Krenov's apprentice and Link also taught wood working at the Redwoods School of woodworking taught me that our eyes are VERY powerful.

He thought that candling is BS and that with the two sides on the bench in front of us and decent lighting we should be able to see an gap, it appears darker and that's how I do it as a result of Link's advice. So one tool for checking fitment is our eyes and learning to trust them.

I don't know a specific number for any acceptable level of not fitting we have never discussed a tolerance here on the OLF in a specific number sense so that's why you are not getting any answers to your precise question my friend.

Just make the mofo fit is what I do :)

Author:  bluescreek [ Mon Apr 15, 2024 1:21 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

why would you need to have a glue tolerance. Glue is about mating 2 sufaces. Glue is not a filler so no need

Author:  meddlingfool [ Mon Apr 15, 2024 1:26 pm ]
Post subject:  Re: Tolerances of to be glued surfaces

Lee Valley Cabinet Makers Glue 2002 boasts being a gap filling glue, to .006 I once read. I used to know some guys that used it for gluing bridges.

Personally, the joints have to be real nice and tight, but…I do use the 2002 for the neck block and tail block as the radius is sanded in by hand without a ‘perfection’ jig, so I hedge my bets. I ran out of TB3 for purflings, and the 2002 has been working well for that task too…

Author:  Alan Carruth [ Tue Apr 16, 2024 8:35 am ]
Post subject:  Re: Tolerances of to be glued surfaces

When I was learning we were taught that a center joint should be very slightly 'sprung'. We would 'shoot' the joint with a plane, taking the finest possible cuts, and then hold the two pieces up together to 'candle' them against a strong light source. If they were perfectly straight there would be no gap that showed light. We would go for contact at the ends with a very slight gap iin the center, such that you could push the two pieces in with hand pressure and close the gap. There are two ways to think about this.

One is that putting on the water based glue would cause the wood to swell a little bit, producing a convex surface. Making it a bit concave to begin with would make up for that. The other is that wood glues tend to have low peel strength; if the joint starts to open at the ends it is fairly easy to 'unzip'. If it's tight at the ends you'd have to, in effect, pull the whole length of the joint apart at once, which is hard to do.

This carried over to other joints too, of course. Thus, when making a stack laminate for a Classical guitar heel you'd flatten the surfaces, and then scrape the center slightly, to show a barely visible amount of light under a machinist's straight edge. Bridges are made to fit the top perfectly around the edges, with the same [u]slight[\u] dip between the front and back. If nothing else, this insures that you don't end up with gaps at the edges.

Works for me.

Author:  Woodie G [ Tue Apr 16, 2024 9:12 am ]
Post subject:  Re: Tolerances of to be glued surfaces

For common build and repair glues, the poorer the fit, the greater the clamping pressure required, and the thicker the glue line, the weaker the joint.

Titebond, hide, and fish will fill very minor gaps over a small percentage of the joint up to about 2-3 thousandths of an inch, or slightly less than the thickness of a sheet of copier paper. Epoxy will fill up some gaps, but really needs something like a structural filler to have much in the way of strength for the long term beyond about 0.005" gap. Obviously, we'd like surfaces to be perfectly mated, but certainly in repair work, we see fit issues that require rework and sometimes the use of something like 315g hot hide to maximize the good areas that remain with something like a vintage bridge joint.

In general:

- Gap of 0.001" or less in clamped joint at moderate clamping pressure for Titebond and animal glues; a gap of less than that is desirable for best bond strength and long-term stability under load. Thick glue lines in Titebond are particularly prone to cold creep over time, which is a great reason to avoid the issue and work in hot hide if possible.

- 0.005" or less for epoxy (with use only on instruments that are so compromised/so cheap that rebuilding the joint is not economically or structurally feasible, or for specialty use in advanced construction techniques such as balsa/CF lattice or cored construction). For gaps larger than that covering a larger percentage of the surface, add a structural filler to the epoxy, or rework the joint with a scab patch to close things up.

If you are not getting those sorts of joint fitments, check your tools and technique... dull edge tools are the very devil for fitting joints.

Page 1 of 1 All times are UTC - 5 hours
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/