Official Luthiers Forum!

I know this is a dynamic topic, but I just need some ballpark numbers for thicknessing tops and sides. First though, just curious what grit sandpaper should I use in my 16-32 drum sander for this process, 80 or 120?

I have been following some guidelines as far as thicknessing tops, back and sides and was just curious if they seem to be acceptable.

For Tops:
Sitka .105-.115
Adi 10% thinner than the Sitka
Cedar .110-.120
Engelmann, not sure, I assume about the same as Sitka, maybe on the thicker end.

Backs:
East Indian Rosewood .110-.120
BRW .115-.120

Sides
EIR .85-.95

I'ld love to hear your opinions on other tone woods as well. I have some Walnut, Cocobolo and Bubinga in my stash.

Thanks,
Kurt

_________________
_____________________________

Kurt Thomas

"There's a first time for everything even if you do it by mistake."

Definitely use the 80 grit and make light fast passes to prevent loading the paper. Going at a slight angle to the grain seems to help too.
As far as ball park top thicknesses go they seem about right. For me the sides are heavy at .095 (really heavy at .95! ) and the back seems heavy to me also.

_________________
Jim Watts
http://jameswattsguitars.com

Those backs seem way too thick to me.

.85-.90 for small guitars, .95 for large for EIR backs.

.85-.90 for sides seems just right.

I found that light slow passes worked the best for rosewood. Or rather my apprentice did, after I told him to take light fast passes which he cheerfully ignored as is often his way...

Ed: You lost your zeros.....!!!
Tom

_________________
A person who has never made a mistake has never made anything!!!

Thanks, I am working on a dred right now with EIR Back and Sides and Engelmann top, I was using 120 grit and it was definitely clogging up on me quite a bit, and I thought the passes were pretty light. Anyway, I have 2 rolls of 80 but no more 120 left so I will switch it up to the 80 then. Also, I will make the sides and back a bit thinner than I have been seems to be the consensus.

_________________
_____________________________

Kurt Thomas

"There's a first time for everything even if you do it by mistake."

If you are thicknessing (a number of) soundboards it's nice to do the final passes with 120 or 150 grit. Poorly graded sand paper can sometimes leave a few long deep scratches when used on drum sanders.

the thickness sander is not a finish sander. 120 grit it way too fine. I used 60 and 80 only
You have to think of heat when sanding. Often people think finer and slower and that is not what you really need. Think coarser and faster feed.
Martin uses about .075 for sides off the sander for production , depending on the era you can find this all over the place. Older vintage instruments are what most of us go after.
Tops came into martin at .117 worked them through production and I have seen them as low as .095
Backs about the same at .095 and sides finished out about .065 to .070. Finish thickness also varied a bit. I had a 39 in the shop the back with finish was at .095 sides varied .065 to .068 the top was .095 around the rims to .105 around the neck area. These were hand sanded and there was a variation . These are with finish measurements.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

Running your IR sides or any sides for that matter through at an angle will help spread heat over the entire drum avoiding paper loading. I do this until I'm close then run them through normally. I thickness my sides to feel which ends up being between .075 and .090". When I run my backs and tops through for the first pass, I run them across the grain. By doing this you avoid leaving a glue stripe in the paper. I thickness my tops to deflection. I run the top through good side first, then to the sanding table where I take all of the drum sander scratches out with 150 grit on my RO sander. From there I run the stock through face down until I'm within 10% of my final target. It's a good idea to check for debris on the conveyor (dents). I usually hit it with the vacuum.

Oh.. I run 80 grit

Cheers, M

I run 80 grit on my sander. Thin,slow passes seem to work best no kickback.


Sent from my iPhone using Tapatalk

What John said. How do you measure them John when the guitar is still a guitar and not in pieces? Hacklinger gauge?

Don't forget you are an individual builder, not a factory. You can continue to voice after the box is closed and before the binding goes on. Don't get too thin from the start.

I have found myself building a little tighter over the years. I tended to overvoice my earlier guitars and they were a little muddy and lacked clarity and tonal separation. I'm probably around .115-.118 for a Sitka top on an OM these days. (Traditional scalloped Martin-style bracing)

I have not gotten into vibration patterns on the top yet but deflection testing and later correlation with the finished product and critical evaluation by professional players I respect has been very helpful for consistency.

_________________
It's not what you don't know that hurts you, it's what you do know that's wrong.

Every one has missed the #1 Thumb rule of thicknessing !! I am surprised ! The #1 Thumb rule of Thicknessing is ...... Do NOT thickness your THUMB !!



Most of my backs I start at .100+ so I have plenty for hand sanding later Sides .80-.90 and tops around .110 But Im a newbie , so as I get better that may change .

_________________
The Shallower the depth of the stream , The Louder the Babble !
The Taking Of Offense Is the Life Course Of The Stupid One !
Wanna Leave a Better Planet for our Kids? How about Working on BETTER KIDS for our Planet !
Forgiveness is the ability to accept an apology that you will probably NEVER GET
The truth will set you free , But FIRST, it will probably Piss you Off !
Creativity is allowing yourself to make Mistakes, Art is knowing which ones to Keep !
The Saddest thing anyone can do , is push a Loyal Person to the point that they Dont Care Anymore
Never met a STRONG person who had an EASY past !
http://wiksnwudwerks.blogspot.com/
http://www.facebook.com/groups/GatewayA ... rAssembly/

Ed: You lost your zeros.....!!!
Tom

Lost my marbles a long time ago...

What about something really small like a ukulele?

this one was in pieces but as a machinist it isn't hard with the proper equipment. A 6 in micrometer helps. I also get to see neat stuff at Martin as an authorized repair center I have had opportunities many don't get.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

Thanks, that's a great tip about running them through cross grain the first few passes to avoided getting the glue line on the paper, which happened on mine. And that put's it into perspective John, not to view the Drum Sander as a finishing sander. I stuck the 80 on it and will speed up the passes a bit. I have the Jet 16-32 with the variable speed control which will slow down on its own if you try to take too much off on a pass. I had been running the speed just below 50%, maybe I will try it faster like 65-70 on the next passes once I get the rosette in. Right now the top is pretty thick, like .130 ish.

_________________
_____________________________

Kurt Thomas

"There's a first time for everything even if you do it by mistake."

However you do it, you need to think about the stiffness of the piece, rather than the species. There's so much variation in long grain stiffness within any species that rules like 'take Red spruce to .xxx" ' don't really make a lot of sense. There are ways to determine the best thickness for a top depending on the way you build: I suspect everybody will come up with something a little different. However, they all depend on some sort of measurement of the actual properties of the wood involved. If you can tell that by 'feel', well and good. Some folks can, but from what I understand most people aren't nearly as good at it as they think they are.

If you want an approximate thickness & do not want to do deflection testing, here is a useful approach that I attribute to Al Carruth.

Al has published experimental data for Young's Modulus as a function of wood density for many species used to make tops. I do not recall the source - this forum? GAL? Maybe Al can direct us to the original data. To a 1st-order approximation, he obtained a linear relation, i.e., regardless of species, the density predicted Young's modulus (with scatter, of course).

Young’s Modulus: E (MPa) = -9000 + 50*rho where rho is the top wood density in kg/m^3 (within the range of tested densities). (I do not recall if this particular equation is Al's fit to the data or if I did a graphical fit to his data.)

Based on his experience, he also defined a parameter (which I call the Carruth Index) to relate top thickness & Young's Modulus:

"Carruth Index": Ic = E*h^3 where h is the top thickness in mm. Ic = 250,000 for steel-string & 180,000 for classical guitars.

Optimal top thickness: h = [Ic /E]^1/3

Using Al's 1st-order approximation, I calculated "optimal" thicknesses for steel string & classical guitars using wood density as the sole determining factor:

Steel String

Density (kg/m^3) Young’s Modulus (MPa) Top Thickness (mm) Top Thickness (in) Mass/Area - kg/m^2

360 9000 3.03 0.119 1.09
380 10000 2.92 0.115 1.11
400 11000 2.83 0.112 1.13
420 12000 2.75 0.108 1.16
440 13000 2.68 0.105 1.18
460 14000 2.61 0.103 1.20
480 15000 2.55 0.101 1.23
500 16000 2.50 0.098 1.25


Classical

Density (kg/m^3) Young’s Modulus (MPa) Top Thickness (mm) Top Thickness (in) Mass/Area - kg/m^2

360 9000 2.71 0.107 0.98
380 10000 2.62 0.103 1.00
400 11000 2.54 0.100 1.02
420 12000 2.47 0.097 1.04
440 13000 2.40 0.095 1.06
460 14000 2.34 0.092 1.08
480 15000 2.29 0.090 1.10
500 16000 2.24 0.088 1.12

Steve, are the densities quoted above dry, or at some stabilized moisture content? Or RH?

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

Stabilized in your shop, ready to build.

The Young's modulus data on the wood was obtained be me using a vibration analysis method. The equation relating density and Young's modulus was an 'eyeball' fit to the chart.

I'll note that about 2/3 of the samples I have come within 10% plus or minus of the predicted line. There can be points that are 'way out, so it's safest to test each piece. Still, if all you can measure is the density, that may be better than nothing.

Both Young's modulus and density vary with changes in relative humidity. Since E rises and density falls as the R.H. goes up, there's a potential pitfall there. If you control the humidity in your shop reasonably well it's probably less of an issue that some of the other sources of uncertainty.

Measurement error is always an issue. This is particularly true when one of the measurements is small, like top thickness. A misreading of a few thousandths can alter the density derivation, and, of course, is a direct variable in calculating the Young's modulus. Here's where the grit of the sandpaper in your machine can figure in. Sanding leaves a fuzzy surface, with broken fibers to a certain depth that can't really contribute to the stiffness, but add thickness and mass. An article in the Catgut Journal years ago noted that different surfacing methods gave different Young's modulus determinations for the same piece of wood.

The equations that convert your measurements into Young's modulus are also approximate, since they've been simplified to eliminate non-linearity. The one I use (Trevor Gore's equation 4-52) is good to within about 10% of the 'real' value (according to McIntyre and Woodhouse in the CAS 'Journal' ), and would be hard to make more precise, even with perfect' measurements. Note that a 10% error in Young's modulus implies a thickness change of about 3%; from, say, .110" to .113. How much do you sand off at the end?

The 'index number' system that I use that SteveT outlined seems to be a usable simplification. Be aware, though, that the exact index number I use is based on my own experience and the way I build. If you use scalloped rather than the tapered bracing I use, you might find that some other index number will work better for you.