Official Luthiers Forum!

How important to you is it, and what does it mean to you?

I think there is a proper cross grain stiffness for any given shape. Really high cross grain stiffness may not be appropriate for a given design. This is, of course, conditioned by the way I work: I use free plate tuning by Chladni patterns, and find it easier to get the patterns I want if I use wood with the correct cross grain stiffness. Basically, that means using the really stiff stuff for wider patterns, like Jumbos, and the wood with lower cross grain stiffness for narrower boxes like 00 and 0.

I used to think that high cross grain stiffness helped to control bridge torque, so that you could leave a top that was stiffer in the cross direction a little thinner. Reading Hurd's 'Left Brain Lutherie' brought me to question that. He uses deflection maps to fine tune tops, and also tries to use them to back out the wood properties of assembled guitars. This seems to work pretty well on new ones, or guitars that have not bellied too much, but on ones that have bellied his cross grain stiffness numbers don't work out well. It lead me to suspect that cold creep in the wood, which happens mostly in shear, was allowing the tops to distort especially in the cross direction, so that, over time, the cross stiffness contributed less and less to keeping the top flat. If that's the case, then you might end up fooling yourself by relying on it, ending up with a guitar that could belly too much over time even if it looked good right off the bench.

Then Mark Blanchard told me about his work, looking at the modes of unbraced tops. His initial idea was that he could manipulate the thickness to get some sort of standard mode shapes, which might reduce or even eliminate the need to shave braces. It turned out to be very difficult to alter the mode shapes of unbraced tops, and even adding bracing didn't have as much effect as you'd think. This is in line with the results of some researchers, who say that bracing only fine-tunes the response. As Mark says, the sound really in in the top.

Again, I'm just speaking from my own experience (what else?). I do know that Serge DeJonge feels that high cross grain stiffness is over rated, and he makes a pretty nice guitar. Some folks, who use different systems from mine, might find they need all the cross grain stiffness they can get, and for them, they may be right.

It affects the cross dipole motion of the top. Less stiffness = more cross dipole (side to side) motion. Some folks believe that more cross dipole equates to a more enveloping, present tone for the player... kind of the opposite of the tone projecting out in front like a cannon. In theory you could use that to steer a guitar's tone one way or the other.

I don't think too much about it... I can't say I am sophisticated enough to tell much of a difference with all the bajillion other variables involved in the final tone of a guitar, so I've always kept my tops near a mid point... not to floppy, not too stiff.

I work much like Alan in that I use a more cross grain flexible top for the smaller boxes.

_________________
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

I test my tops for long grain stiffness, as that is what I think carries the string tension, and let both the thickness and cross grain stiffness be random elements. I don't even record it, but perhaps I should, as well as the freq of he dipole...

Though I do flex by hand and use the floppies stuff for small guitars too...

Average Elong /Ecross for Engelmann spruce from the many samples in my records is 13.6.

Average Elong /Ecross for WRC is 8.7. A significant difference between the two.

So who's treating them that differently? Or what audible differences can be attributed to not treating them differently?

On some other measures the two woods are rather similar; density, damping, average mass of a panel thicknessed to the same vibrational characteristics, sound radiation coefficient....

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

So that's a big 'meh', eh Trevor?

Trevor wrote
'Average Elong /Ecross for Engelmann spruce from the many samples in my records is 13.6.

Average Elong /Ecross for WRC is 8.7. A significant difference between the two.'

I have long felt that the ratio long to cross is more important than just cross grain. With my typical bracing pattern I like to see 15-18 or less for Sitka and Lutz. From that point I just thickness the top by 'feel' and brace.

I suppose there is much more in the way of testing that I should be doing. I'm just floundering here and would like to know more about the subject.

_________________
Joe Beaver
Maker of Sawdust

I've measured it on my last 15 guitars. It varies more than any other specific I record. Have never adjusted anything build wise because of it and have heard no correlation in the final result. I'm not very experienced though, only 20 guitars total.

Tim

Most of the WRC I've measured has significantly lower damping than most of the spruce. I've got a Red spruce top and a WRC one that are identical in every measure except damping, with the cedar much lower (Q values of 197 vs 65 along the grain, and 78 vs 53 across). That's going to make an interesting pair at some point.

Alan, where can I purchase your DVDs? Hope it's ok to ask that.

Tim

In general I would agree, but I mentioned Engelmann specifically because the pieces I've measured approach WRC on average. Maybe I've just been lucky with the Engelmann I've had! Cross grain stiffness is very dependent on how well quartered the wood is, so the cut quality comes into the measurement as well.
Not really. My WRC guitars sound different from my Engelmann guitars when both have the same low order resonant frequencies, same body shape etc.. There's the difference in sound spectrum absorption between the two woods (basically the damping frequency response of the woods, which I haven't found a way of measuring), but I think there's more than just that. The ratio Elong to Ecross might be some of it.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Is there a correlation between low cross grain stiffness and lower cross dipole freqs? Unfortunately, the data I've collected over the past 50 guitars does not include dipole data either long or cross, only monopole, main back, and main air. Which now seems like quite an oversight. I haven't been recording cross grain deflection data either. Time to modify the charts...

Cross grain stiffness is very dependent on how well quartered the wood is, so the cut quality comes into the measurement as well.

Agreed, yet very ironic inasmuch as I concern myself very little with top "cosmetics" at this time but I assure you my tops are as well quartered as they come. There has to be more to it. Thx for the response though.

Tim

I'm gathering that cross grain stiffness might have some affect on the lower frequencies of the guitar. Using it for smaller boxes that need more bass, western red cedar having lower cross grain stiffness and generally bassier than spruce when treated the same, and from some of my own experiences bracing tops to increase cross grain stiffness that is my conclusion.

As has been said, cross grain stiffness does affect the 'monopole-cross dipole' resonant frequency ratio. This may alter the perception of the low frequency response, but I'm not sure it's as simple at is 'should' be.

I've noticed, for example, that Flamenco guitars often have a notably low cross dipole pitch. They frequently make them with only five fan braces rather than seven, which makes the top outside of the bridge wings a lot looser and drops the dipole. From what I can make out, this seems to contribute to the 'cut' of these instruments. I suspect this is because the cross dipole eats into the response of the monopole at a lower frequency than on a Classical. Since the dipole is not as effective a sound radiator as the monopole, the 'main top' peak in the output spectrum is narrower with a low dipole pitch. I'm pretty sure it was Meyers who noted that guitars with a narrow main top peak tended to sound 'bright' or even 'harsh'. You need some of that on Classical guitars, of course, but Flamencos really benefit from it. This points up the fact that our perception of sound quality is fairly complicated: 'brighter' doesn't always equate to more high frequency sound.