Soundports and tuning

[jeffhigh]

Hi Trevor
Have you done any testing on the use of Soundports in the upper bout side and their effect on tuning of the first three modes.
thanks
Jeff

[Trevor Gore]

Quick answer is "No".

I still haven't heard a guitar that sounds better for having a sound port. The audience always seems to get a raw deal; never as much projection. But that is just me listening rather than any more objective testing. Alan Carruth did quite a bit of work. Might be on his web site. Bottom line (iirc) was that most people couldn't hear a difference in a blind test, but that it can act as a "monitor" for the player - useful in some settings. A hole in the upper bout is generally thought to raise the air resonance and I guess that has the usual coupling effects.

I drilled the butt for a jack (12mm) on one guitar and could hear the difference that made!

[Dominic]

Timely question Jeff. This has been going through my head since I got the books because all 6 guitars I am currently finishing have reinforcement in the upper bass bout to add sound ports. But I have not done any yet as I figured it must have some impact on the main resonances. A port that can be closed would make things even more interesting because you would shift them around and if they were perfectly pitched between notes with say an open port, you would be moving closer to notes when closed.

I do tend to bend my neck over the guitar a bit to get my ear closer and am always amazed when I hear others play my guitars and I sit out front. As a player I feel I am missing some of the richness in sound and so will those who play them for pleasure rather than performance.

The blind test stuff on sound ports was analysed in the GAL and it was here that it was revealed that the results of the test showed statistically significant degree of bias among the test subjects, what they called the placebo effect. Where people will hear a difference if they think they should. That is, they knew they were listening for the difference between sound ports and no sound ports and often reported a difference when in fact none of the guitars had sound ports. Same thing with Strad violins. People say they are this and that and worth millions of dollars but blind tests often fail to pick any difference between them and new violins. But a million dollar price tag and the power of that name has a huge influence on people's perceptions of their sound. Its all quite interesting.

And in the John Williams/smallman interview on ABC a few years ago, John was saying that many professional classical guitarists can not hear the guitar either and so are heavily influenced by what they think they should be playing and this results in space for the BS merchants and marketing guys to make up stories. A strad sounds like X and so that is how this Strad sounds. Makes it harder for us. And more interesting.
Only thing for it is to try and do the testing.

Cheers
Dom

[jeffhigh]

Besides any benefit from the soundport itself, the reason I was aking is because I have an existing guitar that I would like to get the air resonance up 2Hz from where it is and don't have room within the rosette to enlarge the soundhole.
I was thinking a soundport could help.

[Dominic]

Jeff, this is one of the questions we need to answer. That is, is increasing the size of the soundhole by a few mm to give an area of say 25mm square the same as adding a 25mm square hole somewhere else. Or is the placement of the additional area important?
Again, don't know so I suppose a test is in order.
Dom

[charangohabsburg]

Or is the placement of the additional area important?

Both, area and placement.

For example, a sound port at "the other end" of the soundbox than where the soundhole sits (e.g. lower bout region) will raise the resonant frequency more than a sound port of the same size in (or closer to9 the upper bout region. So Trevor's observation on the 12mm hole is not a surprise. There is a nice, easy understandable article about these realtionships: Basics of Air Resonances, W.D.Allen, Big Red Book of American Lutherie, vol.1, p.8, or AL #1, 1985.

[Dominic]

Yes, but do you get the same effect on resonances as if you just add the extra area to the soundhole? I'm sure the pressure changes inside the guitar are pretty equal everywhere. The air space would not subdivide into different pressure zones would it? So why does the placement matter to the main frequencies? I'm not challenging you, just asking a question so I understand what is going on. If we are going to use soundports we need a way to predict what will happen so it can be taken account of in the design stage rather than as an after-thought.

The relationship between soundhole size and main air is linear. So I wonder what kind of function defines the relationship between soundport size and main air and then as we move the port around. Do we just get a wide range of linear relationships with quite different constants and elasticities? I would suspect we would get a pretty neat cluster around the soundhole relationship which might make a decent proxy for soundports. But all of this is just a wild guess.

I have the BRB for that year but I'm sure it does not discuss soundports.

Cheers
Dom

[charangohabsburg]

Yes, but do you get the same effect on resonances as if you just add the extra area to the soundhole?

No, not exactly the same.

I'm sure the pressure changes inside the guitar are pretty equal everywhere. The air space would not subdivide into different pressure zones would it?

Sound pressure levels in a guitar body mainly vary depending on the exact measuring location inside the box, box dimensions, box shape and frequency. Not sure if I am right if I assume that a hole in the box could be considered as something like a "part of its shape".

So why does the placement matter to the main frequencies?

It seems to be kind of "the easiest way a sound wave can place its length or half of it" in a perforated box. Read the article in the BRB vol.1 pp.8 I mentioned before, or I had to paraphrase half of the five article pages here. Glad you have the Big Red Book(s).

I'm not challenging you, just asking a question so I understand what is going on.

No problem. But maybe I have not an answer to everything.

If we are going to use soundports we need a way to predict what will happen so it can be taken account of in the design stage rather than as an after-thought.

Good idea. Be the first one!

I have the BRB for that year but I'm sure it does not discuss soundports.

The article I mentioned in my post before discusses air resonances box dimensions, box shapes, one-hole boxes and two-hole boxes. You are right though, the word soundport was not mentioned there, and maybe did not even exist back in 1985, but the holes did. The thought behind the second hole was "Mountain Dulcimer".

[Dominic]

Thanks Marcus, I just looked and I don't have BRB V1. Sound like a good article I should read. I am sure Trevor mentions other research in this area. I'll have to look it up.
Cheers
Dom

[Nick]

Here's the link to Alan Carruth's research paper on this subject using his "Corked Guitar" -http://www.alcarruthluthier.com/Downloads/sidePorts.pdf

[kiwigeo]

Alan Carruth's paper makes good reading. The only thing missing from the study was a person with the corks inserted in his ears

[charangohabsburg]

Depending on playing position soundport placement may be really critical...

[charangohabsburg]

I'm sure the pressure changes inside the guitar are pretty equal everywhere. The air space would not subdivide into different pressure zones would it?

Sound pressure levels in a guitar body mainly vary depending on the exact measuring location inside the box, box dimensions, box shape and frequency. [...]

I attach an illustration of that (the distribution of sound pressure levels within the box at two different frequencies).



These are measurements made by Jürgen Meyer back in the eighties and presented in a book in german: Akustik der Gitarre in Einzeldarstellungen, Jürgen Meyer, Verlag E. Bochinsky, 1985. Until now I couldn't locate an english translation of the book, only some english papers that are part of the book but none of them covers this aspect.

[Dominic]

Thanks again Markus,

Alan's paper goes into this stuff. Pressure zones inside the guitar can be different and even positive and negative during the exhale. I think Allen also mentions that most of these papers are analysing the un-coupled Helmholtz with solid guitar shaped boxes or guitars buried in sand. Nonetheless, at Helmholtz frequencies the pressure rises and falls at the same time at every point inside the box which is what makes these powerful frequency responses stand out from all the others. So not the same but not too dissimilar.

The paper was hinting at ports actually increasing sound output for the audience and the player. This seems a stretch but there could be some combination of soundhole size and port size and location that leads to such improvements. As a summary Alan says that the further the port is away from the soundhole the more effective it is. But a port on the butt or even under your right arm on the lower bout would not be as effective as a monitor as one that faces the ear. So for praticalities sake, the upper bout seems like the best compromise. And that it should be easy enough to make the soundhole slightly smaller and then tune the box using the port which is still a linear relationship.

So with that little bit of theory, I think the thing to do is make a prototype and cut holes in it all over the place and make measurements and then plot the function. Simple. Then use the results to determine the best location for the port and how sensitive the T(1,1)1 mode is to changes in the size of the port at your chosen location for that particular body shape.

Cheers
Dom

[kiwigeo]

Depending on playing position soundport placement may be really critical...

He needs one of these:

[Dominic]

Hey Jeff, what have you decided about soundports? Are you going to install them?
Cheers
Dom

[jeffhigh]

Actually I think I have come to the conclusion that the problem with this particular guitar is that I have underbraced the back
It's giving me 88, 188 and 206Hz on the frequency analysis. and the back peak at 206hz is showing a higher amplitude than the top at 188.
I think it is basically too close and is attenuating the top monopole.
I reckon I may have to pull the back and rebrace unless I can come up with an insitu solution.

[Trevor Gore]

Depending on how you braced the back, it is often possible to do it in situ. You can either cap the top of the existing brace or just put another one in just behind the first one. Usually the place that needs the extra stiffness is only the middle of the span.

I'd be interested to hear your reaction when you have it stiffened up.

[johnparchem]

During a discussion on the effect of increasing the sound hole from 3.5 to 4", a question came up questioning whether a sound port would affect the guitar in the same way. It occurred to me I have a permanent setup that allows me to do spectrum analysis and I have a small guitar with a sound port. My sound hole on this guitar is 3.68" so an area of 10.64 sq-inch; my sound port is a true ellipse 2.816" by 1.373" so an area of 3.03 sq-inch. In the discussion I stated (based on information in the Book) that a change from 3.5" to 4" would result in about a 7-8 Hz higher Helmholtz resonance.

A change from 3.5 to 4" is a 2.94 sq-in difference in area. Coincidentally that is very close to the area I added with my sound port, 3.03 sq-in. Amazingly my test showed a 7.5 Hz different in the first resonant peak, The main top resonance and back resonance the second a 3rd peak also shifted down when I covered the hole. So adding a sound port seems to be equivalent tuning wise (with my sample size of 1) to increasing the area of a sound hole.

sound hole open.jpg (57.39 KiB) Viewed 23555 times

covered sound hole.jpg (51.32 KiB) Viewed 23555 times

[johnparchem]

Matching the difference with area may have been a coincidence, as the Helmholtz frequency should be proportional to the diameter of the sound hole, where area changes exponentially.

[Trevor Gore]

Interesting, John. The Helmholtz frequency is proportional to the SQRT(Area), so it might be interesting to do the comparison on that basis. That's based on a "rigid" vessel, of course. I hate to think how all the coupling would interact!

[johnparchem]

I used sound port area, as I was looking for an equivalent to a change in the diameter of a sound hole. Hypothesis: adding a sound port of a given area is roughly equivalent to increasing the sound hole diameter to achieve the same change in area.

A simple next test would be to make some masks and plot a few sound port sizes. In that case I would think to support the hypothesis the change in the T(1,1)1 would be proportional to SQRT(area). To be a good test I would also have to plot changes in the size of the sound hole. I just used the table in the Book for the change resulting in a larger sound hole.

Looking at the graphs I think that sound port really becomes part of the sound hole as there is not any new peaks that I can see associated with the sound port. It just moves the T(1,1)1,2,3 to the right.

I will make some cutouts for both the sound hole and the sound port to see if what the correlation really is.

This has me wondering, I have a classical guitar that has a T(1,1),2 at 172 Hz, It looks like I might be able to add a sound port and bump it up.