MIMF

I may have asked this but the answer didnt stick,my bad.
Should a sound port be routed before or after bending,and is a one layer backing veneer enough?
I was going to use titebond 3 since its moisture and heat resistanct??

Definitely rout the port after bending and even after attaching the top and back to the rims. One layer of veneer should be enough.

I don't use Titebond 3 for anything on a guitar except maybe gluing side purling to the binding. Original Titebond would be preferred for the veneer, which should also be glued on after bending the sides.

Barry, I have only cut two side ports and neither of those guitars are finished so I have basically zero experience. I'm curious why you advise the port to be cut after the box is closed. I cut both of my ports when the rim was assembled and profiled but before I attached the plates. Keep in mind that I use laminated solid linings so taking the completed rim out of the mold is a nonissue for me. My rationale was that I would have much easier access to the inside surface of the port for a backup caul and clean-up if need be. Also, the second one is bound and I didn't have the skill to produce a binding that fit the hole without a visible joint AND matched the curve of the side in depth. It seemed much easier to put the bindings I. Extra wide the. Flush them up to the inside and outside.

I hope I am not in for a surprise I failed to consider. I did make sure the ports were located such that the binding router berring would not fall in. . .

Cutting the port before or after attaching the plates is not really the issue. A closed box is more stable than a loose rim but it can be done before if you are careful. But cutting the port before the sides are bent is definitely not advisable.

Some folks do, apparently, cut ports before bending, but I'm with Barry in saying that after is best. For one thing, you can get it exactly where you want it. Access from the inside is good, so I do mine after putting in the liners, but before gluing the top and back.

I'd reduce the area, or length, of the F-holes if I made one with a port. Archtops already have a very high 'main air' pitch due to the way the holes work, and the port just raises it even further.

I do it at that point, too. I can see it giving me fits if I tried it before bending.
I make my offset soundhole smaller if I add a side port. Admittedly I don't have a lot of science to back that up, I just want the two holes to add up to roughly the same size as a standard soundhole.

Thanks for the good info. So youll bend sides then glue the veneer in place,then maybe make a jig to support the bent sides and rout the sound port? Or glue the veneer on flat and then bend the side then rout??
Ive never used titebond III ,but another Canadian luthier said he liked it only when gluing the veneer to an unbent side for a sound port. His reasoning was that the make up of that glue would withstand the water,steam and heat needed to make the bend. I could make a gluing caul for the veneer too I suppose and just glue it on after the side is bent and linings are on.I use original titebond for most everything,but have seen that heat will soften it and veneers do creep under it.That of course was in a funiture application.
i like the idea of routing the sound port before the top and back are assembled,less chance of a total fubar.

I glue on the veneer after bending. Doing it before bending makes it difficult to get an even shape during bending.

Ill use this info on my harpguitaar build as i want to put sound ports on the side of the hollow arm instead of the face of the hollow arm.the main sound hole will be an oval also. It will be a while though.

Comments - Allan Carruth's comment about the main air resonance of an archtop being high, and a port raising it is exactly what I realized when I put a side port in my archtop build. It created an annoying resonance that deadened the D note played on the fifth string fifth fret, and to a lesser extent the open D string. Why one was worse than the other I have no idea. I could hum that note into the side port and the box would resonate strongly. Closing the port dropped the resonance pitch by over a full tone. My eventual solution, such as it is, was to leave the sould port open because the guitar sounded a lot better to me as I played, the idea of the port for player experience was fully realized, but I turned the guitar down a full tone to D. I did that to solve other issues (the guitar doesn't like high tension string sets, low tension lighter gauge strings sounded weak, so I put a Bluegrass set on - .012 to .056 - and tuned to D, it came alive) but the dead note went away almost completely. Yet again, I have no idea why. I doubt I will put a side port on my next archtop, but I will also pay a lot of attention to tuning the plate tap tones of the top and back to different notes. On this guitar they were almost the same. Making them different is an experiment that I think will be useful.

I laminated the sound port area fully with a piece of side wood the same thickness as the side. I liked the way the edges of the port looked when finished, and I feel the sides should be as robust as reasonably possible anyway. I hand-cut the port in the same way as I hand-cut the F-holes, since my sound port was supposed to look like a wave.

I have been pondering why sound ports are so popular,if unwanted higher resonances are present. I suspect that builders with way more knodedge and experience have tuned their tops to allow the sound ports to work for them.it is behond me so ill move away from the idea until i fully understand it.

I put sound holes in for the simple reason that it makes paying the archtops I build more enjoyable. My guitars tend to sound very different in front than from the player's perspective. A sound hole gives the player better feedback.

That said, I use a single round 1" hole. I've not noticed a difference in the sound from out front, but it is a much more satisfying experience as a player.

I've seen a trend towards elaborate ornate soundholes nearly as large as the main sound hole. I don't know how that affects the guitars voice, but it never seemed a good idea to me.

The high frequency sound the ports put out is not 'unwanted'. Putting a port into an existing guitar will raise the pitch of the 'main air' resonance some, but unless you get pretty carried away with a big port it will be on the order of a semitone or so. If you're making one with a port from the get-go, you can reduce the size of the main holes a bit, and end up with the 'air' pitch in the usual place.

The lowest notes the guitar makes have wave lengths in air that are a lot longer than the box. As a result, the instrument radiates those sounds about equally in all directions: the player hears as much of the low note fundamentals as the audience. As you go up in pitch, particularly starting at around the open G string frequency, the sound output of the instrument becomes more and more directional: it's going off the top and out of the holes toward the audience. Unless that sound gets reflected back by the walls of the room the player is not going to hear as much of it. In extreme cases this can be very disconcerting (forgive the pun).

When I was in a Navy the carrier I was on stopped over in Athens over Christmas one year. To boost morale they held a talent show, setting up a stage on the hanger deck. I got the first slot, playing solo guitar and singing. The 'sunny Med' still gets darn cool in the winter, and it was about 55 F on the hanger deck. It's a big space with very bad acoustics, the PA was poor, and they had no monitors. By the time I got up to play, I couldn't feel my finger tips, and I couldn't hear the guitar over the background noise of the crowd and the ventilators. I had no idea what I was playing; for all I knew i'd put my hand down a couple of frets up from where it should have been. I'm told it came out OK, but you could not have told it by me.

A port in a place where the player can see it projects some of the high frequencies toward them. It doesn't take a big one: a 1" hole is useful. I did blindfold playing tests with a Classical guitar that had a 2" port at a couple of festivals where the background noise level averaged 72 dB-A; enough to make it hard to hear a guitar well. Blindfolded players tried the instrument twice; sometimes the port was left the same between trials, and sometimes it was changed. When it was the same, they were guessing, but when it was changed they heard it virtually every time.

The bottom line, then, is that for somebody playing acoustically in large, noisy, or 'dead' spaces (the 'restaurant gig') a port can be very useful. Just put it where they can see it, and don't make it too big.

May i pick up this thread?

Alan Carruth wrote: A port in a place where the player can see it projects some of the high frequencies toward them. It doesn't take a big one: a 1" hole is useful..... a port can be very useful. Just put it where they can see it, and don't make it too big.

Do You have an idea how small it may be in order to be useful?
I.e, what to we need physically? Just a point source connecting to the interior of the body? So what about areas even smaller than that of a 1" hole? 1 cm (the nearly unvisible sound port....)?

In my first experiment, the 'corker' guitar, I drilled 20 holes in the upper side of a classical test mule that could be plugged with corks; hence the name. To be on the safe side I got the corks first, and drilled the 15mm holes to match. The holes were arranged in pairs, every 75mm or so around the side from the neck block to the tail. Normally, in testing, I would pull both corks for a particular location, getting the equivalent of a single hole about 20mm in diameter. In some cases, though, I did try it with a single cork, and it still did some good. I don't think I've made a ported guitar with a side opening less than 25mm wide, and usually it's closer to 35.

R.M. Motolla borrowed that guitar to do a test of the utility of ports. He ran 'blind' trials to see if players could actually hear any difference, and came up with a negative result, which he published in 'American Lutherie'. In a subsequent issue the letters column lit up with objections, one of which was that the ports were too small to be useful. Most makers use a port closer to 50mm in diameter, and some are much larger. The thing I picked up in his study was that all of the tests were done in fairly 'live' rooms, where I would not expect a port to be of much use.

The lowest notes on the guitar have wave lengths in air longer than three meters, and the guitar body itself in much shorter than that. Those tones are radiated as from a 'point source', with the power going out about equally in all directions. As you go up in pitch the tones tend to be put out in a more directional manner, eventually coming of the top and out of the hole(s) toward the audience. The player hears the low tones about as well as anybody in the room, but only hears reflections of the high frequency sounds that come back from the room. In a large, dead, or noisy room (a 'restaurant gig') the player may not get much feedback. I've experienced this myself, and it can be very disconcerting (not to push the pun).

To test this out I out together another classical test mule with a 52mm port in the 'right' place that could be blocked with a magnetic cover. I took it to the Montreal festival (twice), a local folk festival, and a GAL convention, and got about 100 'blind' tests. I'd flip a coin to determine whether the port should be open or closed, set it up, and hand it to the blindfolded player. As they had played it a bit I'd flip the coin again to determine how it would be set up for a second trial. I'd take the guitar back, change the port or not as indicated, and hand it back. After more play they'd be asked if it was the same or different. These tests took place in hotel ball rooms with carpet and draped tables, a large 'live' college assembly room, and a cafeteria. In each case the background noise level was about 72 dB-A and there were lots of people standing around conversing.

The results were pretty clear; when the port configuration was the same for both trials people were guessing, getting it right about 50% of the time. When the configuration was changed they almost never got it wrong. I did try it on a few people in the courtyard of the college where the GAL convention was held: a large and quiet outdoor space, and it seemed about equally effective.

When I make a guitar with a port I usually reduce the size of the main sound hole to keep the 'main air' resonant pitch about the same. I've never used one on an arch top, so I'm not sure of the best way to do that.

Just a simple comment - there is an argument some make that the perimeter of the various holes is more important than the area of the holes Even with round holes the ratio is not the same, and it is very different when considering long skinny holes like F-holes. Just a thought...

Brian

Thanks for the interesting observations. BTW: i read Your 2009 article.

Maybe i am able to contribute a little bit: the thinline archtop i just have repaired (mostly...). About 30 years ago, after my 1st attempt to repair it, i noticed by chance that it sounded a lot fuller when at least one of the pickups was missing, and that there was an audible difference between both pickups installed, one pickup installed and no pickup installed. Audible in the near field but also from a distance of a few meters. I checked with some cardboard that this was mostly the hole. That has been my main motivation to leave some holes in the disks closing the top.

I find it quite interesting that it sounded "fuller" despite of the raised Helmholtz resonance.

And that's why i am also interested in having a sound port also in the guitar i am just trying to restore. Given this, i am mostly interested in the near field effect; that's a guitar i'll probably mostly play for myself. I actually bought it in an impulse to save that baby...

@Brian: i tend to give the sound port a carreau shape with concave edges, so these will be pretty long..

And a note to the diameter: 2 cm is roughly the wavelength of 20 kHz. Such a small hole should act as a quasi point source at all frequencies of interest.

A speaker or port starts to become reasonably efficient when the diameter is about 1/4 wave length, so the 2" hole works OK above about 4KHz. A slightly larger one would put it in the sweet spot for hearing; 'normal' ears are most sensitive between 2kHz and 4kHz.

I often wonder about the role of the 'upper cutoff frequency' of the hole. When the wave length of the sound in air equals the hole diameter it becomes essentially 100% efficient at radiating sound. Since the efficiency rises steadily from 0 to the UCF, having a larger hole may actually increase the output at lower frequencies by useful amount. This might account for the more 'open' sound you get from a larger hole. Keep in mind that a set of closely spaced small holes acts like one big one of the same diameter except for the added drag of all the edges. Something like a lute rose doesn't affect the Helmholtz air pitch much, but the drag make it much less active.

Since the perimeter of a round hole goes as the diameter, and the area as the square of that, a larger hole has less perimeter per unit of area. Losses from fluid flow from drag and turbulence are mostly incurred at the edges of the hole, so a round hole of whatever size will have lower losses. I've been told that the equivalent area of a slot, such as an F-hole, equals that of a round hole 1/3 the diameter of the length of the slot, but I can't tell you off hand where I heard that. I suspect the low frequency behavior is more in line with the length of the slot. The drag is, of course, quit high, so the Q value of the Helmholtz-type mode will be low.