Double Bass Instrument construction and phase

[Jon Stefaniak]

I wasn't sure exactly where to put this, but it relates to mic placement, I guess

Recently I was watching a video on the ol' YouTube about loudspeaker construction and it got me thinking about doublebass construction. How analogous are speaker phase (tuned ports, passive radiators) to the difference in phase on inside and outside of a bass?

If the top of a bass is acting like a speaker can the sound from the f holes come out out-of-phase with the top and actually cancel some low end. I never thought deeply about it, but there is so much tonal variation on a bass depending on where you listen... I always feel the bass sound is strongest listening perpendicular to the top about bridge height... But that is where the fholes are. Is some of this tonal variation influenced by phase?

If you miked the inside and outside of a bass top at the same point from the same distance and summed them, would there be significant phase cancellation... I imagine the low end would almost disappear. Has anyone tried this?

Has anyone ever tried applying some of the ideas of loudspeaker construction to the bass... Tuning f holes internally like speaker ports? Passive radiating speakers in the back or ribs? I imagine you could get even more low end out of a small bass. Am I crazy?

 

[Don Hergert]

My DB has an access port in the driver side C. It's large, about 7"x7" square. Large enough that I can listen to the sound of the inside of the DB as I'm playing if I want to, especially if I'm sitting down while playing.

Now granted, my DB is an Alcoa aluminum DB, and even though it has been treated and setup to minimize metallic tone, both its inside and outside tone are going to be at least slightly different than those of a wooden DB. But my observation is that the tone (or sound, or whatever one might want to call it) from the inside of the DB is much more echo-y, boomy, and for that matter, muddy, than the tone from the outside of the DB. Enough so that, albeit interesting, I consider the inside tone pretty worthless on stage; not worth mic'ing or trying to use a pickup with it.

There are other people here who have access ports on wooden DBs and may be able to discuss their experience listening to the inside of their instruments. Hopefully some of these people will speak up here.

 

[Jaygetsreal]

Think I might experiment with this. Two pencil mics taped together and inserted through the f hole so that their diaphragms are equidistant from the front deck, say +5 and -5 centimetres. Record. Play with phase and filters in DAW. Maybe delay too. Sound travels outwards at about 300 metres per second.
There could be a whole microcosm of sound reinforcement here that you've started!

 

[DoubleMIDI]

Even though the basic idea is correct there are a few differences in the construction and function of loudspeakers and DBs.

Your idea of different polarity (phase is only a correct naming in the case of different time traveling which is not the case in an equidistant pickup position) cancelling the sound is correct if there is only a big wall with a speaker in it and not an enclosure.

Speaker cabs are damped internally to damp (at least) the higher frequencies. This is typically needed because the volume of a speaker cab is typically much smaller than that of a DB body and to avoid reflections (of higher frequencies) that may drive the cone too.

The cone of a loudspeaker is constructed to be as rigid as possible, so that the complete cone follows the force of the magnetic field.

The top of a DB differs due to the top also vibrating in modes (less on a plywood one, more on a carved one), so that parts of the top might go in and other out of the steady state. There are also more inner reflections (as mentioned before) that may drive (modulate) the top a bit, but mostly cancel out in air pressure going out of the f-holes.

It‘s mostly the lowest mode (whole top moves in the same direction) that gives the major part of the air pressure going in and out of the f-holes (as well as the speaker port).

So even though the vibration of the top in a small distance inside and outside might be similar, the output through the f-holes is a lot different.

Also don‘t forget that the speaker cone can easily travel because of the flexible suspension whereas the the DB top needs to bend to deliver a similar sound pressure (the main reason why it needs to be that much larger).

Your idea of tuning the low end by the area of the port is correct. You can tune the low end that way, but I assume that the area size of the f-holes in relation to the body size has evolved over the centuries to be close to the optimum.

As with speakers, tuning the port too low will give a „hole“ (attenuated low frequencies above the tuning frequency) and too high a heavy resonance for the higher low end and a big attenuation below.

But the biggest difference is that top and back of a bass are a coupled system by the soundpost that drives the back from a single point of the top.
This is a bit similar to speakers with a second passive membrane, but different enough that the results are quite different.

So the question might arise if a 3/4 bass with a weak low end is badly tuned by the f-holes or just too small to get the same volume as for the higher low end notes.
But changing the size of the f-holes is only possible in one direction and not reversible. So I would first experiment with the soundpost position because it is reversible than to risk to ruin the top.

 

[AGCurry]

The top of a DB differs due to the top also vibrating in modes ...
It‘s mostly the lowest mode (whole top moves in the same direction)

Do you mean "nodes?"

 

[SAlcott]

 

[Ryan in PDX]

Do you mean "nodes?"

He doesn't.


Here are three normal "modes" of a one dimensional string with fixed ends. The points at which there is zero motion are called "nodes." Every vibration can be described as a combination of all the modes up to infinity (it's actually kind of awesome, and a big part of nerd school).

When playing an octave harmonic, you can touch the string at the octave, nut, or bridge, and the sound won't stop, because that part of the string is not moving. "The second mode has a node at the half way point," in other words. The second octave is the fourth mode (n=4, not pictured) and has five nodes (at 0, 1, 2, 3, and 4 quarters of the string length), as another example.

If you stop a note, the same modes occur, but the vibrating length changes, so only one node (the bridge) remains in the same place, and all of the antinodes move. That means close-miking a single string is always going to be a compromise, if you play more than one note.

The top of a bass is a lot more complicated than our hypothetical one-dimensional string, but it will have modes. They will be, for our purposes, impossible to calculate, and (at the risk of assuming, the point @DoubleMIDI was making) almost not worth thinking about. Adjusting things we already know about, like changing the setup, miking from a distance, etc., will get us better results than attempting to modify our f-holes or calculate internal mic placements and phase relations.

 

[DoubleMIDI]

Thanks Ryan. yes I meant modes.

In a two-dimensional vibration of plates and membranes there are are areas in higher modes than the lowest that vibrate in opposite directions, the higher the modes the more areas are there. Since they are not at the same place they radiate in certain direction better than others (where they cancel each other out mostly) due to phase (read short time delay).

You can play some frequencies to a drum skin and put small particles on the membrane. Sweep through the frequencies slowly and you will see the boundaries of the vibrating areas as they change with frequency.

Inside the cavity the pressure out of the f-holes is mostly influenced by the lowest mode that is the complete top area moving in and out. That‘s the reason why a mic should not be placed too close to the f-hole.

Some cheap electret capsules might not represent lows too well (where often the (de-)coupling capacitor is too small to let low frequency frequencies pass equally well), so these might be placed close(r) to the f-hole than full frequency spectrum mics.
But there might be an overly amplified range or an attenuated one if the border frequencies of both are not the same as explained in my last posting.

I did a bit of reading about the acoustics of musical instruments, but I‘m not as informed as a physicist specializing in that area.

Go make it even more complicated the resonant frequencies of the vibrational modes of the top are fixed, but to a lower degree work in frequencies close to it. The radiated frequency is not changed, due to the constant energy feed to it, but may radiate (very shortly because of a high damping) at the resonant frequency afterwards.
The frequencies of the modes amplify frequencies around it which are called formants if they stick out a lot. These formants do not move with played pitch but partials of a tone which are falling in the range of the formants are amplified.
These formants give an instrument is sound character.

If you want to know more about that, get a book or find a website about the acoustics of musical instruments.