Official Luthiers Forum!

The newest Luthier Tips du Jour instructional video is on Youtube. The topic is the Tornavoz and there are two parts. Subtitles are available in both English and Portuguese and you can translate them into any other language by using the CC option on Youtube.
As always, comments and discussion are welcome.
Enjoy!

Thanks for taking the time to post these.

You are welcome! Posting them is the easy part.

Man you are killin me! First you give us no footage of that 8 string guitar being played, (tornavoz or not) and then you give us no audio of any tornavoz action! I must have this!!!!!

This is exceedingly interesting. I did not understand what was going on with a tornavoz before.

I appreciate how hard it is to create and deliver these videos.

Thank you!

_________________
http://www.booneguitars.com
https://www.facebook.com/pages/Stephen-Boone-guitar-builder/488208541257210

haha! I had to edit out the part where Paul was playing. I do have some footage of me playing a cardboard back and sides guitar that has a tornavoz but that is for a different tips du jour.
However, the background music in the video is Paul playing.

Very interesting stuff, Robbie. I keep meaning to try one on my wolfy OM, to either smooth it out or change main air frequency to some 1/4 tone.

Your Portuguese sounds pretty good!

Thanks,

Pat

_________________
formerly known around here as burbank
_________________

http://www.patfosterguitars.com

Muito obrigado Pat! I would be interested in knowing if anyone has ever experimented with the tornavoz on a steel string guitar.

If I'm not mistaken, Mike Doolin put a tube inside a side port on one (or more) of his steel strings. This isn't really a tornavoz but it seems like it would do something similar.

_________________
John Lewis
Wannabe builder owned by 2 crazy dachshunds

Interesting stuff. Not sure why it's called a "Tornavoz" though - perhaps someone can enlighten me as to who/when Tornavoz was. In modern scientific parlance, it's a Helmholz resonator with an extended length port.

The physics/math of such designs is well understood. A guitar box with a normal soundhole is a ported design (Helmholz resonator), with the port length being set to zero (well, almost - it's actually the thickness of the soundboard, but I'll not quibble). Extending the port length by the addition of a tube on the soundhole will indeed extend the bass response. In fact, if properly designed, it should smooth out the bass response as well as extend it. However, there are a few caveats from the world of loudspeaker design that will apply equally to guitar soundbox design IMO. Number one is that it the port needs to be designed properly. I can't stress this enough. The odds of finding just the right port aperture and length to smooth out and extend the bass response in a controlled manner by trial and error are tiny. In order to do the design properly, you need to know a lot about the driving device, that is, the soundboard/bridge/strings/saddle/etc. Unfortunately, without extensive measurements and lots of lab time and equipment, the required parameters (impedance etc.) are not easy to measure. This leaves (most of) us with trial and error. Hobson's choice I'm afraid!

Also, be aware that even when you've successfully designed a really good bass reflex port (usually with a Chebyshev response), it may actually be impractical to build. The port design may indicate the need for too large of an apeture - in which case it would look silly and eat into the structural requirements of the guitar's soundboard/bracing, or the port design may indicate the need for too small of an apeture - in which case the listener may be aware of the sound of the air passing into/out of the port (yes, the port itself can "make noise"!). Another impracticality may be a port deisgn that indicates a port that is too long. If the port comes too close to the back of the guitar, then the math will need to be adjusted, and the port may not be able to feed air into/out of the guitar fast enough, resulting in port noise (just like having too small of an apeture). In bass reflex loudspeaker deisgn (AFAIR) the general rule of thumb is that the length of the port should not exceed half of the depth of the speaker. In all of these boundary cases, the nice thing is that we can go back and adjust the physics of the driver to make the port design practical. That is, change the stiffness of the soundboard, weight of the bridge, etc.

If someone would fund me, I'd happily go back to university and study this on behalf of luthiers everywhere.....

Cheers,
Dave F.

_________________
Cambrian Guitars

"There goes Mister Tic-Tac out the back with some bric-brac from the knick-knack rack"

I'm not sure I would call that a Tornavoz, perhaps another name should be coined for it.

Colin

_________________
I don't believe in anything, I simply make use of a set of reasonable working hypotheses.

tornavoz pequeño de palosanto en vez de bronce

_________________
Build log

Oh, here is one of these contraptions made by Romanillos. It's longer, almost as long as in a Torres but made of poplar or smth.

_________________
Build log

Hi Robbie. Great videos !!
Do you think there would be a benefit to imply this on a steel string guitar to ? . Sounds really interesting !!

Lars.

Nice work Robbie, thanks.

I am not sure a tornavoz is required for the classical guitar. Most of the old good instruments that had one were removed. This suggests to me it is not a necessary part of the charming sound of the guitar nor for projection.

I built a copy of La Suprema and fitted it with a groove to put a tornavoz, yet never got around to it. It sounded great without it. It is mahogany spruce and has more than enough bass response and yet is a small guitar. In fact I am totally amazed by the amount of bass response it has, yet still the trebles are balanced. I give credit to the model more than anything else.

Golbrath said that he think it gives the bass more focus. We should get Al Carruth's input because I am not sure the typical response of a classical guitar in the low end. But I don't think a low A or B would have much fundamental to the note and I am not sure the E does either. Al...? It would be pretty easy to test. Even on a recording of Galbrath playing the same piece on a guitar with and without the tornavoz will tell something. Perform a long time average signal analysis of the same portion of the piece played both with and without a tvoz (LTAS) and then do a fft analysis of them and plot in in 1/12 average signals. It would be best to use the same set up and play with and without the tvoz. I am curious so I might have to put a tvoz in my la suprema copy.

_________________
Ken

The bass of the Leona is certainly the nicest classical guitar bass I ever heard recorded. There are some clips on youtube with Lieske, and his CD is money well spent. Maybe not the nicest guitar overall, here I could write a list, and of course the player, the hall and the recording setup play an even more important role, however, it is pretty hard not to be notice the extra depth of the bass in Leona. While most good guitars sound just normal or at least decent (many modern style guitars tend to have particularly un-interesting bass next to shrill trebles ), the Leona has that something extra that make you go

The fact that Tarrega and Llobet used tornavoz guitars for decades, when they could have picked a non-tornavoz guitar if they wanted to, says something. Simplicio used it extensively too. Hauser used it too, said it's not better or worse, just different. I guess that if done right, it is a good trick to try for a special bass.

_________________
Build log

Hi Robbie,

Thanks so much for the great videos. The tournavoz has been a mystery to me since I first became aware if it. This really helps to understand more about its function. Now I am thinking about doing one as an experiment on a guitar I am building now.

Thanks again.

Best regards,
Max

_________________
Max Bishop
Brighton, Michigan

I've noticed the bass as well, Lieske may be tuning down a 1/2 step

Ah, the frustrations of dial-up!

I've done some work with putting a tornavoz, copied from Romanillos' drawings, in one of my 'test mules'. It did about what I'd expect: dropped the pitch of the 'main air' resonance (through the floor, in fact!), and reduced it's activity (to almost nothing). The tonal result was interesting: it sounded more like a small guitar: a Panormo or some such, but with a bit more power due, perhaps, to the larger and more acoustically efficient top. I'll note that the Panormo I worked on, and a later near-copy of it, had the 'main air' pitch up near the 'main top' frequency, and little fundamental on the lowest notes.

Those in the know protest that you can't just put a tornavoz in any old guitar: the instrument has to be designed for it. OK. Can somebody explain how the design should be changed, and why?

As has been pointed out, the guitar is, in the low range, a 'bass reflex enclosure'. The length of the 'port' amounts to the Helmholtz 'end correction' on the outside, and who knows what on the inside? Since an object outside closer to the soundhole than one diameter will alter the 'main air' pitch, let's say the effective port length is somewhere in that ballpark on the outside, and add what you will inside.

Lengthening the port sleeve drops the pitch of the Helmholtz resonance, and therefore the guitar's 'main air' resonance. This is usually anywhere from about F or F# to as high as A on 'good' guitars. Usually the bandwidth is something like a semitone or three: you get one strong note, or maybe two or three at that pitch. Below that pitch the notes have little or no 'fundamental', above it they usually do, but the power falls off a lot until you get to the next big resonance. This is actually the other half of the 'bass reflex couple', the so-called 'main top' resonance, usually somewhere near the pitch of the open G string.

Using some sort of sleeve to drop the 'main air' pitch below the pitch of the lowest played note makes little sense to me. Granted, the 'main air' pitch is also where one 'guitar wolf' lives, but you should be able to get around that problem by other means. Making it that low simply sacrifices power, and I need all the help I can get there.

It's possible that the makers in the mid-nineteenth century were willing to give up power to get a 'better' sound. The guitar was not as often called on in those days to fill a large hall, and it was being used in many cases to play a repertoire that was written on much smaller instruments.

Most of the folks writing about the tornavoz cite it's 'remarkable' effect on the tone. As I have said, a foghorn in a Beethoven symphony would be 'remarkable': the term in itself does not convey much useful information. If I could get some good information about how to make one 'properly', or, better yet, get my hands on a 'good' tornavoz equiped guitar, I could make some measurements and figure out just what is going on. As I say, the guitars I've seen with extended ports worked pretty much in the way that physics would predict, and I reject the 'leprechaun model'. It's possible that by getting things just right you can make the guitar work better, but a lot depends on what you mean by 'better'.

I've corresponded to at least one of them, they keep their knowledge pretty well guarded.

_________________
Expectation is the source of all misery; comparison the thief of joy.
http://redrivercanoe.ca/

Whether one will like or dislike the sound/effects/recordings of a well-made guitar with a tornavoz, is still subjective.

Doug, you should send an email to Joshua French. He and Colin are the OLF resident experts on the Tornavoz. Either of them could probably answer your questions. Didn't Colin's La Lena have a Tornavoz? I believe it did, and it sounded great.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

Tornavoz means "voice changer" and could be also interpreted as in Tornado and Voz (both voice in Portuguese and Spanish)

_________________
Peter M.
Cornerstone Guitars
http://www.cornerstoneukes.com

On that basis, if we are just using literal translations any part of a guitar capable of changing the voice, bracing etc should be called a Tornavoz. Those 18mm tubes are something which may change the voice, but they are not a Tornavoz. As far as I am aware there are only three (possibly four) builders making Tornavoz guitars in the true old method.

Colin

_________________
I don't believe in anything, I simply make use of a set of reasonable working hypotheses.

I'm currently building (albeit slowly) a copy of the Torres FE09 which I think, in response to Alan's question, would be considered a good example of how the design of the tornavoz guitar is different.
In this guitar, which I think was not just designed to be compatible with a tornavoz but actually dependant on one, there is no (according to Grondona and
Waldner) lower harmonic bar at all. The pressure of the top at this point is supported by pins which brace the lower edge of the tornavoz itself against the back. Romanillos has stated that several of the Torres tornavoz guitars which he has examined (FE 17 for one) have had the lower harmonic bar "replaced" and I think it may be open to debate (in light of FE 09) if they even had them to begin with.



The top on the original of this guitar (the one used by Llobet during his career)
is described as quite thin and I've followed suit.
Since the tornavoz is fastened in permanently, getting the bridge on should be an adventure. It will be interesting to evaluate the sound of this one and hopefully I'll gain a better understanding of the tornavoz.

David, here is the way I attach the bridge when I build my Tornavoz guitars. It works fine, in fact I do it like this for my non-tornavoz guitars as well. You'll see that the cam clamps which just snug down the wings do not touch the guitar back, but bear against a support fixed to the cradle.



Colin

_________________
I don't believe in anything, I simply make use of a set of reasonable working hypotheses.