Official Luthiers Forum!

I'm sure T1,3 is there, but I'm not sure which one exactly. I don't think tapping will get it either as it's a coupled response. I believe blocking the sound hole will probably suppress it, but it's be a while since I've tried it.

_________________
Jim Watts
http://jameswattsguitars.com

Last night I did the exact same test again. I didn't change anything. I was wondering whether or not it might have changed after 24 hours. Now I consistently get these double peaks. I ran the test ten times, same every time. The lack of consistency using is rather frustrating. At least I cannot make any sense out of it. I guess I will bonk it again tonight and see what magical numbers show up.

Mr jfmckenna, maybe you can find something ho the ANZLF which has a section on the Gore book.
Trevor often answers questions, or at least gives hints, and many of the subscribers are getting well up on frequency responses.
http://www.anzlf.com/viewforum.php?f=33&sid=29896d300e4e9d03e2a12efc8ae8c9c2

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

I have found inconstancy when I have the taps too close together or a double tap and also if there is too much gain on the microphone and the input is clipping.

Looking at the earlier cleaner luthierslab app plots where you carved the back braces, it looks like the top and back were very close together (T(1,1)2 and T(1,1)3). As you carved the back brace it looks like you put the back right on the top. Your top is pitched a bit higher (stiffer) than my small guitars. Just for fun put some bluetac or some mass on the top and see if the peaks separate. If you the lowered peak will be T(1,1)2.

_________________
http://www.Harvestmoonguitars.com

As I read it, in your first chart the output peak at 225.4 is from the 'main top' resonance. There are a couple of possibilities regarding the 'main back' pitch. The 'notch' just below the 'top' peak, with the small peaks around 210-215 could be from the couple between the 'main top' and 'main back' resonances, with the 'back' pitch most likely being pretty close to the notch a bit below 220(?). Alternatively, the 'main back' could be the cause of the peak at around 380. As you shave the back braces both of those peaks shift downward. I actually rather suspect that they're both back resonances, with the lower one being a monopole and the higher one being a lengthwise dipole type. This is somewhat speculative, and can't really be resolved without more data.

The most common case is that the 'main back' monopole mode is significantly higher in pitch than the 'main top' mode; high enough that the two don't couple strongly. Moving the back pitch downward, by shaving braces, will enhance the couple between the two, and this is what drops the 'main air' mode pitch. In effect, it's like enlarging the top and/or making it more active.

If you start out with the back monopole lower in pitch than the top, then shaving braces is going to move things in the wrong direction in terms of coupling. The looser back should still move more air, and this might drop the 'air' pitch a bit, as it seems to have here, but it won't be as effective as it would be with a higher back pitch.

A tap spectrum is a useful 'snapshot', but won't tell you everything you'd like to know. You could have gotten a better perspective by doing another test with the mic on the back, which would have given a better reading of what the back was doing. If you have access to a small amplifier you could also use the signal generator module in LL to drive the guitar, say with a speaker at the hole. Driving it at the the peak frequencies indicated in the spectrum chart and touching the guitar in various places would give you some ides of what's vibrating to produce those peaks, even if you don't have a 12-15 W amp that could allow you to produce Chladni patterns. Finally, adding weight to the various spots to drop the resonant pitches, rather than shaving braces, would have helped in diagnosing what was going on, allowed you to hear the result, and been reversible.

You posted your 'double peak' chart while I was typing. Usually double peaks like that indicate a more or less exact frequency match between coupled oscillators; look up 'Butterworth filters'. Things 'relaxed' a bit after you did your brace shaving, and the pitches shifted somewhat, or there could also have been a change in humidity. Again, with limited tools at your disposal it will take some time and effort to figure out what's going on.

This is becoming a real ‘diagnosis at a distance’ scene but it is fascinating. Please keep talking to us as it is exactly how we would like to improve our results after the end of a build.

As Alan says using temporary mass additions in the form of blutac could be very illuminating while being reversible.

Cheers Dave

Thank you all so far for your insights. Continuing on... The thing that bothered me about the chart with the double peaks was that I changed absolutely nothing from the LL charts I posted on the 1st page. I keep the shop temp and RH the same and am very strict about that.

So I did the tests again tonight changing nothing and they went back to looking a bit more normal. Why? I have no idea. But anyway here is what I am looking at now:

Here is the top tap on the bridge wing:


Here is the top with the sound hole plugged up:


Here is the back tapped center just below the waist:


Here is the back with the sound hole plugged up:


What I see in the first chart of the top is a weak peak at 208 and when the hole is blocked it becomes the most pronounced peak. And the one at 224 just goes away so I am not sure what is going on there.

"What I see in the first chart of the top is a weak peak at 208 and when the hole is blocked it becomes the most pronounced peak. And the one at 224 just goes away so I am not sure what is going on there."

It's likely that the peak at 208 is (more or less) the 'real' main top (monopole) resonance. Without the Helmholtz air resonance (due to the blocked hole) that's what the stiffness and mass of the top produces. When you open up the hole and introduce the Helmholtz resonance the mass flow through the hole and resulting pressure changes in the box couple with the top. At the 'top' frequency air is moving into the hole, and the inside pressure is increasing, as the top itself moves inward. The rising air pressure can be considered as an additional 'stiffness' element on the top, and the pitch of the resonance rises from 208 to 244 Hz.

What I am seeing from your charts July 22 is main air 124.8 Main top monopole 223. and dipole 423.8
The dipole is emphasised because you are tapping on the wing rather than the middle of the bridge.

I don't see any contribution of the back in the top tap open soundhole test and this is to be expected since you built it as a small body "stiffly braced reflective back"
Again closed soundhole and back taps are not relevant in the Gore method of modal testing and not really relevant to the playing of the instrument where the back is only coupled not directly driven. And a very stiff back is not going to be coupled anyway.

124.8 is quite high for a main air resonance and is possibly why your bass response is lacking, as is the main top monopole at 223.
As you have found, shaving the back has lowered things but not by much, top bracing stiffness reduction and/or soundhole size reduction could be considered

Given that your back is still not likely coupled to the top, tapping the back with the soundhole open is very useful to get its freq. Gore’s way is not the only way to measure.

223 is a bit high for a guitar that size, but not hugely so, about 10hz.

Were I you, I’d find a hz to note chart and find out what four semitones was above 223 (I’d guess around 280) and hog the back braces til I was approaching that number. Then I’d measure the top again to check the status. Chances are, by taking off that many hz from the back, your top freq will dip below 220hz, one number you do not want your top to be at.

I would also measure bridge rotation before adjusting any top bracing…

That being said, iirc you thought your back was at 380, so you may not be able to drop it by 100hz, but you can probably go a long way. Unlike the top, the back doesn’t need to be ‘strong’…at least not in the lower bout, IMO…

I just got one of those angle gauges so I will measure the bridge rotation later today. I have reduced the lower bout bracing about as far as I care to go and not too much happened.

At this point we are talking about retopping the instrument anyway so I have the liberty of trying everything and I hope that it can be a good learning experiment in getting there, where ever that may be. So I suppose I could reduce the back braces even more and just rebrace when the top is off.

I found it helpful to make a short run of expendable guitars.

Why are you thinking of retopping the guitar?

How tall are your back braces?

At the moment you have a stiff back which is not coupling with the top, and that was a deliberate design choice. It will probably take a fairly large reduction in the lower bout bracing to change this to a live back IF you want to do this.

Do you have access to any of your previous succesfull builds of this design that you can tap to get a comparison.
Try tapping on the centre of the bridge NOT the wing this will give you a better monopole response.

Knowing the frequency of a non live back is just not useful because it does not couple with the soundboard.

But if you are going to use the Gore method of Modal tuning it helps to use the exact methods of testing rather than mixing in others, and if the back is not coupled, it's tapped frequency is not relevant.

But if you are going to use the Gore method of Modal tuning it helps to use the exact methods of testing rather than mixing in others…

Why/how?

The resonances are what they are no matter how you measure them. The way they’re measured doesn’t change what they are. Modal tuning is about placing your resonances intentionally for a purpose, more than it is a method of measuring those resonances, IMO.

Since the back is so high and uncoupled, it won’t show in the top tap graph.

And since he’s trying to get more bass by coupling the back, the only way he’s going to be able to track progress is to measure the back independently.

Well the Modal tuning is all about measuring and modifying the coupled frequencies so closed soundholes and tapped backs are not relevant unless you intend to cover the soundhole and play the back. It's the top being driven by the strings that the tap testing is intended to approximate.

If the coupled back response is too weak to show up in a Top tap then it is not relevant.

JF has started with a deliberately non live back, does'nt say he wants to change it to live. and has used the same design before with good results, It will take a lot of brace shaving to change it to a live back.

The back freq is the same whether measured from the front, or back. Perhaps you are thinking that I’m suggesting he measures the back without strings, or with the soundhole covered?

I’m not. Measuring the back should be done exactly the same as tapping the top. Strung up, strings muted, soundhole open. Just with the back to the mic, not the top. Well, and tapping the back in the middle lower bout and not the bridge.

His closing sentence in the first post was asking about shaving back braces to get more bass. The only way I know of to get more bass by shaving the back braces is to couple it with the top. Perhaps I’m only assuming that’s what he’s after.

Shaving the back braces will lower the air resonance even with a non coupled back by decreasing the stiffness of the Box, not by changing the back frequency enough to couple.
But it is not going to achieve huge changes as JF has already found.
A non live back is quite a valid design choice, Converting an existing design to live is not a trivial exercise.

The top tap test would possibly be more definitive if it had included taps on the centre of the bridge not just on the wings.

JF is also tapping the back "center just below the waist" not middle of the lower bout

Details matter

Here is a top tap dead center on bridge.



So then I made this little stand (Thank you Trevor)


Zeroed out:


Then less string tension:


This is with a set of D'Addario 10's. I probably did my initial measurement with the phone app using 11's and I bet that would push it closer to 2.0 deg. Or it's possible that this device with the little stand is way more accurate then a phone. Anyway I don't think it's prudent to shave top braces on this one.

My own 'theory' as to what is going on here is that these methods work within reason up to a point. In hindsight it probably was a bad idea to use a very stiff torrified top for such a small guitar. In theory it should not matter so long as you do the acoustic testing to determine the top thickness. If one piece of wood is stiffer then another but you thin them out to the same deflection then they should share similar properties right? Well yes true but within reason. I think this top was so stiff that in order to get to the proper deflection (as compared to the other ones of these I made which came out good) the top had to be thinned so much that it lost something. What that something is I don't know but these are my thoughts on the matter at this point.

If I am going to retop this guitar it seems wise to find a low density less stiff top.

Interesting statement, can one get back coupling with a non coupled back by adding mass to the back and lowering its frequency?

I understand shaving the back braces will lower the air resonance by changing the stiffness of the box. It also changes the backs resonant frequency. While shaving back braces I do not see a change to the Top and Air resonance until the back resonance is close enough to the top. I had thought back resonant frequency was a good metric for the relative stiffness of the back. Practically it is for me, but that is because I am working in a design space where that is true.

_________________
http://www.Harvestmoonguitars.com

Well that tap test central to the bridge is instructive 122, 221 and 424ish as definite peaks, Id say this confirms that the 424 is the cross dipole, reduced in magnitude by the central tapping.What the lower peaks between 221 and 424 are i'm not sure?
How is the back braced? 4 cross braces? that may not give you a definitive single lower bout back resonance either coupled or directly tapped.

Well here’s a thought.

If you can’t take enough off the back to couple it with the top, you have another option.

Your top is in a terrible place at 221hz. 220 is a scale tone and you’re pretty much right on top of it, so it should be changed regardless.

Since you’re considering retopping it anyway, you have nothing to lose by bringing your top monopole down to 213.7. You still have 25% of your bridge rotation budget left, and, taking 7hz off the top will bring out a lot more bass, but not likely put you past 2 degrees. And it might open up the guitar well enough to not need retopping.

Yeah no doubt. If anything it will be fun to experiment with.

Wow! It's nice to see people using Luthier Lab. I put a lot of work into it. If you have any questions or suggestions, let me know. You can do that directly from the app or create a topic on the forum and we can all talk about it.