Testing good examples

You can ask questions here about Trevor and Gerard's exciting new book on Luthiery.

Moderators: kiwigeo, Jeremy D

Craig Bumgarner
Blackwood
Posts: 195
Joined: Thu Apr 11, 2013 10:28 pm
Location: Drayden, MD, USA

Testing good examples

Post by Craig Bumgarner » Tue Jun 04, 2013 10:44 pm

I'm going to a festival in a couple weeks where there will be a LOT of good examples of the Selmer style guitars I build and I would like to document as many as I can using the principals in the Books. I've only just read them and begun using the methods so I'm inexperienced, but I don't want to miss this opportunity to gather data even if I am not ready to interpret it yet. Selmer style guitars seem to vary considerably from the models outlined in the books, so I'm trying to establish a base line using known good examples to work from later on. I'll probably have only ~ 15 minutes with each and will have to do it on the spot, or a quite spot nearby, so my kit has to relatively portable. Chlandi patterns for instance, are probably out of the question.

What data would you collect? Here's what I'm thinking:
  • Frequency Response Graphs coupled top, uncoupled top and uncoupled back in the strung up, playable condition. I have a laptop set up with mic and VA, been tapping everything in sight for the last month. I have a variety of 20mm styrofoam sound hole plugs in the necessary shapes.
  • Deflection test so as to calculate Specific Mobility from deflection and FR of uncoupled top.
  • Plate thicknesses via Hacklinger Gauge
  • Quick evaluation of bracing with mirror, light and sketchbook
  • Checklist of various physical characteristics, like scale, size, weight, soundhole size & location, body dimensions, etc.
I'd welcome any suggestions, thanks.
Craig Bumgarner

Bumgarner Guitar Blog

User avatar
Mike Thomas
Blackwood
Posts: 226
Joined: Fri Jan 30, 2009 1:34 pm
Location: Adelaide

Re: Testing good examples

Post by Mike Thomas » Wed Jun 05, 2013 8:32 am

I think angle of pliage, bridge height (and weight if you could get it), and action would be useful.
Mike Thomas


"There are some enterprises in which a careful disorderliness is the true method"

User avatar
kiwigeo
Admin
Posts: 10582
Joined: Sat Sep 29, 2007 5:57 pm
Location: Adelaide, Sth Australia

Re: Testing good examples

Post by kiwigeo » Wed Jun 05, 2013 2:09 pm

Ive often thought of setting up a library of Chladni and tap test results here on the forum. Not sure how much work would be involved but IMHO it would be a valuable resource.
Martin

Craig Bumgarner
Blackwood
Posts: 195
Joined: Thu Apr 11, 2013 10:28 pm
Location: Drayden, MD, USA

Re: Testing good examples

Post by Craig Bumgarner » Thu Jun 06, 2013 6:36 am

Mike, Yes, those are all important physical characteristics I usually document. Break angle of the strings over the bridge is something I also measure as it is functional end result of neck angle, fingerboard thickness and bridge height, and it is quick & easy to do. I usually take a neck section profile, thickness and width in a couple places and this has proved invaluable over time.

I'm pretty comfortable with the physical documentation. What I'm hoping someone will chime in on is if there is anything I can quickly do with Gore & Gilet design principals besides FRCs and deflection/specific mobility to document good examples as a means of dialing in the sound of my own builds.

Martin, yes, a Chlandi library would be handy wouldn't it? I must admit I haven't gotten set up for testing Chladni patterns yet. I need to do this soon as I'm having a hard time understanding my FRC peaks beyond the air body resonance and the top monopole resonance without them.

I'd be glad to share here the FRC and Specific Mobility data I gather on the Selmer types I'll be looking at in a couple weeks, assuming I get something worthwhile.
Craig Bumgarner

Bumgarner Guitar Blog

User avatar
Tom West
Myrtle
Posts: 70
Joined: Thu Jun 25, 2009 7:40 am
Location: Nova Scotia Canada

Re: Testing good examples

Post by Tom West » Sun Jun 09, 2013 8:44 pm

I'm not so sure most folks would like you giving their guitars the once over as you described. But good luck in your adventure, some good data may be acquired.
Tom
The person who has never made a mistake has never made anything....!

User avatar
kiwigeo
Admin
Posts: 10582
Joined: Sat Sep 29, 2007 5:57 pm
Location: Adelaide, Sth Australia

Re: Testing good examples

Post by kiwigeo » Sun Jun 09, 2013 9:40 pm

Tom West wrote:I'm not so sure most folks would like you giving their guitars the once over as you described. But good luck in your adventure, some good data may be acquired.
Tom
I don't see why many of the regulars who post up screeds of photos and detailed descriptions of their builds would baulk at posting up some Chladni and spectrum analysis results. I've found that generally two sorts of people lurk around luthier's forums.....1. people who are happy freely sharing their experiences and 2. people who are often very experienced builders and are happy posting up a few glitzy pics of their builds but then proceed to belittle the "amateurs" like myself and eventually blow their stack and get banned.

Just my thoughts on the matter :D
Martin

User avatar
Tom West
Myrtle
Posts: 70
Joined: Thu Jun 25, 2009 7:40 am
Location: Nova Scotia Canada

Re: Testing good examples

Post by Tom West » Sun Jun 09, 2013 11:25 pm

My point was that someone other then themselves would be doing the data collection and handling their guitars. Tons of folks pass out data on these type of forums and we are all the better for it. Also not sure where the belittling is coming from.......???
Tom
The person who has never made a mistake has never made anything....!

simonm
Blackwood
Posts: 176
Joined: Mon Sep 07, 2009 7:09 am

Re: Testing good examples

Post by simonm » Mon Jun 10, 2013 8:51 am

You will either have to know the folks or be very diplomatic before anyone lets you poke a mirror inside or run a hacklinger over the surface :-) Its not so much that they are worried about you discovering their dark secrets - more that they will be worried about your hand slipping and causing a bunch of damage to the finish or the soundholes. Damage=lost income. Could be worth contacting the "victims" in advance to get them used to the idea that you would like to make some measurements when you meet them.

User avatar
kiwigeo
Admin
Posts: 10582
Joined: Sat Sep 29, 2007 5:57 pm
Location: Adelaide, Sth Australia

Re: Testing good examples

Post by kiwigeo » Mon Jun 10, 2013 9:19 am

simonm wrote:You will either have to know the folks or be very diplomatic before anyone lets you poke a mirror inside or run a hacklinger over the surface :-) Its not so much that they are worried about you discovering their dark secrets - more that they will be worried about your hand slipping and causing a bunch of damage to the finish or the soundholes. Damage=lost income. Could be worth contacting the "victims" in advance to get them used to the idea that you would like to make some measurements when you meet them.
I was aiming more at the builders of the instruments doing the testing and then submitting information. There are quite a few members doing Chladni and spectrum analysis on their builds so the data is there.
Martin

User avatar
kiwigeo
Admin
Posts: 10582
Joined: Sat Sep 29, 2007 5:57 pm
Location: Adelaide, Sth Australia

Re: Testing good examples

Post by kiwigeo » Mon Jun 10, 2013 9:30 am

Tom West wrote:My point was that someone other then themselves would be doing the data collection and handling their guitars. Tons of folks pass out data on these type of forums and we are all the better for it. Also not sure where the belittling is coming from.......???
Tom
Tom,

It wasnt made clear in my OP but I was talking more about members engaged in builds and already doing testing on their instruments submitting the data for collation. I'm far too busy with my day job to be running around gathering data on other people's instruments :)

Re belittling behaviour...if you've been on this forum for as long as myself and a few other members youll have witnessed a few "bigshots" burst onto this forum full of promises to share their knowledge with others. Eventually the tune changes and it becomes obvious they are simply here to show off their superior skills and belittle those of us who dont build for a living and dont have that much experience (Im in that category). I wont make any further comment on the matter as its off topic and simply a fact of life when youre part of a forum. Also note my comment is not directed at anybody in this thread...its simply a comment based on my experiences as a moderator on this forum.
Martin

Craig Bumgarner
Blackwood
Posts: 195
Joined: Thu Apr 11, 2013 10:28 pm
Location: Drayden, MD, USA

Re: Testing good examples

Post by Craig Bumgarner » Tue Jun 11, 2013 1:37 am

Hi,

Just wanted to say that the instruments I would be gathering data on are vintage instruments 50-70 years old and their builders are now departed. To me, to gather and publish information on these does not seem inappropriate. The festival I'm attending is a players festival, not a builder's festival. Yeah, I can imagine that walking up to builder and asking if I could have a "good look under the hood" of his guitars might be over the the top.

This begs the question where the line between simple observation and reverse engineering a guitar of an active, living builder might lie. And the ethics of publicizing this information. If someone were to reverse engineer one of my guitars and blast it all over the Internet, I would be highly flattered and amused, but I can see how not everyone would feel that way, so while I might look & gather when I can, I'll leave it to the builders themselves to publicize.

Craig
Craig Bumgarner

Bumgarner Guitar Blog

User avatar
Mike Thomas
Blackwood
Posts: 226
Joined: Fri Jan 30, 2009 1:34 pm
Location: Adelaide

Re: Testing good examples

Post by Mike Thomas » Tue Jun 11, 2013 5:17 pm

Craig, I would see no issue at all with you taking all the measurements, it is practical to take, of player's instruments. You do have to have credibility, and of course the trust of the players, and you clearly have that. Owners of $10,000,000 Strads don't seem to object, if the violin is being measured by the right people. And, after all, it is in players' interest that modern makers can produce instruments that are equal to what they regard as the best of the past.
Mike Thomas


"There are some enterprises in which a careful disorderliness is the true method"

Craig Bumgarner
Blackwood
Posts: 195
Joined: Thu Apr 11, 2013 10:28 pm
Location: Drayden, MD, USA

Re: Testing good examples

Post by Craig Bumgarner » Thu Jul 04, 2013 1:53 am

Okay, here is most of what I collected on twelve Selmer style guitars at the recent Django in June festival in Northampton, Mass. Five of these were guitars I had built (CB) and brought to the festival.
Mobility, et.al spreadsheet.jpg
Some notes:

My subjective rating, first column, is indeed very subjective and is based solely on what I think about the way the guitar sounds in relation to the requirements of the style: Loud, responsive, cutting, tight woody bass, never muddy. The emphasis is more on the fundamental than the harmonic. These guitars do not sound pretty, they are acoustic swing jazz instruments. Guitars rated in the 9-10 range are among the very best vintage or modern. Guitars below 5 are either common, failures or played out. Most players are very happy with a guitar that rates 6 or more, 9-10 usually induces a fair amount of jaw dropping.

The rest of the columns are pretty self explanatory. If anyone would like to see the FRCs on any of this, let me know, I have them all. JeffHigh, Trevor & I have been talking about measuring Monopole (Specific) Mobility and there is some agreement that my MM numbers generally run higher than Trevor's, so comparison to guitars other than what is listed above is not really fair.

It appears to me that the best guitars have the best mobility numbers. Other than that, it is hard for me to make conclusions. I welcome any and all interpretations. My main interest in doing this was to establish some base lines for Selmer style guitars and it seems this is a start.

Regards the guitars, the seven that are not mine are truly an interesting bunch. The modern builder #1 not a household name, but he is both a great builder and player for Holland. He could have sold this guitar five times at the festival but in the end a friend of mine bought it and so I got to measure it some. This guitar is close in sound of Django's Selmer 503 that we hear on the recordings from the early 1940s than any I've ever heard.

The 40s Mystery (my name for it) is possibly a Busato or a Joseph Dimauro, no label, but clearly from Paris in the late 40s. I have pictures of both George Brassens and Felix Leclerc playing this same model. This guitar weighs just a little over 3 lbs. and is a cannon. I recently put a new fingerboard on for the owner and delivered it to him at the festival.

Busatos are amongst the most desired gypsy guitars. The 1940s Busato Gran Model was very good, but I was not that impressed with either the 1950s Bustato or Selmer #862 (yes, a real Henri Selmer, one of the last built, 1952, rosewood neck, five braces) . Both seemed heavy, overbuilt and perhaps played out.

You might also take a look at my CB #03, last in the line. This guitar is one of my earliest builds. It was okay at the time and is probably better than a 3, but I wanted to put it at the bottom of the list because of the oddly high mobility number. It was so high that I redid the measurements, but this seemed okay. What you will notice is the particularly high deflection. That is quite so and is likely because the cedar top on this guitar is much thinner than most and this deflection is what is driving the high mobility number. Problem is, it doesn't sound like it. It just average in response, bass is strong as one would expect with the thin top, but FRCs are in the middle with all the rest. I'm perplexed. Is it possible that there is an upper limit to MM where going over gives no improvement?
Craig Bumgarner

Bumgarner Guitar Blog

jeffhigh
Blackwood
Posts: 1536
Joined: Thu Feb 07, 2008 5:50 am
Location: Caves Beach, NSW
Contact:

Re: Testing good examples

Post by jeffhigh » Thu Jul 04, 2013 7:12 am

Very interesting Craig
A couple of questions
Were they all small mouth long scale?
The Grand Busato has a wider body? any others non standard width?
Were you able to observe any variations to the top and back bracing?
cheers
Jeff

Craig Bumgarner
Blackwood
Posts: 195
Joined: Thu Apr 11, 2013 10:28 pm
Location: Drayden, MD, USA

Re: Testing good examples

Post by Craig Bumgarner » Thu Jul 04, 2013 7:54 am

jeffhigh wrote:Very interesting Craig
A couple of questions
Were they all small mouth long scale?
The Grand Busato has a wider body? any others non standard width?
Were you able to observe any variations to the top and back bracing?
cheers
Jeff
Sound holes: The Joseph DiMauro (the elder) guitar has what is referred to as a heart shaped hole, more of a triangle with curved sides, about as large as a hole in a standard flatop, ~ 100mm across. It was one of his trademarks, I don't think anyone else used this, even his brother Antoine or his brother's son Joseph the younger who both used many of his other ideas. My #7 & 9 have a middle size horizontal "D" hole, patterned after J. Castelluccia guitars from the 1950's. You can see these on my blog, I love the shape. The size is nice as it gives the player a little more volume to hear himself without getting tubby like the large D holes. The rest are small oval (ellipses actually) holes. Most are the standard 54mm x 70mm. The 40s Mystery sound hole is 60 x 83 and my #11 is 60 x 76.

Scales: all 670mm except for the Busatos and Favino which are 675mm. I originally measured the Mystery guitar at a sloppy 660mm but this was not right. It was actually 670mm spacing everywhere except at the zero fret which was too close to the first fret by 4mm! Similar fret spacing issues happen all the time on these old guitars, even the good ones.

All were 14 fret to the body joint. All with floating bridges and tailpieces.

Yes, both the Busato Grand Models and the DiMaruo are 410mm wide at the lower bout The Favino is 420mm. The rest are standard Selmer size, 400mm. Body depths are all around 95 to 100mm.

Braces & Backs:
  • Modern builder #1: very Selmer 503 all the way. Four major ladder braces, four minor ones (2 under bridge, 2 alone side fingerboard on angle. Solid mahogany back and sides, back has three 10mm x 16mm braces, just like Selmer
  • DiMauro: four ladder braces, with the little cross braces under the bridge. Back is molded into an carved archtop shape, no braces, pretty stiff.
  • 40s Mystery: Three braces, bridge sits about 30 uphill of the bottom brace. Molded, arched back, no braces.
  • Busatos: Five braces ala Selmer, but lighter. Backs both molded into archtop shapes
  • Selmer 862: If you know Francois Charle's plans of Selmer 807, this one is just like it. Five major braces, four minor, laminated back and sides.
  • Favino: Four braces on the top with the little cross braces under the bridge, larger body, otherewise pretty much like Selmer 503.


Top braces on almost all the guitars I have looked at are usually 5, 4 or 3 braces on the top, always ladder braces. The backs are usually laminated, either as flat plates pulled into an arch by braces or molded into a curved arch or an archtop guitar or violin back shape. The molded back often have very little bracing and range from scary flexible like on the 40s Mystery guitar to rock solid like the Busatos. It is hard to say whether one bracing system or back configuration makes a better guitar. In general, the guitars I tend to like have three or four braces and that is what I use. My #9 is three braces, #3, 7 and 11 are four and #1 (my first, which I still have, its my gig guitar) has five ala Selmer.
Craig Bumgarner

Bumgarner Guitar Blog

jeffhigh
Blackwood
Posts: 1536
Joined: Thu Feb 07, 2008 5:50 am
Location: Caves Beach, NSW
Contact:

Re: Testing good examples

Post by jeffhigh » Thu Jul 04, 2013 2:32 pm

Thanks for that detail Craig, there is a lot to consider there.
From those results, what direction will you be heading in on your next build?

User avatar
Trevor Gore
Blackwood
Posts: 1605
Joined: Mon Jun 20, 2011 8:11 pm

Re: Testing good examples

Post by Trevor Gore » Fri Jul 05, 2013 12:00 am

Great work, Craig. Thanks for sharing.

I think we might have discussed this before, regarding the zero count for K and m. K should be 70,071 newtons per metre and m should be 0.043 kg to be in SI units. As you have it, the "errors" cancel and you end up with 18.2 milliseconds per kilogram (18.2 * 10^-3 s/kg) for the monopole mobility.

User avatar
Nick
Blackwood
Posts: 3641
Joined: Thu Feb 26, 2009 11:20 am
Location: Christchurch, New Zealand
Contact:

Re: Testing good examples

Post by Nick » Fri Jul 05, 2013 6:36 am

Thanks for posting results Craig. I'm not into the numbers & maths side of building....yet! but as a little bit of a Selmer Maccaferri builder myself some of your numbers are interesting reading. There's quite a variance in top thicknesses (I build mine lighter than all of your samples :oops: ) this is one thing I notice for a start! Once I've read (and digested) Trevor & Gerard's book all this will make better sense and I'll be able to interpret the numbers better but just wanted to say thanks.
"Jesus Loves You."
Nice to hear in church but not in a Mexican prison.

User avatar
Mike Thomas
Blackwood
Posts: 226
Joined: Fri Jan 30, 2009 1:34 pm
Location: Adelaide

Re: Testing good examples

Post by Mike Thomas » Fri Jul 05, 2013 2:47 pm

Like Nick, I very much appreciate what you're doing here. And like Nick, I know how much more it will mean to me after I've read and digested what Trevor has written. I have, I'm ashamed to say, not yet got the book (but I shall), largely because, having read the wrong things at university, I feel at some disadvantage with my maths skills.
One of the great things about what you are doing is giving those of us who don't have access to good examples of these guitars, some notion of what makes them tick.
Last edited by Mike Thomas on Fri Jul 05, 2013 2:49 pm, edited 1 time in total.
Mike Thomas


"There are some enterprises in which a careful disorderliness is the true method"

User avatar
Mike Thomas
Blackwood
Posts: 226
Joined: Fri Jan 30, 2009 1:34 pm
Location: Adelaide

Re: Testing good examples

Post by Mike Thomas » Fri Jul 05, 2013 2:48 pm

Sorry, double post.
Mike Thomas


"There are some enterprises in which a careful disorderliness is the true method"

User avatar
Trevor Gore
Blackwood
Posts: 1605
Joined: Mon Jun 20, 2011 8:11 pm

Re: Testing good examples

Post by Trevor Gore » Fri Jul 05, 2013 6:33 pm

Mike Thomas wrote:... having read the wrong things at university, I feel at some disadvantage with my maths skills.
Mike Thomas wrote:Sorry, double post.
But surely you can count to two...

:wink:

Sorry, couldn't resist that on a Friday night.

User avatar
Mike Thomas
Blackwood
Posts: 226
Joined: Fri Jan 30, 2009 1:34 pm
Location: Adelaide

Re: Testing good examples

Post by Mike Thomas » Fri Jul 05, 2013 7:49 pm

Barely, Trevor. :?
Mike Thomas


"There are some enterprises in which a careful disorderliness is the true method"

mqbernardo
Myrtle
Posts: 63
Joined: Wed Aug 03, 2011 8:34 am

Re: Testing good examples

Post by mqbernardo » Wed Jul 10, 2013 8:14 pm

Craig, you´re the man! collecting and publishing these data is very generous of you!

thanks,
Miguel.

Craig Bumgarner
Blackwood
Posts: 195
Joined: Thu Apr 11, 2013 10:28 pm
Location: Drayden, MD, USA

Re: Testing good examples

Post by Craig Bumgarner » Fri Nov 08, 2013 6:22 am

Craig Bumgarner wrote:........ You might also take a look at my CB #03, last in the line. This guitar is one of my earliest builds. It was okay at the time and is probably better than a 3, but I wanted to put it at the bottom of the list because of the oddly high mobility number. It was so high that I redid the measurements, but this seemed okay. What you will notice is the particularly high deflection. That is quite so and is likely because the cedar top on this guitar is much thinner than most and this deflection is what is driving the high mobility number. Problem is, it doesn't sound like it. It just average in response, bass is strong as one would expect with the thin top, but FRCs are in the middle with all the rest. I'm perplexed. Is it possible that there is an upper limit to MM where going over gives no improvement?

Just to close the loop on this..... I have this CB#3 guitar back in my shop today. It is being sold and the buyer wanted me to look it over before fully committing to it. I redid the deflection and the top FRCs. The FRCs are the same as was tested last summer, but the deflection is half of what I measured earlier.

My guess is I didn't have the load squarely seated on the bridge. Selmer guitars have archtop like bridges with very narrow edges where the strings go across. Unlike a fixed bridge, there is nowhere else to take the deflection measurement. I set a narrow block of plex across the strings, on the edge to land my dial indicator shaft. My guess is the block was off center and the deflection load was partially being taken by the strings. My bad! I need to cook up a little adapter that will self center on the edge.

I have been "calibrating" my thumb pressure on guitar tops and the first thing I did when I took this one out of the case was to push down on the top. .18mm deflection? No way!

The .09mm deflection measured today yields a mobility of 13.1, much more in keeping with the way this guitar sounds. This confirms its position at the bottom of my list posted earlier, both in subjective assessment and in mobility. Glad to have that little discrepancy cleared up.

I had mentioned the cedar top on this guitar is thin, very thin, 2.0mm and even a bit less in places which is why I initially thought the deflection number seemed reasonable, but in hindsight, it has a overbuilt bracing structure so the assembly is still overly stiff and heavier than need be.
Craig Bumgarner

Bumgarner Guitar Blog

blackalex1952
Blackwood
Posts: 776
Joined: Tue Aug 27, 2013 6:36 pm
Location: North East Victoria

Re: Testing good examples

Post by blackalex1952 » Sat Jul 04, 2015 1:40 pm

Does anybody have any tips for doing chladni tests on archtops and selmacs? Due to the arching, the tea leaves, poppie seeds or whatever tend to roll off the guitar when the top is vibrating!
"Everything I say on the topic is based solely upon inexperience and assumption!"

Post Reply

Who is online

Users browsing this forum: No registered users and 25 guests