Tap Tuning F5 mandolins.

[rsnelson0984]

With new and better equipment for tap tuning I've been starting again building Katy Mandolins. www.katymandolins.com For background: I have a Peterson 490 and a 450 strobe tuner. I'm currently working with three soundboards mounted to the rims/necks installed. I've cut the tone bars out of one to check the pitch of the soundboard itself. The board material is dense, from Old Standard Wood. Red Spruce. With the bars out and the center of the board measuring .200" and a .120 ring around the outside one inch in from the sides the board measured A# +10 cents. It was suggested to reduce the board thickness by R.S. and take it too .160". Did that and tapered to match. Reduce the ring to .110". Well guess what it only dropped to A# -30 cents flat. The recommended tone bar setting that I've worked to has been Bass A flat, Treble A#. If my thinking is correct ain't going to happen with a soundboard sitting at A#. And yes I did cut one bar down to check my thinking, of course when the bars pitch stopped changing it was at A #, no bar much left. I must say the board really rings loudly at the A#, crisp sound. Where to go from here with tap tuning on these boards? I know just build the darn things. I have been working in the past to the L. L. scale of A=431, for the purists out there. Ideas please

[HoGo]

With new and better equipment for tap tuning I've been starting again building Katy Mandolins. www.katymandolins.com For background: I have a Peterson 490 and a 450 strobe tuner. I'm currently working with three soundboards mounted to the rims/necks installed. I've cut the tone bars out of one to check the pitch of the soundboard itself. The board material is dense, from Old Standard Wood. Red Spruce. With the bars out and the center of the board measuring .200" and a .120 ring around the outside one inch in from the sides the board measured A# +10 cents. It was suggested to reduce the board thickness by R.S. and take it too .160". Did that and tapered to match. Reduce the ring to .110". Well guess what it only dropped to A# -30 cents flat. The recommended tone bar setting that I've worked to has been Bass A flat, Treble A#. If my thinking is correct ain't going to happen with a soundboard sitting at A#. And yes I did cut one bar down to check my thinking, of course when the bars pitch stopped changing it was at A #, no bar much left. I must say the board really rings loudly at the A#, crisp sound. Where to go from here with tap tuning on these boards? I know just build the darn things. I have been working in the past to the L. L. scale of A=431, for the purists out there. Ideas please

I am LL nut and purist and spent many (20+) years studying construction and evrything even remotely connected to LL and found out that most of the tuning theories that you can find in books or online are just made up by their authors. I'm strictly scientific guy (math and physics) as well as accomplished musician (judged by others) and I believe there IS connection between all the frequancies and densities and other measurable parameters and quality of tone of instrument BUT in last 100 years (since the tap tuning/ chladni etc. theories came to violin world) NOONE has brought ANY kind of exact method AND PROOF that it works. ACTUALLY no one really defined what good sound is, and even the basic premise (that just about everyone seems to accept without proof!!!) that LL signed instruments are special (sound-wise) group of instruments were never even attempted with scientific methods.
From recorded materials ALL we can assume about "tuning" of LL signed mandolins is:
- LL was fond of the popular tap tuning methods back in the day
- He was also fond of other alternative things for tone enhancement (Virzi etc.)
- Prototypes of F-5 were likely tuned bu LL in some way and specs sheet was created for factory workers. The sheet didn't contain ANY tuning info but rather plain measurements - likely taken from the more succesful prototypes. (Could that be reason there is existing F5 pickguard with serial number preceding the first known F-5 - perhaps from failed/rejected or succesfull disasembled F-5 prototype)
- The rest of the LL signed mandolins were just made to the specs with (somewhat) consistent material selection and methods. The graduations were followed just loosely and are typically quite uneven within one instrument. (think fast workmanship without much measuring)
- Some mandolins show excessive material removal from outside (could be just too zealous sanding OR PERHAPS adjusting for tuning of not so succesfull sounding instrument).
- Some LL mandolins show clear manufacturing defects that were ignored or overlooked during production and may have direct affect on sound - like very asymmetrical inside carving (by machine) resulting in very thick one side of the plate and thin other side of plate.

SO, if you want to re-create LL mandolin you can select materials that are as close as possible to LL F-5 (Red spruce, maple - sugar or red, Ebony, pearwood, mahogany for blocks etc., HHG, stain, varnish). Use work methods and specs that are close to original as you can and your result will be close as well, I BET.

[fscotte]

The only correlation I 've seen betwixt my mando builds and others who have confirmed it here, is a tap note of 520hz to 550hz with top and neck attached to the rim. This tap note is prominent and as long as you tap it somwhere on the top away from the edge, that note should be obvious.

Based on your tap note observations before and after further thicknessing, I would take a gander that your arch is too bulbous, or archy. If the center is too archy, it would require you to go dangerously thin. An arch is a very strong design. Small variations in the arch can create drastic differences in the strength and flexibility of the top. I think a more accurate method to measure the top is using deflection. I use a three caliper system to measure the top's flexibility.


Let me add, I know of at least one very prominent builder (we all know his mandos) who has taken the center of the top down to .145" with very stiff spruce, and this is based off an existing Loar. So I say that to say this, it is possible, if your arch is correct for the thinness of the top, and your break angle isnt too high, and you use a full contact bridge, and you feel ok about it, and your spruce is super stiff, you might consider taking the center down a bit more.

Other areas to consider, how thick are you near the fingerboard and the tailpiece? These areas hold a lot of strength for the top and if too thick, can prevent the top from moving. Some careful sanding in those areas can make a top that is too stiff, to one that is just right.

[rsnelson0984]

Thank you for the very thoughtful replies to my post.

[Austin Clark]

Other areas to consider, how thick are you near the fingerboard and the tailpiece? These areas hold a lot of strength for the top and if too thick, can prevent the top from moving. Some careful sanding in those areas can make a top that is too stiff, to one that is just right.

Or make so it bulges and/or collapses.

Warning! Opinion to follow:
This all has been discussed at great length in other threads. Don't fret too much about tuning to exact notes. (Or any, for that matter). They don't appear to be relevant to anything and most makers I know think it is pretty much a waste of time.
Due respect should be paid to to Roger, though, for trying to quantify a method to recreate quality instruments in a time period that Gibson was not exactly making it's best mandolins.....

Stick with what Adrian (Hogo) said. Keep your top and back thicknesses in the general range of a good plan or measurements from a known instrument and you will be fine. Given time and a lot of instruments under your belt you will find what works for YOU to get a consistently good sounding instrument. And maybe even some great sounding ones!

[billhay4]

Ditto what HoGo and Austin said.
Bill

[sunburst]

I've repeatedly said about all I have to say about tuning parts of mandolins to what I consider to be arbitrarily decided frequencies, so I'll leave it at: I tried it and gave it up because I didn't see any use in it.
What I'm not saying is: "Don't bother with tap tuning, it's a waste of time". Just because I concluded that it is a waste of time for me does not mean it is necessarily a waste of time for you to at least try it, so if you are curious and want to tinker with tap tuning, have at it and form your own conclusions. Keep in mind that we, as humans, have a strong tendency to hear what we want to hear. We have a very short memory for sound so it is very difficult to compare sound when we can't do so within seconds. In other words, I can have two mandolins on the bench, pick one up a play a few notes, put it down and play a few noted on the other, and I can easily hear the difference. If I play one and then come back in 20 minutes and play the other, any difference I hear is most likely imagined, unless there as an obvious and huge difference in sound. What does that tell me? That I can't reliably tell the difference in sound before and after a procedure that takes more than a few seconds. So, if I build a mandolin, then build another, I cannot compare the sound of the two because it took longer than a few seconds to build the second mandolin. If I build two at the same time, I can compare them directly to see if I hear a difference. Suppose I arch them differently, of brace them differently, or whatever, and suppose I hear a difference. Can I attribute the difference in sound to the different construction? Maybe, but how to I know it's not because of something else?

Anyway, what I'm saying is: be aware of your limitations as a human being and don't trust what you hear any more than you would any other human being (You know, the ones who install brass bridge pins in their guitar and hear a huge increase in sustain and volume. They're not lying, they think they hear it, and for the human brain, "perception is reality".).

[Jim Hilburn]

When discussing tuning with top and neck glued to the rim it should be noted that Gibson glued the backs on first.

[HoGo]

When discussing tuning with top and neck glued to the rim it should be noted that Gibson glued the backs on first.

True that Jim!
And I agree with John as well. I'm currently making two mandolin from identical materials (all wood matched - cut from the same part of log) to identical specs with just one tiny modification just to verify if it's worth doing it other way. If I can hear difference I can assume some hypothesis (real proof would require more than just one sample, but carefull measurement of the one sample may show something...)

[sunburst]

I did that, Adrian. First I made two identical (they sounded the same), then later changed one thing (red spruce and sitka spruce). Those sounded the same... almost. The slight sound difference, I assume, might be attributable to the different wood, but as you say, the sample size is awfully small.

[fscotte]

What do you fellas think of top+rim+neck = a tap note 520hz plus/minus some? Around Middle C.

John, I've seen you mention in past threads that a few of your mandos tapped within that range, maybe around 550hz? I've also seen another member here (don't recall immediately) suggest 523 hz (middle C), and Peter Coombe, who does some tap tuning say that his F-models were around 530hz to 550 hz. There are others, private conversations too, I've kept notes...

So no, I'm not a tap tuning advocate and I don't even worry about the ranges I've said above, because I measure flexibility. However, if it looks like a duck and its quacks like a duck, it's probably not a goose. There seems to be a pattern here.

I do in fact tap my plates + rim + neck just to see if they stay consistent with this phenomenon, and they do. They start stiff, around 600hz+ and as I zero in on my flex goal the tap note begins to drop and usually ends up around 520-530hz.

For those wanting to tap tune, I think it's something that could use more data. Data, as in ripping the backs off Loars and tap them.

[HoGo]

I glue backs first, then top and neck comes last so I have no data for that.
Back when I measured everything I did chladni patterns of my tops and backs as free plates in various stages and there is at least half dozen most audible frequencies (tap tones) to measure on tops and backs ranging from 100 to 800Hz.

[HoGo]

I did that, Adrian. First I made two identical (they sounded the same), then later changed one thing (red spruce and sitka spruce). Those sounded the same... almost. The slight sound difference, I assume, might be attributable to the different wood, but as you say, the sample size is awfully small.

In my case one of the tops is "roasted", the other normal. Both are from the same log, split one next to the other... Some folks claim huge things about the thermal treated wood... So I want to know more.

[Timbofood]

Can we surmise from Adrian's notation that TT was deemed superfluous by Mr. Loar aside from initial findings? That makes a great deal of sense as far as production might be concerned. Use the notation for tweaking of production measurements. These were made to meet schedules and demand.
What we now have as "Top Level" builders have a certain amount of latitude as far as meeting a schedule which can allow for individual demands. I think I can honestly say that, unless "we" were there, it would be pretty hard to say what was or wasn't done "in the day".

[hank]

OK, My question is the secret to coupling the top to the back with or without Virzi's to complicate an already muddy subject. Do hard tops and backs do well together or do softer backs flex a bit more for the minimum thicknesses that folks want to try without back failures. Today our luthiers are using Redwood, Red Cedar, Alaskan Cypress, Sitka, etc. in place of Red Spruce from the day coupled with Koa, Mango, Walnut, Rosewood, whatever. How do you know when to stick to specs(as per this thread)and when to deviate from specs to allow for the wood's personality(stiffness, weight, etc.)Soft=thicker plate, hard,stiff=thinner plate????

[Marty Jacobson]

There is no secret.
If you use a floppy back with a stiff top, it sounds like A. If you used a stiff back on that same instrument, it would sound like B. If you use a floppy back with a floppy top, t sounds like C.
Except those variables are infinitely variable.
My approach has been to build instruments with extremely stiff, floppy, deep, shallow, steep, hard, soft, dense, light, and whatever other variants I can think of, and to see what the results are.
At this point, I can pretty much anticipate what the instrument will sound like, when working within my process and using my methods. But for a while, the tone of every instrument was a huge surprise to me.

[Ivan Kelsall]

I suppose that it's conceivable that the first few 'F' style tops & backs were tap tuned, as at that point,there was no other criteria to use. Tap tuning was something that Lloyd Loar would have known about & decided to use as a 'starting point'. It maybe that he also came up with the thickness diagrams by measuring each top / back as it was made, & then co-relating the tone of the mandolins when finished, to the measurements taken. If a particular mandolion (or number of mandolins) were heard to be maybe better sounding than others,then the carving measurements for those mandolins might have formed the basis for the carving diagram that Gibson used - with a few variations due to wood density of course.

The 'evolution' of a carving diagram (plan) for the Gibson mandolins seems fairly straight forward = measure the parameters & note the results. When you find the parameters that produce the best results,you use them as the basis for production - with variations as i mentioned before. It may very well be that the guys working on the mandolins were concious of the variations in the density of varying pieces of wood (it would be hard not to IMHO), & adjusted the carving parameters to suit. That would account for some of the carving variations that have been noted when measuring Gibson mandolin tops.

''Exactly how it was done in the begining'',we don't know,but they had to start somewhere, & a ''known technique'' such as tap tuning might very well have been the starting point - why not ?,
Ivan

[fscotte]

There also seems to be strong evidence that they were sanded further after assembly. How else would you explain backs that are thin near the center? How else would you explain the Crusher mando with its thin top?

[HoGo]

''Exactly how it was done in the begining'',we don't know,but they had to start somewhere, & a ''known technique'' such as tap tuning might very well have been the starting point - why not ?,
Ivan

What we can do is called "educated guess".
We can pretty closely replicate their working methods and materials and even the specs from the sheets.
Loar had no one to learn from the TT of mandolins. He probably picked brains of violinmakers around and tried to apply that on mandolins. The specs are actually almost identical to pre-Loar Era F-4, the only difference on backs is slight change of arching (Loar era F-4 had exactly the same backs) and tops are quite close in thickness to F-4 except the part under fingerboard.
Whatever he did we won't know but I guess that workers in the factory couldn't care less about his work, they just got their sheets and did the best job of fulfilling the daily amount of work... We can safely skip the Loars work as Black box and just take the results like the Gibson factory workers. Otherwise we can speculate forever not being able to tell what he really did. He could as well just sip coffee with his fellow violin maker telling him "...just do it like me... I just tell the folks I'm tuning them to real Stradivari formula tap tones and they'll believe your instruments are special and will buy them... while you can easily make them to common numbers..." (PURE SPECULATION)

The process of building F-5 is quite easy to replicate.
- Machine carved tops and backs are smoothed with planes, scrapers and sandpaper so they are close to numbers on sheets.
- Tone bars attached and shaped to exact numbers. F-holes cut (routed to template)
- Sides are bent and blocks cut
- Rim glued in outside form.
- Rim gets the riser block
- Neck (very likely still rectangular block) gets fitted (dovetail machined on both body and neck, probably not down to final fit, yet, but close), neck is marked with fingerboard template lined up with the edge of riser block.
- Profile of neck is cut and trussrod is installed, neck ears are glued on.
- Pre-inlaid headstock overlays are cut to size (several layers at once to template) and bound.
- Fingerboard is made fretted and bound. (in this order)
- Headstock overlay glued onto neck and the shape cut out. (even the order of gluing the binding pieces was standardized)
- Back is glued on rim and glue cleaned.
- Top is glued on together with 15th fret crosspiece(s)
- Body is routed for binding and bound (inside of the scroll was certainly cut before binding) except the heel button.
- Ivory point protectors were added.
- Neck is shaped and glued to body.
- Fretboard extension support glued to top.
- Neck attached and heel button shaped and bound.
- Fretboard attached
- Nut, bridge and tailpiece added and mandolin strung in white and tested. Possibly top or back thicknesses adjusted.
- Mandolin sanded and stained with water stains.
- Few layers of oil warnish added, I'm not convinced they used shellac as sealer - you risk damaging your staining by hand application of shellac over bare stain, oil varnish is much safer in this regard, you don't see evidence of that on Loars or other mandolins ot the era.
- Finish scraped off the bnding
- French polish applied.
- Mandolin reassembled and setup.
- Final inspection and labels glued in.

That's all (I hope).

[HoGo]

Back thin near center can be explained by thickness of the stock. If the raw back material was not thick enough or pattern carver did not adjust depth of routing correctly you'll end up with flat area in the center of back. Often the flat area was not perfectly blended with arching but just the sharp edges sanded smooth (visible on many F mandolins - not just F-5). Next if you rout the inside to nice curve the very center ends up thinner. Even if you blend the outside there will be wood missing in the whole center of back...
Tops were sanded quite aggressively and you can see abruptly thicker area under fingerboard extension where easy sanding was not possible and scooped areas adjacent to the extension.

[Ivan Kelsall]

From Adrian - "..did not adjust depth of routing ..". If we're talking about top / back thinness of ''Gibson'' mandolins,which 'routing' was that ?. That would be correct if routers were used,but as far as we know,the backs were hand carved.
As for 'educated guesses'' - i know them well !. Re. 'tap tuning' - the name is almost self explanatory & i'm sure that LL was familiar with it enough to apply it to his own work in the begining (maybe).

As Gibson employed LL as an acoustic engineer,with the responsibility of ensuring that their instruments were of good build quality & that they sounded good,i'm fairly sure that LL would have had to keep his eye on what was going on in the workshop. Maybe he'd inspect them on a ''batch inspection'' basis = take a certain number from any batch (quantity) & ensure that they looked & sounded ok. That's still done in industry today, & if the quality is maintained from batch to batch,usually,everything's ok. He might not have needed to inspect every one,the ''pre-signed labels'' indicate that that 'might' have been the case. They were there for the builders to put inside. I'm not quite sure that the builders at Gibson would have had such disregard of LL's work as Adrian suggests. They were working directly for him & i feel sure that simple respect for his ideas would ensure that they took note of them - why not ?. It would have been pretty obvious to LL on trying out a mandolin chosen at random,whether the builders were sticking to the plans.

Whatever - it's all in the past !. Whether the Gibson guys tap tuned them or dropped them from a dizzy height is all speculation. Interesting enough from an historic point of view,but with regard to the builders such as we have today - it's what they do that's important IMHO. Some may use tap tuning,some may use other criteria,but as long as we like them,what else is important really ?.

I've no idea at all what criteria Bruce Weber / Tom Ellis or Jiri Lebeda employed in making my 3 instruments - but i sure as h**l like the results !!!,
Ivan

[Clinton Johnson]

Well this thread is fascinating to me and I'm nothing more than a hobbyist with only #6 (in the white)under my belt. But here goes my two cents worth for those who started with the siminoff book years ago as I did, and relative newbies as myself. I never could make sense with the tap tuning part either until this last build I think I may have gotten a little better hold on it. Up until now I've built with mostly specs, feel and intuition(whatever). But on this one I incorporated a technique that a reputable builder uses. In that after I attach the top plate to the rim assembly and tone bars installed, I carved the tone bars down simultaneously until the assembly was ringing on the "C" note. To the OP, I see your doing as I once tried doing, tuning each bar seperatly. Try tuning the ASSEMBLY to the "C" (523ish). Then ofcourse tune the back down to the "D"(293 range) maintaing correct arching. You'll need to find where to "pinch" the plate as your tuning, for it will register differently in each location. This mando is the best that I've built so far and I will continue with this technique in future builds. As I said this is only what worked for me, and I'm a nobody compared to the awesome builders on this forum. But nevertheless that's my story and I'm sticking to it...then as you know the voicing is done so that the mandolin emits the D-D# note.

[fscotte]

I carved the tone bars down simultaneously until the assembly was ringing on the "C" note. To the OP, I see your doing as I once tried doing, tuning each bar seperatly. Try tuning the ASSEMBLY to the "C" (523ish).

Clinton, did you see my post above in reference to tuning the top on rim to a middle C note as well? I'm pretty sure the luthier you are referring is not part of my "C-note database", but I think I know whom you are referring. In addition, I think we can add Siminoff to the C note camp as well as I've either seen him in a video, or maybe one of my private emails with him where he suggested a C note as well.

This C note phenomenon needs more investigation, without the usual "just build it to specs" obligatory responses. That's fine and dandy for someone who wants to build to specs, but there "is something" going on. My personal database is building indicating evidence that this may be a very nice target for those who need to tap tune.

[Clinton Johnson]

Yes Fscotte I've seen your post and I think you were spot on and I've always payed close attention to any information that you've imparted here on the cafe. Very useful information!
I haven't seen any vids with Mr. Siminoff recommending C note tuning, I've only read what's in his book on seperate tuning of tone bars and target tuning of the back plate. He states that the optimum choice is to adjust the bass bar to Ab(G#) and the treble bar to A#(Bb). For some reason this kept leaving me flustrated and confused. It was definitely easier for me to adjust the unit as a whole to the C note and then tune the backboard. The C/D tuning seemed to induce great coupling between plates( at least on this recent build)

[HoGo]

From Adrian - "..did not adjust depth of routing ..". If we're talking about top / back thinness of ''Gibson'' mandolins,which 'routing' was that ?. That would be correct if routers were used,but as far as we know,the backs were hand carved.
As for 'educated guesses'' - i know them well !. Re. 'tap tuning' - the name is almost self explanatory & i'm sure that LL was familiar with it enough to apply it to his own work in the begining (maybe).

As Gibson employed LL as an acoustic engineer,with the responsibility of ensuring that their instruments were of good build quality & that they sounded good,i'm fairly sure that LL would have had to keep his eye on what was going on in the workshop. Maybe he'd inspect them on a ''batch inspection'' basis = take a certain number from any batch (quantity) & ensure that they looked & sounded ok. That's still done in industry today, & if the quality is maintained from batch to batch,usually,everything's ok. He might not have needed to inspect every one,the ''pre-signed labels'' indicate that that 'might' have been the case. They were there for the builders to put inside. I'm not quite sure that the builders at Gibson would have had such disregard of LL's work as Adrian suggests. They were working directly for him & i feel sure that simple respect for his ideas would ensure that they took note of them - why not ?. It would have been pretty obvious to LL on trying out a mandolin chosen at random,whether the builders were sticking to the plans.

Whatever - it's all in the past !. Whether the Gibson guys tap tuned them or dropped them from a dizzy height is all speculation. Interesting enough from an historic point of view,but with regard to the builders such as we have today - it's what they do that's important IMHO. Some may use tap tuning,some may use other criteria,but as long as we like them,what else is important really ?.
Ivan

AFAIK Gibson emplyed large copy carving machnes for carving shaping etc. There's no question to that.
Loar had nothing to base his TT on so he HAD to make (have made in the factory) few prototypes where he fiddled with tap tones just to get some starting figures. There's no way he just told "lets tune this to D!" out of dark and ther was succesfull F-5. Even if he picked brains of violinmakers, the size of violin and its bassbar is so different that any violin tap tones are useless for mandolin and violin has only one bassbar to tune...
They had lots of large machinery going (there were miles of shaft, belts and pulleys running all day long above heads of driving the machinery from central steam engine even before advent of electric motors) and in the noise any tap tuning during workdays would have to be done in different building or after work hours... not very likely in production batch.
Loar was touring musician and spent lots of time away so there may be some randon batches that he also inspected in some stages during building but most likely not many of these.