Repairing cracks in violin soundboards: humidifying, splinting, cleating, and types of glue used

John Wetzel - 04:38pm Jul 2, 2000
Surveyor, fiddler,tinkerhead

I am undertaking the repair of my own fiddle, after a nominal amount of past luthiers experience, and a lot of reading on violin construction. I have removed the cracked soundboard without incident, and, once repaired, I will certainly re-glue it in the traditional manner (with hide glue).

My question pertains to the top. I have a very clean crack running from the treble F-hole (fingerboard end) to the purfling at the chinrest mounting point. The crack is near (but not at) the soundpost. The crack was the result of a short fall onto grass (IE not due to drying of the wood). Traditionally, such cracks are repaired with hide glue (a glue weaker than the wood it joins), reinforced by hardwood cleats. Specifically, can anyone tell me why I shouldn't use a "stronger" glue on this joint, since, unlike the seams, this joint is not intended ever to be reopened? Also, why do the cleats (if used at all) need to be hardwood?

Right now, my idea is to use thinner (less viscous) than usual hide glue, theorizing that it has passed the test of time and usage, but augment the glue joint with cleats. I plan to use two hardwood cleats near the soundpost (where the top vibrates very little), and use some very small, light and thin spruce cleats at a few other points, particularly where the crack terminates at the F-hole. I am trying to preserve the previous tone of the instrument as much as possible (it was as close to perfect as they come), hence the lighter (and very much non-standard) softwood cleats.

Any light that you luthiers out there (especially those with violin experience) can share would be greatly appreciated.

I also note with interest that the guitar luthiers I have encountered during my lurking seem to prefer titebond and other modern glues, and I wonder what you folks use in the specific case of a guitar or mandolin soundboard crack. All info will be gratefully accepted!

Geeze John, you got me on this one in a way. I have always used real fine grained spruce for the pleats on a violin belly crack, running opposite the grain direction, and glued where the two opposite corners are aligned directly along the crack (for maximum effectiveness...right?) but I read somewhere not long ago - was it here? - I really don't remember where, but somebody was giving some odd advice regarding these little pleats, such as having the grain run in the same direction as the belly and other such nonsense etc... I was taught to use tiny, fine grained spruce patches, mounted as indicated, and to have the edges beveled so as to run smoothly onto the surfce that they were being mounted on to. Also, the fresh, hot hide glue, properly used IS stronger than the wood that it is being applied to, if properly made and applied. I know, I have experimented with it long enough to know that a piece of wood glued to another piece of wood with hide glue will leave a layer of wood when torn away rather than simply let loose at the glue joint, try it yourself and see. I hate to always sound like a know it all, but I hear some really odd propositions every now and again that make me wonder if the people giving you this advice have ever really done a repair? My advice is to use full strength HOT hide glue (do you have any questions about what makes good Hide glue? ) not from the bottle, on very tight grained spruce patches. You are right in that this is a repair that you want to last the rest of the life of the instrument. I have been making violins off and on since 1976 and have run a buisy instrument repair shop for the last ten. Hopefully someone else will post confirming somewhat what I'm telling you. Thanks for listening. Craig S. Tucker

Sweetie and I go by the rule of thumb that you should use the same wood for cleats as for whatever you are cleating-ie, spruce on the top, maple on the back. Therein lies our samnesses, herein lies our differences, since we've been sweeties (a long time) he's prefered tight grained squares (edges neatly beveled and crossgrained to the crack) and I believe that diamonds are a fiddles best friend-the grain on an angle to the crack and neatly beveled. On the oodles of fiddles that we've taken apart, it seems that others are in the same dilemma, some mended fiddles have three or four different types of cleats in them. We've even seen beautifully carved little circular cleats. All of these seem to have withstood the test of time as long as the cleat was neatly fitted. Like Craig, we use full strength hot hide glue.

Thanks for the input so far. Apparantly my instincts were OK but some of my research was insufficient or misunderstood. I feel much better now. I will just use all spruce cleats on the top, well beveled, glued with hide glue. I do understand not to use bottled hide glue, but to get it dry and use it hot.

I am not having any luck actually finding a hide glue vendor. Many supply houses carry tools and varnish, but not glue. What gives?

Also, I am still not sure about the correct viscosity of glue for a crack repair. Obviously I can use glue of "normal" viscisity for the cleats and to reglue the belly to the ribs. In other words, is it safe to "thin" the glue with hot water, in order to prevent the creation of a gap such as would be caused by too much glue (being as you cannot really clamp this type of joint, or at least, I can't). Any advice on glue consistency is greatly appreciated.

My idea on "clamping" for this job: Make a jig that supports the sides of the belly plate, but leaves acess to the underside of the belly for glue cleanup just after application. Then, after application, and cleaning of excess glue, apply very light amounts of weight to both sides of the belly, with the idea that the downward acting weight will slightly compress the belly, since it is an arch. I have an idea in my head for a simple jig that will work. Any comments, warnings, caveats etc. on the "clamping" aspect will also be much appreciated.

I promise to share my secret methods for removing violin soundboards with explosive charges in a future post. This prevents the soundpost from rattling around while you are working on the top removal, but sometimes has the unfortunate side effect of annoying the neighbors, not to mention that reconstruction is made slightly more difficult if excessive amounts of beer are imbibed whilst setting the charges.

If in doubt, always try your clamping set-up DRY , then, if it works do it with glue. You can make a plaster cast of the outside of the top in which you can clamp the fractured piece and guarantee an even surface (not stepped), and should be able to close any gaps. Be sure ro line the form with waxed paper or plastic wrap to assure easy removal.

Well John, I feel that anyone who would try to bribe a violin repair man or woman with revealing the secret behind the old 'Explosive Charge' method of removing the front plates, must surely be desperate and in need of information in a swift and timely manner. Also special efforts must be made to keep the excessive amounts of beer you speak of from sloshing over and contaminating the hot hide glue... Hide glue is available from the Stewart Mcdonald catalog, or from International Luthiers Supply in Tulsa Ok.

Man, there is almost no way I could reccommend clamping or gluing proceedures without being there and seeing the plate. Many different clamps are usually needed. Less if the belly is not split completely apart and the crack tends to close naturally. be careful that the plate edges align perfectly. Also be careful that the plate is glued with the edges flat on a surface like they will be when the plate is re attached to the rib assembly. You will need a belly crack clamp which is a long arched steel rod with nylon edge grips attached to either end and a threaded nut which tightens one nylon end up to push the plate halves together.

There is also another clamp that goes on the edge of the plate that will align the two parts and also draw them together...I don't want to dissuade you form doing what you want, but I'm not sure that I'd want to do the repair without some of these tools at my disposal... The glue must be made up at least three hours ahead of time. (this is my own time period, as I want my glue to be well melted, well mixed, perfectly transparent, etc. and between 145 and 150 degrees F. I keep mine in a small sealed glass container sitting atop a tray inside a presto adjustable temperature pressure cooker that I picked up years ago at a garage sale. I have the adjustable temperature control set by a mark and taped in position. I use a candy thermometer set in the water bath to check my temperature if there is any question.

The hide glue itself I make by mixing according to the instructions, which are uusually 1 to 1 by weight. since this will result in really thick glue I usually end up adding another half to full volume of additional hot water. The consistency I like to use is sort of like thick cream or slightly thinned warm honey, you just have to aquire some experience with all of this to get it right. squeez out is expected in this type of repair. I was taught to always put enough glue to get some squeeze out. glueing with hide glue is always tricky and no matter how careful you are pre fitting and pre clamping, things always seem to react much differently when the glue hits the wood. I also notice this phenomonon occasionly when I am joining perfectly mated plates - glue goes on, plates go together, Boom! a giant gap suddenly appears! you must get all of this done before the glue gels, which is what? five or ten seconds? I have never timed this so really I'm just guessing, a lot depends on ambient temp. I guess. Clean-up should also be done right away especially on the varnish side of the belly.

I am warning you that what may seem like a simple repair can be a real trial for the beginner. Hopefully, for you, all will go smoothly. Let it all dry for twenty-four hours then remove the clamps, clean up the squeezout that still remains on the bottom and then apply the pleats. Do not apply them over a glue bump or across an uneven surface where one plate is higher or lower. I don't clamp mine but apply the glue, adjust for position and then press down with my fingernail until it 'grips', then move on to the next one. some people prefer clamping. On the other hand I have never had one fail on me either....

Some odd thoughts...the end of the crack that is not open still nreeds to have glue worked in to it, this can be accomplished by using silghtly thinned glue at that point only and 'pumping' the glue down to the very end where it is still connected, by 'pumping' the glue in I mean a a slight opening and closing motion of the crack you will see glue appear at the surface right on the crack on both the top and the bottom of the plate. I'm too tired to add anything at this point,(so, please ignore all the spelling errors as I'm not going to correct myself) if you have further questions, please, ask away...And John, best of luck with your repair. Realize, these are only my opinions and they work for me. I am only suggesting what MIGHT work for you, OK?

Craig, well said. That was lovely and brought tears to my eyeballs. One little thing I do is to warm the plates a bit, this seems to give me a few seconds extra working time (also works well on bridge reglues on guitars and other hide repairs)

Thank you Amy...Here's my handkerchief, wipe your eyes, my dear....(yeah, me with a hanky, ....AS IF!) I do the same thing in the winter here in Rosburg! (Roswell, NM). If I have a critical glue-job to do, such as plate joining, and my tiny shop hasn't heated up yet, I will wrap the plate halves in a heating pad and put them in a cardboard box (where I store the heating pad and rags) ten or fifteen minuites later things are ready to go. You are right, heating the wood is an excellent gluing aid period. Hide glue doesn't like cold. It will gel right up. For plate repairs I have a small weak temp. hand hair dryer that I aim at the crack for five minutes or so. It is always hanging within reach.

Speaking of cleating, what is everyone's opinion of using parchment instead of wood.? Have read the pros and cons. Have been tempted to use parchment but never have.

All right, I know that I am about to upset the violin gods, but, here goes anyway. I understand the use of hide glue for all the areas of a violin that may need to be taken apart later, but, why couldn't you use another glue (yellow wood glue or, god forbid, super glue) to mend a crack in a violin soundboard? I have used both of these glues with great success in the repair of many acoustic guitar soundboards, backs and sides, as well as, a number of violin soundboards.

The water thin super glue is especially useful for cracks that can't be opened to get other glues into. I use hide glue when gluing other parts of the violin: back/top to ribs, neck to body, etc. But I can't see any reason why I shouldn't use modern glues for violin soundboard cracks. Especially super glue. I believe that violin makers of old would have used todays modern glues in building and repair if they had been available to them.

As for the cleats. I don't think it matters whether the grain of the cleat is perpendicular (45 degrees) or at a smaller angle (22 degrees to 45 degrees) in relation to the cracked top or side. Judging by old work I've run across, they both seem to work. I don't think you should ever use a cleat with grain that runs in the same direction as the wood being repaired. It will likely split along the grain just like the original crack.

I'm certainly not as knowlegeable as some folks on this thread and maybe the solution I'm about to mention really doesn't apply to this type of repair. I had my violin {1858 vintage} repaired by a guy here in town some years ago. This gentleman was quite knowledgeable and had been at it some 30 or 40 years. The crack that my violin suffered from was more then likely due to dryness and or age. Since it was a much more maticulious job then I felt compenent at doing , I took it to him. The crack was running in a simaliar area of the top plate. What he did is fit a wedge sliver of spruce into the crack. As well as repairing the top it also would not introduce any stress into the top . { not that anyone here is suggesting stress introduction} To my knowledge he did not use any cleats on the crack. I'm wondering that the cleats would effect the tone by changeing the thicknessing aspects of a violin. I also tend agree with Neil that modern glues Ie titebond would be acceptable for this type of repair. I could be wrong but my observations seem to indicate that hide glue is more brittle then titebond and would therefore think that as the top plate is resonating the lower notes , would allow the top to flex more. I'm just another neophyte trying change that statis, so please inlighten me.

Most of the work of restorers is redoing old repairs. No crack gluing job is forever--all you need to do is look at all the broken center seams in instruments, and there's an ideal gluing situation: fresh wood, the maker gets to choose the joint, the time, and the glue. Additionally, I'd say a very large proportion of the cracks I see have been glued improperly by someone who THOUGHT he was doing just fine--but they really should be redone. . . . but they can't if the wrong glue was used in the first place. To me talking about other glues is like reinventing the wheel. For 450 years violinmakers have used hide glue--it's one of the few things they agree on. Why not just go with the flow on this one?

On the superglue issue, the answer is easy: how do you know what's going to happen in 100 years to that glue? Art history is just filled with "restorers" who had great ideas at the time which turned out not to work. And then when that crack comes open, the next person is going to have a very hard time finding a glue that will hold it, since superglue primes the wood around it. The other problem with superglue is that most of the difficult jobs I've seen glued with it (it often works well on tiny cracks but not big ones) are characterized by the presence of very visible a black line. With hide glue it's possible to do a *completely* invisible repair--completely. And if the first person can't handle the job and make it invisible, the second guy still has a shot at it.

Violin work at the best level is characterized by completely invisible repairs, which is why wedges of new wood aren't usually inserted, and why hide glue is used.

Michael I agree with you 100% on the super glue and you are probably right on the hide glue line , done well, is invisible. I'm not clear though on how one would handle a crack that has separated without introduceing a new pc. of wood. It seems to me if you were to force it together there would be stress on the crack and it would fail again.

Many cracks result from low humidity conditions and can be made to close by re-humidifying the instrument, without any force necessary

OK, I'll add my own opinion about this also. I use hide glue because I see no alternative glue that works better, given the particulars of violin work. Hide glue really does have a proven track record as far as longevity and strength. Why try to fix something that "ain't broke?" Then there is the fact that it (hot hide glue) will remelt any old hide glue that remains in the repair, as in regluing a 'tight' loose seam, where it would be difficult to remove all of the old glue without removing the top in order to get at it, the seam just needs a little extra hot hide glue to help it stay shut, etc. Yellow and white glues are notoriously bad glues to try to reglue to themselves, they dont meld into the prior dried glue as hide glue does...this is a critical and necessary property of hide glue in regard to repairs...Then there is the fact that my customers expect it of me. They know that I use hide glue and hide glue only, they have heard me preach the virtues of hot hide glue, and have heard me curse having to work on instruments that have been white or yellow glued...what a mess that always is. As far as those guys that skew their pleats slightly from perpendicular, I don't know as that has any ill effect on the proceedure at all. I was taught 90 degrees which always seemed optimum to me so that's what I do. I have been aware of the elongated diamond shape that some repairmen use and often thought of using that shape myself since it covers more of the actual crack and has less mass, and so I may even do that one day soon (instead of the usual square), depending on the repair...

Okay Craig you convinced me on virtues of hide glue on the crack repair. Perhaps Mike took care of my other query, but lets assume for a moment that crack can't be rehydrated to fully close the crack. what are you left to do, fill the crack with hide glue?

Jay, I agree with the masses on humidifying, it is very important, however I've over-humidifyed instruments just to have them open up again. Now I try to reasonably humidify (not saturate) the wood and let more time pass, in other words I'll put one humidifyer in the guitar (not two or three) and keep it checked for a week or ten days or more if it deems necessary. This is in the shop where we keep the humidity constant at 45%. This seems to work for me and if at the end of a time like that, the crack hasn't closed then I figure that it is not in the destiny of the guitar (or violin or mandolin) and do the absolute best splint job that I can. The customer needs to know that it's not a quick turnaround job.

Jay, sorry about the delay, I often am called away unexpectedly for days at a time... I ran into a problem recently where I used a top that was given to me by a violin maker who gave up on a violin half way through, since it was nice looking wood, and already joined I decided to go ahead and use it (mistake #1) When I had this violin finished and was getting ready to varnish it, an odd thing happened...It (The belly) split apart at the center seam about two inches up from the saddle to about mid-f hole... I couldn't push this crack together because it was still joined at both extreme ends of the crack etc. Upon questioning the maker I found out that he had no idea how old the wood was, or how it was dried, and that he had used tremendous pressure while gluing to 'shut' an imperfect join....usually he said, he had no problems with this method. Using some advice from this forum I did exactly what Amy suggested in her reply, I filled the split with a splint of like wood. DO NOT simply fill with hide glue. The splint, while a real bitch to fit perfectly, can be nearly invisible...I was very happy with the results.

Craig thanx for yours every one elses response.That extreme preasure to close the joint on the top plates sounds like bad news from the start.Sounds like you had to pay the piper for the other guys lack of thought.I shy away from that kind of joint in cabinet work and in instrument building it's deadly. It would seem though that there is a place for splint application though.I agree that splints are a bitch to fit and although I might attempt one these days,when I was younger and with an violin I loved,I didn't feel compitent to attempt it then.Thats why I took it to some else I thought more compitent then me.

Jay, it sounds like you did the right thing. Even today it is a struggle sometimes to say no to a job that I feel uncomfortable with for some reason.....But I have learned to do just that. The alternative is usually much worse... Also, when dealing with violin plates the only consistant way to glue plates that I have found is to use the fameous "rub joint". Popularized, I believe, by Harry Wake. I put a half plate in the vice joining surface facing up. Coat both faces with hot hide glue, put the other half in place on the half that is in the vice, align the pieces and then rub them together lengthwise while pressing down slightly, at a certin point the plates will 'stick' at which time I will align the plates top and bottom edges and the back (mounting) surfaces, and hold the plates flat together until they 'grip' at which time it is time to let them go...Thats it, no clamping or anything.

In discussing violin glues ,crazy glue came up for it’s ability to penetrate into small cracks. Hot hide glue ,thin and hot with a warmed surface can be percussed into the finest cracks with exceptional results. Just tap it in with your finger.

This is an archived file from the Musical Instrument Makers Forum
Index of Archived Files
MIMF Home