Wavy Sides

[Joe Farmer]

I seem to be ending up with wavy sides on the rim when it comes time to route the binding channel. I'm working on my 6th guitar and seem to have this issue just about every time. I don't want to sand down or scrape the sides too much to make them straight since I only have 2mm- 2.5 mm side thickness to work with. Putting a straight edge on the sides shows up to 1/2 mm gap that you can see daylight through. Could be high in the middle and then gap on one side or both to the top and bottom plate, or there may be just a high part of the wave in the middle.
I made some spreaders out of 1/4 " threaded rod attached to blocks on both ends to put in the upper and lower bout so I can put on the top and back while everything is still in the mold and hopefully square and sturdy. Cant really tell if the sides are square after putting the kerffing on. The spreaders come out of the body through the sound hole fine after top and back are glued on. However, it doesn't seem to help much with the wavyness (maybe some). By keeping the spreaders in when putting on the top and back plates, it seems that may be putting pressure on the top and back glue joint since that's not where the rim wants to "live" when you glue the plates on. Not sure how that affects the sound, but it seem the top and back would vibrate more naturally if the rims weren't forced straight before gluing. ????? I don't think this would create wavy sides but I don't know. Wouldn't it be best to glue on one plate with the rim in the mold and the other plate with the rim out of the mold?? I'm not sure what else to try to prevent the wavyness.

Could the sides be forced into a "wave" if too much pressure is used in gluing the top and back on? I'm using a go bar deck for that and I try to put just enough pressure to clamp the joint firmly.

Anybody got any ideas of what I'm doing wrong, or not doing something that I should be doing to prevent these constantly wavy sides?
Thanks!

[ken cierp]

Hi Joe,
In all likely hood the wavy sides came off the bending operation that way. Do you use solid wood side reinforcement material? Using fabric along with a water based glue could cause some cupping.

Its doubtful that forcing the rim into submission has any affect on sound quality --- but in fact it can hamper fit and spring back/distortion throughout the build process --- so my recommendation would be to spend all the time necessary to get the rim to fit the mold with minimum resistance. As I have mentioned before, take a look at the factory videos Martin and others. The rim is assemble outside a mold and even the linings are installed with the rim assembly free standing. To this end I also recommend bending the waist curve first (we use a separate fixture like the factories) and also using style specific waist cauls in the bending press.

[Joe Farmer]

Hey Ken,
Thanks for your thoughts. I use 6mm by 3mm side struts, or bracing. 16 of them. 8 each side. Here's my build process:
I have a side bender and bend the sides on it. I also have a matching mold to the pattern in the bender. The sides fit nicely in the mold after bending. I check to make sure the sides are perpendicular in the mold, put in the spreaders, and glue in the neck and tail block. All good at this point. I radius the top and back (40' and 15'). I glue in the kerffing just proud of the rim edges, and do a final radius. I take out the spreaders and install the side struts. Then I cut the channels for the back braces into the rim. I put the spreaders back in (just snug) and glue on the back. I flatten the top of the rim above the upper bout (like in your manual). I cut the channels in the rim for the top braces.
Now, here's where I've tried different things when gluing on the top. 1) Take out the regular spreaders and put in my threaded bar spreaders at the bouts and glue on the top. 2) Take out the regular spreaders and glue on the top. 3) Put in a long threaded bar spreader between the neck and tail block with no bout spreaders and glue on the top. 4) Put in the long bar spreader between the neck and tail block AND put in the bar spreaders in the upper and lower bout and glue on the top.

This current build, I didn't use any spreaders but left the rim in the mold (as with all of them so far) and glued on the top, back glued on first. Result, wavy sides. Do you think too much go bar pressure could do that?

The next guitar, I'll build the rim in the mold to make sure everything is square, install the end blocks, struts and kerfing and do everything else out of the mold while checking for side straightness. If the go bars are scrunching down the sides, it will be easy to tell with it out of the mold.

Thanks for you input!
Joe

[ken cierp]

I've always used rubber bands or a style specific clamping caul when gluing on the top and back. As I've mentioned many times before I am not a fan of go bar decks.

I still believe the waves/cups are a function of the material reacting to the moisture, heat, flexing and subsequent re-acclamation after the bending process. Flat vs quarter sawn materials can be a contributor in wavy distortion issues as well. The Mahogany's especially ribbon can develop severe cupping. Sapele is full of surprises Bob Taylor mentioned that it is the most problematic to bend.

I certainly could be all wet -- so your future observations regarding crush force of the GB deck (are the side reinforcements bending or popping off as well?) will be of interest. Perhaps other GB deck users will chime in.

[Jay McClellan]

I had a similar problem on the last pair of guitars I made, which had backs and sides of curly maple. I cut the sides from a board that came from a family member, and the guitars were made for other family members, so there was sentimental value in using that wood but it was not really "instrument quality" i.e. it wasn't perfectly quartersawn and the growth rings were irregular which I think encouraged it to become wavy as it was bent. I used Super Soft II to keep the curly grain from splitting and bent the sides by hand over a hot iron, and I'm not very experienced so I'm sure my technique contributed to the problem. Anyway I ended up with no breakage and sides that fit the mold pretty well but with quite a bit of waviness across the grain. I improved it a bit by using beefy maple splints (about 5x12 mm) every 2 inches or so, glued on with epoxy since I couldn't get a perfect fit, and substantial clamping pressure with straight heavy cauls inside and out. That did flatten the sides noticeably but I think I still had maybe 1/2 mm of waviness in some spots. The sides started out about 2mm thick so I felt I had enough material to get them flat or close enough.

Because of the heavy curly figure of the wood plus the waviness I used a random-orbit sander to flatten the outsides. Better men might get good results with a scraper, but for me the RO is the tool. I put a bright but cool CFL bulb inside the guitar and stopped sanding frequently so I could turn out all the lights and turn on the bulb inside. That way I could tell when the sides were getting thin and it was time to stop! I couldn't measure the precise thickness that way but I could get a pretty good subjective idea of where they were getting thin, don't know how well it would work in darker woods though. In the end I still had some waviness in a few spots but it was much less than when I started and with a satin finish nobody notices. These were only my second and third guitars so the wiser and more experienced luthiers may have better advice (like don't use difficult wood until you have the skill to bend it well), but I was able to get a decent result with the combination of heavy splints, epoxy, judicious sanding, and a satin finish.

[John J]

Well said Jay. I appreciate the detail of what you did to get through that adventure.

[Joe Farmer]

Yeah, thanks Jay. As Ken pointed out, looks like everything is pointing to the bending. Using a side bender, and letting it cool all the way down after bending, I guess I've always EXPECTED the sides to be straight, but I never checked. I'll definitely check next time I bend a set of sides and try to post back here. I haven't bent anything exotic. Limba, Myrtle, Ovangkol, Walnut, and Rosewood so never any problems.
My whole issue with this "wavyness" is that when I route the binding channel and scrape the binding down to the sides, it gets thin in places. Primarily on the outside edge of the upper bout. The channel looks perfect all the way around after routing so its definitely the wavyness of the sides causing the scraper to scrape at an angle.

Ken, I think I'm going to take your advise and stop using the GB Deck to glue on the top and back. Though it IS really nice and fast. I did the rubber bands on my very first guitar and it worked fine, but it wasn't a lot of fun. And by the way, none of the side struts popped off. I figured that with them beefing up the sides a GB Deck wouldn't be a problem. I've never felt comfortable with its clamping pressure on the rim though.

Once I find the culprit of the wavy sides, I'll come back and post the answer I found.
Thanks to everyone!! Happy building!
Joe