Back to the drawing board?

[Skip]

Make the top wider than the hole it fits in or the hole smaller than the top side to side , support it at the bridge supports by capping the supports. The down pressure that is causing the buckling is converted to a more sideways pressure, towards the pin blocks. It's like some bridges and dams are built. Or try supporting the top on all four edges without capping the bridge supports. Example, assemble a small box and cut a 'top' out of a milk container. Cut the top so it fits inside top to bottom and is a bit oversize side to side. try pushing the top through the box. It won't go without distorting or buckling. Putting supports in a couple of places will make it even harder to push through. Try the same thing with a 'top' that fits, just side support then with all four sides supported. If the sides of the box flex towards each other enough the 'top' can/could buckle downward unless all four edges are supported.

[mrchips]

You dont need BOTH the side rods and the wood rails. Either one alone will do the job. Get rid of one or the other. The thinking behind gluing the wood side rails to the top is based on the thinking the top wont expand or shrink as much as the rails that will move the pinblocks back and forth so there should be less tuning shift due to humidity changes. The reality is that is false thinking. The stings are anchored to the pinblocks, NOT to the rods or anything else on the top. Ignoring string stretch that you get at first with new strings the tension will still change based on the distance between the tuning and hitch pins. Its that tension change that's the real cause of tuning shifts. As the rails expand the anchor points get further apart and the pitch will go up and vice versa. Where the side rails/rods set doesn't change that effect at all. The only way to stop tuning shifts due to humidly changes is to use something that humidity wont affect for the rails and internal supports.

That being said a wood rail setting on top of the top just past the top edge will not add any sideways pressure like a rod setting IN the pinblock/top joint will if that joint is not glued in place. If the top is glued down there the rod wont force the joint open. If that wood rail is seeing a small bit inside the ledge it can add more vibration from the strings to the top giving a bit more volume and possibly some tonal changes. A rod or whatever setting on the ledge or a pinblock wont have much, if any, effect like that.

Sam Risetta has made some tottaly out of carbon fiber except for the pinblock tops and the bridges. These DO NOT shift tuning at all once the initial stretch of the strings is worked out. The one I had the privilege to bang around on for a bit also has excellent sound. The idea of a tune once and forget HD is fantastic. It has me thinking about trying something with fiberglass one of these days.

[Verduin]

So, I can do one of two things(besides making a new top out of a different material): either use the 1/2" wood rail without the delrin rod as a side bridge(and maybe cut it down to 1/4"), or just lay the delrin rod right in the slight gap between the top and the edge of the pin block where it meets the top, and use it as the side bridge...

[mrchips]

DONT lay a rod in the space between the pinblock and top UNLESS the top is glued down there. The down force from the strings will drive it down forcing the top away from the pinblock on both sides. That is likely to cause bowing. It sounds like you have he rod setting on top of whats basically a half inch high bridge glued to the top at the edge with a delrin rod on top. Removing the rod will lower the down force some BUT there will be a tonal change as the wood doent have the same surface hardness. I was thinking you had essentially 2 side bridges side by side. Lowering that rail while keeping the delrin rod in place will lower the down force but as long as that rail is above the ledge on the pinblock it wont make any noticeable difference in the stress situation. The key here is not the height but where its at. If its not over the ledge It WILL bow the top down. It MUST have some support under the top to prevent that. Above the ledge it has support. If needed, you can glue a strip of wood to the inside of the pin block to create a wider ledge. setting over the ledge with some thin felt between the top and the ledge wont lock the top down. The felt lets the top slip easily and smoothly as needed.

No felt and the top will stick until the stresses build up enough to break the friction lock and then it will actuality move in jumps instead of smoothly if there is any movement at all. Two pieces of wood held together with only enough pressure will actually start to self glue together creating whats basically a real weak glue joint. Glue together a couple scraps pieces of wood with some paper in the joint then see how easy the joint will come apart after the glue is fully set up and you will understand how a weak joint acts. The force needed is a function of how dry the wood is and what kind it is. Softer woods stick with less force than harder ones. Note softer and harder here does not mean the same as hardwood like Maple or a soft wood like say pine here. Its just a way of saying RELITAVE surface hardness. Even within whats classed as hardwoods the actual hardness varies among the species.

The more I am finding out about the issues, the more I'm suspecting that crappy underlayment itself is causing the bowing or the top is friction locking instead of moving as it needs to. There is no way to confirm that outside of making an exact duplicate of the top you now have then gluing down the current side rails with no changes in the same exact spot on the new top. The idea is an exact duplicate of the current one changing only the top material. Making 2 changes at once just fogs up solving the problem.

Its near impossible to make a good guess without getting my hands and eyes on that HD. Eyes, ears, and hands on will always work better than any pictures or text base description but that is all that's possible.

Just for the record I DO use a lot of that underlayment for things its not made for like making what is functionally a super strong cardboard box or as a backing in large picture frames. Ive even used it as a cheap paneling in a couple fishing camp cabins too. NONE of the uses involve cases where great stability or reasonable water resistance such as long lasting yard art is needed. Functionally consider it as super strong and dense cardbord, NOT real wood.

[Verduin]

OK...looks like I have a plan. First - glue a strip of wood to the inside edge of the pin blocks to create a wider ledge for the top to sit on. Second - make a different top; I'll probably have to match up two pieces of baltic birch because I can't find any that is over 30 inches and my top is 18 x 33. Third - like the first one, I'll glue the side rail onto the edge of the sound board. Fourth - I'll place a strip of felt on the pin block ledges to facilitate expansion slippage. Fifth - I'll widen/even out the treble/bass bridge holes with a spindle sander to fix the mess I made of 'em trying to accomodate for some off-center string placement.

[mrchips]

I fully understand the bride hole problem. Thats probally the hardest thing to get right from scratch the first time. EVERY wire on a hd runs across a bridge and through the holes at a different angle. There is no exact duplicates either. You have to think in 3 dimensions all at once and that's not easy at all.

I cheated with my scratch built ones by using auto cad inventor a 3d solid modeling drafting program that can build up some really complex stuff like a compete F 16. All I had to do was create the parts and fit them together then rotate things around where I could look essentially down the center of the strings like they were tubes. After I cranked in the data that defines the physical properties of the woods I used like the force needed to break the wood I had all I needed to run some stress checking on it I knew exactly what was going to happen to the HD stress wise long before I even got any wood. Then all I did was print out the prints for the parts and cut them to the exact dimensions there. No wasted wood from guessing at things or mis fitting parts at all. Those prints could be accurate down to about .0001 of an inch but I set the the limit at a 64th of an inch when i did the prints. One of the guys in the machine shop made one from the CNC control code generated from that program too. CNC mills are neat to watch while chewing out parts.

At the time I had the complete resources of the entire engineering, physics and music departments of a fairly large private college at my disposal while working there. Well over a couple million worth of lab stuff available to run tests with too like an environmental chamber where you can control things like temperature, air pressure, and humidity at nearly what NASA is capable of doing. Its was interesting seeing the effects on what the average conditions at 40 thousand feet has on a HD verses at ground level. I am retired now but I still have some limited access to the same things.

The only thing I couldn't test out was the amount of tuning change at normal earth gravity to what it would be at say on the moon. There was no way to create artificial gravity to do it. If you check out the string tension formula part of it is based on the acceleration of earth gravity that is actually 32 feet per second. On the moon gravity its far less so the tuning would shift all else being equal. Its easy to come up with the values by changing that value in the formula but its neater to actually see and hear it.

That is why I know exactly how a HD acts under the stresses. All it takes is to guess a bit from the values on mine to others. There is a limit to how far simulations can get such as there is no way to predict the exact sound you will get out of an instrument but you can get amazingly close but that involves some really serious math best done on a computer with the right software instead of just a calculator. Things like a bit thicker coat of finish does have its effects..

[Verduin]

Fascinating stuff. Interesting how a hobby can branch out to many other disciplines. As a biologist and safety person, I could probably delve into the depths of the habits of wood-boring beetles, toxic/irritant properties of hardwood sawdust, risk assessments for splinters, cuts, abrasions and respiratory ailments in wood shops....the mind boggles

[mrchips]

yup....

life can be like following links on web pages if you let it be. you never know what something can lead to ..

[Verduin]

IF only a pop-up would result in a brand new dulcimer sittin' on the front porch...heheheh

[Verduin]

Well...drawing board cleaned up

Finished the new top, and re-strung...only had to make two new loops. Here she is, sitting on the stand I made, which is a slight modification of one I saw online, in that the uprights are in-line instead of next to the horizontals, and everything's held together with dowels & glue:

Pay no attention to the cheat sheets ;)

Here's a pic of the top viewed from the high strings looking toward the bottom rail - virtually no bowing at all, and there are no braces glued to the soundboard - amazing what the right materials will do:

[Shape note singer]

Hooray!!! Hooray!!! Hooray!!
Three cheers!!!
! Woot! Woot! Woot!

I've been following the saga of your dulcimer build from the start. I'm not a builder, so have not been able to actually provide assistance. Still, been wishing you the best in spite of all the plot twists! At long last...

Congratulations!

[Verduin]

One disadvantage I have found to light-colored tops - the strings are practically invisible. Until I memorize where the notes are, the "cheat sheets" are gonna be mandatory