View Single Post
  #25  
Old 10-07-2023, 11:44 AM
misterg misterg is offline
Registered User
 
Join Date: Aug 2023
Location: North Wales, UK
Posts: 67
Default

Quote:
Originally Posted by Fathand View Post
Looks like you got the process down. If it helps your confidence, I've heard that Indian Rosewood is the easiest wood to bend.
Thank you, I can well believe it - the rosewood just seemed to turn to toffee when it got hot enough. I was surprised how long it stayed workable, too.

Anyway - humidity isn't playing ball here, and it's still too mild to put the house heating on, so I cobbled together an improvised wood store from a large cardboard box. It has a computer fan and a couple of power resistors inside (running off 12V) and keeps the humidity inside the box at around 50%. It's just big enough to get the top, back, brace wood and a few other bits inside.





(Incidentally, I can recommend the little thermopro TP357 humidity meter - it gives a graph on your phone over bluetooth.)

I'm still working up courage to try bending the sides, so I continued to work on the top instead:

I used my CNC router again to cut a recess for the rosette.



I hogged the bulk out with a conventional spiral flute bit, then used a down-cutting bit to gradually machine away the OD and ID. This shows the difference in the finish and amount of fuzz on the spruce (ID is down-cutting, OD is conventional):



I kept going until I could fit the rosette and purflling in. Having seen people struggle with the wood swelling during glue-up, I went an extra 0.1mm - in hindsight, I don't think it was necessary.



I glued everything in and clamped it up while still on the machine. After the glue had set for a few hours, I used the CNC to deck the rosette down to about 0.2mm proud of the top.



Next day I took it level with the top using a scraper



I then proceeded to thickness the top - the plates had been sanded with something of a wedge, and it took a while to get them to an even thickness. I was using a dial gauge to measure thickness again:



The shavings that came off the top were beautiful - thin enough to read through!





I was aiming for the 2.8mm on the plan, but there was still some unevenness when I got there. Working this out, and a bit of sanding took the general thickness down to about 2.7mm (0.106"). At 2.8mm, the top still had a noticeable tone when held at the nodal point and tapped, but this had pretty much gone at 2.7mm. I have heard someone use the word 'flooffy' in the connection with top thicknessing - I would say that this is the perfect description of what happened to my top. It still very much has the 'sheet metal' sound when you hold it by the edges of the lower bout and shake it.

This is how the thickness ended up (the pencil numbers are the thickness deviation from 2.7mm in 0.01mm units) - there are a couple of very thin areas in the upper bout (roughly 2.6mm / 0.102") but I think this is near to areas that are braced, so I'm not unduly worried about that - the lower bout thickness seems to have come out quite uniform.



I sanded it to 240 grit (front and back) and put it back on the CNC to cut out the sound hole.



It seems to have come out OK cosmetically.





It's back in its box now, stickered up while I agonise over whether I've gone too far thinning it out... (there'd be a smiley here, but I'm at my image quota for this post!)


Last edited by misterg; 10-07-2023 at 11:56 AM.
Reply With Quote