20130426
I glued the braces to the back and start the process of closing the box.
Marking where to glue was definitely a problem.
I started with the center brace, marking where it should glue and glue it.
After that I apply some hide glue to the 1st back brace in the body and close the box onto it.
After some time when I lifted the back the 1st brace come off with it.
I heated with some hair dryer to reactivate the glue and clamp it properly. After both glue are more or less set, I did the same with the 3rd brace (LB)
Unfortunately this back brace fits tightly to the pocket and doesn't want to come off when I lifted it.
I had to glue according to the glue mark.
After the glue dried, I try fitting into the box and it was quite well fitted.
Except the LB brace was skew to the sides, and I had to trim the ends a bit.
Now I notice the back was off center by a bit, I choice to adjust the brace or the pocket, so I chose to adjust the ends of the braces.
I also shape the cross-section of the braces.
Well the back is almost done, I had to glue some finger brace to complete the active back design.
Similarly for 3rd back brace (LB) but this one has a problem it's pretty tight in the pocket so I had to estimate where to glue based on the glue mark.
20130429
I glued the finger braces for the back to complete the active back design as per Trevor Gore.
I did shape the finger brace for a slight curve on my radius stick.
Just use simple weight will do the trick. The base is teh radius stick to impart a slight curve on the back lengthwise
20130515
The box is closed.
The procedure is pretty much the same as previous closing.
But I did do a dry run before carrying out the actual one.
I applied the hide glue to the rims and heat it up with the hair dryer and close the box and apply the clamps.
It was quite fast work; but with the dry run everything worked out smoothly.
Following that, I trim the edge overhang and check out the resonances.
The main top resonance was nice and so was the main back but the main air was higher than I expected.
This was attributed to the wider dentellones I used for this build.
I trim the rim thickness a bit and it drop a few Hz but I can't bring it near my target of G#2. It remain as A#2 for now.
No comments:
Post a Comment