By Jim McCarthy | June 4, 2009
Hi all – it’s been just over a week since my last post, and I’ve managed to put together just a few hours this week to work on the 5 octave marimba project. Progress has been a mix of straight forward “business as usual” work for fast progress, and experimental playing around to develop some exciting brand new techniques. I’ve been thinking about these for years now, and finally this project provides the perfect time to work out the details and get them implemented.
Firstly a little more progress on the frame… After some careful measuring and bar layouts, I worked out the best position for the longer more diagonal horizontal support struts. It is a matter of positioning the bars where they need to be, then finding the best “average line” for the diagonal strut so that it lies under as many of the exact nodes as possible. The process takes quite a while, but the end result is really important – the diagonal struts can be cut, and the outer vertical legs on the big end section can be positioned and attached. Below is a photo of the result.
After that step, I put in some basic grunt work on the frame carpentry. This was cosmetic stuff. Firstly mixing PVA glue with sawdust generated by sanding scrap frame timber to form a similar colour timber putty. This was used to fill any tiny gaps in the joints and the holes left by the countersunk screw heads. The putty needs to be left “proud” to allow for shrinkage when drying, then sanded back. I also put in some time in general with the disc sander and orbital sander to clean up all the joints and surfaces – sanding them smoother and rounding out the hard corners. The result is hinted at in the photo below.
I also found time to start cranking up the next major stage of building – the resonators. In order to do this I needed to finalize some design details, then go shopping! I purchased a bunch of stuff I knew for sure that I’d need, and also a whole bunch of stuff for some experiments I wanted to work on.
An easy and fast step was to cut and tune the resonators for the notes f#3 to C5 – two and a half octaves. These are the easiest as they use just two different pipe diameters and are in the easy medium range. Also they have no bends. These literally took only a couple of hours to cut and tune as I’ve done heaps of these before. Cut to a bit shorter than final length – clean pipe – put on end cap and quickly tune with stoboscope – mark cap position with texta – apply glue – quickly slide cap on and fianlize tuning before glue dries. That’s pretty much it! These resonators can be seen in their basic rough stage below.
The resonators below these will be on a second bank and involve some interesting bends and coupling to bar variations etc… so these are left for now. The top octave of resonators is also not included here, but that is the area where I am experimenting with some new ideas. For those of you who have tried creating resonators for the top octave – C6 to C7 – before – you will know that there are some difficulties. These tubes are hard to tune with any sort of tuning device simply because it is difficult to get them to produce a distinct tone. It is also difficult to simply measure and cut these accurately. Tiny differences in length make a significant differnce in pitch, and these variations are in the order that is common in a cut for the diameter of pipe that matches the bar. In theory it is simple to calculate and measure the perfect length – quarter wavelength minus the “end correction”… in practise however the end correction varies because of the proximity of the bar and varying arch shape… this is part of the coupling to the bar which is critical in the top octave. Whilst it IS possible to tune these tubes by ear and careful listening for sympathetic resonance under the actual bars – I’ve never been happy with the dubious accuracy of that solution. The new idea I’ve been working on to alleviate this issue should make producing accurate resonators possible – whether simply measuring, them or tuning them with a stroboscope or tuner. It does this by taking the variation in “end correction” out of the equation as much as possible.
More on that next post!
Last night I temporarily bolted the frame together, and layed out all the bars on the struts in an accurate way. This involved finding the best average nodal lines under the bars on the support struts and getting all the bars square to the center line of the frame. Then marking the support struts with a pencil between all the bars, and also ruling new pencil lines on the bars for the string holes to be drilled in the bars. This is a critical step because once this is done, the final tuning can be done which will make checking resonator effect possible at the instrument extremities, to achieve the best possible bar/tube coupling.
Coming up this week… I’m hoping to find the time for the following… Drilling the bars and fine tuning them. producing and tuning the top octave of resonators using my new “clustered” design. Finalizing details for the lower resonator banks and tuning the pipes.
All in all, I’m hoping to have te building guide for this 5 octave marimba available now in less than a month from www.makeamarimba.com.