Musical Modes -

This entry describes the events of January 6 to 8 2012.

The Elephant project is an unusual musical project for me.  I have made, and invented, musical instruments, either by myself or in collaboration with other people, particularly my brother Peter and Mr Glyn.  I have also made sculptural items, such as the large metal owl that appears in the background of many of the Elephant pictures.  This is the first project to combine both a large sculptural object and multiple musical instruments.  Furthermore, the instruments I am anticipating for this project are all novel instruments, although I have come to the realisation that I could insinuate a bass guitar into one of the legs.    As, I suppose, a 3d artist, the sculptural Elephant part of this project isn't finished until it is sanded, painted, has a remarkable head, and there is no more refining to do.  A finished and perfected form is the goal.  However, from the perspective of a musician I know that I will tend to stop work on a musical instrument at the point where it makes notes and I can explore the musical possibilities.  If I was commissioning an instrument from a luthier, I would expect it to conform to all the musical requirements, and have smooth edges and a nice coat of paint or varnish.  When building for myself, I want to make notes, hear it sing, and will usually stop at the point of glorious noise.

My first instrument project was in 1984.  A cheap mandolin I had bought self destructed due to the manufacturer cutting through the soundboard bracing to install a pickup.  I found some old wood and brass in my parents garage and made a solid bodied electric mandolin.  I did put some varnish on it at the time, but I had no patience to wait between coats and sand as I wanted to play it.  Finally, in 2010, when I was doing the 8 string bass conversion I put a decent coat of laquer on it.

Electric Mandolin

I've been thinking about and working on the Elephant for a while now, and it isn't making much noise. There is a nice woodblock in one of the legs, and I have spent some time exploring the percussive possibilities, but there are no notes.

As a fabricator I'm excited because, hey, there is a really big Elephant sitting in the conservatory, although I know there is a lot more work to be done. Especially the head.

As a musician I am frustrated.

I need to spend some time, although not too much time, on musical modes.

This project is linked, in many ways, to my earlier trip through Africa 20 years ago.  In the Central African Republic a group of men came up to us attempting to sell us hideous animal heads.  In all likelihood this is what the previous tourists had bought.  They weren't the first group selling such heads, for several kilometres there had been people holding up hideous animal heads in the hope that we would stop and purchase one.  In amongst this group was a guy playing a musical instrument, which I don't think he had any intention to sell, it was just the instrument he played.  Background music for the hideous animal head sellers.

Instrument for sale.

I swapped 2 T shirts (I was moving out of my brightly coloured shirt phase) for this musical instrument, a Kundi (if anyone reading this feels inclined to google image search, please use the term 'kundi harp', or there may be unexpected consequences).

The bush telegraph being what it is, for the rest of our travels through this particular country, the hideous animal heads being held up for offer by people on the road side had been replaced by Kundi's.

I didn't really spend much time learning to play this instrument during the trip, but brought it back to New Zealand with good intentions.  However, MAF quite rightly felt the need to fumigate it and kill off any insects which may have been hiding in the wood.  I think the insects were all that were holding the Kundi together, and since then I have been unable to tune it properly or play it without pieces falling off.

I've been intending to make a new and playable Kundi ever since.

The first step to incorporating a Kundi into the Elephant is to decide where it will fit.  My original plan was to extend one of the hips with a Kundi neck.  Now I have the full size Elephant I can compare the original instrument with the hips to see if this still makes sense, and it does.

The traditional Kundi is made from hollowed out wood.  I am going to make it from cut plywood as it is part of the musical Elephant.

The first step is to form the shape of the neck, which I trace from my original instrument, and cut the shape with the remaining working jigsaw.  The original instrument had hand carved friction pegs.  I bought acoustic guitar machine heads online.  I remove the pegs and tuners from one of the metal plates and use this as a template to drill the holes.

Kundi neck being marked up for holes.

Then I measure the Elephant hip to work up the sides of the Kundi and cut out the appropriate shapes from plywood.  These need to be glued together and left overnight at least before more work is done.

Glued together and drying.

The next morning, the Kundi is half built, but still has not made a sound.  I hope to be able to make some kind of a note soon.  Today, however, I am off to my brother's house to trim the pieces of wood Ivan has donated into shape for the marimba.

I've never made a marimba before, most of the instruments I have built have been stringed, or friction idiophones.  I have done some research on the internet and discovered that a marimba can be built by an 8 year old, from expensive plans, so how hard can it be?

Before we set to the Marimba, we quickly trim up the Kundi on the table saw, to make sure all the angles are square and true.  All it needs is a little sanding, a metal resonating plate, some strings and it will be making noise!

The next step is to cut the wood to thickness using the table saw.  My online research suggests we need thicknesses between 20 and 25mm, widths of 40mm to 65mm and lengths of somewhere between 500mm and 140mm.  This is really too wooly a specification, but I have to start somewhere.  By the end of this process I'll know what I am doing, hopefully, but at this stage I have to make some decisions with limited time and resources.  I did discover a really excellent marimba tuning resource early in the morning but have not had time to read and understand it, so I'm working from several set of partial information.  I'd like to cut a few small pieces of wood, take them away for a day or two, do some experiments and then come back and make the decisions.  That's not the best use of noisy and tricky to set up table saws and planers.  It is better the make a call and run with it.

So, we cut various widths and thicknesses out of the wood we have, leaving one long uncut piece as a spare.

The next step is to try making a marimba bar, and we start with the biggest suggested by my online research, which is a 25mm x 65mm x 500mm bar.  This proves problematic.  The first thing to do is remove some material from the centre, and whichever tool we use this seems difficult.  A note has been struck, and checked against a tuner, but the tone is disappointing.  After a couple of frustrating hours, I realise I need to check a few things rather than waste any more of my, and more importantly my brother's, time.

The most convenient place I know where there is a marimba is KBB on Queen Street, and I will need some marimba mallets, and I saw some other cool percussion things there before christmas, so I head off there to measure a marimba, and buy stuff.  It's 15:30 on a saturday afternoon, should be no problem.  I get to KBB just before 16:00 and they are closed.  I can see people inside as I check the door.  I've been in this situation before, last minute strings, that kind of thing, nocking on the door of the Rockshop just after closing, someone always opens the door, lets me in.  KBB, no, they look up, see a customer trying to come in, look back down, and this is before closing time.

I have a good look at the marimba in the window, go to the rockshop and buy percussion mallets.  Of course, they're all owned by the same parent company, but still.  Don't classical musicians need last minute stuff before gigs?   Or maybe they are all so much more organised than all those other musicians?

So, I return home, make a few measurements, and realise that the big bar we've been working on really is too big.

I return to my brother's with a new specification and recut the wood to 40mm.  If we'd started with this specification we could have made better use of the wood, made less noise, spent less time and had more bars to spare if I mess up tuning them.   I'm not sure how I could have done this better, but feel bad that I didn't do so.

I cut the wood to the lengths I've now worked out to be the only ones that will fit and take them home.  I'm now committed, I have a number of bars, of defined widths, and won't be able to use the noisy tools until at least next weekend.

The next morning I have a better read of the excellent marimba tuning resource I found online.  Having had an attempt yesterday at making a marimba bar, this all started to make sense.  I now know how long my marimba can be, how wide the bars are, and therefore how many bars I can fit into the available space.  This gives me the pitch range assuming different tunings.  Traditional African marimbas are mostly either pentatonic or diatonic tunings, so I decide to go with a diatonic major scale.  This will give me 3 octaves, if I don't ruin any bars, and there aren't any cracks or holes in any of the bars that stop them from working.  From here I can pick an approximate top and bottom pitch and work from there.

I think this technically also means that I am making a Balafon, rather than a Marimba.  A Balafon is a diatonic wooden struck ideophone popular across much of west and central Africa, whereas Marimbas are a chromatic instrument popular in South Africa and South America.   From the point of view of bar  construction this makes very little difference, but it pays to know these things in case of encounters with ethno-musicologists.

My first step is to find the node points for the fundamental notes of each bar.  Tuning for each bar is achieved by carving out a curve between the nodes.  Also, the nodes are where there cord holes will be drilled.  This is all starting to make sense.  Most of the on-line resources I found have formulas to calculate these points but the La Favre documentation suggests that finding the real node point by the following method is a better way.  I don't imagine that musicians in sub-saharan Africa have been using mathematical formulas for the last fifty thousand years or so that they have been playing instruments of this kind, so this method seems like the right one to follow.

Step one.  Sprinkle salt on the bar.  Yes, this line looks like cocaine does in the movies.  You could use cocaine, but if you could afford to use cocaine in this way you wouldn't be making your own marimba, surely.

Sprinkle salt on the bar.

Then hammer the bar with the mallet.  The salt jumps around and converges on the node point for the fundamental.

Mallet, salt and bar.

It soon becomes apparent that this works best when everything is level, so I find a suitable block from an earlier art project, and a leather shim, from another art project, and make sure the platform is level.

I do this for all of the 20+ blocks I have cut.

After a little experimentation I work out a reasonable method to cut the smaller bars based on the tools I have available.  The first step is to mark a curve in the bar.

Marking/checking the curve

I derive this by careful experimentation with the smallest and largest bars, so I can have a reasonable chance of achieving the pitch range I would like.  Then I have to remove as much of the material from the curve as possible as quickly as possible.  For these small bars, and given the tools I have at home, this is to cut lots of small slots to the marked groove using the jigsaw.  For larger bars I will probably follow the curve with the jigsaw.

Cutting slots with the jigsaw

Then remove the excess material using a chisel.

Chisel away the excess

The drum sander comes in handy again to smooth the curve..

Drum sander attachment

Tuning bars

Now that I have a bar which is close to pitch I can use the tuner to check and then either the curved surform or the drum sander to remove additional material.  The drum sander is quicker, but more likely to take off too much material and make the note flat.  

Curved surform.

It takes a while, but by the end of the day I have an octave and a half of approximately tuned bars.  The final tuning will happen after I have rechecked the node points and drilled the holes for the cord which will hold the bars together.

More importantly I have played some simple but important tunes with the bars.  The musical side of me can start exploring the notes, but the bars are still separate and not yet part of an actual instrument, so the builder/fabricator is not hindered in the task of creating a finished musical elephant.