Experimental Development – Nov ’11

Since May ’11 I’ve been experimenting with different approaches to analogue tonoscope design and documenting my approaches,  outputs and conclusions.

I should précis this record of six months of experimenting in my studio with a declaration that I have intentionally taken the ‘high road’…

Although I’m aware that there are several well documented examples of successful DIY cymatic rigs that I could have referred to and emulated, I have a personal agenda:

  •   I want a custom tool designed specifically for the job – a device that behaves the way I want;
  •   I’m trying to develop an intimacy with the device so that my interaction with it becomes more natural and fluid – like a musical instrument;
  •   I want to engage in a design dialogue with the device – allowing it to inform and shape the design process.

But perhaps most importantly, I’m trying to find that small, (possibly) unnoticed detail or ‘never quite configured in this particular way before’ set of circumstances that open up new lines of enquiry and so may lead to fresh thinking.

To help me realise this I’m trying to apply a research framework of my own design – an ‘artistic experimental method’ – based on a handful of artistic axioms that seem central to my work. By settling on working definitions for these creative tenets and thinking about how I might apply them in useful ways I’m trying to embed them into my practice and drive my creative enquiry forward. In the longer term, and when the Augmented Tonoscope actually starts to generate outputs, I will probably be more heedful of notions of beauty, concerns about aesthetics and issues of authorship. But for this stage of the study – the design, fabrication and crafting of a responsive analogue tonoscope – I seem to be far more focussed on shaping process, misusing technology and courting serendipity.


  •   I’m trying to be cognisant  of my own creative process and more aware of its dynamic, pulse and trajectory – to let these factors guide my progress iteratively and in intended rather than incidental ways;
  •   I’m exploring technology misuse – “the use of technological systems and products in ways that were never intended by their manufacturers” – by experimenting with unconventional materials, repurposing tools, modifying commercial products and using ‘found’ items for purposes other than that for which they were designed;
  •   and I’m trying to master “the art of making the unsought finding” and introduce more serendipity into my experiments – not just in the hope of the ‘happy accident’ but to open myself to the possibility of the latent potentialities within my devices and even to be systematic about searching for the ‘unfound’.

Have I achieved this? Read on…

Early experiments – May 2011

There were a couple of notable precursors to the tonoscope prototypes displayed at the State Of Play… exposition at RIBA Hub in July ’11:

  •   a silicon rubber baking tray containing both water and medium grade (size not quality) reflective glass beads sitting directly inside a 15” Celestion Truvox 1525 speaker;
  •   a silicon baking tray containing glycerol sitting directly inside a 15” Celestion Truvox 1525 speaker.

The main problem I faced – since I hadn’t yet discovered the temporary adhesive properties of Museum Gel – was that the silicon baking tray would ‘bounce’ around in the speaker, particularly at low frequencies – hence my efforts to weigh it down with volumes of water and glycerol. Even then, at specific pitches probably corresponding to a natural resonant frequencies it seemed to overdrive and distort. 

Despite these issues, these early experiments showed some interesting effects which deserve to be revisited now that I’ve worked though some of the early design limitations.

The glycerol was pretty much unaffected at anything over 50Hz, but below this, and certainly down to frequencies of 5Hz, the surface of the glycerol began to undulate and pulse and seemingly defy gravity by concentrating towards the centre of the baking tray. While there were none of the detailed surface patterns often seen with water or alcohol the reflection of the LED torch I was using to illuminate the experiment appeared as a distinct Lissajous like figure which shifted with frequency.

I have some rough documentation of the glass beads in the water – and though there’s some natural resonance of the baking tray with it’s associated distortion – the beads do create shifting geometric patterns as the frequency changes – at least in the centre of the tray. I do think there’s something interesting in the idea of a ’suspension’ of glass beads in a liquid medium that warrants more experimentation here.

Studio documentation of early tests of analogue tonoscope prototypes.

For ‘State of Play…’ – July 4-10th 2011

I displayed two tonoscope versions at the State Of Play…’exposition at RIBA Hub, Manchester, UK:

  •   a 15” Celestion Truvox 1525 speaker with a plastic tray ‘glued’ to its rim using Museum Gel and containing fine grade (size not quality) reflective glass beads;
  •   a Rolen Star 30Watt Audio Transducer with a 200mm x 3mm acrylic disk screwed to it’s mounting plate on top of which sat a silicon rubber baking tray containing 99% isopropyl alcohol.

Both sat on custom-made acrylic anti-vibration bases using a mixture of anti-vibration gel and neoprene ring and bush mounts. The anti-vibration base for the far heavier Celestion Truvox speaker actually had two acrylic rings supported by neoprene ring and bush mounts – essentially a floating plate on a floating plate. These DIY anti-vibration bases seemed to be quite effective at absorbing vibration from the speaker and transducer – albeit there was some transfer to the table at frequencies below 20Hz – presumably at the natural resonating frequency of the dampening materials when everything on the table kinda wobbled a little 😉

Both of these prototype tonoscopes were simultaneously driven by the output from my custom Sine Wave Generator via a Sure 2x100Wt TK2050 Class D amplifier.

The Rolen Star set up produced noticeable patterns on the surface of the alcohol – some essentially static – but only at frequencies above 300Hz by which pitch they were quite complex.

While the set up with the Celestion Truvox certainly produced a wide range of dynamic cymatic forms it failed to produce any static modal standing wave patterns. The fine grade reflective glass beads were mobile and fluid and exhibited an almost triple point phase quality – seeming to behave simultaneously as a solid, liquid and vapour – which made them quite mesmerising to watch. 

Medium and coarse grade reflective glass beads produced little discernible pattern when sweeping through frequency ranges.

It seems as if the fine grade reflective glass beads tend to either clump together in fluid globules behaving much like the Lycopodium spores in Hans Jenny’s experiments or – and presumably these are the finest of the particles – accumulate around the maximum rather then the minimum vibrational nodes on the plate –  behaving more like a liquid according to Hans Jenny.

I suspect the plastic tray – which looks as if its been produced through an injection moulding process – is not a particularly good diaphragm, having too many tensions, deformations and variations in thickness in the plastic to behave consistently across its surface. 

I also reckon that an impedance mismatch between the Sure 2x100Wt (@ 4 Ohm) TK2050 Class D amplifier and the 8 Ohm rated Rolen Star and Celestion Truvox 1525 didn’t help.

I concluded from these early prototypes that I needed to experiment with alternative diaphragm materials.

A demonstration of The Augmented Tonoscope at State Of Play…, RIBA Hub, Manchester, UK – an exposition of ‘work in progress’ organised by the PPR Group as part of the PARC NorthWest Carnival.

Documentation in my ‘PhD: The Augmented Tonoscope’ set on Flickr.

For Understanding VISUAL Music ’11 – August 24-26th 2011

Prior to the Understanding VISUAL Music 11 conference in Montreal in late August [although I was actually unable to attend UVM 11 due to a family bereavement] I experimented with a couple of next stage configurations.

  •   15” embroidery hoop with weather balloon latex

At a workshop at the Madlab Music Hack Day in March ’11 a researcher from the University of Salford and University of Southamption Aeolus Outreach Programme demonstrated the use of embroidery hoops to stretch rubber sheet taught. By pushing these rubber diaphragms down over the rim of an empty ice-cream carton fashioned into a crude analogue tonoscope using additional PFET plumbing pipe and components and then singing down the tube, cymatic patterns could be seen on the stretched rubber.

So I bought some 3’ party balloons and a weather balloon with the intention of cutting them up use to the rubber for diaphragms.

Unable to find a commercially available 15” embroidery hoop to fit on the Celestion Truvox 1525 I used the Metal Workshop facilities at MMU to make one out of steel and then used it to stretch a diaphragm of latex rubber from the weather balloon. Despite sticking a strip of insulation tape around the outside edge of the inner ring and inside edge of the outer hoop to try and provide a better grip on the rubber it was difficult to achieve an even tensioning of the latex which was delicate and tore quite easily. But once realised this then sat directly on the rim of the Celestion Truvox and was weighty enough to require no additional fixing.

A sprinkling of fine grade reflective glass beads did produce dynamic cymatic forms – but again no discernible static modal standing wave patterns.

  •   8” speaker with acrylic tube and 3’ rubber balloon diaphragm

Although I’d thought about an alternative approach of fitting the Celestion Truvox 1525 into a clear acrylic tube with a diaphragm stretched over the top, the high cost of an acrylic cylinder of sufficient diameter made this unviable – so I decided to test the principle using a scaled down version of the acrylic tube + speaker combination and with the cheapest components I could find on ebay.co.uk…

I smoothed the uneven machine cut rims of the acrylic tube using the sanding blocks and drawn-air table of the Wood Workshop at MMU and then added 4mm rubber edging strip to the top rim to provide friction for the circle of rubber cut from the 3’ party balloon. I then aligned, marked and cut away the outer edge of the foam mounting strip on the rim of the speaker so that the acrylic tube would sit snugly on top of it. Initially I used insulation tape to stick the rubber to the outside wall of the acrylic tube then pull it tight and stick down the opposite side gradually tightening the rubber over the end of the tube. Finally I laser cut a couple of rings of 6mm acrylic to sit snugly around the edge of the diaphragm, hoping to provide a ‘lip’ that would prevent the glass beads from falling off the edge (though it didn’t work too well). 

Similar to previous experiments a sprinkling of fine grade reflective glass beads did produce dynamic cymatic forms – but again no discernible static modal standing wave patterns.

Documentation in my ‘PhD: The Augmented Tonoscope’ set on Flickr.

Achieving fixed static cymatic patterns now seemed a priority, so I decided to try a switch of visualising material to something altogether different – common salt. I thought that it’s larger grain size and generally cuboid form might add an element of ‘friction’ that better supported the formation of static modal wave patterns. I suspected the performance of the fine grade reflective glass beads was just too ‘fluid’ since they were very small and probably too spheroid – and previous experiments had shown generally poor results with the medium and course grade glass beads.

Post Understanding VISUAL Music ’11

I’d noted from the previous experiment that the rubber from the party ballon had a variable thickness and was also somewhat deformed, so I thought I’d try a different mechanism to keep it stretched taught over the top of the acrylic tube – and settled on rubber bands. Although a little awkward to manipulate (they sprang off the rim several times), once I’d worked out how to use them they allowed me to slowly work my way around the edge of the rubber sheet gradually tensioning it more and more. I’ve since bought thicker tent guy rope rubber bands to do the same job but a little more neatly.

Using salt I did manage to produce some of my first distinct static cymatic patterns – though it don’t occur to me at the time to systematically log those specific frequencies they occur at as documented in this video on Vimeo.

But irregularities in the patterns suggest that uneven tensioning – partly due to the irregularities in the rubber from the ballon but mostly because there was no mechanism to realise this systematically – affected its behaviour as an effective diaphragm.

Using salt I did manage to produce some of my first distinct static cymatic patterns…

Testing the principle of an acrylic tube + 8″ speaker combination with rubber from a 3′ party balloon stretched taught over the top of the acrylic tube using rubber bands.

For Seeing Sound 2 – 28-30th October 2011

  •   400mm clear acrylic dome

While looking for a 380mm diameter acrylic tube I came across the far more affordable vacuum formed 400mm clear acrylic dome. Measurement of the Celestion Truvox 1525 suggested it would just fit inside. Also the dome’s 1cm wide ~3mm thick flange seemed an ideal mounting rim for a diaphragm. The dome also turned out to sit perfectly well – with the addition of 4mm rubber edging strip – in the anti-vibration base I’d originally made for the Celestion Truvox 1525 too.

Somewhat disappointed with the response and delicacy of the weather balloon rubber – as well as realising its natural latex colour made it a poor contrast for the glass beads or salt – I bought some .45mm black latex rubber sheet. I then stretched this over the top of the acrylic dome using a mixture of mini ratchet clamps and short sections of the rubber edging strip to hold it in place. 

While I did manage to create sufficient tension to produce distinct static cymatic patterns – and of similar form to the 200mm earlier prototype but at significantly lower frequencies – it just wasn’t possible to tension the rubber sheet properly. Also the salt, being hydroscopic, quickly became damp and ‘sticky’ in the studio’s humidity making the effects difficult to reproduce.

I did try to address the problem of creating sufficient tension in the rubber – and conceived of a fairly complex system for tensioning the latex as a diaphragm across the top of the acrylic dome. 

This involved glueing a couple of reinforcing rings of rubber onto the outer edge of a circular sheet with latex adhesive, then using a special tool to insert a number of regularly spaced 4mm eyelets around the outer edge. I then planned to laser cut a ring of 6mm acrylic that just cleared the diameter of the dome but not its flange and which also had regularly spaced holes matching the eyelets in the rubber sheet – so that short M3 machine screws could be bolted onto it with their threads pointing downwards. I could then evenly tension the circular latex rubber sheet by stretching it over the rim and slipping the eyelets over the screw shafts underneath. 

I even envisaged a number of regularly spaced wing nuts glued onto threaded shafts passing through rivet threads in the acrylic ring with flanged ends on the opposite side that would allow me to gradually separate the ring from the flange of the dome and provide even more control over the tensioning of the latex rubber… 

While pondering how to realise the mechanics of this system – and with input from John Hyatt slowly sinking in – it suddenly dawned on me that humanity has been successfully stretching and tensioning vibrating diaphragms over resonant chambers for millennia.  I didn’t need to ‘reinvent the wheel’ – I just needed to buy a drum.

  •   16” floor tom 

A 16” floor tom seemed an ideal choice – the diameter of the shell being almost exactly equal to the diameter of the acrylic dome. 

For my initial testing I didn’t want to fix the dome to the bottom of the drum so just removed the bottom drum skin and ran a strip of adhesive backed neoprene rubber along the bottom edge of the drum shell. The acrylic dome with Celestion Truvox 1525 speaker mounted inside it sat on an anti-vibration base in a 60cm square black plastic potting tray and the floor tom stood on its three legs immediately above it. By adjusting the height of the legs the floor tom rested on top of and created a snug fit to the flange of the acrylic dome – and with further careful adjustment to the height of the legs I managed to get the top skin – now replaced by a black suede Remo drum head – almost perfectly horizontal.

I do plan to make this a more permanent set up by fixing the dome to the bottom of the drum using the existing metal mounting ring – the inside diameter of which clears the widest diameter of the dome but not the flange. By drilling six holes in the flange of the acrylic dome to match the holes in the mounting ring I reckon a set of longer lugs will fix the dome to the bottom of the drum via the existing shell fittings. 

I also decided to return to the medium grade glass beads as medium and tried to generate distinct, static cymatic patterns on the drum skin – which I finally achieved and with a level of success I feel confident enough to call a “proof of concept”…

I made a series of short documentation videos showing how:

  1.   I could use the tuning lugs on the floor tom to tighten and loosen the drum skin – essentially adjusting the tension of the diaphragm –  and so affect the detail of the cymatic pattern. This not only confirmed that control over the tensioning of the diaphragm could result in a high degree of fine tuning of the visible cymatic patterns and forms – but also indicated (to my satisfaction) that the Augmented Tonoscope, like any instrument, would require tuning before use;
  2.   I scanned through the frequency range selecting those pitches which resulted in a distinct cymatic pattern on the drum skin. I then programmed these frequencies back into the Sine Wave Generator (SWG) as a rough and ready ‘cymatic scale’, playing these tones back out via the Softpot Linear touch sensor in its ‘ keyboard’ mode to see the corresponding shift between distinct cymatic patterns on the skin;
  3.   I selected two distinctive tones and then demonstrated how the various ‘tweens’ generated by Andy Brown’s Arduino easing functions animation library changed not just the way that the SWG moved between these frequencies in a series of distinctly different portamentos to the ear – but also as a visible difference in the transitions between the corresponding distinct cymatic patterns. This modest experiment was actually a significant milestone in realising the subtleties and nuances that I hope the Augmented Tonoscope should be able to effect.

Using the tuning lugs on the floor tom to tighten and loosen the drum skin – essentially adjusting the tension of the diaphragm – to affect the detail of the cymatic pattern.

A rough and ready ‘cymatic scale’, playing tones from the DIY Arduino/AD9835 Sine Wave Generator via its ‘keyboard’ mode to see the corresponding shift between distinct cymatic patterns on the drum head.

Demonstrating how the the various ‘tweens’ generated by Andy Brown’s Arduino easing functions animation library changed not just the way that the SWG moved between frequencies in a series of subtle but distinctly different portamentos to the ear – but also a visible difference in the transitions between the corresponding distinct cymatic patterns.

I now plan to make a smaller but more refined table-top version of the16” floor tom prototype… so I’ve since bought a second-hand 13” piccolo snare and new black suede Remo drum head and ultra thin clear bottom batter to test which might perform best as a diaphragm for creating cymatic patterns. I intend to drive this via a second-hand Vibe Slick 12” 1200 Watt woofer and FLI UNDERGROUND 720 Watt 4 channel amplifier.


 Download this as PDF from my public WebDisk or read via Scribid