On Super low pitched Marimba notes - (discussions with Michael Sparrow)

-----Original Message-----
From: Michael Sparrow
Sent: Wednesday, 12 May 2004 7:42 AM
To: jim@percussionclinic.com
Subject: Marimba Resonators

I recently discovered your website and have found your advice on instrument building to be of great value. I'm working on a Theatre project for which I'm planning to build a single bass marimba bar with a frequency somewhere around 24-26Hz. The thing is that the bar will be mounted in the deck of the stage (level with the stage) with only a few inches below to accommodate a resonator. I thought about using the open space under the stage as a cavity resonator but I don't think I'll get a good result. I'm planning for the bar to be 23cm wide by 120cm long and I'm hoping to use 15cm PVC for the resonator. I plan to use an L-shaped PVC joint just below the bar to run the sound into a 1.7m section of PVC that will have a U-shaped section linking it to another 1.7m section of PVC with an adjustable end cap. I'm worried that one 15cm resonator won't create enough amplitude to fill the theatre so I was thinking about having two identical resonators side by side. My questions are 1) will two bends in the resonator drastically reduce the amplitude as opposed to a straight 3.4m tube? 2) will having two identical resonators boost the sound or will they take sound away from one another?

Lastly, I've heard conflicting reports about where the fundamental nodes are in a marimba bar and I'm guessing that that has something to do with the cut-out below the center of the bar. I've read three different percentages saying that the nodes are at 20%, 22.5%, or 25% of the bar. Which is correct? Should I tune the fundamental before deciding exactly where the nodes are?

Thanks so much for your help,
Michael Sparrow

Making Marimbas, Xylophones & Vibraphones is now Easy

Get These Comprehensive Plans & Instructions with Video!

A project like this is problematical to say the least - but also right up my alley! You probably came across one of my beasts online called GUS - a similar 1-note marimba with a frequency of around 25Hz. Now in many ways I made Gus as ideally as possible, and whilst it worked, there were still many problems getting an acceptable result in a performance situation - even with amplification. (amplifying 25 hz is not that easy)

Firstly your bar:-
The dimensions are ones you might JUST get away with assuming you are using the proper timber (one of the traditional rosewoods or African Padauk) I would even then tend to make it wider. The width of the bar is what determines amplitude (amongst other things) With Gus - I mounted it vertically because you need as much free vibration as you can, and with a bar that heavy, and vibrating that slowly, you tend to dampen the vibration quite a bit with the bar's own weight, and even create a slight difference between the speed (therefore frequency) of the upwards movement an the downwards movement. With horizontal mounting, it will come down to having a bar which vibrates really easily (usually meaning lighter and not too thick in the middle) and a really good suspension system for the mounting. You don't want the suspension too flexible as then this gets its own frequency - but soft enough to allow good vibration and prevent the clanking sounds. Is the bar mounted horizontally so it can be danced on or similar? If so - careful, as it will be fragile in the middle, and a rosewood or Padauk bar of that size will cost a small fortune.

2ndly - your resonator/s:-
L joints or even U joints are not that big a deal - they don't effect the amplitude much at all although you do want to keep them as far away from the open end as possible as the air is moving more at the open end, so that's where the friction is - sounds like that will be a problem for you - but try - it shouldn't be that big a deal especially with big diameter pipes. The main problem with joints is that they make the length measurements trickier. The length doesn't change, its just harder to measure round corners. In theory you think its not a problem, but just try it - especially with bigger pipes. Having two resonators will work better than one in this instance - or even go for three or four. Better still go for bigger pipe. 150mm is really not likely to be up to the task. Think about it like this: What's louder - a violin or a trombone? ok so if we make it 2 violins which is louder? You need quite a few violins before you start to compete with the trombone. Same with the resonators - one big pipe is more efficient than multiple small ones. The more air moving in resonators the louder the sound, but BEWARE! the note will actually be shorter as well as you are taking more energy away from the bar. In practise, with a note that low and long - I don't think it will be an issue. You will also need a fair amount of potential adjustment in that end cap and/or the joins - just to allow for air temp. changes let alone tuning the thing in the first place - don't forget to allow for end correction in your calculation of pipe length. Also make sure your joins are all air tight especially that end cap - some PVC tape or gaffa tape is good for temp sealing after tuning for the night. The end cap also needs to be pretty rigid - remember that it is forcing a nodal position in the pipe, so you don't want the air to be able to move at all - which means no flexing - any flexing of the end cap is air movement lost at the open end.

All of the stuff I've said so far is about getting maximum volume from your note - believe me you will need every bit you can get. I'm not sure how much knowledge you have about this kind of thing, but if you didn't already know - a marimba bar's note has pretty much all its energy at the fundamental frequency (ie. 24-26Hz in this case) Well that's right down there where human hearing is just not that sensitive. To make people hear it you need just truckloads of sound energy - just to hear it at all. Its easy to get fooled by other instruments that produce low notes, but other instruments make their sound with a more complete set of upper harmonics which are related to the fundamental. For example the lowest note on a piano is around 27.5 hz fundamentally - but the human ear doesn't actually hear this - there's no way that small string is radiating enough energy at that frequency. The human brain is very clever though because it hears all the upper harmonics that the string is also producing and it calculates the ratios between them and puts back in the fundamental for us inside our imaginations. I other words we think we hear it. This is part of the psycho-acoustic effect. A marimba note pretty much just relies on the fundamental inside the resonator - and in fact the upper partials that we tune in the bar are tuned so that they DON'T also resonate in the resonator. All of that being said - a great deal of how well this kind of thing works is dependant on the room it is in, and the placement of both instrument and audience. Its sometimes hard to predict the effects too. A theatre should have enough size, but ideally there should be lots of reflective surfaces, and not so many parallel ones to introduce standing wave problems. Also be aware from a musical or compositional perspective, that it may take a bit of time for the sound to propagate. We usually need to hear at least 3 or 4 full waves before we recognise its frequency - at 25Hz that's getting to over a second.

Certainly there's lots to think about - but its all good fun stuff.

The nodes for the fundamental vary a little with lots of things - probably accounting for the varied reports you have heard. Timber type and bar thickness can make a bit of a difference, and the exact position can actually move a bit after tuning, so yes don't drill the holes until after a rough tuning has been done. You will need to do the final tuning after the drilling though. The best way to find the node is to get the bar horizontal and suspended a bit - even if you just get a mate to hold it from the sides or roughly rig up a couple of horizontal ropes for the bar to sit on. Sprinkle some of the sawdust that you are likely to have after rough tuning over the nodal areas and start gently tapping the bar in the middle to get it vibrating. You might also need to blow gently. What tends to happen is that the dust starts to gather at the nodal points because that is the only part of the bar that is not moving and causing the dust to dissipate. You have to play with it a bit, bit you get an exact node by doing this. Do it for each end, because they might be slightly different measurements from the ends. With a one-note instrument you don't have to worry about matching up the holes with the next bar, so you can make it ideal.

Hope this helps..
Keep in touch - and if you have the time/inclination take some photos along the way - I'd love to see how it all works out.
Also feel free to ask if you have any questions.

Jim McCarthy

For more help on instrument building you can email Jim.

Back to ARTICLES list