sights and sounds

Like all my pieces, ‘Whistle’ started off as an idea – the sound of steam engine whistles echoing around the landscape and making a journey from town to wilderness. Making it happen is a very different thing. A lot of my work seems very simple from the outside. I like to keep the ideas clear and fairly minimal at heart. Yet that simplicity is often the hardest thing to do. To achieve the illusion of simplicity takes lots of planning, a little bit of technology, a splattering of engineering and an acute attention to detail.

The sound of the whistle had to be just right. On one level the piece sought to trigger something in the memory of those who remember steam engines up the line. That audio memory is very specific, so I had to make sure the whistle would do just that.

First up was finding how that specific whistle worked. The whistle was known as a ‘hooter’ due to its relatively low pitch. The first engines to run on the line had hooter style whistles, but I was keen to replicate the ones used within memory of folk in the village. There’s little detail around about the specifics of ‘hooter’ whistles and engineering drawings were proving to be elusive. Luckily however, a real whistle off an engine of the type used on the line came up at a local auction. It’s a beautiful piece of brass for such a functional object. The chamber and bracket at the base is cast in brass, the top body being turned from another casting. This all makes the piece really weighty. Originally these would have been made by the apprentices in the loco works – simple bits of engineering but ensured that no two were exactly the same.

original whistle

original 1930’s ‘Stanier Hooter’ whistle

The first test up in Glen Ogle back in January used carbon dioxide to sound the whistle. This was powered by a small cylinder from a paintball gun. The cylinder was adapted to get a good volume of carbon dioxide out as the whistle required a lot of puff to sound at all. The first test worked well, and even at the relatively low pressure of the CO2 and in a mild wind and sleet we still managed to get a good echo off the hills.

While trying to find a fitting to connect my 1930’s original whistle to a 21st century paintball cylinder, I stumbled across a great engineering company. Not only could they fabricate a fitting for me (in under an hour), but they had history with steam engine whistles having previously made whistles for preserved steam locos all over the world. Besides, there’s something about having steam engine parts made by engineers in Darlington, with its weight of railway history behind it that I think adds to the narrative. The next site test was going to use compressed air – the carbon dioxide idea just wasn’t powerful enough. Besides, the little cylinders only gave me 3 seconds of whistle. I needed bigger cylinders and much more pressure.

cad drawing

Auto-CAD drawing of the new whistles

The basics of steam engine whistles are pretty straightforward. There’s a volume of air which passes through a chamber and then split into two thin streams. These streams are then passed over an opening and split by a sharp edge (called a fipple), as it does so the air vibrates and makes a sound. The pitch is determined by the length of the whistle body and the volume by the size of the hole. At its basic level it’s the same mechanism as I built for the ‘Curlew Machine’ piece, but at a different scale.

That said, there are so many other variables n the whistle design which make each one very different in many ways. In order to get the right whistle sound we needed to replicate the  Stanier whistle as close as possible, while making it quick, cheap and easy to build. While the original is solid brass, we were using cheaper steel for most of it.


the parts making up each whistle

Once the design was worked out the replica was tested against the original at high pressure to match the tuning and tone – small adjustments were made to the hole to tweak the sound – making it more ‘breathy’ and rounded.


Obviously steam engines use steam instead of compressed air and this in turn makes a difference in the sound. Warm air and steam raise the pitch significantly. The first test on CO2 was a very low note as the gas was expanding liquid gas, and so very cold. I found some recordings of Black 5 engines and some spectral analysis to get the prime note pitch and overtones.

Steam engine whistles have a characteristic rise in pitch as they sound – this is caused partly by the increase in pressure as the regulator is opened, and partly the steam warming up the air in the whistle. rather than shortening the new whistles, a baffle was inserted inside the top on a screw thread which enabled each whistle to be tuned individually.


the interior baffle used to tune the whistles


The second site test was to determine the reach of the radio transmitters which trigger off each whistle, and make sure they worked in the rocky terrain. Blessed with a very still and sunny day, it also gave us a chance to hear how the whistles reacted to the landscape.



With the second test working well, another half a dozen whistles were built (with some small modifications) and a chain of control boxes for a much larger test.

So, a couple of weeks ago we did the first of the big tests – six whistles over two miles of the railway path up Glen Ogle. BBC Radio Scotland came down for the day and did a feature on the Out of Doors programme the following weekend:

That particular stretch of line has the most problems, in terms  of access, terrain logistics so if we got it working there, the rest of the line would be relatively easy. Doing a test over two miles also allowed us to test the logistics of working over larger distances. I needed to know how to deal with issues that were over a mile away. This was by far the biggest single installation I’d done to date, so was a big learning curve. So, it didn’t all go to plan. There were issues with cross-talk with the radio triggers which meant we couldn’t run the whole sequence automatically so unfortunately I still haven’t heard the cascade that I’m after. However, the great folk at Radiometrix were just a phone call away and are already working on a solution.

The logistical problems of working over large areas became more apparent than I’d anticipated – communication methods had their issues too, and it was clear that the compressed air cylinders we had weren’t lasting as long as we needed.

What we did get was a real sense of how the piece strengthened with scale. The sequence of sounds echoing certainly carried way out over the surrounding countryside. The subtlety of the repetitive calls blurring the lines between succession and echo, and most surprising of all for me was the delicacy of the piece – considering just how loud and brutal the whistles are at close quarters. It was also apparent that this was going to be a really difficult piece to document. Although the line it draws in the landscape is very visual, you can’t photograph it. The subtleness of the very distant sounds pushes the limits of what works well in a recording and despite some great audio clips of the test, they don’t quite capture the very real experience of being there.

Compounded with the fact that the piece sounds so very different where ever you are – up on the line, down in the village, as a passing motorist (my favourite bit was drivers in the glen tooting back). the misty morning was also very different to the clearer afternoon – so each and every difference makes a new experience.

Just driving up the A84 from Callander you get a real sense of the vast scale of the piece. It’s really going to be something you just have to experience for yourself. Spend time with it. explore it. contemplate it and the landscape around it.

As a work of art I’m really excited by how it works – its layers of narrative and real experiential nature. Although some of the themes of the piece – landscape, scale, narratives, history – are things I’ve used before, the method of exploring them, the scale and ambition is all really new to me – and that’s a great thing to take away from a residency. It’s also got the whole community talking about it. Whether it’s good or bad is not so much the point as being a topic that intrigues.

On my first evening at the pub in the village the other week,  when the owner found out I was the artist doing this, came over with a list of questions. One of the previous guests was a railway buff and interested in the project. Apparently, his parting remark was “I bet they get the whistle sound all wrong”.

Detail, detail. It’s all important.


So, this part of the residency is sadly at an end. It’s been an amazing journey for me, and one which I hope will continue as we look to realise the final ambitious installation later in the year.

About stevemessam

Artist, designer, curator and rural activist
This entry was posted in history, landscape, research, sound and tagged , , , , , , , , , , , , , , . Bookmark the permalink.

1 Response to sights and sounds

  1. Pingback: Whistle round the Toon. part iii (Whistles) | Mud 'n Art

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s