Project 'Hose'.
Building the first hose.
This project was the brainchild of Frank W. Watlington, Senior Scientist at SOFAR Station in St. David’s, Bermuda. The design and construction of the original hose spanned late 1960 through mid-1962. The photo below indicates that 'Project Hose' lasted several years, until at least 1967.
‘Hose’ was an apt name for this project, because the major element was a hose, very similar to a garden hose – about the same diameter (perhaps slightly larger) - and about 80 feet long – I have forgotten the exact length. The outer surface was exceedingly smooth, because it was going to eventually be trailed behind a boat, and we did not want any extraneous friction noise of water ‘rubbing’ against the surface of the hose. This necessitated some very careful handling of the hose during the preparation of the final product, and during transport and deployment.
Another component of the rig was 25 to 30 (again, the exact number escapes me) hollow, porcelain cylinders, about 2” long and ¾” outside diameter, which were coated on the outer face and the inner face with a piezoelectric coating, leaving the ends of the cylinder uncoated. The fascinating property of these cylinders was that the slightest change in pressure on the outer or inner face of a cylinder would generate a small, brief electric charge, which could be measured between the inner and outer faces of the cylinder using an oscilloscope or sensitive voltmeter.
Sound is nothing more than vibrations in the air or water, and vibrations are just rapid changes of pressure. These changes of pressure would cause the faces of the cylinders to change from positive to negative charge, back and forth very rapidly, and we would record these electric variations. We could then play back these vibrations through an amplifier and speakers to hear what we had recorded.
Short pieces of electrical wire needed to be soldered to the inner and the outer face of the cylinders, a quite delicate operation, because excessive heat would crack the porcelain, ruining the cylinder.
Once this was completed, each crystal needed to be tested to see whether it was an ‘outer’ or an ‘inner’ – whether pressure on the outer face of the cylinder would generate a positive charge (an 'Outer’),or a negative charge (in which case the inner surface had a positive charge) and therefore was deemed to be an ‘Inner’, and the cylinders were appropriately marked ‘O’ or ‘I’.
The cylinders were eventually to be placed inside the hose at quite precise distances apart, and connected together in a specific sequence using a special 8-pair multicable which also had a quite strong (steel or copper?) strain cable to stop the whole thing from stretching when it was being towed.
The SOFAR machine shop had manufactured the front end and the back end plugs for the hose from aluminum stock. These fit very snugly unto the hose, and would be sealed in with epoxy at the appropriate time. The front plug had a hole through its length the exact diameter of the multicable, through which the multicable would pass.
The next part of the operation was to remove about 80 feet of the outer covering of the multicable, to reveal the 16 electrical wires (8 pairs), and the strain cable. These were passed through the hole in the front plug and about 1 foot of addition covered cable was also forced through the hole (That was a tough job, using lubricants, pushing, pulling, grunting, groaning, and even a little swearing!)
The wonderful machine shop at SOFAR had built a trough about 3’ off the ground and about 80’ long, down a gently sloping hillside, in which we laid the cable, with the front plug at the upper end. At the lower end of the trough, there was a large, wide, sheave, over which all the cables were placed, and a weight (about 20 pounds, I think) was attached, so that all the cables were held taut.
Since there were only 8 pairs and there were about 30 cylinders, each cable pair would have to have several cylinders attached to it. A pattern had been worked out for the spacing of the cylinders on each cable pair, so each cable pair needed to be marked with the exact location where the cylinder was to be connected to that pair. An arbitrary point was chosen and marked near the front end of the rig from which all distances would then be measured. The cable pairs were colour coded – there was a red wire, and a red wire with a white stripe, there was a green wire and a green wire with a white stripe, etc., so it was easy to mark the correct pair at the correct distance from the start point.
Another task that needed to be done was to ‘paint’ the strain cable with ‘liquid rubber’, so that it would be less likely to damage the cylinders, or create a chafing noise. This was accomplished by raising the strain cable up and placing a few pieces of wood across the trough, all down the length of the trough. The 20 pound weight kept the cable nicely taught, so it was easy to paint all around the cable. Once dry, touch-ups were required to cover the spaces where the wood had been, and finally, it was decided that a second coat was warranted – ho hum, do it all again!
Once the strain cable had been painted, the tension was removed from the lower end, and the connecting of the cylinders began. It was decided that the strain wire and 2 pairs of wire would pass through the centre of the cylinders; the rest of the pairs would lie outside the cylinders. The cylinders were thusly run up the rig to their approximate positions, with pieces of rag between them and the hard rough metal trough. Then began the laborious but delicate task of connecting the cylinders to the correct pair, always remembering that the wire connected to the outside of the cylinder was to be connected to the solid coloured wire if the cylinder was an ‘O’, and to the white stripe wire if the cylinder was an ‘I’. This would ensure that there would always be a positive charge on the solid coloured wire when pressure was placed on the outside of the cylinders. Once this was completed, and all the connections were tested to be sure they worked, and none of the ‘I’s and ‘O’s were backwards, all the solder joints were liberally painted with liquid rubber, and then taped with lectricians tape. Each cylinder was then wrapped in rags, and the entire staff of SOFAR (well, not quite, but quite a lot of them) assisted in turning the entire rig end to end, so that the front end plug was at the bottom end of the sloping trough. In hindsight, we probably should have built the thing that way around from the start – I can think of no reason we decided to do it the way we did.
Time to unpack the hose! It came with a pull string through the entire length of it, which was a great help. We had accumulated a bunch more rags, which we would use for the hose to rest on as we pulled it along the trough, removing the rags tied around the cylinders as we went along. The end of all the wires – the strain wire and the 8 pairs – were tied to the pull string, and the string held taught, while the hose was moved forward and the wires and cylinders were fed into it. This operation also used quite a few people, because the hose had to be straightened out (held above ground by willing hands) in order to get the pull string to move inside it. We were done in an hour, which was good, because just as we finished, it started to rain - cloudburst – we remembered to cover the open ends of the hose first!
The next job was to securely push the hose into the front end plug, and waterproof the joint with epoxy. Then we filled the hose with very pure castor oil, allowing a couple of days for all the bubbles to run up the tube. The end plug had an eye (as in ‘hook and eye’) on the inside end, to which all the wires would be attached. We had marked the point on the wires where this attachment was to be made, and had to compress the hose a bit to make the connection. Once this was done, the hose
was given a final top up of castor oil, and the end plug was inserted into the hose. Inevitably, there was a small air bubble, but probably not a big problem. This joint could not be epoxied, because epoxy and castor oil don’t mix, so several hose clamps were used. Finally, a 20 foot long piece of rope was spliced to a hole at the very end of the end plug, which was to provide some drag to keep the hose straight as it was pulled through the ocean, and to stop the end of the hose wiggling like a kite tail (the end of the rope was going to do the wiggling.)
The whole thing was then placed back in its nicely padded shipping crate, with the attached reel of multicore cable nearby, ready for transport to the boat.
‘The boat’ happened to be a 70 foot two-masted schooner, the ‘Grace’, which we were using because we didn’t want to record any engine noise. We were going to record the noises that we heard as we sailed along. The tape recorders had only 7 channels, so we used two tape recorders, powered by
battery. We put pairs 1 thru 6 on one recorder, along with WWV, and pairs 4 thru 8 on the other, along with WWV and a voice channel for any commentary.
We set off from Dolly’s Bay, accompanied by the Buoy boat. Once we were in reasonably deep water, and on a steady course, we deployed the hose, trailing it about 50 yards behind us. The Buoy boat then steamed in very loopy, elongated circles around us, dropping light bulbs tied to a sash weight every so often, and the occasional a ¼ pound of TNT, just for fun. When a light bulb is dragged down into the ocean by the sash weight, eventually the increased pressure on it causes it to collapse, or ‘implode’, with a nice crack, which can be heard underwater for a couple of miles. We had no radio contact (life before cellphones!!), so everything was done to a preplanned schedule. Meanwhile, our position, and that of the Buoy boat, was being recorded at the B. E. stations at Mount Hill and Paynters Hill, with bearings every 20 seconds, alternately on the mast of the ‘Grace’, and the mast of the Buoy Boat. They used WWV for the time marks.
By early afternoon, we were done; we retrieved the hose, and headed for home.
The next day, we listened to the tapes, and noted the times that we heard the implosions and explosions – the sound sources. We then used the data from the BE stations to plot the positions of the Buoy boat and the ‘Grace’ and the heading of the‘Grace’, in order to know what angle the sound was coming from relative to the line of the hose. The sounds that were recorded when the source was at right angles to the hose were quite different to the sounds recorded when the source was in front, behind or at an oblique angle to us. In addition, some of the channels showed a significant decrease at some angles. Strange as it may seem, these were approximately the results that were expected.
This project was funded by ONR, and all the data and tapes were soon bundled up and sent off to them. (I don’t know what happened to the hose itself* – I had returned to University shortly after this operation was complete.) I will do a simple write-up on the physics behind this once I have figured it out myself i.e. dredged it up from the depths of my memory. The outcome of this project (following further development by the US Navy and others) became standard equipment on submarines, and on surface vessels. Google ‘SURTASS’ if you don’t believe me!
One of the differences between doing research back then compared with doing research today is that, back then, if you needed something, you had to build it yourself. Nowadays, most of the components can be bought ‘off the shelf’ and assembled together with interface cables.
Bruce Hallett.
November 11th 2013
* Since writing the article above, Bill Andrews has added the picture (below) which answers my question about what happened to the hose! It probably went over to McCallans wharf for storage until further funding was available for more experimentation to be done. I am hoping that Bill will be able to shed more light on the later use to which it was put.
Bruce Hallett
November 23rd 2013
‘Hose’ was an apt name for this project, because the major element was a hose, very similar to a garden hose – about the same diameter (perhaps slightly larger) - and about 80 feet long – I have forgotten the exact length. The outer surface was exceedingly smooth, because it was going to eventually be trailed behind a boat, and we did not want any extraneous friction noise of water ‘rubbing’ against the surface of the hose. This necessitated some very careful handling of the hose during the preparation of the final product, and during transport and deployment.
Another component of the rig was 25 to 30 (again, the exact number escapes me) hollow, porcelain cylinders, about 2” long and ¾” outside diameter, which were coated on the outer face and the inner face with a piezoelectric coating, leaving the ends of the cylinder uncoated. The fascinating property of these cylinders was that the slightest change in pressure on the outer or inner face of a cylinder would generate a small, brief electric charge, which could be measured between the inner and outer faces of the cylinder using an oscilloscope or sensitive voltmeter.
Sound is nothing more than vibrations in the air or water, and vibrations are just rapid changes of pressure. These changes of pressure would cause the faces of the cylinders to change from positive to negative charge, back and forth very rapidly, and we would record these electric variations. We could then play back these vibrations through an amplifier and speakers to hear what we had recorded.
Short pieces of electrical wire needed to be soldered to the inner and the outer face of the cylinders, a quite delicate operation, because excessive heat would crack the porcelain, ruining the cylinder.
Once this was completed, each crystal needed to be tested to see whether it was an ‘outer’ or an ‘inner’ – whether pressure on the outer face of the cylinder would generate a positive charge (an 'Outer’),or a negative charge (in which case the inner surface had a positive charge) and therefore was deemed to be an ‘Inner’, and the cylinders were appropriately marked ‘O’ or ‘I’.
The cylinders were eventually to be placed inside the hose at quite precise distances apart, and connected together in a specific sequence using a special 8-pair multicable which also had a quite strong (steel or copper?) strain cable to stop the whole thing from stretching when it was being towed.
The SOFAR machine shop had manufactured the front end and the back end plugs for the hose from aluminum stock. These fit very snugly unto the hose, and would be sealed in with epoxy at the appropriate time. The front plug had a hole through its length the exact diameter of the multicable, through which the multicable would pass.
The next part of the operation was to remove about 80 feet of the outer covering of the multicable, to reveal the 16 electrical wires (8 pairs), and the strain cable. These were passed through the hole in the front plug and about 1 foot of addition covered cable was also forced through the hole (That was a tough job, using lubricants, pushing, pulling, grunting, groaning, and even a little swearing!)
The wonderful machine shop at SOFAR had built a trough about 3’ off the ground and about 80’ long, down a gently sloping hillside, in which we laid the cable, with the front plug at the upper end. At the lower end of the trough, there was a large, wide, sheave, over which all the cables were placed, and a weight (about 20 pounds, I think) was attached, so that all the cables were held taut.
Since there were only 8 pairs and there were about 30 cylinders, each cable pair would have to have several cylinders attached to it. A pattern had been worked out for the spacing of the cylinders on each cable pair, so each cable pair needed to be marked with the exact location where the cylinder was to be connected to that pair. An arbitrary point was chosen and marked near the front end of the rig from which all distances would then be measured. The cable pairs were colour coded – there was a red wire, and a red wire with a white stripe, there was a green wire and a green wire with a white stripe, etc., so it was easy to mark the correct pair at the correct distance from the start point.
Another task that needed to be done was to ‘paint’ the strain cable with ‘liquid rubber’, so that it would be less likely to damage the cylinders, or create a chafing noise. This was accomplished by raising the strain cable up and placing a few pieces of wood across the trough, all down the length of the trough. The 20 pound weight kept the cable nicely taught, so it was easy to paint all around the cable. Once dry, touch-ups were required to cover the spaces where the wood had been, and finally, it was decided that a second coat was warranted – ho hum, do it all again!
Once the strain cable had been painted, the tension was removed from the lower end, and the connecting of the cylinders began. It was decided that the strain wire and 2 pairs of wire would pass through the centre of the cylinders; the rest of the pairs would lie outside the cylinders. The cylinders were thusly run up the rig to their approximate positions, with pieces of rag between them and the hard rough metal trough. Then began the laborious but delicate task of connecting the cylinders to the correct pair, always remembering that the wire connected to the outside of the cylinder was to be connected to the solid coloured wire if the cylinder was an ‘O’, and to the white stripe wire if the cylinder was an ‘I’. This would ensure that there would always be a positive charge on the solid coloured wire when pressure was placed on the outside of the cylinders. Once this was completed, and all the connections were tested to be sure they worked, and none of the ‘I’s and ‘O’s were backwards, all the solder joints were liberally painted with liquid rubber, and then taped with lectricians tape. Each cylinder was then wrapped in rags, and the entire staff of SOFAR (well, not quite, but quite a lot of them) assisted in turning the entire rig end to end, so that the front end plug was at the bottom end of the sloping trough. In hindsight, we probably should have built the thing that way around from the start – I can think of no reason we decided to do it the way we did.
Time to unpack the hose! It came with a pull string through the entire length of it, which was a great help. We had accumulated a bunch more rags, which we would use for the hose to rest on as we pulled it along the trough, removing the rags tied around the cylinders as we went along. The end of all the wires – the strain wire and the 8 pairs – were tied to the pull string, and the string held taught, while the hose was moved forward and the wires and cylinders were fed into it. This operation also used quite a few people, because the hose had to be straightened out (held above ground by willing hands) in order to get the pull string to move inside it. We were done in an hour, which was good, because just as we finished, it started to rain - cloudburst – we remembered to cover the open ends of the hose first!
The next job was to securely push the hose into the front end plug, and waterproof the joint with epoxy. Then we filled the hose with very pure castor oil, allowing a couple of days for all the bubbles to run up the tube. The end plug had an eye (as in ‘hook and eye’) on the inside end, to which all the wires would be attached. We had marked the point on the wires where this attachment was to be made, and had to compress the hose a bit to make the connection. Once this was done, the hose
was given a final top up of castor oil, and the end plug was inserted into the hose. Inevitably, there was a small air bubble, but probably not a big problem. This joint could not be epoxied, because epoxy and castor oil don’t mix, so several hose clamps were used. Finally, a 20 foot long piece of rope was spliced to a hole at the very end of the end plug, which was to provide some drag to keep the hose straight as it was pulled through the ocean, and to stop the end of the hose wiggling like a kite tail (the end of the rope was going to do the wiggling.)
The whole thing was then placed back in its nicely padded shipping crate, with the attached reel of multicore cable nearby, ready for transport to the boat.
‘The boat’ happened to be a 70 foot two-masted schooner, the ‘Grace’, which we were using because we didn’t want to record any engine noise. We were going to record the noises that we heard as we sailed along. The tape recorders had only 7 channels, so we used two tape recorders, powered by
battery. We put pairs 1 thru 6 on one recorder, along with WWV, and pairs 4 thru 8 on the other, along with WWV and a voice channel for any commentary.
We set off from Dolly’s Bay, accompanied by the Buoy boat. Once we were in reasonably deep water, and on a steady course, we deployed the hose, trailing it about 50 yards behind us. The Buoy boat then steamed in very loopy, elongated circles around us, dropping light bulbs tied to a sash weight every so often, and the occasional a ¼ pound of TNT, just for fun. When a light bulb is dragged down into the ocean by the sash weight, eventually the increased pressure on it causes it to collapse, or ‘implode’, with a nice crack, which can be heard underwater for a couple of miles. We had no radio contact (life before cellphones!!), so everything was done to a preplanned schedule. Meanwhile, our position, and that of the Buoy boat, was being recorded at the B. E. stations at Mount Hill and Paynters Hill, with bearings every 20 seconds, alternately on the mast of the ‘Grace’, and the mast of the Buoy Boat. They used WWV for the time marks.
By early afternoon, we were done; we retrieved the hose, and headed for home.
The next day, we listened to the tapes, and noted the times that we heard the implosions and explosions – the sound sources. We then used the data from the BE stations to plot the positions of the Buoy boat and the ‘Grace’ and the heading of the‘Grace’, in order to know what angle the sound was coming from relative to the line of the hose. The sounds that were recorded when the source was at right angles to the hose were quite different to the sounds recorded when the source was in front, behind or at an oblique angle to us. In addition, some of the channels showed a significant decrease at some angles. Strange as it may seem, these were approximately the results that were expected.
This project was funded by ONR, and all the data and tapes were soon bundled up and sent off to them. (I don’t know what happened to the hose itself* – I had returned to University shortly after this operation was complete.) I will do a simple write-up on the physics behind this once I have figured it out myself i.e. dredged it up from the depths of my memory. The outcome of this project (following further development by the US Navy and others) became standard equipment on submarines, and on surface vessels. Google ‘SURTASS’ if you don’t believe me!
One of the differences between doing research back then compared with doing research today is that, back then, if you needed something, you had to build it yourself. Nowadays, most of the components can be bought ‘off the shelf’ and assembled together with interface cables.
Bruce Hallett.
November 11th 2013
* Since writing the article above, Bill Andrews has added the picture (below) which answers my question about what happened to the hose! It probably went over to McCallans wharf for storage until further funding was available for more experimentation to be done. I am hoping that Bill will be able to shed more light on the later use to which it was put.
Bruce Hallett
November 23rd 2013
Bill Andrews and Jan Card operating the 'Hose' recording equipment on Sir Horacr Lamb. September 1967.