|Table of Contents||Previous page||Next page|
Riding the "Leading Edge"
By 1967 the group had acquired three more development physicists, Peter Berry from the U.K., Dieter Donhoffer from Austria and Chris Thompson, a very young and very bright New Zealander. When Chris first arrived we were quite unprepared for his hilarious sense of humour. One day someone asked him if he spelt his name with, or without a 'p'. The answer came back quick as a flash in his heavy down-under accent; "...with - but the pee is silent - as in swimming..". We ranged in age from twenty two (Chris Thompson) to mid thirties and we were all totally involved, obsessed perhaps, with staying on the forefront of the latest developments.
The aim was to be one of the most advanced groups in Canada, if not in the world, in developing industrial applications of radioisotopes. This meant making analysis techniques like X-ray Fluorescence and Neutron Activation, more attractive to industry. One of the impediments was the lack of reasonably compact equipment at prices that medium sized institutions could afford. A case in point was the need for an inexpensive method for sorting the energies of the gamma rays and X-rays that were generated in these analyses, so that the elements that had produced them could be easily identified and their concentrations in unknown samples measured.
The catalyst for progress was the availability of integrated circuit logic and we took full advantage of it. The Radiation Protection Division of the department of National Health and Welfare, had a requirement to monitor the ingestion of the radioisotope Caesium-137 by Eskimos living in northern Canada. This came about because of the fallout from the atomic weapons tests in the 1950's. Caesium-137 is one of the many radioisotopes produced by the explosion of an atomic bomb, and traces still remain in the enviroment because of its relatively long half-life of more than 30 years. The concentrations of it in soils is now barely measurable and eating vegetables which absorb it from the soils is not considered to be a health hazard today.
In 1965 (nearly a half-life ago) the levels were twice as high as they are now and Caribou herds in northern Canada grazed on lichens over hundreds of square miles of territory. These absorbed the fallout directly and tended to retain the radioactive caesium, causing them to have higher concentrations than vegetation which grew after the fact and derived their radioactivity from the contaminated soils. As a result, the caribou had substantial concentrations of Caesium-137; and guess who ate the caribou.
What was needed was a compact portable instrument which could be taken up north to monitor the Eskimos by measuring the radiation from their bodies. The detector would be a scintillation detector, but there was a need to distinguish between the Caesium-137 radiation and that from the three naturally occurring radioisotopes, Potassium, Uranium and Thorium. That requirement translated into the need to count pulses coming from the detector which had amplitudes in a predetermined range, corresponding to the energy of the Caesium-137.
The conventional approach to that problem was an analogue one, involving the setting of the required amplitude limits for the pulses with two volume control type knobs on the front panel. It was fine in theory but it was very difficult to be sure that the settings were accurate and repeatable. Tolmie wanted to produce something based on digital techniques which could be calibrated easily and reliably, and which would provide consistent and repeatable readings.
The ultimate solution was to use an Analogue-to-Digital converter (ADC) to sort the pulses into as many as 128 different amplitude slots or "channels". There were laboratory instruments available which used such converters, and they were called "Multichannel Analysers". They had magnetic core memories to store the individual tallies of pulses falling in each of the 128 channels (which we did not need), were hideously expensive and definitely not in the portable category.
By this time the Fairchild semiconductor company was offering not only digital logic in the form of the new microcircuits, but complete analogue circuits, particularly devices like level comparators and operational amplifiers. This was a truly breath taking leap forward, because the equivalent circuits using individual transistors and resistors required circuit board areas of at least three square inches apiece. We realised that with these things it might be possible to design and build an ADC small enough to incorporate into a portable instrument, thereby leapfrogging the existing state of the art.
Because I was the only one in the group who had any experience in designing ADCs (the one for the digital readout of peak gun pressures), I was given the task of developing one using the new devices that would meet the requirements of our proposed instrument. It was a pretty frustrating experience. The two devices that Fairchild now offered (and which were critical to the implememtation of a miniature ADC) were the model 702 operational amplifier and the model 710 level comparator. They were the world's first commercial production of such devices made as micro circuits. They were also pretty expensive.
The first ones we got were $70.00 each, which in 1965 was equivalent to ten times that in today's money. I had to be super careful to check every connection while I was experimenting with them, because I knew it would not take much to zap them into oblivion. They were both what can only be descibed as "user hostile", and burst into uncontrollable oscillation at the slightest provocation. The 710 comparator was particularly exasperating and I remember saying to Tolmie at the end of the umpteenth session of trying to tame the thing, that as far as I was concerned the only good 710 was a dead 710.
These frustrations were the price of riding on the leading edge. We were very much in uncharted waters with integrated circuits in those days (and so it turned out was Fairchild), but the payoff was clear for all to see if only they could be made to fulfill the promise which they held. I learnt a whole lot of new tricks to circumvent the problems in the hard school of experience and finally came up with something that, although not as good as the expensive laboratory versions using discreet components, was more than adequate for what we needed and light years ahead of the conventional analogue solutions.
This portable "multi-channel analyser"
was ideal for field applications
There was much midnight oil burnt in solving the many bugs and glitches that turned up in bringing the project to fruition. That again was the price of being on the leading edge. The most powerful laboratory aid available in those days was the oscilloscope. But oscilloscopes were not designed (as they are now) for dealing with the vagaries of digital electronics. Futhermore no one had much experience in the things that could come to get you in trying to track down problems with complicated digital logic. For every choice or option that was added as a button on the front panel, the mumber of possible combinations and permutations of what could go wrong in the sequence was usually doubled. Until that time in the electronics design business, if something didn't work it was a pretty safe bet that it was either a faulty component or a wiring error. Now however the sophistication of the multifunction logic elements was such that malfunctions were more likely to be the result of an unforeseen possibility in the logical sequence than anything else.
Eventually the bugs were ironed out and we had a working prototype that met the requirements. As we worked with this instrument it began to dawn on us what the other possibilities were. Thanks to Tolmie's foresight, we had indeed succeeded in shrinking an expensive laboratory instrument into a portable box which could be reproduced at a fraction of the cost of the ones currently on the market. The first test was the one which had prompted the project in the first place.
It proved its worth, the only problem was one which would never have occurred to us in our wildest dreams. When it was actually used to monitor the Eskimos for their Caesium uptake, the procedure was to push the scintillation detector against their stomachs in order to trap as much of any Caesium radiation as possible. That part was fine, but when the mechanical counters on the instrument front panel began to clatter, they were absolutely terrified.
It was a classic case of assuming that just because people were not literate, that they were ipso facto stupid. The whole problem had been explained to them with the sort of beguiling and misleading simplicity that is used to convey upleasant truths to six-year olds. The problem was that they were not six-year olds and were perfectly capable of connecting cause and effect, even if they did not understand all the ramifications. As far as they were concerned, the clickety- clack of the counters began as soon as the thing on the end of the wire was moved close to their bodies, which meant that there was something in their bodies which was bad for them.
How right they were, some of them had ingested quite a bit of Caesium-137, although in absolute terms the accummulated radiation dosage that they were getting from it would never be as much as the massive accummulated dosages that I myself had already received as a kid, shoving my foot into an X-ray "Pedoscope" in the 1940's and watching fascinated as I saw the bones in my toes wiggle as I moved them back and forth.
It was in the autumn of 1966 that we had a visit from a Dr. Arthur Darnley who had just arrived from the U.K. to join the Geological Survey of Canada. He had spent some years with the British Geological Survey and was interested in the possiblity of making airborne radiation measurements to map the distribution of the naturally occurring radioelements Potassium, Uranium and Thorium. Most people know that uranium is radioactive, but not very many know that potassium, one of the most abundant elements in the earth's crust is also very slightly radioactive. The radioactivity is easily measurable in concrete structures because one of the constituents of concrete is aggregate made from gravel which comes from granitic rocks. Granites are the primary source of potassium uranium and thorium, but the eternal weathering processes of wind and water gradually erode the original rocks and redistribute the radioelements in the form of sand, sediments and conglomerates.
The major sources of radioactivity are the original rock formations, some of which are exposed but most of which are under a layer of overburden consisting of soils, clays, sediments etc. Darnley wanted to see if the different rock types could be identified by airborne measurements on the basis of the amounts of the three radioelements they contained, because if they could, then systematic airborne surveys would be able to cover large inaccessible areas and provide rudimentary geological maps. Conventional mapping is done by sending teams of geologists into the bush every summer where they set up camp for three months or more and trudge around on foot taking samples of the rocks and mapping the rock formations that they can see at the surface. Needless to say this is very expensive and time consuming in a country like Canada with three and a half million square miles of territory and only twenty five million people. It is nevertheless important that it be done, because Canada is still potentially one of the richest nations on earth in terms of mineral wealth. Accordingly any new methods that show promise for speeding up the process are always much sought after.
Darnley had managed to convince the managment of the Geological Survey of Canada that the idea was at least was worth some extensive experiments and he had been hired on to set the ball rolling. He had been told that our group at AECL was the one most likely to have the expertise to undertake such experiments and that is why he had come to visit us. The first order of business was to see how big (and therefore how expensive) the scintillation detectors had to be to field enough radiation at a reasonable altitude (say 400 feet) to identify the potassium, uranium and thorium components. The earliest experiment consisted of flying over the conveniently adjacent Gatineau hills (immediately across the Ottawa river from Ottawa and part of the Precambrian Shield with massive granite outcrops) in a twin-engined Beechcraft.
Arthur sat next to the pilot to guide him while I lay prone on my tummy in the back of the little aircraft, operating a version of our Caesium measuring instrument, now programmed to record the count rates from the three radioelements that we had needed to reject in the other application. We flew out over the flat plain of the Ottawa valley and across the river, with little or no clatter from the mechanical counters recording the radioactivity picked up by the array of three small scintillation detectors. We approached the escarpment formed by the granites of the Shield and as we flew up and over it all of a sudden the counters clattered away to beat the band. We flew around over different formations and saw correspondingly different readings of the three radioelements. It was only a qualitative experiment but it was a very encouraging first step and indicated that it was worth proceeding to more ambitious studies.
A National Medicare system for Canada
In 1966, the government of Saskatchewan, under its crusading left-wing Premier, Tommy Douglas, was the first one to institute a government-run health service in Canada, along the lines of the British National Health Service. This was quickly followed by similar schemes in other provinces and eventually the Federal government set aside funding for all the provinces with basic standards defined nation-wide. The whole enterprise ran into implaccable and bitter opposition from the Canadian Medical Association, and I saw an almost exact replay of the contest that I had seen in 1948 between the post-war Labour government and the British Medical Association, when Britain established one of the first nation-wide health plans in the world.
There was one disturbing difference however and that was the blatant intervention by the American Medical Association. It provided massive funding to the Canadian Medical Association in an effort to try and snuff out the possibility of any sort of government operated health care scheme springing up in North America. As far as they were concerned if it succeeded in Canada, it posed a dire threat to the American medical community and they decided that they would do whatever had to be done to prevent it. It was a flagrant violation of Canadian sovereignty as far as I was concerned and it was music to my ears when despite the amount of money they poured into the battle, they lost hands down.
Canada (along with every other nation of any significance on the face of the planet) now has a government operated nation-wide health care scheme which every one takes for granted and which no one is prepared to see dismantled, as politicians who have tried, know to their cost. The Americans by contrast, still lurch along with a fragmented market-driven system having all sorts of administrative duplication and other expensive overheads, which still leaves thirty seven million of its poorest citizens without any coverage at all.
1960 Riley one-point-five
Such conditions are unheard of in Britain, but are the norm in Canada, on and off, from late December to the end of February in the course of the average Canadian winter. The British Motor Corporation (BMC) as it then was, made no concessions to either the cultures or the climates of their export customers. This was a source of intense chagrin to me as I felt that I had to apologise personally when colleagues or friends who had bought British cars had been disappointed. One such was the New Zealander Chris Thompson, he had bought a Morris 1100, the much touted front-wheel drive successor to the famous "Mini" which had captivated the world in the early 60's. He had had all sorts of trouble with it and had been getting the run-around from the BMC dealer that he had bought it from.
I saw red and got on the phone to the wretched dealer and spent about twenty minutes reviling him and BMC for being so stupid as to alienate one of the worlds largest potential markets for British cars by cavalier treatment, sharp practice and downright incompetence. He was utterly astonished at such a diatribe being levelled at him from a non-customer on behalf of a third party and didn't have a lot to say for himself. It did me a lot of good to vent some pent-up steam on that subject, but I don't think it cut much ice with him.
Electronics had no place in cars in those days (apart from radios), partly because car makers had no expertise in electronics, as opposed to conventional electrical technology, relays, motors and so forth, and partly because the transistors of the time were much too fragile to survive in the dauntingly fierce environment that is to be found in the engine compartment of the average car.
One obvious problem area which was tailor made for a hi-tech solution was the standard automobile ignition system. It had been invented in the early twenties to allow a battery to be used as the power source and consisted of an induction coil, "the coil" in auto mechanics parlance, and "points", which were a pair of electrical contacts opened and closed mechanically at the right moment to provide a jolt of current to the coil from the battery. Each time the jolt occurred, the induction coil generated a ten to twenty thousand volt "lightning bolt", which caused a spark across the spark plug, igniting the petrol-air mixture and causing the "internal combustion" which drives the car forward.
At low speeds it was entirely satisfactory, but at higher speeds the coil did not have time to charge up between jolts, with the result that the voltage available to produce a spark fell off drastically, which in turn reduced the engine efficiency as the speed went up. Reduced acceleration was one very noticeable consequence of this, which tended to make passing another car at highway speeds a long drawn out and hence knuckle-whitening business for all but the most powerful cars.
A brand new device called a "Silicon Controlled Rectifier" (SCR) offered the possiblity of updating the old system by using it to empty the charge on a hefty condenser (charged to about 400 volts) into the standard coil, thereby producing the twenty to forty thousand volts needed for the spark. The SCR had a trigger electrode which could be actuated by the standard points. The advantages were (a) that the condenser could be reacharged between jolts in a fraction of the time needed to recharge the coil in the conventional system, (b) that the points never needed replacing, because they no longer carried the heavy current needed to charge up the coil, and (c) that because the original components of the old system were still used, the original configuration could be switched back in if the new system failed on the road.
I thought that it would be an interesting spare time project to design and build one to perk up the Riley and it turned out to be more than interesting. I did some careful measurenents on the spark voltage and found that beyond about fifty miles an hour, the performance dropped off very drastically indeed, far more than I had realised. Designing a solid state ignition system as described above was not difficult to do, so long as it only had to work on a lab bench at room temperatures. Designing one which would work reliably from minus thirty to about a hundred and fifty degrees Fahrenheit and be totallly immune to the electrical noise generated by the average car electrical system, was quite a different kettle of fish. This was when I found out just how fragile and temperature sensitive the much vaunted new solid state devices were. It took much blood, sweat, toil and tears to come up with a design which survived the rigours of a temperature conditioning chamber, which I used to cycle the many prototypes through the temperature extremes which the Riley could be subjected to in the Ottawa climate.
One of the major problems at that time was cold weather starting. Batteries were not as good as they are now and at minus thirty degrees Fahrenheit they were barely good enough to turn over a warm engine, let alone one in which the oil was like treacle. Add to that the fact that the petrol-air mixture was more of a spray than a vapour at those temperatures and one can begin to understand why a successful start under such circumstances was little short of a miracle. To top it all off, the spark voltage available from the battery was sharply reduced because the battery was killing itself to turn the engine over. Thanks to the freedom which the clever (but all too fragile) little integrated circuits gave me, I was able to incorporate a couple of wrinkles into the design to help things along a bit on cold winter days. The first was a regulated power supply which kept the spark voltage constant no matter how inadequate the battery voltage was. The second was a scheme to supply multiple sparks to each plug during its appointed firing time as long as the starter button was pressed.
The first test was a total disaster and gave me cause to marvel at how a little design error in something as tiny and fragile as integrated logic circuits could practically wreck something as solid as a car engine. It turned out that the multiple spark scheme was sending sparks to the plug as the piston was coming up the cylinder, well before they were supposed to occur. The result was that when the petrol-air mixture fired, it tried to drive the engine backwards, jamming the starter completely. It took me about an hour underneath the car to dismantle the thing and reassemble it and a lot more time to find where the error was in my timing logic, correct it and re-test it on the bench.
When I finally got it working it was worth the effort. I took the Riley out on the Ottawa Queensway, the new motorway across Ottawa which was then only partially completed and still under construction. The acceleration was phenomenal and I enjoyed the stunned faces of the drivers of the big powerful V8 engined American cars as they watched this little British car suddenly pull out from behind and streak past them at well over 90 miles per hour. They weren't the only ones who were stunned, I had never managed to push it to anywhere near that speed until then. The petrol consumption was much better and the improvement in cold weather starting, the main reason for embarking on the project in the first place, was quite dramatic. Enough to raise the probabilty of starting on an extra cold day from "a good probability" to a near certainty.
Some more airborne experiments
Following the encouraging results of the Beechcraft flight and some other related work, Arthur Darnley decided to try some more quantitative experiments. After a lot of consultation we arrived at a plan which involved using a helicopter to hover over specific locations (where the composition of the rocks was known), recording complete gamma ray spectra at different heights. This was a pretty ambitious project; we had to come up with an assemblage of equipment, some of it off-the-shelf but most custom-designed, which would survive the notorious vibration generated in helicopters and successfully record the complete spectra that we wanted.
The payoff however would be well worth the effort, because the results would allow us to calculate the various parameters needed to determine just how large (and hence how expensive) the array of scintillation detectors would have to be to make it practical to do any sort of airborne mapping with available aircraft. Ideally we needed one with all the characteristics of a balloon, something which could traverse over the terrain at treetop level, (to be as close to the source of radiation as possible), and move slowly enough to field as much radiation as possible from each acre of ground.
First airborne gamma-ray spectrometry experiments
using a sikorsky helicopter at Bancroft Ontario, 1967
It certainly was an interesting experience. We flew in shifts of about half an hour at a time over an area near the town of Bancroft, a resort area in Northern Ontario. The reason that it was a resort area was its situation in the rocky Canadian shield, with crystal-clear lakes and streams shimmering against a backdrop of rugged granite escarpments, visible as outcrops here and there through the dense conifer and deciduous forest cover.
This scenic roller coaster landscape was (and still is) prime and much sought after "cottage-country" for the well-to-do; and as motel and provincial park camping-hiking-canoeing country for the less well-heeled younger set. The reason that we were doing our experiments there was because of those very granites that made it such a rugged and scenic area. They contained the three naturally occurring radioisotopes in concentrations that were representative of the large unmapped areas of Canada where this technique would be very useful. The data that we would get here would therefore be a reasonably reliable indicator of whether or not the method was likely to be generally successful.
Hovering to acquire natural radiation spectra
From front to back: Chris Thompson,
the author and Dieter Donhoffer
|Table of Contents||Previous page||Next page|