Deaths and near-death experiences...

Press announcement about the project to make atmospheric measurements

During the first few months of 2000 Chris continued to go though periods of being seriously unwell. When feeling OK she was still able to be active and regularly went to the CAB in Dundee to work there as a volunteer, advising people who had problems. On the 23rd of January she also took a train down to London. The nominal reason for the trip was to attend a meeting organised by the Child Poverty Action Group. But she also took the opportunity to spend a few days visiting friends and relatives in and around London. She then travelled on to Wilmslow for a couple of days to visit her Father. I didn’t go with her because I was anxious about getting work done after the Xmas holiday – mainly urgently required for undergrad coursework and exams. Overall, Chris enjoyed the trip, although her London hotel was noisy and uncomfortable. Spending time with her Father also worried her because she discovered that he was finding coping with everyday life increasingly difficult.

In terms of research, my main concerns at the time were to get work funded in two areas . The first was developing scanners for volcanology. Up to this point, all our applications to the Natural Environment Research Council (NERC) to build a scanner for volcanology has been rejected. But – despite Duncan Robertson having left the University – I continued to press on and put forward a fresh proposal. This was in co-operation with a number of volcanologists who were keen to see a system supported, and be used as a means for providing them with data.

The second concern was to produce a system which could be used to assess the reliability and capacity of mm-wave radio data links operating in the frequency range between 30-40 GHz. Links of this kind were beginning to be installed to handle the rapid rise in the use of mobile devices and the internet. Higher data capacity was clearly needed, but at the time engineers had almost no reliable data to determine the extent to which atmospheric effects would limit the capacity of these links. It was known that bad weather – particularly rain – could severely attenuate the signals and thus degrade data capacity. But very little was known about the statistics of how much impact this would have when many links were in daily use. So we set about devising a method to observe the atmospheric effects in more detail over a long period.

I had continued to keep in contact with Duncan. He’d moved house and no longer lived in St Andrews. So most of our exchanges were via occasional emails. But we also got together when he visited St Andrews. It became clear as the months passed that he was becoming disillusioned with his situation at Racal. It was fairly routine compared with the R&D done by the mm-wave group, and he also wasn’t as confident as he had been that the job was secure and offered prospects of being able to make progress within the company. In contrast, after years of nagging, pushing, etc, the University had finally set up an arrangement named the Photonics Innovation Center or ‘PIC’ that was able to take on ‘commercial’ work on a properly funded basis. This could then operate in a more business-like manner than was usual for academics. As a result, doing work that wasn’t supported by the research councils was now at long last, ‘respectable’. The PIC had been officially declared ‘open’ by Trevor Bayliss – the inventor of the clockwork radio. Duncan was invited to the ceremony, but was at that time still too irritated by his previous treatment by the University that he didn’t attend!

To a large extent the creation of the PIC and its choice of name came about because the main research groups in the physics department worked on lasers and their applications. The University and Government had finally realised the potential value of this research as a basis for many commercial applications. So the head of department and other professors had worked to get the PIC established, based in the Physics building. Given that ‘Photonics’ was the key term used I was then able to cheerfully make the comment: “Well, you get far more photons per Watt at mm-wave frequencies than in the visible!” And from then onwards we could work in conjunction with the PIC.

A real bonus from my point of view was that the PIC was also then able to serve as a route for Duncan to return to work with the mm-wave group in St Andrews. Not only could it undertake a more diverse range of work than the mm-wave groups. It also could offer him the prospect of being freed from some of the limitations that had been imposed when he had been employed as a University researcher. In particular, the way this had prevented him from generating and being seen to be responsible for his own projects. Having chatted with him about this, we set about smoothing the way towards him being offered a new job, working for the PIC. Initially I operated as a sort of ‘go between’ to find out if the department and PIC would be pleased to see him return and offer him a suitable position. This was because he was wary of finding himself back in the same ‘box’ which had frustrated him and caused him to leave in the first place.

Although the attempts to obtain a NERC grant for volcanology scanners had failed we’d been more successful in attracting interest in developing a suitable instrument and carrying out measurements on atmospheric effects upon mm-wave communication links. This had grown out of earlier contacts between Duncan Robertson and the Radiocommunications Agency (RA). They were the Agency who advised the UK Government on deciding what parts of the radio spectrum could best be used, by whom, for what purposes. Hence they needed a technical basis to inform their decisions and plan spectrum use. Which included understanding environmental factors like the effects of weather, etc. Duncan had joined an RA working group on propagation anomalies and suggested St Andrews become involved. This led to a number of discussions with Dave Eden who worked for the them. John Goddard of the Rutherford Laboratory was also on the same working group, and the Rutherford usually carried out measurements for the RA, so we also linked up with them and the weather radar systems they developed for purposes of this kind.

As a result we’d formed a plan to put two complimentary instruments on the roof of the St Andrews Physics building. One would be a weather radar system, supplied and operated as an outstation by the Rutherford Laboratory. The other would be a 3-Port spatial interferometer. This was similar to the 3-Port systems the mm-wave group had already developed and used for other purposes. But in this case it would operate at a frequency in the 30 - 40 GHz to match the frequencies coming into use for newer data links at the time. Given our existing contacts with DERA at Malvern and their interest, also, in mm-wave communications, they also became involved in the project.

Map showing the eventual transmitter location and the line of sight chosen for the atmospheric measurements.

The initial proposals we’d made had generated a positive response from all concerned, and during early 2000 I was trying to get the University to deal with what would be required from them in order for the work to be possible. In particular we needed the University ‘Estates and Buildings’ (E&B) department to survey the roof of the Physics building so we could establish what building works would need to be undertaken to mount the weather radar, interferometers, and associated equipment. Alas, as time passed it became clear that actually getting E&B to do anything was like trying to squeeze blood out of a stone!

From the point of view of an academic physicist, some undergraduate students can be regarded as failing to obey Newton’s Laws of Motion. This is because you may set them going in a given direction of study. However when you stop pushing they don’t continue to move forwards, but slow down and stop! During 2000-2 the St Andrews University E&B department would exhibit the same behaviour. For example, they had been asked to survey the roof and then tell us what work would be required, and specify the cost. Weeks passed and we had no response. And whenever we asked, we were told in effect that they would get around to it sometime. When they did visit the building they would usually come up with a reason why what we wanted was ‘impossible’. In some ways it reminded me of the old Civil Service maxim that “Nothing should ever be done for the first time.”...

By the start of 2000 Chris’s condition had deteriorated. In addition to having fits she also was going though periods of severe anxiety or panic combined with confusion, very odd behaviour and forgetfulness. Increasingly, this was accompanied by becoming severely depressed to the point of feeling suicidal.

At first these states tended to last for a few days, then disappear for some weeks or even months before returning again. But gradually, these ‘down’ periods were getting longer, more intense, and closer together. In addition, she started to experience problems like an arm or leg jumping or twitching every minute or two during the night, even when asleep, and this would continue for anything from a few minutes to hours. Her condition was usually at it worst in the early morning. She would wake up at 3 or 4 am and be unable to get back to sleep because she was so frightened, anxious, and confused in addition to experiencing the physical symptoms. This would also wake me up and we’d try to have a conversation about the reasons she felt extremely anxious. The added difficulty was that we were both tired and anxious, and could be almost impossible to have a sensible conversation – and this then made her feel worse!

She would then tend to stay in bed and not eat any breakfast. Sometimes she was able to sleep, but still might feel too unwell to get up or eat any lunch. At its worst she might or might not get up until about 6pm and might eat some dinner then – or not. Then she might sit up and watch TV with me for a while. The general pattern was that she gradually felt better as the day passed. By the late afternoon or evening we could usually talk more easily and would tend to feel much more positive. She would often then be apologetic about being so negative that morning. Alas, all too often we’d repeat the process starting in the early morning of the next day. She would then dismiss as a fantasy that she had felt better the previous evening and refuse to accept that we might have come to any useful agreements at the time. It was as if going to sleep pressed a ‘reset’ button each night. But after a number of such days the behaviour would fade away – unfortunately, only to return in the future.

To a large extent, however, only Chris and I experienced these problems because she wouldn’t go out or see anyone when feeling bad. If anyone came to see her she would often stay in bed and say she couldn’t speak to them. I’d have to apologise and say she wasn’t well enough. Most of these times she was almost literally a different person to when she wasn’t experiencing the symptoms. When ‘OK’ she would enjoy getting out and attend various public meetings, going and working at the Dundee CAB, etc, and be active and very positive. This was the Chris most people saw, and it was a pleasure to see her at these times because she was happy and doing good work to help others.

As the ‘down’ periods got longer, more frequent, the effect on me became increasingly severe. In later years I came to describe it as “like living in a war zone” because almost anything during the difficult periods might cause Chris to become very upset, even things she’s said she was pleased about a day or two before. Because I was worried about her, and a lot of my work involved writing or using a computer to do analysis I’d increasingly taken to working at home as much as possible. This let me be near her more of the time and keep an eye on her and do things like cook meals. It was also because I was increasingly worried about her rising tendency to talk about wishing to be dead, and suicide, when she was feeling very anxious and confused. But even when sitting at a computer working I would suddenly find that Chris was shouting at me because I was trying to get some work done and hadn’t turned round when she came in un-noticed. In effect I had to be ‘on call’ instantly all the time, sometimes having to engage in circular conversations that led nowhere when she was very confused and frantic. If I failed to respond instantly and perfectly, the consequences were stressful for us both. This was in addition to having to be alert for odd noises - or even silences! – that might indicate she was having a fit and I needed to get to her in case I was needed to prevent a fall or some other problem. But I also had to work and do the job I was paid to do.

Although I didn’t realise it at the time, the result for me was a state of stress which I came later to describe as a minor version of ‘Combat Stress’. The feeling that you have to be ultra-alert all the time. Whatever I was doing, no matter how much attention it might require, I had to also stay alert for the slightest sign that I had to instantly attend to Chris in case something bad might happen. This also meant that when I did go out when she was unwell, I’d try to make the time I spent out of the house as short as possible because I worried about what I might find when I got home. Against this, when I did go to the department to give a lecture or run a teaching lab I was often able to ‘switch off’ the anxiety for a while. So in a sense it was a brief diversion, but the instant this ended I started to worry: Would Chris be OK, or had something bad happened while I was out? As a result I came to dread walking up the path and opening up the front door, for fear of what I might find.

Snippet from an ‘Acorn Publisher’ article on using !EasiWriter.

During February 2000 Mike Williams, the editor of ‘Acorn Publisher’ invited me to write some articles for his magazine. I was pleased by this because it was a professional magazine and paid well. I had already written a number of articles for other ‘Acorn’ computer magazines, but all bar one of these was on a free/amateur basis – and the one exception had been the article for ‘Acorn User’ which they published but never paid for! (I learned later on that many others had experienced similar treatment by the then-editor of ‘Acorn User’. Although I did at least finally get a letter from him formally acknowledging that he and his magazine had no rights regarding the material due to their failure to honour their contract.) Fortunately, ‘Acorn Publisher’ was a totally different experience and they published and paid well for a number of items I produced for them over the following few years. The AP articles focussed on topics like how to use RISC OS software to create webpages, etc. The illustration above comes from an article in a series on how to use the !EasiWriter/!TechWriter programs.

On April 3rd I came home from the Physics Dept and found Chris in bed after she’d had a bad fit. I discovered that she had badly burnt one of her fingers. At the time she might have three or four fits per day during her ‘bad’ periods. So we took a taxi to the A&E of the St Andrews Memorial Hospital to have it examined and treated. She had been holding a saucepan when she had the fit and the hot metal must have been in contact with her finger for a while. During a fit she often has no awareness of what was around her, and would not sense any pain. Unfortunately, she had another fit whilst at the A&E. This involved an ‘automatic’ phase. I was accustomed to these, but the nurses there hadn’t experienced the behaviour it produced. Her standard ‘automatic’ behaviour after a fit was ‘go home and go to bed’. However because she was confused and not really aware of where she was, etc, this could have strange results...

In this instance she kept trying to leave the room at A&E where her burn was being treated and dressed. Two nurses had to struggle to stop her leaving as she tried to fight her way past them. At one point she was trying to get into a cupboard and – I assume – thought it was a doorway that would lead her home or to our bedroom. This went on for about a quarter of an hour and then she gradually became more passive and aware, able to follow simple instructions. The nurses were able to finish treating her finger and we came home and she went back to bed to sleep for a while.

During April I had a number of discussions with Duncan and the PIC about arranging for them to be able to offer him a suitable position. The head of the Physics department and others involved keenly supported this happening. On the 28th we finally got an official agreement that we could install a weather radar system on the Physics building roof but still encountered an, erm, lack of urgency on the part of E&B. On the 12th of May my eldest step-brother, John, died. I’d not seen him or my other step-brother for some years so this was an unpleasant surprise that made me wish I’d tried harder to stay in contact with the family.

At the beginning of June 2000 Prof Harry Pinkerton (Lancaster University) was in the process of putting together yet another NERC application for a grant that would pay us to build and use a mm-wave radar/radiometer for volcanology. As time had passed some of the original academic volcanologists had moved on, but others had begun to join and support this project. As result, despite previous rejections we were becoming more confident – or perhaps more stubbornly determined – that this project would eventually succeed! Dav Macfarlane who had been my PhD student was on track to work on the scanner, and all being well we hoped that Duncan would join the PIC and then also become involved in developing mm-wave scanners.

I also started arranging with DERA Malvern that they would participate in the project to make atmospheric measurements and that one of their workers, Mark Evans, would help build the required receiver system. He also became one of my Ph D students using this work as the basis of his thesis. We held a meeting on the 9th of June to help arrange the work he would carry out. I then promptly came down with flu’ and Chris experienced another ‘down’ period during June. There were two items of excellent news though. One was that on the 22nd Duncan Robertson accepted a job working for the PIC and we knew he would be returning to St Andrews. The other was that the RA (Radiocommunications Agency) agreed to provide over £200k to fund the building of a suitable system and carrying out the proposed set of atmospheric measurements. A snag being that the University E&B were still failing to actually carry out tasks that they’d been required to perform, so were holding up progress.

During the first week of July, Duncan Robertson and I went to DERA Malvern to attend a meeting with the British Airports Authority and discuss possible provision of mm-wave scanners at airports. In effect, looking for another application area for the technology the PIC and mm-wave group were developing. However during the following few days Chris went though another period of multiple fits, panic, and confusion. This lasted from about 10th July until about the end of the month. The anxiety about Chris began to have two apparently physical effects upon me. One was bouts of pain in my lower abdomen. The other was that I found that my eyes were becoming very sensitive to bright light levels. As a result, when I went out I was increasingly finding that my eyes hurt and I would struggle to see unless I wore a pair of clip-on sunglasses in addition to using lenses which darkened in sunlight. I also got into the habit of always wearing a hat with a brim to shadow my eyes from direct sunlight. At the time I assumed these problems were purely physical. Only later did I realise they had probably been caused by stress. During this period Rob Hunter dubbed me, “The man from De Monte” because the hat I wore reminded him of the man in an advert who wore a similar type of hat!

At the end of August Chris again went into a very bad ‘down’ period that lasted though September and into October. She had many fits during this period and was often confused, terrified, and/or seriously depressed. On the 1st of September she’d gone to Dundee to work at the CAB, but was too nervous. Called a taxi to get home and had a fit on the way. The driver knew what to do because he’d driven Chris home before on more than one occasion when she’d had a fit. (He remembered this, she didn’t!) In fact over the years the drivers of our local taxi firm – Williamsons – have been very helpful on many occasions like this. On the 5th Chris felt too confused and anxious to get up in the morning and was beginning to panic. I phoned and a GP came to see her. After this I managed to persuade her to get up at lunchtime and we did go for a walk down to the harbour and back in the afternoon. However on most of the following days Chris felt unable to get up in the morning because she felt too confused or frightened, and did not go out. Although on some days she was able to come out to do shopping, often this was left to me, as was cooking meals.

So far as possible, I would stay home with Chris, and do what work I could using the computer at home when not with her. Chris would wake up in the very early morning and want to talk about her anxieties when in a confused state. I’d eventually get breakfast although she often was unable to eat hers. On the 8th I did get to a PIC meeting and then came straight home. On the 14th Chris’s CPN (Community Psychiatric Nurse) came to see her in the afternoon. On the 16th we went shopping together in St Andrews but Chris became frightened and was trembling with fear when we were in the supermarket, although she was unable to say what was making her so anxious, but she just wanted to get home and lay down. On the 18th she did manage to go to Dundee and pop into the CAB before buying some things in the nearby M&S and coming home for lunch.

This unpredictable pattern continued throughout the month. Sometimes she felt able to get up or out, but more often did not. She also would tend to twitch or jolt whilst sleeping, or when laying awake worrying in bed at night, and would also get up in the night for a while, then come back again. This disrupted both our sleeping patterns making coping more difficult. However the occurrence of the problems gradually reduced as September ended and October began. I also began seeing a (different) CPN regularly to help me cope with the difficulties were were having.

During the same period Mark Evans of DERA had sent us his overall design plans for the receiver to be proposed for atmospheric measurements. I wasn’t happy with all the details, so discussed the potential problems with him over the phone and via emails. On the 29th of September we got the formal confirmation letter that Duncan Pryde was now employed as a postdoctoral researcher to work on the atmospheric measurement project, starting 1st October.

Illustration showing how the 3-Port system operated.

The basic method of operation of the 3-Port system can be explained using the above diagram. The signal from the distant transmitter is received at three locations place in a line roughly perpendicular to the path from the transmitter. The output from the center (‘C’) port is used as a reference to determine the frequency and phase of the arriving signal. The Left (‘L’) and Right (‘R’) ports are slightly to either side. The system then compares what these collect with the ‘C’ input. In an ideal situation all three signal levels are almost exactly the same. However there will be a very tiny delay in the times of arrival at the ‘L’ and ‘R’ ports causing their signals coming via slightly longer paths.

Two examples are shown above. The one on the left represents what would happen if the system is ideally aligned and the atmosphere is assumed to have no effect – i.e. is perfectly uniform and does not attenuate the signal levels or add extra delays. In this case the phases and powers observed for the ‘L’ and ‘R’ reception are the same. The example on the right illustrates a patch of damper air drifting into the line of sight of the ‘R’ port, upsetting this perfect balance. In practice the presence of water vapour in the air attenuates the signal’s power, and raises the air’s refractive index. Hence the arrival of the patch of damp air causes the ‘R’ signal power to drop, and the measured phase of ‘C’ to be delayed.

We can now imagine what happens as the damp patch continues to drift across the view from right to left. As this happens it moves into the path between the transmitter and the ‘C’ port, leaving the ‘R’ path. It now attenuates and delays the ‘C’ output. That has the effect of delaying the reference we are using to work out the relative phases of ‘L’ and ‘R’. So the result is that the ‘L’ and ‘R’ signals seem to both arrive ‘early’ compared with when the air was uniform!

Diagram of 3-port responses.

The diagram above shows four different ‘drift response’ patterns we might then observe as a small patch of either relatively ‘damp’ or ‘dry’ air crosses between the transmitter and the 3-Port system. The top left pattern is for the case when damp air drifts across from right-to-left. Note that having moved on from being in the ‘C’ path the damp patch then affects the ‘L’ path before leaving. Using the same argument we can predict the patterns for the different kinds of air inhomogeneity and drift direction.

In reality the atmospheric variations are more complicated and the explanation here is simplified for the sake of clarity. However the explanation shows that the system can distinguish both the direction of movement and the degree to which any such air humidity variations depart from uniformity. Hence the instrument can allow us to investigate the causes of fluctuations and their effect upon a communications link at the chosen frequency. When the project was started it we had intended to eventually use the 3-Port to simultaneously observed a number of transmitters placed at various locations in its field of view, but operating at slightly different frequencies. The receiver electronics was designed with this in mind, and by using many transmitters it would have provided more information on the atmospheric variations. However the initial work was to be based on using just one transmitter as a way to try out the system and provide initial results. Even this was felt to be important because the results would help communications companies predict more reliably the effect of weather upon the ability of their mm-wave links to carry information.

On the 27th of October Chris had a fit on the bus, coming home from Dundee after a session working for the CAB there. She came to and found herself walking along the dual carriageway that leads down to the Tay Bridge on the Fife side of the Tay. Apparently the bus driver had stopped and let he get off the bus while it was still on the dual carriageway– which has no pedestrian pavement or safe area! She tried to flag down a car to ask for help. The first four cars zipped by, ignoring her. Fortunately, a couple driving along in their car then saw her, realised she was in a distressed state, and stopped. They then gave her a lift home and to safety. I do not know who the couple were, but if they ever read this, I’d like to thank them for their kindness. It may have saved her life as she would have been very confused and vulnerable! At the time I was furious at the bus driver and it is probably lucky that I had no idea who they were. But I guess that Chris may have been in a confused state, banging on the bus doors in a panic to “get out and go home” having no real awareness of where she was. And that the bus driver probably then just stopped and let her out because of her behaviour, without realising that she was out of contact with reality. That said, I hope drivers since have had better training to recognise and cope with situations like this.

During November Duncan Robertson and I set out the details for what we started calling ‘AVTIS’ which stood for: All-weather Volcano Topographic Imaging Sensor. This combined radar with thermal imaging and could be used to simultaneously map alterations in the shape of the rock surfaces and their temperatures. And would do so though a thick haze of dust and hot gases. At the time I was suggesting an ‘FMFMCW’ ranging method as this had the potential to avoid some of the practical demands of established radar modulation techniques. But in the end Duncan adopted a more conventional method once he’d determined that it could be made to work fine for AVTIS. However in November 2000 we were still a long way from having gained a grant to build the system, and our previous proposals had all been rejected.

On the 20th of November I wrote to the Profs Wilson Sibbet and Alan Miller keeping them informed about my state, and Chris’s because it was clear this was affecting my ability to work. I also informed them that I was now seeing a CPN. In addition, my GP also sent them a letter regarding the problems. A few days before I had experienced a hallucination where I saw Chris quite clearly in the hallway at home, but then a few moments later realised she wasn’t actually there. I was also getting symptoms like headaches and stammering at the time. In part this was probably due to lack of sleep when Chris was ill, but it was adding to the worry.

Wilson and Alan were both very sympathetic and helpful about the difficulties. However towards the end of November, Chris started going into another ‘down’ period that then lasted though December and into early January. This was similar to previous periods and involved loosing sleep, anxiety, spending most of the day in bed trying to recover/relax, etc. Often unable to come out or see anyone, but on some days she’d come out shopping or be more active despite continuing to be frightened and confused. She managed to go to the St Andrews Community Council meeting in the evening of the 18th of December, but didn’t say anything as she felt too confused and unsure. The next day we went out and bought a small Christmas Tree. We both saw our CPNs a few days before Christmas. In addition, we had each started seeing a support worker from the Fife Family Support Project (FFSP). This was/is a charity whose aim is to support people who are having difficulty caring for someone else. On the 22nd Chris did go to the CAB in Dundee, but found it a struggle as she felt too nervous. As a result of Chris being unwell, I took most of December ‘off sick’ from work.

Our Christmas turkey was delivered on the 23rd. As usual, we’d ordered one in advance from Murray Mitchell’s. They were an excellent butcher’s shop in town and we’d established an arrangement where they would cut a turkey in half and deliver both halves to us a day or two before Christmas. We would then put one half into the freezer to cook and eat at a later date because even a small turkey was too much for us to eat over Christmas. Because Chris was anxious, I agreed to do all the preparation and cooking for our Christmas dinner. Perhaps that helped her because on the day she was actually relaxed enough to do some of the work, and so we did it together. Christmas day itself went well, but Chris became very depressed and anxious again during the last few days of December. She then felt better once we were into the new year.

By the start of 2001 I’d written a number of articles about the use of ‘Acorn’ computers and some specific items of software. These had appeared in various magazines. While I’d done this I had also started to correspond with the authors of some of the software I used. This had led to my forming a friendship with two of them in particular. One was Bob Pollard who worked on developing the !TechWriter / !EasiWriter document processor I’ve continued to use to write all my magazine articles, webpages, etc. I felt then – and now – that for a technical writer who needs to include complex diagrams and mathematical equations !TechWriter is the perfect tool for creating technical documents for publication. The other was Rob Davison who wrote and developed a program called !Composition, or “Compo” to its users. What this does is harder to describe beyond saying that “It does what it says on the tin”. i.e. it lets the user combine and manipulate graphical items to ‘compose’ them into a resulting appearance. Thus it was, and is, very handy for creating images to illustrate webpages. (All the illustrations on the webpage were generated using it.)

Both Bob and Rob were superb programmers as well as very helpful and friendly people. They would send me new versions of their programmes to ‘alpha test’ and I’d comment on any problems I might find. I also would suggest or request new features or abilities, and often found they’d added these and sent me a new version including the change in a matter of a day or two! Rob lives in New Zealand and one result was that I’d often suggest a feature one day, and he’d implement it ‘overnight’ and I’d get an new version of Compo the next morning that included what I’d suggested! This ability to interact with the programmers was a feature of the Acorn software scene that was probably a world away from any individual user trying to ask, say, MicroSoft, if they could add a feature to their main programmes or OS!

At the time Rob was creating a scripting language that could be used to automate a series of actions which Compo would then carry out at the click of a mouse on an icon. This enabled the program to do far more than simply compose still images. It could also be used to generate animations, analyse images, or create them from data sets! So it was well on the way to being a very valuable tool for anyone wanting to work with graphics. As a result, I wrote a number of magazine articles about this, and about !TechWriter, explaining how they could be used for various tasks.

At the beginning of 2001 Duncan Robertson and I were working with both Harry Pinkerton and Geoff Wadge (Reading University) to put together another NERC application to build a scanner. From discussions with them it had become clear that our earlier proposals had failed partly because of a lack of understanding on the part of NERC committee ‘experts’, but also to some extend because there were already some clashes between volcanologists, and in we’d partly fallen victim to this. However given the support from Harry and Goeff and some others we were now more confident that a new application would succeed.

During February it was planned that I would go to a meeting that had been arranged at DERA Malvern. But I had to cancel this because Chris became seriously unwell again for a few weeks. At about 3am on the 2nd she’d woken up in a panic and was unable to relax or sleep. She started to take pills and refused to stop. So I phoned for an ambulance because I was terrified she’d kill herself. The ambulance came and we prevented her taking more medication. A GP was called out and eventually she calmed down. The ambulance and GP then left. But I had to call them again on at about 10am the 9th when she went through a similar state. Again after waking up in a confused panic at about 3am. She did then talk to the GP about wanting to commit suicide and feeling hopeless. After the GP had been with us for about a hour, Chris had calmed down and the GP left. These were particularly bad days. But the pattern on many other days at that time was similar. Just that it was usually sufficiently less severe that we managed to cope without my feeling a call for an ambulance or GP was urgently needed.

Chris also remained very worried about her Father because he was clearly in a muddle whenever she phoned him. For example, he wasn’t able to say if he’d been shopping that day or not. Over the phone he complained that the “radiators aren’t working” and said he was wearing a coat indoors because it was cold. She bought a cardigan to send to him, but it reminded her that the previous summer she had sent him some clip-on sunglasses when he’d said he found the light too bright. However when she’d asked about them a couple of weeks later he had no recollection of getting them. It was increasingly clear that he was often unable to remember what had happened even a day or so earlier. Chris therefore wrote a letter to her Father’s GP, asking them to go and check on him because she felt he wasn’t well and needed help. At the same time I was becoming worried about my Mother because when talking to her on the telephone it became clear she wasn’t well, either. She said that they’d had to move her bed downstairs to the living room because she had bad ulcers on her legs that made getting up or down stairs too difficult and painful.

My Mother was admitted to Southend Hospital during the night of Thursday 8th March. I got a call from Arthur telling me about it that evening, but it wasn’t clear how serious it might be. Chris and I did talk about if I should travel down to get there quickly. But we decided to leave any decision until the next day. On the Friday I considered what might be the best way to travel and I did book a flight for the Sunday morning as this was the quickest arrangement I was able to make by that time. And I was very worried about leaving Chris by herself. However at about noon on Saturday I got a call from the Hospital saying that I should come to see my Mother as soon as possible. They would not say if what they really meant was that she was about to die, but it was obvious this was what they were telling me. Although we did not know it at the time, she actually then died at about 2pm. Arthur told me that she was quite peaceful and he with her when she died.

I’d known she was unwell, but neither she nor Arthur had ever let on that it she had cancer and that this was expected. After the event, the signs should have been obvious: When we’d last visited them she had been wearing a wig because her hair had fallen out. And having to move the bed downstairs because of her legs. I’m sure now that she hadn’t want to worry me, knowing we were so far away and could do nothing to help. This was a bitter mirror image of the way I had never told her or Arthur how ill Chris had become with her bouts of anxiety, depression, and suicidal feelings. Because I had also not wanted to worry her when she was too far away to help us.

On the morning of 16th of March I had to call an ambulance again because Chris was suicidal. A GP also attended again. The day was largely a repeat of what had happened a few weeks earlier. When talking to medics, etc, she would often say in an apparently calm and reasonable way that she simply wanted to be dead. Fortunately, once again, by the evening she was far more relaxed and became apologetic about what she had said or done earlier, although on other occasions she would simply forget what had happened. Throughout this period she would also often have a number of fits per day. Both of us were short of sleep. I was unable to go to my Mothers funeral – partly because I felt awful, and partly because I was frightened to leave Chris by herself for more than a few hours at a time – particularly in the mornings – for fear of what I would find when I came home again. I spoke to Arthur over the phone and he said the funeral went smoothly. Given what had happened I then explained to him why I’d not come to see Mum or attend her funeral, and apologised as best as I could.

Ever since I have felt ashamed that I did not react more quickly and go to her as soon as I was told she was in hospital. I felt it was serious and feared that she might die. But I was, myself, feeling so stressed and anxious, and was intensely worried about Chris being OK that I simply made myself think it wasn’t urgent and that my Mother would probably be OK, or there would be more time. ...Until that final call from the nurse telling me I should come and see her as soon as I could – and it was already too late. Ever since, I’ve felt that I should have dropped everything when first told. As it was, I let my own anxieties delay me until it was too late. To make things worse, I then felt so upset that I couldn’t even face going to her funeral. It also made me admit to myself that stress was having a serious effect on my ability to cope. Perhaps I could have got there sooner, perhaps not. But I should have tried harder. Instead I went into a state of simply hoping all would be well because I couldn’t face the likely reality or bear leaving Chris. At the time it felt like I was in a sort of automatic “this can’t be happening!” state of mind whilst, inside, a part of me kept screaming at me that it was all too real!

On the 23rd Chris did manage to come to Dundee with me on the bus. This was a step forwards as it meant she could pop into the CAB while I went to M&S to get some food shopping. Unfortunately, she then had four fits overnight which made her feel worse again. But we managed to go to – and enjoy – the dancing at All Saints Church Hall on the Sunday afternoon and we walked home having had a good day together. On the Monday Chris went to the Dundee CAB by herself and managed to do some volunteer work there, but found she was having to be helped because she wasn’t always sure what to do despite having done work of this kind for many years.

On the 29th we had a phone call from Sally (her sister-in-law) to tell us that Chris’s Father had been sectioned under the Mental Health Act into a care home for the purpose of assessment. The GP had visited him and decided this was necessary. When the GP went inside she’d found a pile of wood in the middle of the living room floor. He was planning to use this to keep the house warm because the electric power and gas supply had been cut off! It emerged later on that the electricity, etc, at her Father’s home had at times been cut off because he had forgotten to pay the bills.

Although Chris was still anxious and confused we went the the Sunday dancing again on the 1st of April and enjoyed it. Chris’s condition improved again and she resumed being able to go out. On the 1st of May we were informed that Chris’s Father had been diagnosed as having severe Alzheimer’s disease and would have to remain in a care home. Roger and Sally set about sorting out his affairs and finding a suitable long-term care home. Sad as this was, it was a relief for Chris because she’d been worried about him being left on his own.

On the 16th I contacted the ‘PDO’ Compact Disc factory regarding some discs I’d bought which had started to show signs of a manufacturing fault which caused the internal reflective layer to deteriorate and go brown. (The problem gained the name ‘Brown Rot’ for this reason.) The factory had changed the way the discs were made. This had seemed fine at first. But a few years later it gave rise to the problem, which eventually could render the CD unplayable. Their response was to promise to replace any affected disc, free of charge. So over the following year or so I sent batches of CDs back to them, and they then send me replacements.

By 2001 I had produced a few specialist websites and it occurred to me that it might be a good idea if I also created one to document the history, people, and products of the old ‘Armstrong’ Hi-Fi company I’d worked for many years before. So I set about re-contacting some of the people I’d met and worked with there. I enjoyed this and it was an emotional escape from the things in my immediate life that were alarming and difficult to cope with. I also felt – and still feel – that a lot of good work was in danger of being forgotten. So wanted to make the information known and available before the possibility of doing so was lost. Perhaps in this I was particularly prompted by my Mother’s death and Chris’s Father essentially losing his memories. However via a mix of letters, phone calls, and emails I recontacted people and started building a set of webpages on Armstrong. Sadly, many of those involved are now gone, usually along with any photos, documents, etc, they had.

It was particularly sad that Ron Sheppard died on the 6th of May, although I only learned about this when Ted Rule told me in an email on the 19th. Ron had joined the company much earlier than any of the others still alive at that point. And was for many years the helpful “voice on the phone” who dealt with callers asking for information. He was also a real ‘gentleman’ in the way he treated everyone and I’d enjoyed talking with him when I worked for Armstrong. I did manage to have a few more conversations with him, via phone or letter, and get some recollections before he died, and I’m grateful for that. But he was a real loss. On the 26th Sally rang to say that Chris’s Father was now accepted by a nursing home for long term care.

During late May onward, Duncan Robertson and I worked on the new AVTIS proposal with Geoff Wadge and Harry Pinkerton. I wrote various analysis documents, etc, to flesh out the case and show what it would be able to achieve in terms of resolution and sensitivity. Geoff and Harry concentrated on the impact this would have on what could be observed and monitored and how valuable it would prove. Chris had started taking an additional anti-convulsant medication sold under the ‘Keppra’ trade name and the dose of this had been increased during May. At first it seemed that this had reduced the number of fits she was having, so she continued to take it. However as time progressed she developed ‘puffy’ and painful hands which we suspected might be caused by the Keppra. As a result we varied the dose in the hope this was a side-effect that could be avoided whilst, hopefully, getting some reduction in fits from the medication.

At the end of May we were told that, yet again, our NERC application for AVTIS had been rejected! Following the previous rejections we had been arguing with the relevant committee that they were failing to correctly understand the basis of our proposal. Put simply, they refused to believe that the instrument could work as well as we said. The problem being that they were essentially volcanologists, not mm-wave radar/radiometer engineers. We’d put a sensible case together – again! – but were hampered by two restrictions. The first was that there was a strict limit on how many pages of explanatory text we could submit. The second was that they refused to enter any kind of dialogue that would allow us to discuss and clarify the points they tended to misunderstand. To make things worse they now had someone they regarded as an ‘expert’ on radar who flatly disagreed with us and said what we proposed wouldn’t work!

Alas, by then I was virtually unable to travel away from home most of the time because I was so stressed and anxious about Chris. This was a problem because I often wasn’t able to go to meetings with various people involved and and clarify the case we were making. Understandably, this was very frustrating for Duncan Robertson and the others, and I was worried about it, but simply could not face being away for more than very brief periods when Chris was relatively OK. So it was practically impossible to plan and make any long journeys. Instead, I had to rely on emails, phone calls, or having Duncan go instead. He was quite capable, but it must have looked bad that I was a ‘no show’ as if evading facing people. Whereas, if I had been able to travel it seems quite likely that we could have resolved the disagreement and made progess. As it was, my inability to travel probably handicapped progress on more than one occasion.

In addition, at the time a reasonable ‘strategic’ question had arisen. People accepted that radar could be valuable as a tool – but would it be better to use an airborne or satellite based system, not one using a ground based scanner? Considering this question made good sense. However our view was that each method had its advantages and drawbacks, but if you wanted a continuous monitor of both the shape and temperature profile of a volcano then the ground based system we proposed was the optimum approach. That said, if the volcanologists preferred an airborne or orbital system, we’d have been quite happy to develop them!

Following the committee’s decision we were sent a short report explaining their reasons. This included some information on their reasons and a reference to a paper which they had been told showed we were wrong. We looked at this and found that it actually supported our case! The good news was that Geoff and Harry were rather more ‘well ken’t faces’ in volcanology than Duncan or myself. And they were now determined to push what we proposed. We also did try to get the committee members to agree to read more material we sent to them, or meet with us for discussions. So we fed back our feelings and I wrote more material to explain the basis of what we proposed, etc, to feed into the argument for our case. In effect, we asked the NERC to review their decision and to allow us to provide them with material to explain why we felt this was required.

The above shows the weather radar and 3-port receiver system once they were all installed on the Physics building’s roof. Duncan Pride stood in the rain so we could use him to give a sense of the scale!...

During June, preparations were under way to place the Rutherford Labs weather radar on the roof of the Physics Building. E&B were still dragging their heels and either failing to do promised work on time, or finding obscure reasons why ‘nothing could be done’ – or simply going into a state of apparent hibernation. In essence, they simply ignored anything I said or wrote, and I had to keep asking the head of the department to kick them into action. By the end of the month the actual radar dish and its support and front end had been placed on the roof, but the process didn’t go smoothly... The Rutherford Lab had been told that the roof was ready to have the radar placed there. They arrived on the agreed day with the radar system and a crane, ready to lift it into place. E&B then said some of the required support structures weren’t installed, despite what they had previously said! Fortunately, this was then dealt with fairly quickly, although the episode clearly made the University look shambolic.

View of the radar and 3-Port receiver from out in front of the building. This photo was taken before the ‘patriotic washing-up bowls’ were added to provide some stylish colour!

Alas, once physically place it still wasn’t able to operate because E&B – despite promises – still had not done the work needed to run cables down from it and install the associated electronics into the room on the top floor assigned for them. This equipment would allow the Rutherford lab people to run the radar remotely and collect the data it provided. Even putting the electronics into the room didn’t happen until the last Friday in June, and it wasn’t connected and working until some days later. To make things worse, E&B had allowed someone else to place a large cooling fan right where we had arranged – with them! – that we would place the 3-port interferometer once it was ready to use. So it seemed that the E&B people didn’t even to talk amongst themselves, let alone anyone else! The irony being that they had suggested the location for us in the first place.

At the start of July the Institute of Physics Press agreed to publish a paperback second edition of the undergraduate textbook I’d written, ‘Information and Measurement’. On 3rd July we held a meeting with everyone involved on the atmospheric measurements project. Overall, people were happy with the progress Duncan Pryde and I were making, but the Radar still was not working. So the people funding the project were unhappy about the delays caused by E&B, but willing to press forwards as soon as possible. Duncan had made a presentation on what we were planning at an international conference and it had apparently generated a lot of interest. The work was now actually a large part of a bigger ESTEC project.

During July Chris picked a lot of raspberries in our garden, and I then made another batch of raspberry jam. At the end of July there was a meeting with the NERC committee to consider our request to reconsider their rejection. Again, this failed as a result of the ‘expert effect’. Duncan Robertson did attend the meeting and said when he returned that he was amazed that their ‘expert’ simply seemed not to understand what we were saying and rejected it out of hand. But we refused to give up.

Chris’s Birthday Ceilidh

Chris had decided that she wanted to celebrate her birthday by organising a Ceilidh based on the Sunday Scottish Country Dance group we were attending quite often at the time. She really enjoyed these and they made her feel much better. I enjoyed them as well, but did struggle with the dancing. She arranged for some cakes to be provided by our excellent local bakers, Fisher and Donaldson and it made more of a party atmosphere.

On 21st August I was shocked to find that over £9,000 had been deducted from the mm-wave group’s accounts for the work done to install the weather radar! It had been made crystal clear long before that E&B should bill the Rutherford Lab directly for the work they – eventually – did. Not charge either the Physics dept or us. The charge had then been paid without my approval or knowledge. Which should simply have never happened because I’d not authorised the transaction. It then took some time to recover this money. During August Duncan Robertson and myself also set about contacting some genuine experts on mm-wave systems and radar that we knew. The aim being to use these as counters to the tame ‘expert’ NERC had been unwisely relying upon.

During September Duncan Pryde was also finishing off his Ph.D. Thesis to hand in a few days later, ready for his viva. John Goddard of the Rutherford Lab had agreed to be his external examiner. Duncan also was carrying out some preliminary measurements on the completed 3-port system. We’d previously tested the signal transmitter for this, and the three ‘front end’ receiver boxes and they’d all been OK when used individually in the lab. But we discovered that the behaviour of the complete system kept varying unpredictably. The variations didn’t correlate between channels or seem related to any weather effects. Following more tests it became clear that there were a number of practical problems with the system Mark had built at DERA Malvern. This manifested in use in apparently random variations in the gains and noise levels of the three input channels of the system. Which seriously impaired performance as an interferometer.

Having taken the system back into the mm-wave lab, Duncan carried out a much more in-depth investigation. He found that there were various flaws in the way the system had been put together. For example: in a number of places the power supplied to items like amplifiers or convertors was provided from a voltage regulator – but via a series ‘dropper’ resistor because the unit required a lower voltage than the regulator provided. This did allow one regulator to power a number of items, but is not a good idea. The problem with it is that it relies on the powered item to draw a steady current that is just right to enable the drop across the resistor to provide the powered item with exactly the correct voltage it requires. Unfortunately, many of the items used didn’t draw a steady current. And many were also temperature sensitive. Hence checked briefly in a lab the behaviour might seem OK. But placed out on a roof in a box exposed to the weather, the current and voltage provided to the items would wander around.

Once we realised what Mark had done, it was obvious why the performance was so variable. Until the electronics was delivered we’d relied on functional diagrams that didn’t detail all the practical electronics, and we hadn’t expected such problems to be assembled into the actual electronics. Sadly, one of the front end sections was also damaged by this before we realised what was happening. But that was replaced. Another problem was with ‘rf loops’ and internal signal leakages because the individual items weren’t always connected together in the best way. The result can be stray signals from one part of a system degrading the behaviour of other parts. Under investigation a number of problems showed up, each having consequences like reducing sensitivity, increasing the noise level, or causing unpredictable performance variations.

Experienced RF engineers are familiar with problems of these kinds and wouldn’t have built the system as Mark/DERA had done. Finding and fixing all of them was to take Duncan some months, and significantly delayed the system coming into reliable operation. Strictly speaking, the arrangements we made meant that we could assume we could rely on DERA to build a satisfactory system. However I’d accept that I shared any blame because Mark was at the time one of my research students, despite doing all his work at Malvern out of my sight. He’d also phoned me on a number of occasions because he had difficulty understanding the operational principles of the 3-port interferometer’s operation. Perhaps that should have set off warning bells in my head. However in the past DERA had always built excellent equipment when we worked with them, and they had developed the circuits used as the front end of the receivers which had outstanding sensitivity and low noise.

To be honest, so far as I was concerned, doing something like using a ’dropper’ resistor in this way was the kind of elementary mistake I’d took for granted no one with any knowledge of electronics would make. I’d assumed the guidance Mark had available at DERA would allow him to put together a reliable system, and that they would be supervising this work as a part of their contribution to the project. Against that, he was my research student, and thus I also had some responsiblity for supervising his work. Unfortunately, I was having to do this remotely via telephone calls, letters, etc, because my personal difficulties at the time made being away from home very difficult. As a result the scrutiny of his practical work fell between two stools! So we shared the blame, put up with the annoying delay, and the eventual system worked very well. But it diverted Duncan’s effort for some months and seriously delayed the start of reliable data collection.

On the 12th of October we agreed a press release about the project to make atmospheric observations and fixed the 1st of November (a Thursday) as the day for Duncan Pryde’s viva. During October we also worked on the next proposal we planned to submit to NERC to try and get funding for AVTIS. By this point we were also planning to suggest that once we’d succeeded in extracting the required dosh we should appoint Dav Macfarlane as the postdoctoral researcher to work full time on building the system. He had already been working on mm-wave thermal scanners, etc, in the lab with this in mind.

Duncan Pride’s Viva celebration at the Vine Leaf restaurant.

At the start of November I was told that copies of the second edition of my “Information and Measurement” book had now been printed and were with the publisher. In addition to having some copies for myself, we also arranged to send a copy to various academics who said they’d recommend them for the courses they gave. The radar scanner on the Physics building roof was now in operation and the 3-port system had been installed. Duncan’s viva went without any hitch, and we then had a celebration meal in the ‘Vine Leaf’ restaurant. The press release the University sent out to announce the project appeared in newspapers during the first week of November. (The result as it then appeared in one newspaper is illustrated at the top of this webpage.)

On the 1st of December I was told that Bob Pollard had died. I was very sorry to hear this because I’d got to know him very well, despite not ever meeting him face-to-face. He was a superb computer programmer and a very friendly, helpful, and smart person. For years I’d enjoyed chatting with him via email and trying out new versions of !TechWriter. In the middle of the month Chris went into another ‘down’ period. As had happened in previous years she became anxious about cooking the Christmas dinner even though I’d said I was happy to either help and share doing this, or simply do it all, she still worried about it. How much this was a case of worrying about the cooking causing the anxiety, or feelings of anxiety leading to feeling unable to cook, it was almost impossible to say. More like a case of a negative-driving feedback loop that any feeling of anxiety would trigger, and then caused a rapid downward spiral. The now-familiar symptoms – fits, unable to sleep, confused conversations, and some loss of memory, etc – also reappeared. Fortunately, Christmas day actually went well despite the anxieties, and her condition improved again during the following few days. So by the second week of January 2002 she had largely recovered.

During January of 2002 I was analysing some preliminary measurements using the 3-Port system to receive signals from a transmitter we had briefly set up on the top of Dundee Law. (This is a tall hill in Dundee which gives a good long distance view of the direction towards St Andrews.) At the time we were still explaining to Mark and DERA the nature of the various build problems with the 3-Port system that Mark had built. For obvious reasons they wanted to present these as only requiring “minor modifications”. However I had to explain that although – once identified – each change was “minor” the real problem for us was that Duncan was having to spend a great deal of time testing and searching for the causes of some of the performance limitations, even though fixing each one when identified was relatively simple. Thus the ‘fault finding’ process was generating a long delay. To some extent this kind of thing is normal for a new system, but in this case the number of “weevils in the biscuit” was much larger than usual! For my part this process required me to write, and reply to, a number of long detailed sets of discussions to clarify the situation. The good news was that Duncan was making progress and the 3-Port’s behaviour was improving, albeit slowly.

Chris had continued to experience problems with her hands and wrists. Her fingers would sometimes swell up and become uncomfortable. And – particularly when typing at her computer – her wrists would also become painful. This was contributing to her feeling depressed and anxious. The GP and a specialist had decided that she suffered from ‘carpal tunnel syndrome’. On the 15th of February we were told that she should have both her forearms operated upon. We still suspected that the problems with her arms/hands might be a side effect of the Keppra she was taking. But the medics dismissed this possibility saying that it “wasn’t a known side-effect”. Which in my mind did prompt the thought: “Well if other GPs have simply dismissed it for that reason when a patient has suggested it, then none of them will have reported it, will they?!”

On the 22nd we were able to update Dave Eden of the Radiocommunications Agency and let him know that we now had a 3-port system which operated reliably with low noise and high sensitivity via all three of its input ports. Until that point measurements had shown that two of the three ports generated much more noise than the items were meant to generate, and this was degrading performance. A problem made worse because the noise levels and gain varied for other reasons of the kind I’ve already mentioned. This had finally been solved when DERA supplied some new front end electronics that met the required specifications.

The above images show Duncan using the transmitter for initial tests from Craigowl Hill, and the mast he climbed at a later date to attach the transmitter for continuous use.

By this time we had settled on Craigowl Hill, north of Dundee, as the location for the transmitter during the 3-port atmospheric measurements. Duncan had visited the site which was what communications engineers tend to call an “antenna farm”. i.e. a location that works well enough as a place to put transmitters and receivers that it attracts a number of them, for a variety of purposes. The site was owned and run by NATS and the CAA (National Air Traffic Service and Commercial Airports Authority) so we had set about obtaining their permission and a transmitter license. Some preliminary measurements were made with our transmitter on a tripod, firing out though the chain-link metal fencing. These sufficed to establish that we were able to get a decent signal level at St Andrews.

Photos viewing up and down the road leading to Craigowl antenna farm in winter.

A snag during January and February was, predictably, the weather! In particular it kept snowing quite heavily on Craigowl Hill. As a result some of the initial measurement sessions with the transmitter on a tripod were abruptly cut short due to the snow. One day, Duncan had set the transmitter operating and came back to St Andrews to operate the 3-Port receiver. However he then got a call from Craigowl alerting him to a snowstorm there. So he had to go back and rescue the transmitter from a blizzard – and then escape again to avoid being snowed in at the top of the hill!

Engineers are familiar with accepting that “Sod’s Law” acts as one of the unavoidable laws of nature. i.e. that whatever can go wrong, will go wrong. This had repeatedly cropped up as we worked on the equipment for the atmospheric measurements project. One example was that the 38 GHz source we were using at the transmitter abruptly failed during February. We had been powering it with a home-made supply, but had decided to replace that with a commercial supply that was sold as being higher quality. Alas the commercial supply then duly failed, and blew up the transmitter! Fortunately we also had a 39 GHz source as well, so changed to using that with our own supply. But this meant we no longer had a spare/backup.

During February, Duncan climbed one of the transmitter masts at Criagowl and installed our transmitter there, leaving it to operate continuously from then on. It was now well clear of the ground, and he wouldn’t need to rescue it – or himself – when it snowed. He could now concentrate on the receiving end of the system. Of course it also snowed in St Andrews, but the lab was warm and dry. So once we had run data cables down from the roof into the mm-wave lab, recording the output from the receiver didn’t require a trip up into the open. But Sod’s Law did manage to strike again later on...

The front end receivers for the 3-Port system were built into metal cabinets which were sold as being weatherproof to ‘tropical storm’ levels. i.e. it was assumed that they were perfectly watertight even in the Scottish environment. Each cabinet had a door with a lock. This was so the items inside could be adjusted or worked on without having to dismantle the box or take out their contents. Given the guaranteed weatherproofing we had, understandably, arranged to have the doors on the top of each box to make any in-situ work on the equipment more convenient.

As the days past we found that the performance still varied more than expected. We opened the boxes... and found water inside them! We checked the door seals and they seemed to be in place. So we dried out the boxes and put some drying compound inside each box as a short term measure. Over the following few weeks it became clear that the boxes leaked and, eventually, we established that water was getting in via their lock mechanisms.

View looking over one of the receiver boxes towards Craigowl. You can also see the offending lock mechanism on top of the receiver box!

Fortunately, George Radley our group technician came up with a brilliant technical solution. He checked the size and shapes of the boxes. Then he went into one of the cut-price shops in town and bought an economy pack of plastic rectangular kitchen sink washing-up bowls. He cut a slot into one end of each bowl. He then turned them upside-down and placed a bowl over each of the input boxes. The slots allowed the signals from the transmitter to reach their receiving lenses. The bowls then prevented rain from reaching the doors or locks of the boxes. From this point on anyone looking up at the front roof of the Physics Building would have seen three brightly colourful items – red, white, and blue peering out over the railings, lined up beside the large radar dish! They worked perfectly and we had no more problems with water getting into the electronics. Cheap washing bowls, 3; ’waterproof cabinets’ nil!

During the last week of February Chris entered another period of being unwell and was becoming very anxious and confused again. In part this was triggered by worrying about the proposed operation on both of her arms. Her condition worsened during the first week of March. Despite her being in pain some of the time we both had doubts about an operation being needed and still suspected that the real problem might be the ‘Keppra’ she was taking. Despite this, her GP recommended increasing the daily does of Keppra as they dismissed the possibility that it was causing problems.

On the 12th I was able to report to Dave Eden that the 3-Port system was now collecting data and observing the Craigowl transmitter on an almost continuous 24/7 basis. We set up a system which allowed us to use an Acorn RiscPC machine to continually record the data into a series of files on a hard drive that was placed in a removable ‘caddie’. By having a set of these drives we could then allow the system to run for a week or so without a break. I (or Duncan) could then halt the process, remove one caddie and insert another with a ‘blank’ drive. Then restart the data recording process to run for another week or so. This only took a minute or so. We could then load the data-filled caddie into a second RiscPC and use that to process the raw data down into a more manageable format.

The results were then written onto a set of dated CDROMs, copies of which were sent to Dave Eden and John Goddard for their own reference and analysis. I would also take copies to do some simple analysis of my own to check the system was working and showing interesting effects, etc. During the first few weeks of operation we relied on a large battery to power the transmitter at Craigowl. This should have kept the transmitter operating for long periods because the transmitter’s power requirement was quite modest. It was actually only generating a tenth of a Watt! (For those who know about radio transmissions; it was used with a feed horn antenna with a gain of about 20dB. This gave a clear signal as if having a 10 Watt ERP when seen from St Andrews.) However the battery proved less reliable than expected, and it was replaced with a mains-powered supply which then allowed the transmitter to run without any interruptions. This procedure then ran for many months and collected a very large amount of data.

During March I did a a ‘quick look’ analysis of some of the early data and made it available to Dave Eden and John Goddard via a set of webpages I wrote with some illustrations showing how the measured ‘L’ and ‘R’ signal phases and powers varied with time. By a quirk of fate that old set of webpages is still actually available, so if interested you can find a link to them at the bottom of this page! However one interesting/curious result is shown here:

Curious phase changes before coffee time!

The above shows some initial results obtained on the 17th of March. At about 10am the phase of the ‘L’ channel starts to change quite quickly in a negative-going direction and it does this before the ‘R’ channel seems to alter. Shortly afterwards the behaviour alters and both phases show a positive-going change. During this time the power levels fall, indicating the air is becoming much wetter. The power variations indicate that the atmospheric attenuation by the time we got to just after lunch was much greater than in the morning. So it was probably quite a very wet day – and one that would have had a severely degrading impact on the data capacity of a 30 - 40 GHz communications link! Many other similar atmospheric events would be recorded over the following months.

Now the system was in operation the main problem we had was that Duncan Pryde’s funding was coming to an end! The commencement of continuous data collection had begun many months later than originally planned because of all the initial problems in receiver system he’d needed to fault-find and fix.

During the second half of April Chris went though another ‘down’ period. As a result I mostly did what work I could from home during this time. I was preparing a talk to present to the Radiocommunications Agency people in London on the 1st of May which showed that the 3-Port system was now working well and producing a great deal of data. This could then be combined with the output provided by the Rutherford Lab’s weather radar to determine how atmospheric variations affected transmissions at frequencies in the 35-40 GHz region of the spectrum. So the project was clearly successful.

The really excellent news was, however, was that just before the end of the month we were informed that the NERC had – at last! – awarded us a grant to build AVTIS! So all the persistence and arguing had finally borne fruit. The Principle Investigator was Professor Goeff Wadge at Reading University. But the bulk of the £240,000 grant would be spent at St Andrews as we built and tested the radar/radiometer scanner. The grant supported a postdoctoral position which we awarded to Dav Macfarlane who had been working on scanners with this in mind.

Press release on AVTIS as it appeared in The Herald.

During April Rob Davison had made a trip to Europe from his home in New Zealand. We’d arranged for him to come and meet Chris and myself and we’d intended to spend the day with him and have a meal together. Unfortunately when he arrived on the 26th Chris was unwell and could not face going out or do anything. So I spent some time with Rob in town, and he did briefly visit us both at home during the day. But the planned meal for the three of us in town was abandoned. It was good to see him, but the day was rather less enjoyable than we had hoped.

I flew down from Dundee to London City Airport on the 1st of May and gave my talk at the Radiocommuncations Agency and discussing the project with the people there. They said they were pleased with the project and we talked about how work would now proceed. I then flew back home again to reach Chris again by dinnertime. I did not want to be away any longer than necessary because she still wasn’t feeling very well. She was improving by the start of May, and she wasn’t in a serious condition during the following few weeks. At the end of May we reduced her Keppra dose as we felt as things were it might be doing more harm than good. Duncan Pryde left the group because his funding had ended. This meant that I was left to do the weekly task of saving the data, processing it, and sending copies to people while the 3-port system continued to operate.

At the start of June Chris’s condition deteriorated again quite markedly. This time the situation became very serious and persisted for a much longer time than it had before. Her condition became quite bad and on the morning of the 17th she became very frightened, depressed, confused, and attempted suicide. As on previous occasions the desperate state started at about 3am, and she became increasingly frantic. At 7am I called an ambulance, the police attended, and she was then taken to the St Andrews Memorial Hospital for assessment. For a while it was thought that she would need to be admitted to Stratheden Hospital. But she calmed down and became more rational during the day. So she came home with me late that evening.

During the rest of the month her condition varied unpredictably from day to day. Most days she was too frightened, depressed, and confused to go out or do anything. So she tended to spend a lot of time at home, often in bed and not eating meals, and unable to see anyone. During the early mornings she often experienced her arms or legs involuntarily jerking or twitching, which clearly also upset her and made it hard for her to sleep or relax. I would do most of the shopping and cooking. However on other days she would come out with me and feel better. On the 22nd she did manage to go to the CAB and do some volunteer work there as I’d said I’d go with her, and did some shopping whilst she was in the CAB. Her condition gradually improved during the first couple of weeks of July and things seemed to settle down for a while.

Unfortunately, things deteriorated again in August and from about the 12th onwards Chris became confused, frightened, and was experiencing involuntary jerking of her limbs again. As had become the norm, I worked from home as much as possible. But on the 20th I went into the Physics Dept for Bal Kumar’s PhD viva. Bal worked for GEC but had also become my research student, working on topics related to some of the DERA funded projects we had been carrying out. The by-now-usual mix of good/bad days for Chris continued until the later part of September. We both continued to see a CPN and a support worker from the FFSP to help us cope. I was also now seeing the University Occupational Health nurse, but it remained a struggle for both of us. During all this I did as much work as I could. Since Duncan Pryde had left this included going in to the Physics Dept about once a week to collect and process a week’s 3-Port data and send copies to Dave Eden and John Goddard. I also did some preliminary analysis of the results to see how things were going and produced some webpages to help others involved to interpret the results. However it was becoming increasingly clear to me that things could not go on as they were and that I might have to retire or work part time because most of my time and concern was devoted to trying to help Chris.

During the second half of 2002 Chris had also been transferred from having one Neurologist as a consultant a Ninewells to another. Our first meeting with the new consultant hadn’t gone well because Chris was very confused and anxious at the time and the consultant hadn’t seen her before or knew what to expect. (Anyone familiar with the NHS knows that no-one ever has time to read the case notes before actually seeing the patient. Indeed, quite often, no-one knows where the notes are because some other specialist has kept them and forgotten they have them!)

Over the months Chris and I had come to feel that her treatment wasn’t being correctly directed. Until then it had been assumed that the key problem from a medical POV was her epilepsy. However as her suicidal depressions had become worse, and more common, Chris and I had begun reconsidering this. In practice, her periods of depression, confusion and anxiety seemed more of a threat to her life and well-being than the fits. When we’d asked about treatment and medication for depression we’d been told more than once that this was inappropriate because that the medications for depression were actually counter-indicated for someone with epilepsy.

However I’d gradually become increasingly sceptical of some of the things we had been told by GPs or consultants. As an academic I was used to doing my own research, and fortunately, Chris’s brother had also been a professor of clinical pharmacology. So I started to do my own investigations. This allowed me to discover that, in fact, some medications were used as an anti-convulsant and anti-depressant! So we began asking the consultant and GPs about this.

During October Chris’s condition deteriorated again for a few weeks, then improved again. I had also noticed another daily repeating pattern develop when she was unwell. This was that each day she seemed not only to forget some of what had happened the day before. She was also mis-remembering what had happened. In particular, day after day she might have a bad morning of feeling anxious, suicidal, etc, then calm down during the day. We’d have a conversation during the afternoon or evening to discuss things and come to some agreement that made she felt resolved the reasons for her anxiety, etc. The day would end with her being quite happy and saying that tomorrow would be better. But then at 3am or 4am the next morning it was ‘Groundhog Day’. She would then either insist that her being happy the previous evening was a ‘fantasy’, or that it simply hadn’t happened. This cycle kept recurring and it was as if something was ‘editing’ her recollections, making them seem unreal, or quite different from what had actually occurred. However if I said this during the day before she was feeling better, she would angrily reject it.

However on the 29th of October I persuaded her to let me make an audio recording of our conversations so we could, the next day, check who had accurately recalled what we’d said and done. The following day we played back the recording and she was shocked to find that she wasn’t recalling what we had actually said or done! This wasn’t an immediate help, because she still would go round the loop each day until she came out of the series of such days. But when she came out of the daily loops again it was useful in helping us to get common understanding of the way she was experiencing memory problems that tended to make her depressed and anxious.

After our repeatedly suggesting the change, her GP and consultant finally agreed that Chris could cease taking Keppra at the end of the month. Arrangements had been made to operate on both of Chris’s arms to deal with her ‘carpal tunnel syndrome’. However Chris and I had decided to postpone this – partly because it seemed crazy to me to render her unable to use both arms at the same time – but also because we strongly suspected that the problem was a side-effect. And lo! over the next few weeks the pains in her wrists and swelling of her fingers faded away... So the operation wasn’t ever needed.

Dav Macfarlane’s PhD viva was held during November. He then became a postdoctoral worker on the project to build AVTIS. The other good news from my POV was that I’d found someone with a set of Armstrong 700 range amplifiers who was willing to sell them to me for a decent price. I’d been wanting a second set for years, but only knew of a few, and most owners wanted to keep them! I was delighted to get the second set which were in excellent working condition. I’m still using both sets quite happily. I had also got into contact with Keith Howard about some work he was writing about for ‘Hi Fi News’ magazine. One thing was leading to another and I was now planning to write an article for them.

During the last couple of weeks of November I’d caught some kind of virus, developed pneumonia and felt quite ill. During that time Chris was very helpful and supportive. I recovered, but the bad news then was that during the last week or so of November Chris went into what would prove to be the worst episode of anxiety and confusion we had until that point ever experienced. And on the 3rd of December she made a deadly serious suicide attempt. It was a day unlike anything I’d ever expected to experience in my life. More like the kind of thing I’d assumed only happened to people in the most scary TV dramas.

The day started like many others during the period. Chris was unable to sleep after about 3am and instead lay awake, frightened, confused, jerking, and trying to have a conversation with me about how she felt and what to do. She then lost control and started to take as many of her medication pills as she could in an attempt to overdose. I managed to get the pills away from her and she went downstairs. I followed and again tried to talk with her. I then decided to dial 999 and ask for an ambulance because she was still trying to find a way to kill herself. She came into the hall, determined to stop me from phoning any for help. We ended up struggling together on the hall floor near the phone as she tried to stop me using it. She then managed to pull the phone chord and ripped the connection from the wall as we rolled about the floor.

Having decided this meant I’d be unable to call, she then got up and went to the drinks cabinet and tried to swallow as much alcohol as possible, although she found this difficult as it made her cough and retch. Normally she wouldn’t drink any alcohol because it might cause her to have a fit, and interfere with her medication. But she assumed it would combine with the pills to increase the chance she would succeed in killing herself! I’d realised that I could get the phone reconnected as she’d actually only pulled out the plug from the wallbox. So I dialled 999 and called for an ambulance. As I did this I had to sit on Chris while she lay on the hall floor in order to stop her from trying again to destroy the phone.

Once I’d made the call I got up and she then ran into the kitchen, opened the knife drawer and was choosing a large knife, clearly intending to use it on herself. I grappled with her, grabbing her wrist, trying to pull the knife out of her hand. We then fought as she kept trying to use it on herself. When I managed to take away the knife she ran back into the hall then tried to throw herself down the stairs. I’d followed her as she went up the stairs, so she fell onto me, instead. We then struggled on the stairs and in the hall until the ambulance arrived. Chris refused to go with them and the police were called. When they arrived I was still in my underwear while they tried to control her and calm her down and I was trying to explain what had been happening. Once they’d agreed she must go to hospital I quickly got dressed so I could go with them in the ambulance.

Chris continued to struggle with the police and ambulance crew and repeatedly made clear that she wanted to be dead, not to go to hospital. When we arrived at the St Andrews Cottage Hospital she refused to get out of the ambulance. The police explained to me, regretfully, that if she would not co-operate they would have to arrest and handcuff her in order to legally then be able to get her into the hospital. I explained this to Chris, but she still refused, so it was done and she was admitted at about 7am. It required two police officers to do this because she struggled. The police were, however as careful, considerate, and indeed apologetic as you could wish for in such a truly awful situation. After an initial examination she was then sectioned and was taken to Stratheden Hospital, with me accompanying her.

Looking back on it, the most frightening part of the day is when I’d struggled with her to take the knife away, while she frantically tried to use it on herself. On reflection, either – or both! – of us could have been killed or seriously injured. But at the time I was so desperate to get the knife away from her that I didn’t think of the risk that I might be injured or killed. I only knew I had to stop her harming herself. And only later on did I wonder what might have happened to me if she’d managed to kill herself with the knife at that point. To the police it could easily have looked like I’d attacked her and murdered her with the knife, and everything else was a ‘cover story’!

Having spent most of the day with her at Statheden I came home. The next morning I went back and spent most of the following day there with her. However when I got home on the 4th I found that the house was dark because there was a power cut due to a fault out in the road. This made ‘dinner’ an even more depressing experience that evening, but the electricity came back on again at 11pm.

Chris spent a few days in Stratheden before they decided it was safe for her to come home. On the 9th she tried to write a letter to her Consultant Neurologist at Ninewells about the episode. However she was still quite confused and anxious and couldn’t write coherently. In fact during her second attempt to write she had a fit, perhaps as a result of finding it distressing and difficult. She then asked me to summarise what she wanted to say. So I typed out a letter for her, which she signed, and we attached her handwritten attempts to indicate the state she was in. Perhaps the most significant points she made were that:

• She had no recollection of the police being called, or what had happened that morning. Although she realised it must be true because she’d been sectioned and the police and staff at the St Andrews Hospital confirmed it.
• That we both now wondered if these episodes were due to something like ‘Temporal Lobe’ epilepsy which we had come to think could cause these strange almost delusional states and behaviours. In effect, her state would become like a long weird hallucination which she then tended to forget or misrecall later, but repeated on a following day as if fresh.

We’d both come to feel that her current treatment or diagnosis simply wasn’t addressing the real issues. The specialist who’d examined her at Stratheden had recommended that she be seen by neuro-psychologist, so the letter to the Consultant asked if this could be arranged. However the reply we got told us that she could not seen one.

I asked my FFSP worker about this and he wrote to the Neurologist. But promptly got a rather short and dismissive response. The wording might be described as either ‘brisk’ or ‘arrogant’ depending on your disposition. It was perhaps a hint of some of the later ‘treatment’ Chris experienced from the Neurology department at Ninewells over the following years. Behaviour which a public report decades later came to call ‘dysfunctional’. However at the time although it was advised by Stratheden, we simply were told it was ‘unavailable’. Another aspect of the rather disorganised behaviour of Ninewells was that their letters from Consultants, etc, often had strange dates where the date the content was written was sometimes days, weeks, or even months different from when it was datestamped or sent! In addition, arrangements sometimes clashed. For example, at the same time as the proposed arm operation another letter from the hospital had said she would be admitted to another ward for observations. But she clearly could not be in two places at the same time! One ironic result of the December episode turned out to be positive. She was in Stratheden on the day which had been set for both her arms to be operated upon. However, having stopped taking Keppra the problems had vanished, so we didn’t need to worry about the pointless operation anyway!

Chris’s illness during December meant I didn’t go into the Physics dept for a few weeks. I spent the time initially at Stratheden, and then with her at home. However the 3-Port system ran and collected data throughout, so I was able to collate and sent off the results on the 20th. Compared to the beginning of the month, Christmas itself was fairly quiet although Chris remained anxious, experienced a mix of panic attacks and confusion, and had trouble sleeping, etc. All mixed in with fits. Christmas that year wasn’t wonderful, but it was far better than it might have been, or the start of the month!

My thanks to Duncan Robertson, Dav Macfarlane, and Duncan Pryde for various photos on this webpage.

Footnotes:

I have tried to describe what Chris and I went through in some detail because we both hope that doing so may help others who face similar problems. Even if their GP, etc doesn’t understand or seem to help. A key problem for us for many years was a failure by the various medical people involved to make the correct diagnosis. Instead, Chris was ‘labelled’ in ways that self-misdirected the medical people involved and blinded them to the real causes of our problems. As a result 2000-2 turned out to be just the start of what became a truly horrific period in our lives. And worse was to come...

Eventually, however, we did manage to establish the correct diagnosis and get appropriate treatment. Her condition improved and things since then have been far better for us both. I will write more about this for pages covering the times after 2002. Some of the events in later years may will seem shocking to some readers, but I fear they are more common than most people realise, and can have deadly consequences for the unwary.

So if you or someone you know is struggling with similar problems, please know it is possible to get though them and things can get better once you get the appropriate diagnosis and help. You are not alone. You are not the only person who has been misdiagnosed or mistreated, or even abused by the medical system which is meant to help you. You can get though it. We did.

By the way: You can still find the ‘quick look’ 3-Port results pages here if interested

https://www.st-andrews.ac.uk/~www_pa/Scots_Guide/MMWave/atmos/38ghz.html