Radio, Radio

...the radio is in the hands of such a lot of fools... –Elvis Costello

Dymar logo on Lynx 850 radiotelephone

It's hard to imagine a world without mobile phones, but at the end of the 70s, they were still just a gleam in a yuppie's eye. Indeed, it's hard to imagine when mobiles were thought of as yuppy status symbols rather than just another gizmo we all can't live without. When I left school (1978), I went to work for a company called Dymar Electronics. I was a dreamy schoolboy and sort of fell into the job by default. I was interested in electronics as a hobby, and as it turned out the apprenticeship I more or less got by accident turned out to be the best education I ever had. Dymar built car telephones. The sort that you might find in a taxi nowadays (and back then too, of course), though even taxi radiotelephones have gone high-tech now and have digital displays of queued jobs and credit-card processing facilities.

Back then if you were a businessman you could have a very expensive car telephone system that connected to a base station that had to be within 20 miles or so. You spoke to an operator, always. You shared the channel with every other subscriber in your area, and so had to keep your very expensive calls very brief. Privacy, such as it was, was created by a system called selective calling, where the audio channel was blocked unless opened by a special tone sequence. It was all very primitive. Still, Dymar had a fantastic reputation for quality, and working there helped me understand just what that meant when manufacturing electronics. I eventually ended up in the research and development department, having been spotted as a sort of "teenage prodigy" by the Technical Director. For my tender age (17), I was given some pretty demanding design work, which I loved. Work became my hobby, since basically I just sort of tinkered with radios all day. This led to a number of interesting projects, which this page is all about.

CB

Let's face it, CB was naff. But it had a definite lure for any teenager in 1979, due to its illegality in the UK. For me, it was another aspect of my radio hobby, so naturally I was an early "adopter". Initially, CB was a mere curiosity - something my fellow college students and I had heard rumours about, but had never used. That quickly changed when one day I walked into work - at that time I was in the service department - and one of the guys said "have a look at this crystal". He handed me the crystal (the device that establishes the channel frequency in a radio set) and asked me what it was. It seemed like a pretty ordinary common-or-garden high band crystal to me, and I said so. The way they worked was, the crystal would oscillate at its fundamental frequency, which was stamped on the can. The circuitry in the radio would then multiply that frequency by some number - typically twelve - to create the transmit and receive frequencies that the set operated on. For a receiver, the resulting frequency was 10.7 MHz higher or lower than the desired channel, to allow for the intermediate frequency or IF. Well this crystal seemed to be a 162.000 MHz receiver crystal to me; its frequency was 12.6083333 MHz (thus, x12 +10.7 = 162). A very common crystal. So what? "Now try multiplying by three and subtracting 10.7". I did... 27.125 MHz. A-ha! The penny dropped... this crystal could be used to receive the illegal CB channel 14, the "calling channel" where most CB contacts were initiated. "So if someone were to retune a receiver in a suitable way..." he went on. I needed no more encouragement than that.

A Dymar 830/930 Mobile Transceiver designed around 1975

A scrapped Dymar Lynx 930 AM mobile was recovered from the service department skip, and I set to work. Retuning the receiver was pretty easy. The multiplier was made to multiply by three instead of twelve, and suitable tuned circuits were made up to retune the front end to 27MHz. The rest of the set didn't need to change, since once converted to a 10.7 MHz IF, it was completely common. On connecting it to an antenna, I was amazed - there were loads of people out there using CB! I managed to dig up more common crystals that equated to some of the CB frequencies, and loaded up the set, which had only crystal slots for 6 channels. I got quite excited by the voices on the air - much more ordinary and accessible than the normal professional users of our equipment. I took the set home and rigged it up to a wire aerial strung across my bedroom. I picked up loads of chat on CB in my local area. I decided I had to get the transmitter going, too.

The 930 had been scrapped because its transmitter was completely non-working, so this was more of a challenge. It was a high-band set originally, meaning it normally operated in the 146-174 MHz band. To get a transmitter to put out 30W of power at those frequencies required relatively exotic transistors, but at 27MHz the ones used in the 930 had way too much gain and were completely unstable. In the end I opted to remove one entire stage altogether, and use a single transistor output that was a low-band device (66MHz, closer to the desired 27MHz band). However this was still very touchy, and getting the transmitter to work at all took some doing. The driver stages weren't much of a problem, just the final output. In the end I got it working with about 15W of power, but it was very marginal - the transmitter would sometimes "take off", where the final stages started to oscillate on their own and the drive stages lost control. This meant that the set wouldn't stay on channel, but wander up and down at will, creating god knows how much interference! It was pretty bad, especially hooked up to my completely random length of wire antenna. Still, it was enough to get "on channel" with a few of the local CB-ers. My home-built hacked "rig" soon became a local legend, since the transmitter instability meant that staying in contact was hit-and-miss at best! It often became a case of "one click for yes, two for no" (since the interference it created when keyed was usually detectable!). On the other hand, I had a fantastically good receiver, since a radiotelephone is built to a much higher standard than any US-built CB. Six channels were a problem though, so I soon ditched the unstable 930 for a crappy but compatible 40-channel rig just like everyone else was using.

CB Rig Number 2

When CB was legalized in the UK in 1981, the service wasn't compatible with the US AM sets that everyone was using illegally. Instead it used an odd channel offset and FM. Technically it was superior but of course what counted by then was being able to chat with your mates. So I resisted FM for a while, but gradually people did start to switch over, and as an enthusiast I definitely preferred using FM. So it was time for another rig hack. This time it had to have the full set of channels and work properly! Since a few years had passed since the 930 hack, I had gained a lot of knowledge about transmitter design, and also frequency synthesisers - a technique for generating many channel frequencies from a single crystal. Dymar's sets were going over to this technology at this time, so there was plenty of expertise around. However, Dymar's synthesisers usually had only 10 channels, selected using a single thumbwheel switch. That was no use. Besides, CBs had digital displays! Level meters! Microphone sockets! They looked like consumer electronics, not the weird-looking black boxes that went into taxis. On the other hand, the Dymar Lynx 2000 range had a few good things going for it - it had professional circuitry for one thing, so it was good quality. The set could be plugged into a cradle in a car installation, taken out and put into a battery-powered transportable pack, or into a desktop base station unit. That was a very attractive feature to a CBer - use one rig in all three situations. It also had a front-panel mounted speaker, so audio clarity was usually better than the average CB.

A Lynx 2000 mobile. That is one ugly front panel!

I set to work hacking up another scrapped set donated by my mates in service. This time I had to be more radical. The entire transmitter was redesigned so that it would work properly at 27MHz - though the use of FM made its design much easier. The receiver was straightforward. I couldn't use Dymar's synthesiser design so I designed a new one completely from scratch that had a digital display. Putting this on the front panel of the Lynx 2000 meant that space was at a premium, so for a channel selector I used two buttons - one for "up", one for "down". This was a bit compromised, as getting to another channel was a bit tedious. Pressing both together would jump to 14 though, which helped. I also wanted a level meter like other CBs, so an LED bargraph was included. Circuitry for the synthesiser was constructed on a veroboard and installed into the set, hooking into the modified Lynx circuit boards. This thing worked great! The receiver was top class, with professional levels of sensitivity and selectivity - most other CBs were never that good. The transmitter had about 10W of power - much more than the allowed 4W, not that I cared, and above all, was stable with excellent spurious emissions. The only problem was that the synthesiser VCO was very sensitive to mechanical vibration, which would show up directly on the transmitted audio. Engineers call this effect "microphony", and this set had it in spades. It was something that Dymar had wrestled with on their own sythesisers. Their solution was to mount critical components on special pins away from the circuit board, and to shield the VCO with a mechanically stiff shield. This only half-worked, synthesised Lynx FM sets always had this problem to a degree. My own veroboard-mounted VCO didn't take this approach. I took a leaf out of the Japanese manufacturer's book and simply dumped hot wax all over it. When it set, it cured the microphony totally! I even pointed out this solution to my boss, but they decided that it was too unprofessional for a Dymar-built set.

Anyway, this was a great rig. The modified front panel looked good - far better than a stock Lynx, and the swappability between car, backpack and base station a great feature. There's no doubt it would have made a great product! Dymar definitely weren't interested in CB though. I wish I had this set to this day - not that I have used CB since about '82, but I would like to have kept it because of the effort I put into building it. So what happened? Well, the front panel was a difficult thing to make, so I enlisted the help of a friend who was a complete whizz at metalwork. He made the panel up for me on the basis that he could have free use of the set when it was built. That was fine by me, and he did a great job on what was a truly fiddly part to make. Unfortunately he seemed to think that 'free use' meant that the set was his, and shortly after it was finished, he sold it! I was absolutely insensed! That was mine, it took hundreds of hours to design and build!!! It wasn't just any old CB set. Unfortunately it was one of those classic 'sold it to a bloke down the pub' situations, so it was gone without a trace. It still annoys me to this day.

Professional Radio Design - the Norgas 1995

1995 pre-production prototype shows general design features and very solid chassis

The CB sets above were very much unofficial projects - they got worked on "after school" so to speak. In my professional duties, I worked on other things of course. One of these was an interesting radio set called the 1995. I want to mention it here because as far as I can tell no-one's ever heard of it, and unlike many Dymar sets (even the obscure clandestine set ART177) there is absolutely nothing about it on the 'net. Many old radiotelephones are great for amateur use because one of the most popular amateur bands - 2 metres or 144MHz - is right bang in the middle of the private mobile bands. The 1995, should anyone ever unearth one, would make an ideal amateur rig.

The 1995 was developed specifically for Northern Gas ("Norgas") for use in their vehicle fleets. The specification called for a set that had 100 programmable channels, AM and FM automatically selected, 30W of power, selective calling, a 15-year service life, local or remote control vehicle mounting, and to be very tough and robust. Note that the spec went out for tender in 1980, so with a service life of fifteen years, it would retire in... 1995. Hence the name.

The design we developed which won the tender was basically a "super Lynx 2000", though also taking some of its construction techniques from the much earlier 830/930 Lynx series which was built a lot tougher than the 2000. The control head could be plugged directly into the main unit for local mounting, or via an extension cable to a boot-mounted main unit. This clever design was unique to this model. The main transmitter/receiver was on a single PCB, with the synthesiser and selcall separate boards. The chassis was diecast with some folded sheet parts. Because the set worked with both AM and FM, the receiver had dual demodulators feeding a common audio output, and with a common front-end and IF. The transmitter used low-level modulation with feedback - a technique more or less exclusive to Dymar in those days and pioneered on the Lynx 2000. This technique basically modulates the drive stages of the transmitter only, and relies on detecting the resulting output and differencing it with the incoming audio signal. The transmitter is actually driven by the pre-distorted difference signal - the resulting distortion in the transmitter "undoes" this distortion so the result at the output is clean AM. This works, provided the antenna is very accurately matched to the transmitter. One problem with the Lynx 2000 AM sets is that this isn't always the case, and the result is very distorted modulation. The 1995 sought to fix this using a stripline detector rather than a crude diode detector. It was better than the 2000, but still prone to the problem for poor VSWR.

1995 control head. Badly scratched unfortunately

The 1995 is a good quality solid set. Dymar shared some of the development costs with Norgas, and built the first 100 units. After that, there was some shenanigans. The full production run ended up being handed to another company, Maxon. Some Dymar engineers and one director jumped ship to Maxon, and Dymar were basically diddled, unable to recover their investment in the R&D. It was one of the nails in Dymar's coffin, though not the only one.

End of Dymar

Surely the writing was on the wall when the government published its first white paper about cellular networks, 1981. It seemed to me that Dymar's powers-that-were must have been out of touch, since obviously people didn't really want to have to go through an operator when making calls, share channels with umpteen other users, and say "over" at the end of every sentence - they wanted to dial a telephone number and get through! That seemed obvious to me at any rate, but Dymar didn't seem bothered by the coming of the mobile. There may have been long and earnest discussions around the board room table for all I knew, but it never got translated into action in R&D.

Meanwhile, Dymar seemed hell-bent on divesting itself of its instruments division. Dymar Instruments built high-quality test gear for radio work - naturally we all "ate our own dog food" within the company too. This was coming under pressure from microprocessor-based instruments from the likes of Hewlett Packard, Fluke, and numerous others. It's fair to say there was little expertise in processors in the company at that time, but there were young 'uns around like me very keen to get into that. Anyway, Dymar Instruments got sold to Farnell. As far as I can recall, Farnell kept selling just one item in the range - the RF Power Meter.

So, No to cellular, No to instruments, definitely No to CB, the 1995 going to Maxon... what were we supposed to be doing? There was work on a design for the Home Office called the M3001, but while I did spend an awful lot of time developing it, I don't think that went anywhere. I guess sales of the Lynx 2000 range were doing OK, since we seemed to be making plenty of them, but surely the bottom would fall out of that once cellphones were introduced? I did catch a glimpse of a very hush-hush project that was for tracking security vans using radio (somehow - the technology didn't really seem to be there then and the prototype I saw was a physical mockup only. I never knew anyone developing actual working innards for the thing. I do know the mockup had a Z80 CPU in it though!).

Dymar Lynx 850 Mobile with Selcall option

One day we all came in to work to have it announced that Dymar would be merging with Burndept, its erstwhile competitor. Talk about a sell-out. Burndept were based in Erith in Kent, and after a visit to the place, I certainly wasn't interested in relocating there, which was the next thing that came up. That's when I left. Others engineers clung on, but in the end very few were relocated and most were made redundant. The company continued to trade as Burndept-Dymar, and the Lynx 2000 continued in production for another year. In the end, Burndept was swallowed up by another conglomerate and an era came to an end. Still, at least we could get "proper" mobiles by then! Interestingly, the technical director left to head up Motorola's cellular division in the UK. Seems that, unlike some, perhaps he saw the way the wind was blowing.

The curious story of Mr. Assali

In 1981, I got involved in another radio project outside of Dymar altogether, which is worth mentioning, if only for my lucky escape. Remember I said that what people really wanted from their mobile radios was to be able to dial a number, and get straight through? Well a friend of mine thought so too. He worked for an odd little company in Borehamwood, Herts. His boss had told him he was looking for just such a system, and so my friend set up a meeting between us. It turned out that his boss was a guy called Hassan Assali, and his business built burglar alarms. He was obviously quite well-off, with a big car and expensive suits and so on. He wanted to be able to put calls through from his car to any number he dialled. He already had a pair of radiotelephones, so what he was after was a way to dial numbers over the air. We thrashed out a design in his office in July of 1981 - I remember it very clearly because it was the day of Charles and Diana's wedding, which was showing on a large TV there. Anyway, I designed a system to do it, based on a DTMF keypad and a whole bunch of logic at both ends (in those days DTMF dialling wasn't available on the UK telephone network, so some of this had to do with DTMF to pulse conversion). It wasn't all that tricky, but hardly as usable as a modern mobile, especially as it was tied to his own specific base station.

Development went OK, but I needed to buy stuff out of my own pocket. In the meantime, my friend fell out with his boss because he kept not getting paid! Eventually he left, at which point he had to warn me that doing business with Mr. Assali might not be such a good idea after all. He especially told me to get cash up front because likely as not I'd never get anything for my work. So anyway I went to see Mr. Assali to ask for some cash to buy parts to continue with the work. He agreed and even took me out to dinner, and tried to persuade me to go and work for him. I was happy to continue with the dialler but I didn't want to work for him. I also noticed he had a whole lot of very expensive instrumentation in his office, just unpacked. Things like £20,000 spectrum analyzers from HP. It was the sort of kit I didn't even get to use at Dymar, so I was drooling a bit. He got very cagey and wouldn't tell me why they were there, not that it was my business. Anyway, work progressed, slowly.

Next time I saw my friend he told me that the instruments were part of an embargo-busting scheme, where Assali would buy the equipment from the US, ship it to another agent in Vienna, who would then sell it on to the Russians. It got around the US's technology embargo against the Iron Curtain in force at that time - highly illegal. I was getting very nervous about working for this character, and other stories about his fiery temper (he apparently shot a dog because it was barking and keeping him awake) didn't help. I'd run out of cash again, so I rang and explained that. By then I was hoping I could get myself out of the whole thing really, which actually was easy, since no contracts had been signed or anything, it was all just on a handshake. Well, that conversation didn't go so well. He basically wanted to see some results before he gave me any more cash, and I needed more cash to produce working results. So it was stalemate. That was the case for a while, then he rang me one weekend and said that it was all off and he was sending someone round to collect all of his belongings. Meaning the parts and a few bits of minor equipment I'd borrowed. I was happy about that, it let me off the hook.

What I didn't quite expect was a long black limo to pull up like in a bad gangster movie. Two big burly chaps got out and strolled up to the door. I think they expected resistance from me or something, or at least an argument. At least that must have been how Assali had briefed them - it was all very odd. I just handed over the kit in a cardboard box and they left! And that was that - I didn't want any more to do with him, and since my friend had left in troubled circumstances there was no obligation there either. Phew!

But that wasn't the half of it. It turned out that about 18 months after that, Assali's company was raided and he was arrested on terrorism charges. He was convicted for supplying bomb timing devices to the Libyans (and by that route to the IRA) and sentenced to 9 years. A Google search for Hassan Assali will unearth material on this, for example here. The conviction was quashed in 2005, because the expert witness's evidence has been called into question. Presumably the paranoia of the 80s has given way to something more considered now, so maybe he wasn't a terrorist. He certainly was very, very dodgy however. As I said, a lucky escape. Oh and if MI5 are reading this, that should explain why my name and address were probably in his rolodex!

Other Shenanigans

A 1978 vintage 850 mobile. This one lacks Selcall, but look how well-made it was!

In-car telephones were useful, even in the crippled form that took in the 1970s. So useful that I quite fancied one myself. The way it worked then was that there were a very limited number of base stations, which divided the country up into a number of large areas, maybe 7 or 8 for the whole network. These massive "cells" needed powerful base stations and powerful in-car transmitters, which is why power of 30W for the mobile and many more for the base stations were used. Each area was operated (entirely manually) by a different company, each with its own list of subscribers. If you travelled outside of your home area, you could still make use of the service because of a sharing agreement between these companies - at the end of each month all the "foreign" calls were billed to their home bases. The base station knew which area you were from because the mobile transmitted a hard-wired "area code" along with your selcall ID. And that gave me a handy loophole! The actual caller ID was made up from a 3-digit number that was hard wired using a little plug-in card on the front of the unit (the same card wired the area code), and of course such cards were easy to come by if you worked for the company! So I had a small handful of randomly chosen numbers, all with area codes away from the local area. A rescued and resurrected 850 mobile from the (where else!) service scrap bin served as my in-car unit. I could make calls through the operator of the local service as if I were a visiting "foreigner", and they wouldn't find out for a month. At which point the number was usually blacklisted and they wouldn't return the call. So then, it was a simple case of swapping the card for another and getting another free month! Looking back I was sailing very close to the wind on this one. If I'd ever been found out as a Dymar employee I probably wouldn't have been one the next day. I was also a bit paranoid about calling people up in case they'd record their numbers and either find me out that way or hassle my friends. So in practice it was a brief passing fad that I only bothered with for a few months. I probably used it in earnest twice, when my car broke down and I needed help. Mind you it's because of that that I knew the service was going to get vapourized out of existence when mobiles arrived, and so it proved.

Interestingly, modern mobile networks work in principle exactly the same way. Then you'd ping the base station with your caller ID. An operator would respond and you'd ask for the landline number you wanted (it was also possible to route to another mobile user). When the number was connected, the operator would tell you which channel to go to (1 of 6), and connect the call. Now of course there is no operator, but essentially the same thing happens automatically.