Dr Rod Crawford of Legend AcousticsAdapted from an interview with Chris Green in “Meet the designer” Australian Hi-Fi 1997
Dr Rod Crawford is the founder and designer of Australian loudspeaker company – Legend Acoustics. But he didn’t pursue the traditional speaker designer’s career path of refining successive speaker generations in the back shed and struggling for recognition over trendy Imports. He started his hi-fi career at the top: as Project Leader for the loudspeaker design team at the famous Linn Products establishment in Scotland.
Leading up to this golden opportunity is what Rod describes as “an interesting life”. Rod was born in the Sorell Bush Hospital Tasmania and was educated in state schools in Tasmania. He won the Electrolytic Zinc Company Scholarship that started an illustrious academic career with a first-class honours degree in Applied Science at Melbourne University, majoring in physics and metallurgy. Winning the University’s Travelling Scholarship took him to Oxford University where he picked up his Doctor of Philosophy from the Department of Materials Science for his world-dass experimental and theoretical research.
All the while, Rod (now Dr Rod) was nurturing his deep love of fine music and his fascination with hi-fi. He was raised in a musical family and attained levels of proficiency on piano and saxophone that are hard to gauge from his humorous and self-deprecating manner. “Someone once said that hi-fi was a refuge for failed musicians [What utter nonsense! … Editor] but I don’t really regard myself as a musician, failed or otherwise. I don’t play any more but I still listen a lot, my taste tending more and more toward smaller group works’.
Back in Australia, Dr Crawford completed a Diploma of Education at the University of Tasmania, majoring in Maths/Science teaching, which led to a four year tenure as the physics, mathematics and electronics instructor with a community college in Hobart and subsequently a position as Head of Physics and Technology, leading a team of instructors for 8 years.
Returning to the UK in 1985, he turned his eclectic skills to writing and delivering self-leaming packages for a Yorkshire electronics company. He had earlier met and married Yorkshire lass Liz.
Joining Linn Products
Around this time Rod picked up a copy of Hi-Fi News & Record Review, which he didn’t customarily read, and saw a job advertisement for a speaker designer. It had been placed by Linn Products. “I sent off an application and didn’t hear anything for about four months. Then came a letter asking me up to Scotland for an interview. Although I was perfectly frank about never having designed a loudspeaker before, I got the job.” Rod’s first day at Linn was somewhat inauspicious. “They’d had the worst snowstorm in years and, after pushing someone else’s stranded Jag up a hill with my Subaru 4WD, I was two-and-a-half hours late. But I was still one of the first to arrive because my 4-wheel drive Subaru could plough through the snowdrifts!”
Ivor Tiefenbrun was away in Europe at the time and at the end of his first week, Rod met him in the company cafe. Rod recalls: ‘A shortish, roundish fellow approached me and asked who I was working for. ‘You’ I told him. He said: “I thought we’d decided not to employ any more f***ing Australians.” In retrospect I think the problem with Australians is that when the boss says “jump” they tend to say “why” rather than “how high”! Rod and Ivor initially, however, hit it off and, having productionised an upgrade to the Sara called the Sara 9, Rod was plodding along quietly designing his first new Linn speaker, the Nexus LS250, when there was a panic. Rod explains, “The sales guys got cold feet about the external power supply for the Linn turntable. They’d actually gone into pre-production of the power supply before they decided to stop production. The result was huge pressure on me to bring forward my speaker by three months. Ivor used to hover around our elbows saying things like, ‘Crawford, if you don’t have that speaker ready by the time I come back from the CES… you’re fired.’ And I used to say, “Ivor, I’ll just do the best I can. If that’s not good enough, then fine … you’ve seen the end of me.”
With the panic bells ringing, Rod and his colleagues worked most nights till midnight for about two months. ‘We made a good team, with an extremely good industrial designer and two excellent model-makers. I could scribble a design on the back of an envelope and two days later they’d present me with a shiny new box. This really helped accelerate the process. We didn’t have a formal project team at that stage but we had really good purchasing people and excellent technical staff. Our design brief direct from Ivor was to build a speaker that would beat a Mission ported loudspeaker that was selling very well at the time – to compensate for the inroads CDs were starting to have on turntable sales at Linn. Although I was subsequently accused of undermining Linn’s sealed-box philosophy it was actually Ivor’s instructions to make a ported speaker – and in a blind-listening test where we plugged and uplugged the port the sales guys could not tell the difference”!
Like any loudspeaker designer, Rod found his hardest task was choosing the drive units. “When I joined Linn had sourced most of their drive units from KEF, but I evaluated drivers from about eight different companies, and the bass unit I most preferred was from Audax. It turned out to be too expensive, so the choice was a very close run between Tonegen (Japan) and Vifa (Denmark). In the end, we went with Tonegen; they were responsive, quick, and the quality of their product was really top-class. They were also very happy to produce one-off modifications exclusively for us. I think they would have been willing to make a loss on the drivers simply to land this project with Linn.”
One of the major driver qualities Rod was looking for was speed of response. ‘My previous preference had been for the BBC sound which emphasised good tonal balance whereas the Linn sound emphasises being very fast/foot-tapping ability, so we needed a light, stiff sort of driver. The down-side of this, however, is that light, stiff cones tend to ‘break-up’ chaotically at the top end if you’re not very careful. Anything rigid, once it starts to resonate, gives you a very strong resonance so we had to use fairly high-order (fourth-order) crossovers. And as with all speaker designs, it turned out to be a ‘swings-and-roundabouts’ affair. We ended up with, a very tight, fast bass & midrange but a slightly unnatural join with the top end.’
Because the currently-used Hiqhphon tweeter was too expensive for his budget Rod was on a budget for the tweeter as well so, again, the tweeter he selected was given a high-order crossover to get rid of what he calls ‘the bottom-end nasties.’ “The Nexus was a quite a good speaker given that it was designed in about four months and on a tight budget. The process involved lots of listening. We had a basic listening panel of five or six people which included Ivor, the head of marketing Charlie Brennan, and two sales people. I was quite happy to take criticism but the sales guys tended to like a particular sound that they had gotten used to selling. Anything that differed from that, they tended to reject.’
Rod experienced difficulties accepting Linn’s philosophy that he should use only Linn components in his listening system, which he found somewhat limiting. ‘They had a policy of changing only one component at a time in the listening tests. The trouble with this, which I only discovered later, is that hi-fi and speaker design is a multi-dimensional plane on which there is a series of peaks. Although you hope you’re going higher and higher, it’s very easy to get stuck on a local peak. What Linn had been doing, over the previous ten years – at least in my opinion – was to optimise the chain of Linn Sondek, amplifiers and speakers for one another and so get stuck on a local peak. “I had major battles trying to get additional reference equipment. Linn had some Naim amps left from their days of collaboration, and in the end, I convinced them to buy a pair of Epos lls, which had come out at around that time. My opinion was that it was a big and varied world out there, and consumers were running a huge diversity of equipment. Fortunately Linn had dealers in Glasgow. I found I could take my prototypes to their stores and compare them against other brands as well as use other front-end equipment”.
Rod also had a problem with Linn’s theory of the hierarchy of audio components – that the source is the most important link in the hi-fi chain. “My theory is that you should spend about 40 per cent of your budget on your loudspeakers, around 20 per cent on your amplifier and the rest on whatever sources you want. It’s true that if you degrade the signal at the front end you can never recover it, but that’s also true anywhere else in the hi-fi chain. So you may have the ultimate signal coming from your turntable but if you put it through an inferior amplifier or a pair of substandard speakers, your ears will hear that. You really have to balance the whole system – a chain is only as strong as its weakest link. But it’s true that you can see the worst botches in hi-fi occur at the electromechanical ends of the chain. I see the difficulties as being the mechanical problems at the turntable end and at the speaker end. They’re the hardest parts to get right because there are no neat feedback loops to solve your problems, and the real-world effects can really get away from you. It’s relatively easy to design electronics, and in terms of components, everything except the power supply and the metalwork are relatively cheap to buy. And its harder to control mechanical problems such as resonances in speakers than in turntables because the signals are so much larger”.
The Nexus was released about eight months after Rod started at Linn. But, according to Rod, he lost over two months when Linn decided to change a supplier. “The purchasing department decided to get a new cheaper supplier for a dress ring that went around the drive units. It didn’t work out and by the time we’d switched back to the original supplier, we’d missed our intended autumn launch. The Nexus was supposed to come out in September. It was a hard-learned lesson, but hold-ups like this have occurred in just about all the production runs I have been involved in since”.
The Linn Nexus was a two-way bookshelf speaker with a single, 203 mm bass driver. It received mixed reviews when it came out and I don’t think the sales guys really pushed it. This didn’t please me a lot, particularly since they had agreed to the design during the listening tests. They had compared it against the competition and the imports, also judging it beat their existing Kan and Index hands-down. The Nexus went through three or four modifications in the first year of production, and the modified version received a good response from the overseas dealers, the Americans and Germans in particular. It sold around 4,000 to 5,000 pairs of Nexus a year for around four years, even though it didn’t win any awards and wasn’t reviewed again, even after the modifications which was unfortunate. In the end, it sold more than twice as many as Linn’s previous best-selling speaker. Ivor was happy … so happy, in fact, that we had a free weekend at the Edinburgh Festival.”
Rod remains impressed and intrigued by the public persona of Linn’s founder and managing director. “I was led to understand that he grew up in Pollock near Clasgow’s notorious Gorbals area and was supposedly kicked out of school when he was 14. He’s difficult to work for – in the time I was there we went through three personnel managers. Some others who joined the company left within two or three months when they discovered they weren’t allowed to make any decisions. I think there are limits to how long you can keep total personal control, and that’s why Linn stopped growing when the company reached around 150 people: it was too hard for lvor to hold all the reins. But I admired him. He was a great man in many ways.’
Rod’s next task was to design the Linn Helix, and by that time he’d been promoted to Senior Acoustic Engineer. “But I wasn’t told. I only found out when I saw a photo caption in Linn’s magazine. So I immediately went in for a raise. I got it.”
His design brief for the Helix LS150 was to design a cheaper Nexus LS250. ‘The numbers after the “LS” were the target sales prices – for the Helix LS150 this was £150. Like most of the LS range it still went a bit over budget. I kept the raw materials costs down this time, but thought the on-costs were a little high. When it eventually came out, it cost slightly more than the Mission which Ivor regarded as the main competition. It took me literally three weeks to design the Helix from ‘go to whoa’, because everything came together perfectly it just worked. At that stage we had a guy who was developing a new computer program for crossover optimisation. He’d just got the program running, so I plugged the numbers in, it threw out an answer and the result sounded brilliant! I didn’t have to go around in the usual design loop. For a few moments I imagined I could design a speaker a week, but unfortunately it doesn’t always come out that easily!” The Helix was well-accepted by reviewers, and gained a glowing review from Alvin Gold (What Hi-Fi) which Crawford was very pleased about. ‘Gold was not generally Linn inclined as a rule, but he reckoned it beat the B&Ws and the KEFs at that time.The Helix also received good reviews in Hi~Fi Choice, of which Rod is proud. “They’re all blind listening tests, judged with no reference to brand or reputation, and the judges on the panel were my peers from companies such as B&W and KEF. The design was praised and given a ‘Recommended’ rating.
My next design for Linn was the Index 2. This was actually a totally new loudspeaker design, but was called the index 2 because for “marketing” reasons, Linn wanted to retain the name. It received Linn’s first ‘Best Buy‘ for loudspeakers,’ said Crawford.
Isobarik, Kaber, Keltik, Tukan….
Then came an upgrade of the Isobarik to try to extend its life while we worked on the new Isobarik LS2000 (subsequently called the Keltik). ‘Later we also upgraded the Kan which was well liked. Then we brought out the Kaber LS500, which was a three-way floor-standing speaker. The Kaber LS500 came out in 1988 and its dual-woofer approach was quite innovative at the time. Now everybody’s doing it. It’s a good way around the need for large, wide cabinets, and the floor-standing idea dispensed with the need for expensive stands. Phillip Hobbs [a recording engineer who won a Gramophone award for his recording of the Tallis Scholars] and I came up with that design by ourselves and didn’t tell anybody what we were doing until we’d got a fair way through the process because we just wanted to see if it would work. This is when I found out about Linn’s internal politics, and that I wasn’t supposed to do anything without any instructions from above. We got into terrible trouble – and they set up a bureaucratic product-management structure to control us!’
According to Rod, ‘his’ Linn Kaber is still selling well [in 1997 when this interview was made]. ‘In relation to price points, my design work at Linn started in the middle then went down and then up. Which is what I intend to do here in Australia with my own Legend loudspeaker range. The top-of-range replacement for its Isobarik loudspeaker, the Keltik LS2000, took a long time to develop and was supposed to sell for around £2-3000 in its passive form (but I think ended up being marketed for much more, around £10,000 in active only form). Another loudspeaker, code-named the LS1000, was supposed to go between the Kaber LS500 and the new Isobarik LS2000 (Keltik). The LS1000 had gotten through all the design hoops, including listening tests etc, until it was stopped by the sales guys in pre-production, the final step before full production. I didn’t find out until after it was cancelled and I was absolutely ropeable. For me, that was the beginning of the end, though I actually lasted for another 12 months, basically bringing out the Keltik, which he finished late on Christmas Eve.’
Rod takes up the story. ‘It was the day I left Linn – the night of the Christmas party. I had helped develop a new ceramic dome tweeter. Tonegen provided us with the original units, which used an aluminium substrate with ceramic sputtered onto it. These sounded quite good, so we modified them to sound even cleaner and asked Tonegen to do the same modifications and provide samples. These arrived late, only two days before Christmas, and I spent those two days frantically doing last-minute adjustments. I had already put my resignation in, but I had to follow it through to the end. It was a very good speaker actually, and good value at its proposed £2-3000. The elliptical bass driver was very much based on KEFs B139 driver that was used in the Isobarik DMS model. It was again made by Tonegen, but had a lighter cone, and sounded much quicker than the previous B139. I used the same Tonegen mid-range driver that had been well received in the Kaber – and because I was under instructions from above to minimise the number of drivers used overall in Linn’s loudspeakers to minimise stock levels in the time of of economic difficulty for Linn. In fact the Keltik was mainly designed to be a Kaber but with better bass. It was never designed as a statement loudspeaker – that was supposed to occur after I had been given resources to go back to to square one and do some more basic thinking on loudspeaker design – remember I had a PhD from Oxford Uni. However, it did have an innovative front baffle with random dimples that reduced diffraction effects from the baffle – and some unusual internal construction to minimise standing waves inside the cabinet. The Keltik easily beat the then top-of-the-range Isobarik DMS in the obligatory Linn listening tests’ and subsequently got a rave review in Positive Feedback magazine as part of the Linn reference system at the time.
In addition I left behind the design of an upgrade to the Kan loudspeaker which I called the Tukan, also as a play on words from a bird of the same name we had seen up near John O’Groates. It was just as well I did not call it the Kan 2 – or Kantu! The Tukan used the same mid and treble drivers as the Keltik i.e. a Keltik without the isobarik bass! It was put into production after I left and given a very enthusiastic review by Ken Kessler – an admitted Linnophobe!
Rod was with Linn for four years. “It’s an exciting company to work for, the R&D team were great to work with and we loved living in Glasgow. We could have lived there forever if things had worked out differently. In the end, I found the environment too stressful and control-freak to be creative.
The final straw came when Linn broke an agreement made at my beginning there that if I designed a new range of speakers to replace their aging line-up of Isobarik, Sara, Kan & Index loudspeakers using current technology, as I did with the Nexus through to Keltik, then they would provide the resources to develop some new technology using my research skills.
The near trebling of Linn loudspeakers sales while I was there in no small way helped the company survive the slowdown of turntable sales with CD starting to dominate the playback market but while I fulfilled my part of the bargain they reneged on theirs. I would have liked to have stayed to use my materials background to develop new drivers – and to continue collaboration with Malcolm Hawkesford at Essex Uni using DSP in loudspeakers.”
Interestingly DSP in loudspeakers was taken up by Linn over 25 years later in their ‘Exakt’ loudspeaker systems – and nearly 10 years after Legend Acoustics started using digital crossovers and speaker correction DSP from the Australian company DEQX.
After Rod left Linn, he sold up and came back to Australia, very sad to leave Glasgow, but happy to be back his beloved Australia. The first thing he did on arrival was to embark on his MBA at the Australian Graduate School of Management in the University of New South Wales, specializing in strategy, marketing and human resource management. “It was partly because I felt there must be better ways of running a company than I’d found at Linn. In fact I learned that that Linn was fairly typical of a company stuck in the ‘start-up’ mode of its dominant founder and that change management in an organisation is always very difficult. After my 2-year MBA I got a job with the Australian Industrial Property Organisation (Patents Office) in Canberra. I worked initially in a subsection which examines applied chemistry and metallurgy patents then later in a subsection concerned more with physics/engineering inventions. ” As Dr Crawford points out, Albert Einstein did some of his best work, including Special and General Relativity theory, in the Swiss Patent office!
The ambition to make top-quality speakers remained with Rod after his split with Linn – to make listening through loudspeakers be like hearing a ‘live’ event. “My actual designs for Legend speakers started about 18 months ago, when I founded the company. What finally crystallised my thinking was when Electronics Australia (in conjunction with ME Technologies) brought out the IMP analyser. This gave me access to low-cost computer measurement instruments very similar to the standalone Hewlett-Packard devices I had become used to using at Linn. Even though I spend at least 90 per cent of my time listening, as we did at Linn, I still need that 10 per cent of the time measuring to make sure I am not off track. The sales guys at Linn may tell you I spent my whole time measuring, but that’s just what they wanted to think. I think loudspeakers must be designed for the ultimate observer (the listener), and so it’s the complex interactions of psychoacoustic perception that must be the benchmark, not just the easily-measured laboratory findings.”
The fact that Legend’s first model, the Kantu, is a homonym for Kan II is a complete coincidence, according to Dr Crawford. He says Kantu is one of the aboriginal words for kangaroo. “I’m thinking of calling the whole range the Kangaroo Series, and using different aboriginal dialect words for kangaroo for all the other model names.”
-The Legend Kantu was launched about five months ago, after a year of travelling around sourcing components, during which time he visited Audax in France where he was impressed with a carbon-fibre and gel composite cone woofer (Aerogel-on which Audax has taken out patents). “When I started research at Oxford back in 1966 I learned about carbon fibres but it has taken about 20 years to discover how to use them properly. Stiff cones suffer badly from break-up modes, but the gel adds a gooey sort of damping to the stiffness. It’s like a liquid glass colloid that is always flowing”.
-Rod found his of choice tweeter more difficult. “I evaluated different models from Vifa, SEAS and ScanSpeak. All measured well, which means you then have to rely on your ear to sort one from the others. After well-chosen good drivers the main problem with modern speaker design is then joining the drivers, which makes the crossover crucial. The drivers are joined with a combination of electronic and physical means. It’s an important part of my design approach to keep the crossover simple and design everything else around that, whereas when I was at Linn, I was forced to use complex fourth-order crossovers because of the low-cost drivers needed to keep within project budgets and high overheads.’
“I wanted to get away from those fourth-order crossovers because I think they can degrade the sound, but to do this, you need to use expensive drivers. I found the Aerogel midrange to be very smooth and very flat. And, when it does finally break up, it does so in a very controlled way, so you don’t need high-order crossovers to control it.’
Australian HI-FI asked Dr Crawford if he felt pressured by the fact that Linn would in all probability be keeping an eye on what he was designing for Legend, with a view to looking for patent infringements. According to Rod, it’s not a question of patents, but the potential for other intellectual property disputes such as ‘trade secrets’. He claims to be one of the few engineers to have left Linn who has been “slapped with a confidentiality clause.”But I do have some new and interesting ideas, particularly on the materials side,” he says. “It’s like an artist painting a picture. You have to start from what you know, and you know it’s no use being radical just for the sake of it: producing new ideas but not forgetting about the basics. In my time I have seen many awful loudspeakers that have focussed on just one ‘new’ idea. There’s no magical solutions “.
The drivers in the Kantu are off-the-shelf units modified by Legend, and much of the other electronic componentry inside was created especially for them. Rod describes the Kantu driver array as “overall time-aligned” because he says he’s suspicious of speakers that have been purely distance time-aligned as the crossover also causes time delay. “The ultimate test of time alignment is whether the drivers actually add together on-axis. If they do, the phase difference is zero and therefore they’re time-aligned.”
The Kantus have an efficiency of about 90 dB/W @ 1m and minimum mpedance of 4 ohms. According to well-known Sydney retailer Gary Sellers, whose store HI-FI Junction has been championing the Kantus since their introduction, they sound best with at least 60 watts of power behind them. “It’s not a question of efficiency,” Sellers told Australian HI~FI Magazine, “it’s just to do justice to their potential dynamic range.”
The Kantus is the middle of the Legend range and at the time of our interview, were the only model that existed in stores. The new Kama should be available by the time you read this. It’s a two-way version of the Kantu: somewhat shorter but tall enough not to need stands and can also be used as the rear speakers in home theatre. And the Kurre is a new bookshelf speaker.
The attractively different Legend logo was designed by one of Rod’s colleagues at the Patents Office, Doug Thwaites. “Doug’s design brief had three basic elements: I wanted it to be Australian in flavour, I wanted it to have some musical connotations, and I wanted it to be visually ‘softer’ than the usual high-tech badges. The name Legend has connotations of the Dreamtime and, because it’s also a buzz-word amongst the kids, it has some of the spirits of youth about it too.”
“I am building the production models myself but sourcing the cabinets elsewhere,” says Dr Crawford. “There are a couple of very good cabinet-makers around Canberra, making furniture for Government contracts. I work with them as a team, and when they have good ideas about how to make things better or more, cost-effectively, I am more than happy to accommodate them.” Rod recommends that those who buy his Legend Kantus use them with their grilles on, primarily because the surfaces of the bass drivers are deliberately made slightly sticky. “I have adjusted the balance of the tweeter so it’s flat with the grille on.”
Rod has positioned his production models with some of Australia’s most prestigious hi-fi retailers, including Audio Adventures & Hi-Fi Junction ( (Sydney), Contemporary Sound Centre (Melbourne), and Duratone & Acoustic Images (Canberra).
Rod is pleased that HiFi Junction has been so successful with his Kantus. “Gary’s guys have been very enthusiastic about the Kantu, rating it above several pricey European models … but then I’m used to competing head-on with companies like KEF and B&W on their own territory. My overall plan for Legend is to develop a good local market first and then expand into Asia, and possibly South Africa. Then I’d love to move into the American and British markets. If the Canadians can do it, there’s no reason why Australia can’t. I think Australia is very remiss in not providing the kind of support that Canada does for its speaker companies. It’s our small mind-set. This is one of the reasons behind my ‘Proudly Australian’ logo. Over the years I could have saved myself a lot of hassles by getting British citizenship while in Oxford. I’m not jingoistic, but I am proud to be an Australian, brought up in the bush.”