Paul White and Dave Lockwood talk to Marshall's Technical Director, Steve Grindrod, who has overseen the company's new designs and developments for the last 20 years.

Virtually every rock and pop musician on the planet knows about Marshall amplification, but what most probably don't know is that Jim Marshall, the company's founder, was a drummer, and his first marketing enterprise was a drum shop. In the 34 years Marshall has been manufacturing amplification, the company has grown from a cottage industry to an award winning market leader exporting to 65 countries worldwide. The current Milton Keynes factory covers a staggering 150,000 square feet, employing a workforce of over 400 people — it even has its own 500 seat theatre for product testing and demonstrations — and every week, around 3,000 Marshall products roll off the production line. Indeed, the name Marshall is a part of rock history, but what is the secret of the Marshall sound, and who actually designs the amplifiers?

For more than 20 years, Marshall's designs have largely been the responsibility of electronic engineer Steve Grindrod, and although Marshall keeps Steve extremely busy with its ever expanding product range, he still finds the time to do the occasional gig. Having lured Steve out of the factory on the pretext of a curry and a quiet chat, we asked him how his interest in amplification started.

"In 1961, at the age of 13, I bought my first electric guitar, then asked my dad if he'd buy me an amp, to which he promptly answered 'No'. He was an engineer, so I asked him if he'd build me one, but again the answer was no. Instead, he said that if I wanted an amp, I was welcome to use his technical library to teach myself and build my own. He probably assumed that this would be enough to make me sell the guitar and continue learning classical piano. However, I destroyed the old steam radio and built myself a valve amplifier from the parts. It actually worked, and I didn't kill myself — both remarkable facts, all things considered. Being the early '60s, pop music was very fresh, and I'd had several years of junior classical music forced down me, so I got into rock'n'roll, taught myself to play the guitar, and joined a school band."

At some point, you must have decided to take electronics seriously as a career rather than simply as a way of producing guitar amplifiers for your own use?

"Yes, living in London and going to school near central London, I used to go up Charing Cross Road and Denmark Street, and devour all the information about gear that I could. I got totally involved in both playing music and in the technology behind it, as basic as it was then. There was nothing else I really wanted to do, and to this day I still enjoy performing.

"Music never seemed to be a very secure way of making a living, however, so I went to college to study electronics. At the same time, I continued playing in bands and doing session work. I did try setting up to build my own equipment, not very successfully, although I did a lot of servicing and customising, which taught me a lot about how different amplifiers were designed and how they performed. I learnt a lot of my trade through the late '60s and early 70s, and I did a lot of work on AC30s. Ironically, I used to do a mod on AC30s going back to 1969, which probably made it the first high gain, master amp around. In fact, some of those ideas have gone into the Vox AC15, which we've designed for Korg who now own the Vox name.

"I moved up to Milton Keynes in 1973, because I couldn't afford to be a musician in London any longer, especially as I was then married with two children. In Milton Keynes, I worked for EMI doing Simms Watts amplification, then the recession hit and EMI closed us down. That meant I was stuck for a job, but on checking the local paper, I noticed a job going as a test engineer with Marshall. They had also suffered from the recession, which gave me some cause for concern, but I thought it would do until I found a decent job.

"At that time, there was no R&D department — the previous engineers had left to set up their own businesses — and the product range was relatively small. My previous experience put me in an ideal position to do R&D as well as testing the products, but this started off very informally, as the odd modification, then I went on to updating the printed circuit boards to make them more production friendly. At that time, we'd just started to do the original solid-state amps, which were fraught with problems, and again I had to work on the design to make it more reliable and easier to build."

MARSHALL LAW

Is there a specific Marshall design philosophy?

"I suppose it's to keep things as simple as possible, both to manufacture and to operate. The people who like vintage amplifiers like the tone, and the reason they had a good tone is that they were simple. Furthermore, if you keep a design simple, the less fragile it is and the less there is to go wrong. You don't confuse the customer, and the customer doesn't end up paying for unnecessary frills — light up control panels or whatever. It may look nice, but it's really irrelevant to the product and may cost the customer a lot more money.

"When we build a new product, we don't think of the price first, then build the product to the price — we do the best job we can, add up the cost of the components and the time it's taken to make it, then we arrive at a selling price. We don't add any money because it says Marshall on the front — it's the cheapest price we can reasonably sell it at and still meet the specification — and there's as much pride as we can put in it."

What is the secret of the Marshall sound? If you look at the circuit diagrams of the various valve amps, they're all remarkably similar, and most have their beginnings in old Fender designs, such as the Bassman.

"Where does the character come from? I suppose the simple answer is me — I design amplifiers that are satisfying for me to play through. They have to sound right and feel right from a guitarist's point of view. I sit down with a box full of components and a guitar, but if you were to ask me to describe the Marshall sound, I couldn't. The Marshall sound is different to everybody who hears it. There are many amps that make a great sound, but they don't 'feel' right, then there are those that feel wonderful, but they sound dire. I try to achieve the best sound I can get, and that makes you want to sit down and play it."

SIMPLE SOUND

What technical parameters affect the feel? Is this something to do with low level responsiveness?

"The compression of the power supply and the pre and post-voicing play a large part. Also, the way the power amp reacts with the preamp, the way the speakers react with the power amp — there are a lot of variables. With a valve amplifier, as the power supply sags, the gain of the valves changes, and that produces the compression effect. If you get that wrong, it can be just as bad as if there's no compression, and then it feels as stiff as anything.

"In many ways, the guitar amp is an odd thing to try to design, because the original rock guitar sound was created by shortcomings in the technologies available at the time — sagging power supplies, distortion caused by players turning the amp up too far, limited response speakers and so on.

"When you think about it, very much of the early amplifier technology was down to Leo Fender, and he made amplifiers that sound great overdriven, but the last thing he was trying to do was create overdrive. He didn't like it when guitar players in the '50s started turning his amps up to get distortion. Again, his philosophy seems to have been simplicity. He wanted to make simple things work, which I think he did very admirably — by keeping technology away from amplifier design, by not trying to cure the inadequacies; that's what makes the sound we still enjoy to this day."

The passive tone control also seems to be an important part of the valve amp design.

"Yes, the inter-relationship of three capacitors, a resistor, and three pots is absolutely wonderful, and as soon as you start going away from that, it doesn't work the same. A lot of people have criticised me over the years for not giving Marshall amps enough tonal range, but when I ask them if they'd use the extra range if I gave it to them, they usually say no. Keep it simple, make it sound good.

"Over the years, I have experimented with guitar amp EQs and come up with very radical ones to perform different jobs, but generally, they still rely on simplicity. I think the only time I've made a good active EQ for the electric guitar is for the JMP1 — because it was programmable, the EQ had to be active, but there's a lot of voicing around it to make it work."

"... if you were to ask me to describe the Marshall sound, I couldn't. The Marshall sound is different to everybody who hears it."

TUBE TOLERANCE

Do you find it easy to secure a constant supply of good quality valves for your products?

"We have now. Over the past few years, we've been working with Svetlana, a part American, part Russian company, and through them we've now developed the Svetlana EL34, which is based very much on the old Mullard design — my favourite EL34. Svetlana want to make a quality product, so through working with them, we've come up with the right quality of the valve.

"For the preamp, everybody has to use the Chinese ECC83 and that's been consistently good — not the best sounding tube that's ever been made, but good nevertheless. When I'm designing, I tend to optimise the circuitry around these valves, although now the Yugoslavians have started making valves again — and to my mind, of the valve companies that were around in the 80s, the Yugoslavians made the best sounding ECC83s. I try to design the amplifiers to be as tolerant as possible of different valve types, although there will be some change in sound if you change to a different type of valve as they all have slightly different characteristics."

TONE VERSUS TONE

There also seems to be a lot of discussion over what valves to use — Americans like 6L6s, while in the UK we tend to settle on EL34s.

"There are differences between the two types, but how big the difference is depends on the external circuit design. I like them both. The main difference between a 6L6, a tetrode, and the EL34, a pentode, is that a pentode is more affected by impedance shifts, such as you get from a loudspeaker. Also, the operational knee has a different slope. If you take an American amplifier and a British amplifier, there are bigger differences than the choice of valves. For example, they make output transformers differently to the way we make them.

"If you take a typical American transformer, the high frequency shift is not so pronounced as you'd get from a British iron-cored transformer, and also they tend to use an awful lot more inductance than I would. That means you don't get the very wide impedance swings you get on a Marshall, and the way the speaker reacts with the output stage is an important part of the sound. Most of the Fender transformers take what I would call a hi-fi approach, which makes the Fender bottom end different to a Marshall bottom end. I like the nice loose-coupled sound."

Is that loose-coupled sound down to a low damping factor at low frequencies?

"It is — a speaker goes from its nominal 4 or 8 Ohms up to 50 or 100 Ohms at resonance, and at that point, it flaps nicely."

SOLID-STATE ALTERNATIVES

Is this a major problem when you're trying to design a solid-state amp to pack the same low end punch as a valve amp? I've heard one or two current feedback power amp stages that seem to get very close.

"The Valvestate range relies on very loosely coupled current feedback designs, and I think it performs very well. They sound, and feel, very much like valve amps, but when designing solid-state or valve amps, I do take very different approaches, because you can do different things with the two technologies. I just try to get the best sound I can using the available technology."

Is the headroom limitation of solid-state circuitry a problem when compared to valve circuitry? After all, guitar waveforms can have very large peaks associated with them.

"Obviously, most op-amp designs give you a maximum of plus or minus 15 volts, whereas with a valve, you've got as many volts as you care to put on it up to the maximum the valve can tolerate. That is one of the biggest differences — being able to re-create the same gain level while maintaining the headroom. There are ways around it, and I do have a patent on one method, but it involves discrete components rather than opamps. There's no way you can get around it just with op-amps. If all pick-ups were equal in output, you could balance it up, but as pick-ups range from beautiful single coils up to very high powered humbuckers, it makes life very difficult. I've tried to keep the input stage as clean as possible, then I process it afterwards."

How much of the magic overdrive sound is down to the valve characteristics, and how much is a function of the pre and post overdrive voicing?

"Let's put it this way: it's easier to make a bad overdrive sound from valves than it is to make a good overdrive from solid-state devices. The overdrive we like as guitar players is based on symmetrical clipping, and as soon as you go to asymmetrical clipping, you get some very peculiar things going on."

Doesn't a typical valve amp distort asymmetrically in the preamp and symmetrically in the push-pull power stage?

"Most of the more primitive designs do, and when you push them too hard, they sound as though they're going to break down — you get all that horrible intermodulation on chords. We make sure our valve designs clip symmetrically, but of course with solid-state circuitry, it's very easy to make it clip symmetrically. It's the application of filtering and other design parameters that make it sound good."

Do you go so far as to replicate power supply sag in your solid-state amps?

"I try to — I use various means of adding compression to the circuit. The black art is listening to the design, then tweaking it until it's right."

"Where does the character come from? I suppose the simple answer is me..."

MASTER VOLUME

Were you responsible for the first Master Volume Marshall amp?

"I was involved in the design, although it was underway prior to me joining the company. My role was to make it sound smoother and nicer, but there was no R&D department at that time, it was just whatever we could throw in during the odd five minutes between testing."

The promise of the Master Volume model was to be able to get the same sound at any playing level, but of course, that isn't what happened. Is it possible to achieve this now, or will there always be a difference in sound as you turn up or down?

"I think that as the years have gone by, myself and all the other amp designers have improved preamp distortion characteristics to go some way towards emulating the sound of the output stage, but I don't think that by any stretch of the imagination you're going to get exactly the same sound. You make sounds very close to it, but it's not the same sound, and I think it's better to make the best sound you can in the preamp and leave the rest to the power amp. The whole chain works together, from preamp to speaker, but we have found ways to make it sound better — extra stages, and balancing out the gain levels, headroom levels, and the voicing of each stage to get a more coherent sound.

"The original Master Volume amp was a good design at the time, and it took us through from 1976 to 1989."

How much of the sound is down to the way the speakers are stressed at high playing levels?

"It's not just the speakers that are stressed — so are your ears, and as the level changes, you hear things quite differently as predicted by the Fletcher Munsen loudness curves. Change from a Celestion and put in an EV, you'd get a completely different sound from the power amp. It's such an unquantifiable situation, and when you play louder, you also have an acoustic feedback loop between the loudspeaker and the guitar strings. The coupling effect is significant, and reflected sound also makes a difference. For example, if you have a small amp in a theatre, it's a lot harder to get a good sound out of it than it is in a small room, and vice versa; a big amp is hard to tame in a small room."

FLEXIBLE TONES

Marshall amplifiers are known and loved for their characteristic sound, but does this work against you in discouraging customers who either want a different sound or who want more versatility? Without overcomplicating the design, can you offer people alternative voicings that provide the styles of other leading amplifier types?

"I think the biggest problem is perception. Most people associate Marshall with big rock amps, yet they are used by players from death metal to country, from jazz to big band, with no problem at all. I even came across a Greek folk band, in a remote mountain village, using 100 Watt JCM800s and 4 by 12s — and not only did they sound great, the players loved them. Even so, over the years, I've tried to make amps more versatile.

"If you take the 30th anniversary amp, for example, which came out in 1993, you have changeable voicing characteristics, and we're going further down that route with some models we're showing at Frankfurt [Music Fair held this month]. I think people will be very surprised at the variation in sounds that we can achieve. Essentially, for the past year, we've been working on a new range of valve amp heads and combos that will be coming out over the course of this year, and I'm paying particular attention to what I think are the best sonic characteristics, not only of Marshall amplifiers, but of what I hear other people want. So, if you want more of an American sound, you can get it. At the same time, I haven't relaxed on maintaining a simplistic approach.

"Because of the number of parameters that need changing, the design used switches rather than variable controls, and in the case of the more deluxe version, it means building an entire extra channel. The design has three channels, nominally clean, crunch, and lead, and I've tried to switch valve parameters rather than switch different valves in and out of circuit. So far, the field testing results have been jaw-droppingly good."

A DIGITAL FUTURE?

Looking further ahead, does the fact that a valve amplifier involves so many interacting parameters mean that accurate digital emulations will be incredibly hard to design? Digital amplifiers are just starting to come onto the market; the designers obviously think they are the way forward.

"I haven't even sat down and thought about how you'd try to emulate an analogue amplifier digitally, and I don't see how you can get both the same feel and the same sound. If the amplifier doesn't feel the same, then you won't play the same. The future will obviously involve digital amplifiers, but I don't think the future is here yet."

Would it be a fair comment to suggest that current digital amplifiers offer very good approximations of different amplifier sounds and that they would appeal to players, who need a wide range of sounds and are prepared to sacrifice a little feel and authenticity in exchange for versatility?

"Without any disrespect to the designers of these amplifiers, I think that's true. They're doing a great job, but as with any technology, it needs time to develop. It's not something we're looking at doing in the immediate future, but in the longterm, for the sake of our company and our employees, it's something we will have to look at eventually."

If DSP-based digital amplifiers are a long way off, what comes next? Will you be continuing to refine solid-state designs so that we can have the sound without the weight?

"Obviously, we have gone down the route of the Valvestates, and that effort will continue, but at the same time, I'm working on new valve amp designs. The supply of valves is guaranteed for more than just the immediate future, and there's such an enormous interest in valve amps that the situation is much healthier than it was ten years ago."

Marshall Amplification Limited, (Contact Details)