Howard Popeck speaks at length to people with interesting things to say: Colin Wonfor (Inca Tech, Magnum, Proac-Magnum, T.O.C.A, Naim and Tellurium Q) sets the record straight (Part #2)

Hello again Colin. I guess you hold a couple of patents, possibly quite a few. in layman’s terms – if indeed such a thing is possible – can you explain what these are and perhaps their significance in the audiophile world?

Crickey Howard, some of those got back a bit and now and are out of date. Oh well here goes.

Moving Coil Headamp

This was the Moving Coil stage on the Magnum amps circa 1979. It,used two constant current generators with a reference voltage of zero. It was based on Kirchhoff’s law. This basically says that any current flowing into a point must equal the current leaving that point. This I achieved using matched complimentary PNP and NPN transistors. The bandwidth was OK but the low level noise was what we needed. Many people tried lots of parallel transistors and even a few OP-Amp designs but none sounded right to me. To reduce the noise even lower I used a common reference diode a Green LED to both channel.

The Low Noise Solid State Relay

This was a patent I did for a Solid State Relay. In itself it had very little use in audio but one of the spin-offs were a solid state high power switch for speaker selection and control, and eventually to the TQ patent used in the “Waveform” series of digital cables.

For many years ..... well since the 1960’s .....  thyristor and triac were and are still used in high power solid state relays. The problem arises when the switch or SSR has to pass EMC and RFI, the non conduction part for the thyristor is about 1.4V - 0 -1.4V as the current goes through zero. The normal, but costly solution, is a large filter at 100 Amps.

So as the current approaches zero, the Thyristor tunes OFF. This causes a large spike and in inductive loads this can be very large. My solution was to keep the device conducting continually through zero and therefore no spike. The circuit is the basic design, U1 is used to detect Zero Crossing, the FET bias voltage is set by the photovoltaic control device and note the MOSFETS are back to back to allow them to conduct in both directions. The SCHOTTKY Diodes are to reduce the Gate Cathode voltage to a low level in reverse bias and therefore reducing noise.

I remember you telling me previously that you and your friend Alan Elsdon designed and built the Dreadnought range of large power amps. Rumour has it that you are working together again. If these rumours are true then what are the plans and what particular skills does Alan bring to the collaboration?

My dear friend Alan and I met when he was a programming ICL computers and I was a service engineer for ICL fixing switched mode power supplies in the day when the PSU was about 200Kgs and the 1M HDD was the size of a washing machine.

At that time I had started doing MIH amps and Disco lights for my mates like Andy Williams for fun and to scare the neighbours - as one does. Alan and I both loved music but not necessary the same music. Nonetheless, we got to chat a lot and he knew Tom Fletcher of Dais fame and had lots of goodies he had made. At that time Alan lived in London and I was stranded on Canvey Island. Alan had a lovely Dais Turntable plus a few others and he taught me a lot on how to listen. He knew Tom well and one day we went to see Tom in Nottingham and I brought one of my 1KW amps, Tom played it and encouraged me to keep bashing on. I also met the clever Glen Croft and was very impressed at what he was doing and I still am.

Alan and I played and one day after designing an electronic load for a nasty CUK converter I had this idea about a SECA but not the normal transformer or capacitor coupled designs coupling the amplifier to the speakers in the tradition manor which were based on Mullard and Lindsey Hoods work. This just caused more problems with low frequency rolling off and poor damping factors.

Alan I chatted and discussed and we built one. It was only a few Watts but ..... it sounded good and at this time MOSFET’s were ..... to be honest ..... crap compared to today’s versions.

We went on to build a larger version and got Martin Colloms to have a listen ..... plus the poor guys at Audio T in London. It was mainly for fun ..... and to learn ..... and we did.

Alan came up with the phrase TOCA a “Touch Of Class A” at this time; still a great name. We then designed the Dreadnought Amps. These were real monsters ;200W and a 300W/ch. Time and family then took over and we saw less and less of each other ..... until we lost contact.

Well Alan noticed what I was doing with TQ and what they were doing to me and found me. We are both now much older but wiser ..... I hope. So with my mind productively occupied we started to build amps again, for fun.

Alan brings to me a control for tidiness and slows me down and keeps me on track without me going on space trips and inventing new ways to ride a bike. He has patience beyond belief and a grasp of all things we talk about. He's a truly good friend. We've known each other now for 40+ years or so ..... and he understands the lunatic in me.

Do not get me wrong; I still have crazy ideas like the Boat Engine without moving parts and it worked.

So for Alan, thanks for being there when I needed a friend.

LTC spice can be downloaded from:-  http://www.linear.com/designtools/software/

What distinguishes music from noise?

Noise a random generation of electromagnetic or audible disturbance without rhythm or timing.

What are your design goals as an electronic engineer? What are you trying to do with your designs?

Normally the goals are to build or design equipment to meet specification at the lowest possible price to make a profit, normally governed by what the Marketing People think is a good idea. We see loads of them and most are just junk re-boxed at a lower price and thus ..... apparently ...... value.

But mine are not. Yes I would love have a little money and see my friends in Madeira every weekend but that has never happened. I hate rip offs. I have worked with them. GREED seems to grab them and they think YOU - the Public - are and always will be fooled by a nice coloured box. But .... don't be fooled.  Behind your back hands are being rubbed.

But my goals are basically to enjoy what I do and produce kits that lasts 40yrs or more like the Claymore Amps have. I like to share my designs but .....  that's always done me wrong. But .... I am not going to change simply because some stupid sod did the rip off Colin again trick.

I want people to enjoy and get satisfaction and say with their heads high "I made that I put it together with a little help but I made that." I have seen it now a few time and I love the warm glow it gives them and pride. Some of the things I make can't easily be done by the noble amateur bit I dearly wish they could

What is your response to the statement ‘the vast majority of amplifiers and loudspeakers don't distribute the energy correctly.'There is enough energy in the bass frequencies, but this energy is somewhat lacking through the mids and the highs. However, the human ear is used to the large amount of energy that musical instruments can produce – like a trumpet, for instance. Most hi-fi systems can produce the tone of the trumpet, but not the necessary energy.

All amplifiers are dynamic power supplies designs with luck to drive a load that meets the parameters needed to produce the correct driving force to reproduce a noise or music.

Energy 1W AC at 50Hz will produce the same energy as 1W AC at 100Kz i.e. heat a resistive load. Now if the load is a reactive load i.e. a loudspeaker and all the filters to tune the speaker wherever possible to react in a flat response like a resistor ..... then there will be problems. As you can see with this simplified formula as frequency goes up and a capacitive load the more current is used and crossover have lots of a capacitance. So the poor old amplifier has to work harder.

A 100uF cap at 10V
Frequency Hz 10 20 50 100 200 400 800 1600 3200 6400 12800
Load Z
159.1343 79.56715 31.82686 15.91343 7.956715 3.978358 1.989179 0.994589 0.497295 0.248647 0.124324
Current Amps 0.06284 0.12568 0.3142 0.6284 1.2568 2.5136 5.0272 10.0544 20.1088 40.2176 80.4352

and for 10uF a typical load on a amplifier.

A 10uF cap at 10V
Frequency Hz 10 20 50 100 200 400 800 1600 3200 6400 12800

Load Z 1591.343 795.6715 318.2686 159.1343 79.56715 39.78358 19.89179 9.945894 4.972947 2.486474 1.243237

Current Amps 0.006284 0.012568 0.03142 0.06284 0.12568 0.25136 0.50272 1.00544 2.01088 4.02176 8.04352

Now if we have a two speakers a Tweeter and Woofer the Woofer and the Tweeter need different power level to move the same amount of air. The Woofer could use 1W and produce 90dB in the audio range of 50 - 500Hz were the Tweeter at the same power input level may be only 84dB to produce 500Hz - 15KHz in the audio range.

Now the next problem is neither of them have a flat response and the Tweeter needs more power to drive it to the same sound pressure level as the Woofer and this is before we start adding filters and their losses. Now the box, this absorbs sound so we have another loss of energy, so to sum up.

Losses :- Boxes, Crossovers, wire, packing , not a nice load, and the all the phase changes this causes in the sound we get and the load impedance and phase changes the poor amplifier has to manage current and voltage out of phase. It's a nightmare load. No sane designer would design a load like a speaker.

So to get it right is often a miracle. It takes years of hard graft, unless you are lucky first time. I designed a pair once. They were OK but ..... never, never again.

So ....?

So many of my prototypes ended up dropped of Southend Pier.

It fun being a designer but I do wish I stuck to banking like my Mum said.

To be continued