My To-Do List:

So many ideas - so little time... I have lots of ideas in mind, but I'm sure most of them won't come to fruition.

Power amp:

Plans for the "monster" amp are quite well formed in my mind, and I've got a reasonably clear development plan as well. Basically, I'm after 150-250 watts per channel to drive the ATC's, clearly a bit much for a GainClone! So, I started thinking that I'd build a Self Blameless amp, using two output pairs per channel in a CPF configuration. Then I worried about CPF, and read a lot of stuff on the web that suggested that it's hard to get stable and sounding good. I read a lot about the Leach amp - and found nothing but praise for it. Curiously, I've found nothing but complaints about Blameless amps, but these all come from non-engineering subjectivists (discussion forums are fascinating!)

Last year I bought Randy Sloan's book because I wanted some more info on symmetrical VAS stages, but while there's lots of ideas and schematics, there's little in the way of real analysis. I found it covered much the same ground as Self, but in a more approachable (read: wordy) style. And don't even think about using the PCB designs - they are amazing! For years I'm been meaning to write a page offering guidance on PCB layout, and I'm hoping to get permission to use extracts from those designs as they illustrate perfectly how not to do things!

But there was a spark of insight on the section regarding EF vs CPF stages. Sloan suggests that the characteristics of the driver and output transistors are key to ensuring stability - and I've never liked circuits that rely on transistor parameters for anything!

The Leach amp got me thinking about Triples. Then, some development work published in the third edition of Self demonstrated that triples are indeed a good way to go, especially if combined with multiple output devices.

So my current thinking is to build a universal triple output stage, using 2 or 3 pairs of MJ21193/MJ21194 or MJL1302/MJL3281 pairs. Drivers, pre-drivers, thermal compensation, protection and all other ancillaries would exist on this main board. The small signal stages would be on a plug-in module, and I can build a number of these - eg, Self Blameless, Sloan Opti, Leach, even the CFB used by Arcam in their A85 amps. Then, if there really are subjective differences, I ought to be able to hear them. As long as I standardise on a VAS standing current, the modules should be plugable with no adjustments. This is intended to be a prototype - proper optimised PCB's could be made once (or if) I decide which is best.

I've even got a few ideas for the chassis, but that's a long way off!

Audax GainClone and Subwoofer:

Inspired by the success of my Audax mini-monitors, I'd like to build an integrated amplifier for them in a matching birch-ply case. It will be like those cheesy mini-systems you can buy, only with slightly better performance. It only needs to be 20-30WPC, so there's no point using anything other than a GainClone here, besides the small case dimensions preclude using anything else really. I've got a vague plan to incorporate an FM tuner, which will involve butchering an FM walkman or an old car radio. It will be remote-controlled, probably using the Alps Blue potentiometer that I bought years ago for the preamp (although it's a bit big for this!).

As it's dedicated to the Audax mini's, I can incorporate some active EQ to extend the LF response. Something like a Linkwitz transform, perhaps. And I can easily incorporate provision for a sub-woofer output - I'm thinking of ways of making a tiny subwoofer using matching birch ply. It will be a sealed ELF effort, and to be honest, it will probably be more of a "woofer" than a "sub".

Clock Radio:

Since building my powered speaker, I've been using it in conjunction with my Sony SRF-48 RDS walkman as a bedside radio. I've been really impressed at the quality achievable from such a little speaker, and I'm inspired to try and make something that's a little more convenient and doesn't look like a piece of lab gear (which, of course, is exactly what the powered speaker is!).

So, I set a limit of using a 3" speaker, and I'm trying to find the best one for this application. Unusually, I'd like to give people a fighting chance of being able to make their own, so I'm only picking well established, readily available drive units. My first candidate is a Visaton FR8-4 (about a fiver from CPC), but having auditioned it, I'm not convinced it's as good as the no-name driver I used before.

For the tuner, I've got a couple of TDA7000's to play with. These are a complete FM tuner on a chip, and seem to require hardly any external components. At this stage, I've no idea what the performance is like, but they were cheap enough from Rapid Electronics. If they don't work out, there's no shortage of cheap radios that could donate their FM sections, and there's even a Velleman kit that might be OK. Remember, this is a non-critical mono-only project, so we don't need perfection.

For the clock, I've got a number of digital clocks I could use, or I could even use a PIC. But we've got a nice analogue alarm clock made in Germany by Braun that apparently has an unreliable alarm (sure that's easily fixable). I like the idea of an analogue clock because it fits neatly with the analogue tuning of the FM radio. It's nice and quiet, attractive in silver, and is the same diameter as the required opening for the 3" speakers. It could easily be installed into a birch ply box along with the rest of the radio, and I can even see how I could add some LED's for illumination!

I should point out that I had this idea long before I saw the Tivoli bedside radio!

Speaker design setup:

Inspired by the results of the Audax mini-monitors, I've just bought a Behringer ECM8000 measurement microphone, and I'm planning to build a low-noise preamp and a speaker measurement jig. In order to learn the principles, I intend to attack my old Musical Fidelity MC1's - a rather average design from the early 1990's that uses decent quality drive units in a cheap chipboard box with a poorly designed crossover. For as long as I've owned them, I've been convinced that I can make something better from them, so with the help of the measurement setup, I should be able to improve the crossover in the first instance. This might even help anyone out there that still has a set of these, as it would be an easily reversible tweak. Next, I plan to rebuild them into a proper box, perhaps with sealed loading or different reflex tuning to improve the one-note bass.

Audax centre speaker:

I'm planning an active centre speaker using the same Audax AP100Z0 and TM025F1 drivers that I used for the mini-monitors.

Subwoofer:

Another long-term plan, I'm torn between an ELF sealed box, or an IB design - the decision depends on the layout of the next house. For more "purist" use, I'm also considering a pair of smaller subs to work with the ATC's - having read about the 10" Peerless woofer that Frank Johnson used in his active 3-way design.

Purist preamp and op-amp listening tests:

After the complexity of the 6-channel preamp, I'd like to do a simple stereo preamp with the emphasis on absolute sound quality. I hope to be able to make use of the small enclosure that I built years ago, and perhaps the motorised Alps Blue. This would provide me with an optimised test bed for auditioning different op-amps - there is a lot of talk on the Internet about the audibility of these, and while I don't get too hung up about it (because I've seen the schematics of plenty of mixing desks!), it would be nice to see if there's any truth in the matter. This might lead me to try alternatives to the much-hated (although not by me!) NE5532's that I used as output buffers in the 6-channel preamp.

Headphone amplifier experiments:

I recently noticed that the headphone amplifier stage of my preamp was being discussed on the Headwize forums, and people were saying lots of positive things about it. This is nice to hear - certainly I thought that it performed very well, but I didn't have as much time to devote to it as I would have liked. With this in mind, I'm planning to do some more experimentation with it - one day ;-)

WM8816 tests:

One volume control IC that I didn't try was the WM8816 - since building the preamp I discovered that Rapid Electronics sell them, so I ordered a couple. This IC is interesting because it doesn't use an internal op-amp, leaving you free to select your own - and lot's of people believe that the internal op-amp is the weak point of the PGA2310 chips. Hopefully I'll be able to audition it as part of the above work...

New workshop:

Here I'm just dreaming, because we haven't even found a house to buy yet! But I've got lots of ideas of stuff I'd like to incorporate, and thanks for the recent BBC move in Norwich, I've got no shortage of skip-fodder that will make it into the new 'shop, including a full-height 19" bay! The main worry is whether I'll be able to afford a house with a suitable garage, knowing the price of property where I'm moving to.

The website:

Not to mention the upkeep of this site, which seems to take up most of my spare time these days!