Some mysteries

 

I will readily agree that many controls offered in the past were superfluos, sales gimmics mostly, but let's not overdo it. Tone controls are not bad in themselves, they came out bad because of the features game, because people really went overboard with them, turning them from a very useful thing to a tool by which to murder sound in general. He who actually NEEDS +/- 20 dB boost or cut really needs new ears and definitely a new system. But a calibrated +/- 9 dB lift/cut, in 3 dB steps, individually for both channels, backed by discrete resistors, as offered by say ReVox A78 integrated amp (which I had for some 10 years), is a most useful feature, and even so, I don't remember ever listening to it with a change greater than +/- 3 dB, i.e. just one step.

 

A balance control is even more basic or fundamental than any tone control. Its absence can literally harm an otherwise quite decent, even good device by economy of saving on that one pot, on precision resistors in critical places, resulting in even more channel imbalance, etc.

 

All this puts the consumer in an unenviable position of actually gambling on many factors he cannot control. Not to even start on the subject of personal taste, hearing, equipment, room acoustics, etc.

 

The argument will go: aha, but that's another piece of junk in the signal path, that's a no-no. If you think it's junk, then use a better quality pot, Bourns (to name but one) has very high quality single pots (so you'll need two in form of L and R gain controls) for some MOST reasonable prices, just under 7 euros each. Is 14 euros, or about 9 quid, REALLY too much to pay for the added pleasure?

 

You don't want a switch in your signal path. Fine, use 1 transistor (almost any will do, I use either BC546B or MPSA06 simply because I have bags of them), 1 small signal diode (1N4148 will do nicely), two resistors, a high quality, sealed relay with gold plated contacts (single contact, low power, initial OFF state, switch sets it to ON) and a small switch, any will do, can be the junk of junk, because all it does is to let the transistor switch on, and in turn, it switches on the relay. Upon activation, the relay connects L and R channels and you get a mono signal. The aberrations in the midrange as described by Graham come free, as an added bonus or feature, but in terms of adjusting relative channel balance, they practically don't even exist. Adjust, disengage the switch, it will cut off the transistor, which will open the relay, and presto! - you have a nonexistent mono switch in terms of the audio signal, two ends hanging loose in the air. ZERO influence on sound - but adjusted sound now!

 

Why am I developing Graham's products?

 

But seriously, I have thought much about this, since I plan to produce three levels of headphone amps myself, hence these thoughts of somewhat sacreligious nature. I have included all I have mentioned in the prototypes I am fooling around with, and let me tell you, this ain't no theory, it's hard practice.

 

Oh, by the way, Graham? My all discrete, fully complementary prototype hit 1 MHz delivering 10.8V into an 8 Ohm load ... At 1 kHz and full blast, it deliver 14,87Vrms to the same 8 Ohm load. What? What was that? That's 27.6 WATTS into 8 Ohms? Yeah, it is, so what? Easy when you use a SEPP stage with Motorola's MJE15030/15031 trannies.  And +/-24 V regulated rails. Then, things like an overall (versus the industry's input stage only) Slew Rate of 250 V/uS come naturally.

 

Can I interest you in a professional version, in a 19" rack 2 HE case, with two individual headphone amps, each with its mono switch, channel gain controls and master volume (ALPS Blue), with discrete voltage regulation, for an easy sum of 400 quid? With a custom 100VA toroidal transformer with four secondaries for a true dual mono configuration all built in, just one IEC power connector sticking out? German-made Mentor all aluminium knobs? Heavy duty gold plated brass RCA Cinch jacks? XLR connectors with trusty old AD op amps converting balanced to unbalanced signal? Van den Hul wiring inside? German-made Fischer & Tausche capacitors?

 

In fact, it will drive, albeit with some degradation, an 8 Ohm in parallel with 2uF load. At 10 kHz, there is some square wave overshoot, and some dallying in settling down I don't want to see, however, it is still a prototype, almost finished, but not quite.

 

My main concern was phase response. In the best of British practice (well, at least I think of it as British practice, but then, I have always held BBC labs in very high regard), I'll probably leave its natural response at 1 MHz full power, but will limit the input at 400 kHz or so. At this point, my phase shi(f)t at 20 kHz should be less than 1 degree, I reckon, and that's a value I can live with.

 

At more realistic output levels (14.8V is hardly realistic for headphones), corresponding to power levels sufficient to drive the majority of cans, and even make their owners quite deaf, the results are even better, of course. My measuring gear is fully reliable down to 0.01% THD, and stays dead in the 20...20,000 Hz range, and registers 0.02% distortion only when dangerously close to clipping.

 

Only one serious issue remains to be resolved, and that is the output impedance of the whole gig. On the one hand, I'd just love to have a say 0.1 Ohm output impedance, giving me a true damping factor of typically 300:1 with 30 Ohm cans across the audible range, but on the other hand, I realize this may be opening the door to all sorts of interferenece from radio based devices, such as mobile phones, wireless telephones, radio, TV, etc, as this is no high power amp, but in fact a low power amp (in real world use). At this time, I have no idea what I'll do, except that I will obviously have to experiment much with it.

 

But the very fact I sat down to make them is all your fault - you and your chit-chat about production costs, heavy taxes, Gordon Brown and the other Muppets. You really set me off, plus we have been discussing doing something together for several years now. I have been very busy on that front, I -> WILL <- see a joint product if it's the last thing I see, that much I owe both of us. We'll discuss it further once I have the beta units with me. You know my motto, which I adopted from general Omar Bradley from WWII - hard things we do at once, for the impossible, we need three days.

 

NOTHING is truly impossible, Graham, mostly it's a question of our current knowledge, the willpower to do it and preseverance to last out to the bitter end.

 

One - just for the record, there will always be people who will want the best there is, no matter what. Since I was referring to the top model in a small portfolio, I don't think it's an overkill by any means.

 

Two - let's do some economics, that's what I have a B.Sc. in. Believe it or not, all my calculations tell me that using my current sources, the cost of the materials should not surpass the grand total od 200 euros (app. 130 quid). To that, I need to add manufacturing costs (VERY low in Serbia at this time), company maintenance costs, and of course, my own profit. Manufacturing and assenbley should not cost over 50-60 euros, packaging say 20 euros, odds and ends say another 10 euros, and company costs cannot exceed say 70 euros. Worst case, I am now at (200+60+10+70) 340 euros. Say I am greedy and want 50% as my profit rate, I am at (340x1.5) 510 euros - that's still around 320 quid. UPS or someone will want some 80 quid for shipping it to UK, and presto, I have my top model on your table for roughly 400 quid.

 

Graham, that's in a 19" rack case, with a rather good black piano finish. How much would your repackaged existing product cost in a case like that?

 

And again, that's the top model. Single amp in same package would cost around 50 quid less, lower price home models still less at approximately 300, 250 and 200 quid.

 

Look, I am not trying to ruin your day, rather I am trying to put you back on the track of our discussions some years ago. In the meanwhile, I have done my homework, as you can see for yourself, and have found everything I need to be able to manufacture for both myself and possibly you. I expect you will want to continue with your own designs, which is just fine by me - let's see if we can make it at a lower price, and let's see what that might require. For example, if you change the case from current to fully custom which I can have made locally for me, and if that allows for proper, multivoltage (say 110-120 / 220-240 VAC) PSU using a custom toroid with a GSPAUDIO logo on it, and if all that allows for a nominally cheaper product WHILE MAINTAINING YOUR PROFIT RATES, would you object? Do you think your customers would object?

 

Somehow, I don't think they would. Nor do I have any intention of going bankrupt - quite to the contrary.