Steve Deckert
|
WHY TUBES, CABLES and even boutique parts mean less to me than others…
Tonight during the final voicing of a particular product — I transformed a good but struggling sound into something wonderful. By modifying the value of a part and adding a second part and finding the magic value for that I took a tube stage into legendary status. The changes were electrically very subtle., but the difference was well beyond going from a cheap input tube to a rare NOS jewel that cost a small fortune. The difference was greater than most cable upgrades you might make.
During this process of voicing I’m not using a Spice simulator to model the values and watch what happens, rather just using common sense to pick a starting point and then slowly migrate the value up until it starts sounding bad to my ears and then migrate the value down until the same thing happens. That’s the window. On the high side lives sterility and hardness. On the low side lives richness, weight, and distortion. I swim around in the middle until I find the magic values which is a very narrow window, perhaps two or three tops. In this zone, you have none of the sound of either end of the window, just bliss. Is the bliss exactly where your Spice model put it... almost never.
It takes a long time, because you have to build it and listen to each change. So in this case 10 value changes defined a window with a single part. If you use two parts like I did, the time can double. Actually as is often the case there is no perfect value and I usually get stuck between two values which are equally good, depending on the music and the speakers or headphones. In this situation a silver switch is used to give the user both values at his/her fingertips.
You quickly understand using this methodology that you can make most circuits sound any way you want. You have to first hear it in your mind, and then find that sound with the circuit. When you design audio gear with a calculator, you can’t listen to the windows you test, if you even go that far. Most I suspect will just calculate the “correct” value and test it on spice followed by building the actual unit exactly that way. Then if they don’t like the sound, they blame it on the parts and start putting in 100.00 capacitors and exotic resistors and connectors and wire and so on trying to get that sound. It never really happens... only for them.
Never create a circuit using boutique parts and perfect playback gear. Always inject a full armament of handicaps so that WITH those handicaps you can make it sound right. It’s not hard, they are there before you voice it, and if you can’t get the transparency or the frequency balance or focus you are after, then adjust the circuit until you get it. Then and only then and after it has burned in and you’ve lived with it for a good while, do you systemically upgrade parts. Now you actually hear what those boutique parts are actually capable of to their full potential instead of half their potential.
Once you have lifted every handicap to see how far the design peaks out on the wow meter, you may want to tweak the voicing just a touch, but most of the time not.
As a consumer the closest thing to this process you can experience is tube and cable rolling. Changing one changes the other.
-Steve Deckert / DECWARE
|