They have had humorectomies and did not understand you. (Chirurgische Entfernung aller Sinn für Humor)
Last edited:
Thanks for the input, Martin. We have tried to use 'quiet' caps for many years. It is obvious with experience that many caps have a high mechanical 'Q'. This is one major reason that we often use polystyrene or polypropylene film/foil caps in critical locations. It is not 'magic' just careful engineering.
Polyester / Ceramic Caps
I have obsereved over the years that some older polyester/Mylar caps do indeed have a wide range of quality. I have encountered older "green Chiclet" style caps as well as some large value ceramics which failed by:
1) Becoming noisy.
2) Oscillating at both audio and RF with a small amount of DC bias.
3) Making popping noises at audio peaks, despite being used far below their voltage rating.
Some of these failures could be made to change by merely heating the leads or squeezing them. Like the rest of y'all, I have totally avoided using any polyester or ceramics in audio since the late 70's for these reasons. I have not seen these failure modes in polycarbonate, Teflon, polystyrene or polypropylene caps for some reason...maybe lower mechanical Q as John said?
Howie
Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
1st on the Internet
I have obsereved over the years that some older polyester/Mylar caps do indeed have a wide range of quality. I have encountered older "green Chiclet" style caps as well as some large value ceramics which failed by:
1) Becoming noisy.
2) Oscillating at both audio and RF with a small amount of DC bias.
3) Making popping noises at audio peaks, despite being used far below their voltage rating.
Some of these failures could be made to change by merely heating the leads or squeezing them. Like the rest of y'all, I have totally avoided using any polyester or ceramics in audio since the late 70's for these reasons. I have not seen these failure modes in polycarbonate, Teflon, polystyrene or polypropylene caps for some reason...maybe lower mechanical Q as John said?
Howie
Howard Hoyt
CE - WXYC-FM 89.3
UNC Chapel Hill, NC
www.wxyc.org
1st on the Internet
Howard: I believe that it has much more to do with the way connections are made and the quality/reliability of the sprayed zinc commonly used on the ends of the rolled polymer. Additionally, the aluminum metallization often degrades at the ends of older caps, increasing ESR and, when severe, causing the sort of intermittents that you've observed.
In the last several decades, the processes and materials (including encapsulants) have gotten a lot better, and I certainly wouldn't recommend that people use old stock caps- new, high quality capacitors are inexpensive and readily obtainable.
In the last several decades, the processes and materials (including encapsulants) have gotten a lot better, and I certainly wouldn't recommend that people use old stock caps- new, high quality capacitors are inexpensive and readily obtainable.
The 2SJ103 is being discontinued as we speak. I can't remember if the deadline was April 1, May 1, or June 1. We put in an order for 500,000, which should be a fifty to one hundred year supply. I'll be dead long before that happens.
Just to warn everyone, ALL Toshiba TO-92 packages are being discontinued next spring. They will make some of the parts available in SMT packages but the power handling is way down. Most of the TO-92 parts are rated at 300 or 400 mW, while the SMT parts are 150 mW. I never run a semiconductor at more than 1/4 the rated power, so that means ~40 mW. So if you have 6 mA bias, you are limited to less than 7 volts across the part.
The only way to get that to work is to cascode it. But what will you use for the cascode? SMT parts obviously won't work unless you go to a TO-252 package. Gee, that saved a bunch of room and money, didn't it? Replacing one TO-92 with a SOT-323 + TO-252...
Bottom line is if you like the 2SK170's, you have less than a year to buy more. They will become extinct like the rest of the JFETs.
Boy, I hope Murphy is paying attention. To properly follow his law as soon as you take delivery a new better part should appear!
That's obvious, Erno also did. I thought we were finding parts for input complementary diffs.
Fairly quaint thing to do, commiting to the same design paths forever.
Gee, Scott, we wouldn't have to if Analog Devices made any good sounding parts. The only one that is even close is the ancient AD844 (you were kind enough to measure some resistor values on it for me a decade or so ago), but they moved it from a 4" fab to a 6" fab and now 20% of the parts are noisy.
And where are the ADI parts with complementary differential JFET inputs? PMA did an analysis on this very thread (or was it the previous one?) that showed an unsurpassed performance from this topology.
BUT it's too much trouble for the big IC companies to make. So they just make the stuff that's easy. The biggest driving force in IC design right now (besides lowering costs!) is to reduce power consumption. Does ANYBODY on this thread care if their op-amp draws 2 mils or 20 mils? NO! But Apple does. And as Apple goes, so goes the world...
That's why there is mid-fi and a separate category called "high-end".
And where are the ADI parts with complementary differential JFET inputs? PMA did an analysis on this very thread (or was it the previous one?) that showed an unsurpassed performance from this topology.
BUT it's too much trouble for the big IC companies to make. So they just make the stuff that's easy. The biggest driving force in IC design right now (besides lowering costs!) is to reduce power consumption. Does ANYBODY on this thread care if their op-amp draws 2 mils or 20 mils? NO! But Apple does. And as Apple goes, so goes the world...
That's why there is mid-fi and a separate category called "high-end".
Nah, he's sleeping on this one. We already bought 500,000 2SJ74's three years ago. Still waiting for Linear Systems to deliver their copy -- at 5x the price...
No buyer's remorse on this one. How would you like to be sitting on a stock of a half-million each of the best audio tubes built in the '40s and '50s? You could sell them. You could build the best amps in the world. You would be in pretty good shape. Oh, I suppose that there are people that would argue that the Russian 6H30 "supertube" and the 6C33 power tube are good, but are they really better than the best tubes of that era? Will a 6C33 compete against a 6336? I doubt it...
Actually I am sitting on some great 50's NOS tubes. Still able to buy them for pennies! Everyone keeps using the tubes they know. They made some very nice triodes for portable radios. They seem to have gone unnoticed, a purpose built end of tube design era audio triodes!
But it seems you have bet well. If they stay king of the hill, you win. If something better comes along you and your customers still win!
Will either compete against a 6528? (says the man with boxloads of them)
- Status
- Not open for further replies.