Hi Bob: I enjoyed your columns on the K2-W amplifier (Electronic Design, Jan. 6, p. 20 and Feb. 3, p. 22). I knew I had a couple of Philbrick items in my collection, so while in the crawl space this morning I dug them out—a K2-P and an SK2-B. Both have the original boxes and the packaging material.
The K2-P still has the original data sheet with it. Unfortunately, someone spilled something on the large tube shield for the SK2-B. I'm not sure that's the original shield, anyway. (The 7119s did have a heatsinking shield to help them get the heat out. It's not critical. I still think an SK2-B or K2-B1 mounted horizontal would get the heat out better than when vertical. /rap)
I used K2-Ps to build a controlled-potential electrolysis system while at RPI for my MS in chemistry. Along with an HP dc amplifier, it made a nice setup that produced some interesting chemistry results. I grabbed these two because someone was throwing them out! Glad I saved them. (Good man! /rap)
• Jon Titus(via e-mail)
• Pease: Hello, Jon. Take good care of that old junk! K2-Ps are slightly uncommon. I got a couple. SK2-Bs, which I helped design, with the 7119 output tubes are quite rare. Heck, nobody ever bought those. But we had to be prepared to sell them. If you ever want to see them put to work, I could show you how. An SK2-V (or K2-XA) in the middle would do it!
Dear Bob: Well, your schematics look a lot better these days! (Some do—some don't! /rap) Now we'll have to work on your black iron-stove drawing ability. (I think that it's perfect! I hereby quit improving. /rap) Trouble is, my drawing ability is nil so I'll just say your stove looks a lot better than I could ever do. (It's adequate. Hey, you did recognize it as a stove. If not, then I have lousy skills. /rap)
On another topic, it looks like the Mil and Hi-Rel boys are discovering the ugly side of Moore's Law lately. The EDA industry has continuously shrunk down through the micron barrier into nanometer-size features. The 90-nm barrier is the latest, and 65 nm is next according to the VLSI boys.
Unfortunately, someone forgot to tell Mil COTS and long-life Hi-Rel boys about electromigration. This is the new wear-out mode that gets worse as we descend the nanometer ladder. (What's this "we" stuff? I vote against "continued improvements" that actually bring more grief and less improvement. Was it Yogi Berra who said, "Progress was okay, but it went on too long."—?? /rap) The famous long life of integrated circuits is now a false myth. Today's microprocessors and embedded ICs will wear out in three to seven years. That's not the expected 25-year operating life of the telephone industry, or the 40-year operating life of, say, the U.S. Navy AEGIS program.
After three to seven years, your new car's embedded microprocessors may not be a problem—if you use it intermittently. (Do you know how many 'processors are embedded in my new car? I'm talking about my new 1970 Beetle? Yeah, you guessed it—none! Our linear ICs will not poop out for many kilo years, typically. /rap) Maybe someone in the semiconductor industry should tell the future victims about their misfortune. Isn't Moore's Law great? My middle name is Moore. (Sorry about that. /rap)
• Dave Weigand (via e-mail)
• Pease: Gordon Moore says there are 1018 ants in the world, and every year now we make more transistors than that. Maybe if he gets to 65 nm, more transistors will die every year than ants!
Hi Bob: Have you ever been to Pease Pottage in Sussex (http://freepages.genealogy.rootsweb.com/~villages/sussex/pease_pottage.htm)?
• Fred Stevens (via e-mail)
• Pease: I was pretty darn close to it last summer, coming up the A24. We were just four miles west of Pease Pottage. But we had no reason to go through it, as it was slightly out of the way, and we were trying to beat the traffic into London. Maybe next time I'll have a chance to stop and see this village. Thanks for the reminder!