Electronic Design

Bob's Mailbox

Hi Bob:
During heavy rain, water drips from the underside of my dashboard onto my left leg. I often cannot move my leg out from under the drip as the leg is busy operating the clutch during my commute in traffic-congested downtown Boston. Very annoying.

So where's the sneak path in my '93 Jeep Wrangler? Well, I've traced the path of the water up to the windshield frame. Sometimes the frame accumulates so much water that when I take a sharp turn, it sounds like a small babbling brook is flowing through my dashboard! I haven't been able to figure out how the water is making it into the frame, though. The windshield has been replaced twice due to stone chips, but the seal looks good, and it passes the garden hose test. What next? There also are windshield wipers and roof hardware attached to the window frame. Perhaps I'll have another go at it with the garden hose.

Anyway, I have 21 or so years and $250+ dollars to spend to find the root cause and still be a step ahead of you!
Steve Rideout
via e-mail

Steve, here are a few approaches:

  1. Ask the Jeep guys what's going on. They may know.
  2. Ask the windshield guys. This is not rocket science. It's their specialty. They have to know what's causing the problem.
  3. For free (on warranty) or for $$, either way, ask them to fix it. They can glue the BEJEEZUS out of it to end that problem. I put goo on my old Beetle, and that fixed it pretty well for a couple of years.
  4. Drill a hole in that frame and siphon all of the water out!!—RAP

Hi Bob:
I would like to add to what Mike Cozza from Fender had to say about the sound of tube amps versus solid state (electronic design, May 29, p. 120). I think that it all depends on whether you're trying to produce or reproduce a sound. If you're reproducing a sound with your hi fi, then yep, I think solid state is the way to go. If it sounds the same as the original, then it sounds right. But if you're producing an original sound via an amp/guitar combination, then a tube amp is often the way to go. The combination of the two is the musical instrument, not just the guitar by itself. If you get the sound that you want, distorted or not, then it sounds right.
Graham Pratt
via e-mail

Of course! Very well put. Thanks for writing, Graham.—RAP

Hi Mr. Pease:
I was referred to you by someone who thought that you might be able to answer my question. I'm looking for the producer of the first operational amplifier, and the specs for this chip. He thought that Bob Widlar of National might have designed it. If you happen to know—or know where I might be able to find it—I'd appreciate any info that you could offer. Thanks a lot!
Tara Calabrese
Catholic University of America
via e-mail

Hello, Tara. The first 300 solid-state operational amplifier designs were not "chips." They were op amps made out of 10 or 20 discrete transistors. I designed several of them. Those designs were made by Zeltex, Burr-Brown, Nexus, Philbrick, and over a dozen other companies. Hybrid designs were crafted by Fairchild, National, and a bunch of others that are hard for me to remember now.

For about 20 years before that, there were op amps made out of four, eight, or 14 triode vacuum tubes—or pentodes. Also, some were even built from the parts of a cheap Japanese seven-transistor radio. I can document that—7 GERMANIUMs.

The µA709 from Fairchild was the first real op amp made as a chip, in around 1967, long after the other 300 designs. It was designed by Bob Widlar, but not when he was at National. He was at Fairchild, then. You can look up the chip's specs on the NSC LM709 data sheet, which is located on NSC's Web site at www.national.com. Just ask for LM709.

The µA702 from Fairchild was the first "op amp" made as a chip, in about 1966. But, it was a special-purpose junker that could only run on +12 V and −6 V. Its specs were awful. It had almost no output drive. The input base current was a lot of microamps. Maybe 20? It had no specs to speak of, and they're out of print, as the '702 has been out of production for over 10 years.

It was characterized very thinly. The µA702 went to the moon. Every time the audio systems went silent, the guys who designed the '702 into the lunar landers gasped, because they were terrified that the 702s were DYING—again. It wasn't a very sturdy or robust IC. In fact, its lousy robustness was exceeded only by its poor performance.

There were some TI amplifiers, such as the SN5224 and SN5226. (I know those aren't the correct numbers, but they were something like them). Yet, while TI called them op amps, they were not. They really were differential-in, push-pull out (differential-out). So, they were barely able to make an output swing of 1 V p-p. Plus, they too had poor ZIN and poor IBASE.

The µA709A actually had a base current of 0.5 µA, and it sold for $75 in 1968. Its gain was 25k. We made good op amps at Philbrick, with IB of less than 0.005 µA and gain above 200k, that could put out ±>20 mA at 1 MHz. But they cost a lot to make. As soon as Fairchild got their yield up, the days of the discrete-transistor op amp were numbered.—RAP

All for now. / Comments invited!
RAP / Robert A. Pease / Engineer
[email protected]–or:

Mail Stop D2597A
National Semiconductor
P.O. Box 58090
Santa Clara, CA 95052-8090

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.