Damage to semiconductor device junctions (Keith)

Rich Paton · From: Santa Maria, CA.,

Your comment on device junction damage brought to mind the virtues of applying "overkill" when specifying or chosing components (quality vs. reliability vs. cost)in an electronic design, when it is economically feasible to do so.
>
In the early 1980's I was involved in an engineering project as part of upgrades to the NORAD "Ballistic Missile Early Warning System".
>
I had to design and implement some relatively simple, but as absolutely reliable as possible circuitry and hardware to allow simultaneous operation of the sites with both the old ( ! ) IBM 7090 computers and the new CDC Sigma 7 systems replacing them.
>
We had an in-facility office of "Mitre", a "watchdog" agency, independently monitoring ALL of our goings-on for DOD.
>
One "Quality Engineer" type from that group was a hard-core "What If" sort of guy, and began bugging the heck out of me about the integrated circuit components slated for inclusion in my "Integration Test Tools". Devices".
>
Part of the equipment in question was simply a six channel, bi-directional data buffer using twelve CMOS CD-4049B gated hex inverter chips.
>
I finally decided to order "Class S", space-rated, radiation-hardened chips at a mere $100 a pop, vs. five bucks or so each for standard mil-spec chips.
>
These special-order, highly documented & certified units were at the time the finest devices of their type available on the planet.
>
When this approach failed to muzzle his incessant rambling, I finally told him he needed to sit down, read, and try to understand what the big pile of paper was trying to tell him, since "things had changed a great deal since what he was doing was taught in college, back in the "Fred Flintstone Era" of Electronics Engineering"!

That, plus the fact I basically refused to talk to him anymore finally got him to back off.
>
I considered my "Applied Overkill" approach to be completely economically correct, since it saved us countless man-hours of testing, evaluation and documentation, as well as sparing the life of quite a few trees.

Jack Stoner · From: Kansas City, MO

Good thoughts. I worked on NASA's Manned Space Program for a while. NASA's specs, in many areas, exceeded Mil Spec's. In those cases, with people's lives on the line, it is warranted. Commerical devices, that will be marketed to the general public, has to find a happy medium. If you build something to a price it with be a piece of crap. If you build it to Mil Spec (or NASA spec) it will be too expensive to market. Manufacturers of consumer items have to find a middle ground and a product that is reliable but won't break the bank.

Keith Hilton · From: 248 Laurel Road Ozark, Missouri 65721

Rich, I bet the guy never did understand what the big pile of paper was telling him. That is really funny!
Hartley Peavey told me that in buiness you go from one crisis to the next. It is just a series of crisis situiations you have to solve. The one who keeps solving them is the one who stays in business. The one who gives up is the one who goes out of business. Look at CUSTOM-out in Kansas, the guys who used to cover their stuff in rolled and pleated material. At one time Peavey and CUSTOM were fighting it out. I really thought CUSTOM might out last Peavey, but they just gave up.
Don't think that I have not had difficult problems to solve building steel guitar stuff. The most difficult problems for me to solve are mechanical problems, not electonic. The most difficult problem to solve concerning my new pedal was the clutch. It took me almost a year to get it right. I got a clutch now that will last forever, never needs adjusting, is noise free, and feels like butter. The pedal stays where you left it, when you take your foot off.
There isn't much fame and glory in building electronics. It boils down to a lot of hard work. Right now, my work days are about 16 hours long, and have been that way for a long time.

------------------

Keith Hilton · From: 248 Laurel Road Ozark, Missouri 65721

Mark has just e-mailed me with a correction. It should of been KUSTOM, not CUSTOM. Remember, they were out in Chanute Kansas. Does anyone know what happened to their design information, or the Company in general? For their day, they made good quality equipment. I suppose they just gave up. Does anyone know?

------------------

Jack Stoner · From: Kansas City, MO

Keith, the guy that started Kustom (I think his name is Ross) has started a "new" Kustom company. They are putting out Kustom amps again, but they are not the same electronically as the originals. I saw one of their two piece bass amps in Quigley's Music in KC about 3 years ago. I think Musician's Friend is also selling them.

Jerry Hedge · From: Norwood Ohio U.S.A.

Kustom now is owned by Davitt & Hanser Music here in Cincinnati. The eletronics are assembled in China and the amps are assembled here. The design engineer is a friend of mine. One of Kustom's execs is a PSG player. I'm supposed to get a couple of the bigger amps for evaluation. I'll let y'all know what they're like. Currently I'm playing thru one of their 20 watt amps for practicing at home. Sounds nice.

Ron Dodd · From: North Myrtle Beach SC

Hi Guys

Just wanted to add a little info on
Kustom. The company referred to in
Chanute Kansas is in fact the original
Kustom Audio folks. They sold the audio
division some years ago and are now one of
the leading manufactuers of speed measureing devices. I spent a week there for training
on Kustom Radar Units.

Ron Dodd
Mullen D-10
Nashville 400

Rich Paton · From: Santa Maria, CA.,

"Rich, I bet the guy never did understand what the big pile of paper was telling him."

So...THAT was his problem! It figures.

I was just too busy trying to produce a good unit to waste time worrying TOO much about it!....(What If?, What If?)

Now I remember (a dirty trick?) giving him the phone # of the National Semiconductor Co.
salesman I ordered the chips from, to hash out the details with.

I bet they had quite a chat! :>)

Keith Hilton · From: 248 Laurel Road Ozark, Missouri 65721

Rich, I suppose I am totally paranoid about 2 things. #1 Electronic failure protection. #2 Noise protection. I know people building electronic stuff for steel guitar that put in no failure protection what-so-ever! These same people put in no noise or hum protection. I really don't know how they can get away with it. Hartley Peavey pounded one thing in my mind a long time ago--PROTECTION, PROTECTION, PROTECTION. Hartley used to tell me that if I didn't do it, it would "BITE ME IN THE REAR!" I learned my lesson the hard way! When you have to start sending people new replacements, you learn quick that you can't ever have enough protection. I used to think that a lot of this protection reduced sound quality. It does take time to figure out the right values and the right parts. I am proof you can have great sound quality with great protection. All companies have failures. Ford had gas tanks that leaked and burned up people. Burger King made thousands sick with rotten hamburgers! Static electricity shot from my finger and burnt up a 3 month old Sony VCR. Wal-Mart would not do anything about the VCR!
I sleep good at night, knowing I have done my best to protect my electronics from failure, and from noise. I have a fellow wanting to contract out production of some boards for me. I suppose I have driven him crazy talking about what "COULD HAPPEN". I suppose it did get on his nerves. At one point he said, "We could make the electronics better, I suppose, if we drove a copper rod in the ground for earth reference." How would you like to play a steel guitar pedal, connected to a 8 foot copper rod drove into the ground? I suppose you would know then you had a good earth ground reference. Just kidding! I knew then he thought I was going a little overboard. I suppose it was me reading big stacks of papers, and calling the chip makers. I talked to one applications engineer at Analog Devices, and he thought I was worried about nothing. Nothing, until I said, "What happens when the output junction sees 4,000 volts from static electricity?" The guy then said, "I see what you mean." Hartley Peavey tells me this, "If your device sees the outside world, no wonder what will be plugged into it." Keith Hilton

------------------

Rich Paton · From: Santa Maria, CA.,

Keith, I do admire your philosophy and dedication.
As with EMI/RFI and bonding, grounding,and shielding, ESD (electrostatic discharge) and TVSS (transient voltage surge suppression) are huge technical topics. I've found that in the TVSS area, I have to read between the lines to derive some useful data. For example, the common IEEE test method for evaluating TVSS components doesn't blare out the fact that the tests for clamping response time specs are conducted with the grounded conductor of a supression device having a maximum length of One Inch to ground. Typically a unit sold to protect a main electrical service comes with an 8" to 12" grounding conductor. See the problem? One must sometimes dig deep for the most useful information!

Keith Hilton · From: 248 Laurel Road Ozark, Missouri 65721

Yes, Rich, I agree! You can read big piles of papers on a particular situation,and that only eliminates obvious things. Then, I don't care who you are, Hartley Peavey, a applications engineer at Burr Brown, or anyone else, it is trial and error. Experience is a painful teacher. It all boils down to a lot of hard work. Just about the time you think you have a product perfect, you get some un-explained failure. This usually sets up a chain of events that takes a lot of work, and a lot of time. The good thing is that you usually learn a valuable lesson from the suffering.

------------------

Rich Paton · From: Santa Maria, CA.,

Keith,

This harangue addresses your comment on the trials & tribulations of TRIAL & ERROR...

First, who says "you can't teach an old dog new tricks"?

I've been involved with an extremely structured, formal training program in a new position, which involves the setup, calibration, and readying of all sorts of recording devices such as the "real-time" computer program, hard disk packs, mag tape recorders, a 16 channel strip chart recorder, zip drives and outbound data modem channels of real-time data, all on a complex, old, & very confusing radar system at Vandenberg AFB.

When all that is properly configured, the mode configurations, beacon timing, receiver and transmitter paramaters must be set.
Once that stuff is accomplished, a "track" is done (or not done!) on an object such as a satellite or missile launch. Any personnel errors resulting in a loss of a or reduction in the quality of the data means the company will be penalized a big chunk of change $$.

I've been struggling mightily through these ordeals, and because I was behind the time line, last week the department head himself became my weekly evaluator. Friday's eval was 2 hours on the site's receiver theory (no notes), and three hours of grilling on operations.

So, I figured he would come in swinging the BIG axe, and it would be all over. (INSTANT unemployment...and facing more job interviews, having to explain why I was canned after just one month!) Not fun, not fun! Bummersville.

But...amazingly, as it turned out this situation was the best thing to happen to me!

After doing fairly (And I mean fairly, no better) well up until the last two hours, I pulled several several potentially disasterous gaffes.

OK...so here's the point, and the "new trick" for this old dog:

When I had set a control improperly, I was asked:

"Now, exactly what would you do now, to recover from the mistake and get the setup back on track and done within the alloted time"?

I managed to "pull it all out of my @ss", by grabbing the huge tome filled with all the Site's SOP's, Standard Operating Procedures. I luckily found the pertinent section right off, but began faltering immediately under the pressure, in reading and comprehending the pertinent procedure.

So...finally, the "Crux of the Biscuit", or "Gist":

He explained that the normal human reaction is for the person to try to recover the error by trying to remember what had been done wrong.
But he explained to me to forget all that, and carefully work through the procedure from Step One, and also consider what was accomplished by performing each step, and why it is done (the functional result) within the system.

I immediately came out of my brain "lockup" and fixed the mistake. And I passed the eval, not because I already knew what to do, but because I was (somehow,???) able to correct the problem through calmness and the proper use of the reference materials.

That was a HUGE lesson on the methods of "thinking"...and ever since thebn I find it works for little and big things that pop up at home and about town as well, and it just makes absolute sense...but I had never been taught how to think that way, nor had I ever blundered upon it by chance !!!

So, the old dog's new trick was a huge new insight into the concept and methodology
of.......TRIAL AND ERROR!

There's a lot of little, but critical things in life like this, that aren't taught in 13 years of public education. (That's probably around 18,000 or so hours of total classroom time).

But...the human brain is an inherently lazy device and will automatically avoid any uncomfortable, unfamiliar stimuli, by its very nature.

I find it all very interesting, but also quite frustrating when viewed in trhe light of our perfect human 20/20 hindsight!

My apologies, on the protracted utterance, but it is really a big deal to me at this point.

"doc"

[This message was edited by Rich Paton on 26 March 2000 at 10:24 AM.]

[This message was edited by Rich Paton on 26 March 2000 at 10:42 AM.]

[This message was edited by Rich Paton on 26 March 2000 at 10:43 AM.]

Keith Hilton · From: 248 Laurel Road Ozark, Missouri 65721

Rich, we both could probably use a few days off from electronic work.

------------------

Forum Index > Electronics		Next oldest topic :: Next newest topic

All times are GMT - 8 Hours	Jump to: