DIYHiFi.org

Talking about TO-126 transistors, and how those of us in the USA shun stuff from Europe, gave me an idea.

I know that some of you out there are Googling where I used to work. (I still don't know why you want to know.) But that gave me some ideas about some of the more moronic things that happened at one of those places.

The reason I say "The Cage" is because at one time I really did work in a cage. No, not because of my repellent personality, although that may have been a consideration. "The Cage" was an area that at one time housed all of the spare parts for the entire company. It then held all of the test equipment. Then it held some other spares of some sorts. They needed a place for the Failure Analysis Dept., and since it was a one-man department...............and considering who that department was..............yeah, who else would you stick in there.

So, I got stuck in The Cage. Seems fitting in retrofit.

On to the war stories.

Jocko

_________________
"Because, I hate you guys. I hate you guys so very, very much.

Yours,
General Cartman Lee"

Hard as this may seem, as recent as the late 70s, I was not allowed to use plastic transistors. A lot of our engineers came from Collins Radio, and they had a military-gear view of the world.

So, not only did we never see "fer'n" transistors, the ones we used had to have that metal-can military feel. It wasn't just where I worked. I once went to some sort of local nerd gathering, and one engineer loudly complained that because we had to make stuff to look like it was Mil Spec, that we were stuck using "antiquated crap like TO-3 transistors". This guy really hated them. I did, but had not as of this time seen any of those plastic Japanese output devices.

So.........back to The Cage........

We had this 70 MHz IF amp that would crap out when it got hot. (Which it did, and it was in an 8' high rack. Full of gear.) We figured out (I had help by this point) that it was a certain transistor. It was like a 2N918, only it had a 900 MHz vs 600 MHz f_t. (Maybe it was 1200 MHz vs 900 MHz. Too long ago to remember.)

Well, couldn't find those parts. TI stopped making the metal cans, and only National still made them. Cost $5. If you could find them. No one stocked them.

So, kvetched to the boss. His choice:

Plastic transistor, at $2.

"NO! None of that plastic crap. Understand?"

Uh...........ok...............how 'bout 2N918. Similar, cheap, and easy to get in metal cans.

"If Collins wanted a '918 in there, then they would have designed in a '918."

So, what to do? Easy, boss had me turn over my request for the metal can ones over to our Purchasing Dept.

In Wash., DC. You know, up north. People with attitudes worse than mine.

Wait a few months, and something showed up. You could see the the Nat Semi logo had been rubbed out, along with p/n. But not enough to know what it was.

Uh, oh............this portends poorly............

To make it worse, there was now a cheesy rubber stamped imprint on the top. P/n only. No indication who "made" it.

Things are looking worse.

Could it get any worse?

YES!

Every damn part in the bag had opens, shorts, or both. IOW, some were not even diodes, let alone a specialised RF device.

Time to call the schmuck who bought them. But not until after I clued in the VP I reported to.

"Hey, bub.............these transistors that you bought for me...."

"Yeah, glad you got them. How did they work out?"

"Uh, they don't."

"What do you mean? I got those for under $2/ea. The place you sent me to wanted over $5/ea. for them. How can you justify spending that much when I found them for much less?"

"The reason why you got them for less is because they aren't transistors.(You gotta be firm with those folks up there.) They don't work. They are unusable. I can't stick them in my ham radio. I can't even stick them in my "hi fi". They don't work. Period. I can't use them. You bought some fallouts from National that one of your East Coast buddies that runs some schlock house that sells re-branded crap. (Nice tact, eh?) You can even see where they tried to remove the National logo. So, what are you going to do about it?"

At this point he becomes defensive, and tries to stall for time. (Wouldn't you? With me on the far end, giving you grief.)

"Well, I don't care what you do with these damn things. You better find a way to get me the right ones before my boss wants to know why I still have a stack of IF amps that don't work. You do realise that my boss is Sully, don't you?" (He wasn't my direct boss, but he was the local VP, and I had a direct line to him when I needed it.)

"So, Jocko, if you had that sort of access, why didn't you bypass your boss and get the 2N918s or the plastic version? Wouldn't that have fixed your problem?"

No. Because he also came from Collins. And would ask my idiot boss why I was all wound up so tight. Nope, could not have done that. But get him to light a fire under Purchasing up north.................yep, no problema.

"So, how did the issue resolve?"

Well, funny that you should ask.

We kvetched to Collins about a lot of other stuff. Rockwell owned Collins at this point, and their idea of quality and innovation was to put a new coat of paint on increasingly thin sheet metal chassis.

So, after who knows how many months, they showed up for a dog and pony show. Trumpeting all their solutions to problems that should have never made it past Manufacturing QC.

And their solution.....................drum roll, please............

Use 2N918s in place of the funky ones.

Which is what we said months earlier.

And the problem:

Seems there was metal migration in the silicon. Not much, but enough to make all those 70 MHz IF amps a bit wonky when hot.

Jocko

_________________
"Because, I hate you guys. I hate you guys so very, very much.

Yours,
General Cartman Lee"

La cage aux failure? Wasn't that a movie?

One of the places I used to work had an "energetic" CEO. There was a smart engineer there who somehow had made a mistake on a schematic that got carried over to pc board layout which got carried over to the prototype pc board which caused a problem. It might have been a bypass cap that didn't get connected to ground, now that I think of it. There was a short "spirited" discussion between the CEO and this engineer, which was much shorter than we all expected. The CEO walked off. He came back in about five minutes with the guy who did building maintenance. The maintenance guy then proceeded to remove all the cubical walls around the engineer's desk. And then he moved the desk into the middle of far end of the room. Kind of like Siberia. The engineer was relegated there for six months. The only good news is that he wasn't required to wear a dunce cap.

Then at another place there was a VP of engineering who really wasn't quite the right guy for the job. You'll see why in a minute. But, he made up for it by being drunken, crooked, and arrogant. (Famous quote from him: "I don't feel creative this afternoon; I'm going home...") We had a problem with crosstalk between modules in a chassis. I pointed out that the metal work, although very expensive and nicely fabricated, was machined and that the covers were only held in place by a screw in each corner. You could easily slip a manilla folder between the edges where the thin sheet metal cover bowed up when the screws were tightened down. The guy got mad at me, a condition I was used to. He argued, and I tried to explain about slots and all that. He wasn't very receptive to my ideas. Eventually he got so mad that he stomped out of the room with one of the modules. Great, I thought; problem solved for now. A couple minutes he stomped back in with the module and a digital multi-meter. "See? What the %@** are you talking about?" He was holding one lead of the Ohmmeter on the module and one on the cover. "See? It reads zero Ohms. Look at that! ^%$# stupid @$$ kid." Now, what can you say to *that*?

I guess he was sensitive because he conceived of that very chassis design. One of the unique features was that the heat sink for the rf amplifier and voltage regulators was on the front panel so that the cool air from the front of the racks in the CO's would reach the heat sink. But, since he thought that might obstruct the view of the LEDs from the side and because it looked cool, the heat sink was recessed to be flush with the surface of the front panel. Think about that one.

This whole chassis was machined, as I said. It was designed by a contractor whose expertise was tooling design. He freely admitted that chassis design was *way* outside his experience. But, he was a great tooling designer and a good guy to boot. He was just doing exactly what he was being told to do by the guys paying the bills. Anyway, the guys paying the bills thought that Ma Bell, as pretty much the only big long distance guys were called, wanted something really "solid" so they'd know that they were paying for quality. They required the designer to design to his usual tight tolerances. So, he did. The modules fit so tightly in the chassis that they would often get stuck, and we had to disassemble the chassis to get the modules out. I finally came up with a solution for the problem. I went to the grocery store and bought a spray can of Pam (cooking lube). Before installing the module, I'd spray the contacting surfaces. When we wanted to remove a module, we'd undo the retaining screws - like they were needed - and I'd get someone to stand in front of the chassis. I'd walk around the back, and whack the back of the module with a mallet. The thing would pop out he front and my "assistant" would catch it in flight. Just like something out of that Three Stooges episode where they were plumbers. It even made the same "foop" sound. I got into trouble for that one, too.

Next time: The VP of Engineering who knew all about rf modems.

Mount the heat sink fins flush with the front!!!!!!!!!!! Hey, had to look cool. There is a future for him in High-end.

Any openings at that place down the street? They could use him.

Jocko

_________________
"Because, I hate you guys. I hate you guys so very, very much.

Yours,
General Cartman Lee"

It turned out that this guy and the marketing guy were in cahoots. They worked it out between themselves to get our company to pay for the development of this system (a 70 MHz thingus, BTW), and that later on they would offer to buy the product design from the company when the sales didn't meet the inflated expectations that the marketing guy had created and had screwed up the sales on.

Amazingly, they not only pulled this off, they got one of the executives of our parent company to be the prime investor in their new venture. You can't make this stuff up.

Here was the catch. I was the only one who understood any of the electronic and optical design. And they hated me because I'd caught on to their scheme. So they were forced to contract with their old company to get me to explain how it worked. I went down to give this tutorial and noticed that the marketing guy was no longer with the company. He had already been caught scamming his expense account in a big way. After that visit, I lost track of them.

The group really may have gone on to greener pastures in high end audio.

Most of the local high end audio firms have either moved away or farmed all their products out to some place in China.

Before going on to the modem story, let me add one more about the fabled heat sink designer.

At the time when he was king of the engineering world, the IBM PC XT had just come out. I think the ONE we had wielded a 5 meg hard drive.

Around that period someone had the bright idea that the company should also be playing in the security electronics business. Why not? Video and other analog over fiber, alarm systems - it's all the same thing, right? So the engineering VP got the company to hire one of his buddies to run that group. You never can have too many managers around.

These two guys sat on that computer all day every day. Some of the other people actually needed to do some work on the computer for an hour or so combined every day or two, but they couldn't shove these clowns off. One day, when they were out to lunch (I mean physically out to lunch, not the way they normally were) one of the poor schmucks who had to get some work done came to me to see if I could help her get access to the computer. Huh? These guys had placed a security system on the computer that you couldn't get past.

About a week later, another one of the poor schmucks who needed to use the computer less often, came to me and asked why he could only boot the computer using the floppy drive. It seemed that he had stumbled across the obvious idea of starting the computer not from the floppy and running his software program from floppy as well. I found that I could get past the security system by booting from the floppy and then accessing the hard drive. I located the security system and found I could defeat it once I was into the computer. It was probably the same system used at the CIA for all I know. I think it was called Watch Dog security or some clever name like that.

So, what I did was to write a simple program in IBM basic that ran at boot up from the hard drive. It displayed the same screen as the Watch Dog thing, except I called it Mad Dog security. You just had to type in the password, and you got past the screen and that was it. If you didn't type the right password, the computer screamed with a siren sound and locked up until you re-booted. Then you came upon the same start up screen. I told pretty much everybody, except those guys, what the password was so they could use the computer.

So... These guys came back from their usual 18 martini lunch and one sat down to hog the computer. He started the machine, read the screen, and obviously didn't digest the message. He typed something in and you could hear the siren around the entire floor. Of course, everybody just knew this was going to happen, so they were all waiting for it. The guy must've tried this maneuver at least five times before he got his buddy to try it. He was good for a couple sirens before he just stared at the computer.

Eventually somebody ratted me out and the general manager made me take the Mad Dog security system off. There was no way to delay that trouble comin' everyday.

Finally, the modem story.

At the company with the energetic CEO, the VP of engineering finally had enough and got a different job. Too bad, because he was really a good guy. I guess the foot prints across his forehead just got to him after a while. The company hired a new VP, fresh out of some telephony supplier down South. In fact, this fellow was from outside Dallas, Texas as he reminded us at least 4 times a day.

One of the projects that I had been working on part time was a simple rf modem that was used to get status and health telemetry from the remote CATV nodes. It wasn't anything special - the speed was something like 9600 bps.

This guy dragged me into his office to tell me what he thought about this project. He went on for at least half an hour ranting and raving about modems and how they were very hard to deal with.

I spent about five minutes explaining to him what I was trying to do, and what the challenges were.

He responded with another half hour of whatever he was saying.

Finally, I got what he was on about when he uttered this: "We had nothing but problems with those AT commands. On some modems they worked differently than others. I don't want to go through that again."

The problem solved itself in a couple of weeks. This guy was running some sort of meeting when the CEO came in. The CEO must've said something, and the VP told him that it was "his" meeting. A few minutes later, the CEO spoke up again. The VP said something like, "This is my meeting. If you're going to talk, I'm going to leave." So he left. And he went home. He came back the following Monday thinking all would be good - in fact, he kind of swaggered in suggesting by his stance that "I showed him, huh?" He found that his personal stuff from his office was packed in a cardboard box and there was a note sitting on top to see the HR person to get his final paycheck.

Anyone that I might know? Seriously.

Jocko

_________________
"Because, I hate you guys. I hate you guys so very, very much.

Yours,
General Cartman Lee"

Maybe. First name was Al.

Keep it up guys, this is much more interesting than anything I have to report from my day-job.

..though I once witnessed a contractor carefully building the fold in a drawing in fair-face blockwork (after a 1:200 layout had been zapped down to A3-size by someone careless in the site office, and handed to some other clown out in the mud...).

Oh , and preiodically one gets to trip over steels about knee-height across stair landings when someone at the fabrication yard forgot to check for latest layouts

See, I told you:



That's me... behind the camera. Specified fix was (1) always check for info version control, and (2)apply the gas axe.

Great picture!

^- hilarious picture.

My workplace is relatively drama-and-idiot free, thankfully. But we've had a couple of moments.

When I first started working here in '02, the electronics assembly part of the company was still stuck back in 1980. Everything we built was through-hole, stuffed by hand and run through a wave solder machine. We still had fairly innovative products at the time despite the manufacturing process we had, but you can thank the PLCC socket for that.

But we had to keep up on the technology/feature thing going or get our ass kicked in the market, so in R&D we proposed a new series of equipment with DSP's, FPGA's, LCD screens, etc. Which required SMT assembly.

Everyone was gung-ho to do this product line except for the production manager. Who freaked out. We weren't set up for surface mount, they didn't know how our employees going to place those tiny parts on a board by hand, customers weren't able to fix anything if it breaks, bad design, throw it out, blah blah blah. "Can't you do the same thing in through-hole?"... Yes, because you can buy 500MHz DSP chips, 133MHz SDRAM chips, 1M-gate FPGA's, etc... in 40 pin DIP's that run off 5 volts.

They didn't know a single thing about SMT assembly, and flat out refused to learn anything. And they got increasingly paranoid, singling out R&D people individually and expressing their opinion. "This SMT stuff is really bad news. Do you know what tin whiskers are? Do you know you can't even see under a BGA chip to see if it's soldered correctly?"

Thankfully, they didn't last long after that.

Guys,

I interviewed at Keithly when I was a senior and the guy said that he would offer me a job right away but his boss would have nothing to do with another ONU graduate.

Seems a few years before an ONU grad came in designed some piece of test equipment that blew the industry away with it's specifications. The problem was their was a critical opamp that the prototype happen to use that they had to screen with only a 5% meeting the minimum.

Oh well would have had to move to Cleveland so maybe it was a blessing.

Later,
Gordon

_________________
Gordon
Chief Scientist
Wavelength Audio

Yes, there really are worse places to be, and you managed to avoid one of them.

So, blame the hot-shot graduate, and not the lousy op-amp manufacturer that misleads folks with their data sheets........no, they would ever do that!

Jocko

_________________
"Because, I hate you guys. I hate you guys so very, very much.

Yours,
General Cartman Lee"