Consumer electronics "war stories"

Sort of sad somebody messed up a carousel. The cartridge based changers were infuriating.

Anything that requires extensive soldering and screwing around with that medical type tape to open up, like portable tape/CD players and now cameras suck too.

Reply to
Cydrome Leader
Loading thread data ...

arts from the stock room, onto a protoboard like breadboard.

ime , no problems.

s that they could not get to work.

c cap with a 10 printed on it between the two stages.

=

bleshooting a new box design. The picture was black and white and they co uld not figure out why there was no color. Looking into the box I saw a c rystal marked 3.579545.

s your problem and walked away. :-)

Lawsuits over the years have shown their companies to be just as liable occ asionally, too (like with Nomura, Sumitomo, Mitsubishi, Toyota, Honda, Mazd a, Hitachi and others...) I think that things are done less purposely with regards to American markets, though.

Reply to
mogulah

At college I had a lab with the same thing. We designed simple circuits and built them and took measurments on them. There were boxes of parts that were suspose to be the same parts. Some of the parts were either bad or out of spec. Not on purpose, they just got that way over the years. Me and a person I was with usually could locate the bad parts and get our project going first. Got to be a joke that the ones that got theirs to work had the lucky box with all good parts for that design.

Reply to
Ralph Mowery

It's definitely not consumer electronics, but here goes.

I used to work at a company that made flight simulators. My official job was writing code to make the navigation instruments display correctly, according to the simulated position of the aircraft, which frequencies the radios were tuned to, the position of circuit breakers, etc.

Some of the laws of nature at that particular job included:

1) Techs string wire.

2) Programmers write programs.

3) Techs never make mistakes when stringing wires.

4) No programmer could ever possibly understand the intricacies of stringing wires. Therefore, any complaints by programmers about the wiring can usually be disregarded.

5) Any persistent complaints about the wiring can be remedied by telling the programmers to work around it in the code.

Never being one to obey the laws of nature, I brought in my own meter and checked things out when the sim didn't seem to work right. My boss knew that I had something of a clue; I had been giving "how to read a schematic" lessons to a few of my cow-orkers (in software) who were also new hires or just wanted to know how. He basically told me that he didn't mind me doing my own checks as long as I was careful with the sim hardware, and that I should expect the techs to get mad at me if they ever saw me doing it.

I couldn't get the right DME indicator to light up in one sim. (It was a box in the instrument panel with three 7-segment displays and a couple of buttons. It was supposed to display how far the airplane was from a particular radio station.) After a few rudimentary checks of my code, I wrote up a "right DME inop" trouble ticket. The technician wrote back, "Wiring checks to print, DME sent for repair". Sure enough, there was a hole in the panel. When it was again filled, I tried again...no joy. I swapped the left and right indicators - hmm, the problem stayed with the socket instead of following the indicator. I broke out the wiring diagram and my own personal multimeter and started chasing around behind the panel - no wires or pins for power on the right DME socket. (It was something like a DB25 or DB37, with individual "crimp and poke" contacts.) I dutifully re-opened the ticket, and the tech dutifully wrote "wiring checks to print" and closed it again.

About this time, the sim was shipped to the site, even though it was broken. The standard process was to completely build the sim at the factory, test it out, ship it to the site, certify it, and put it in revenue service. For sites that were far away from the factory, this practice was generally followed, because it was expensive to ship people to site to finish working on the sim. However, the site that was closest to the factory (~3 hr drive) was notorious for the following:

Factory: We will have the sim done on $DATE.

Site: No no no! We've already sold time on that sim to customers on $DATE-30days and we can't reschedule! We *must* have it here sooner!

Factory: Why did you do that? The sim will not be done on $DATE-30days. It will be broken and unusable for training.

Site: We don't care.

Factory: If we ship it at that time, it will suck.

Site: We don't care!!! Ship it shipit SHIPIT!!1!

Factory: OK.

(time passes) Site: Well you got the sim to us on time but it sucks! Everything's broken and we can't put the customers on it! Fix it fixit FIXIT!!1!

Factory:

So I get to the site and the site manager is bugging me about the right DME indicator. I walk in to the site maintenance shop and tell the techs there I need some connector pins, the crimper tool, some wire, and a bench power supply. They are extremely wary of this as they have experienced "programmers with screwdrivers" before, but they give me the requested items and follow me into the sim, probably in hopes that my body will shake and jerk in interesting ways as I electrocute myself. I put the pins on the wires and put them in the (still vacant) slots on the right DME connector. Wires run out under the panel to the power supply, which temporarily gets the co-pilot's seat. Fire up the sim, hit the power supply, and whaddayaknow - DME love for all. I disappointed the techs, but the site manager was very happy.

I pointed out the relevant page in the wiring diagram book, so the site techs could get it wired in correctly. I figured they had a lot more experience than I did in fixing factory screwups. They were satisfied with this, and I got to go home.

Matt Roberds

Reply to
mroberds

Why does that sound so horribly, horribly familiar?

Brrr. Memories I'd rather forget.

--
 (\_/)   Tyson Fury: #homophobe #bigot #throwback #missinglink 
(='.'=)              #neanderthal #misogynist #redneck #dickhead 
 Click to see the full signature
Reply to
Mike Tomlinson

Wow. Lots of times I'm too lazy to read to the end - but this was great! Loved it!

mz

Reply to
Mark Zacharias

How about another slant on "War Stories" ?

The least competent, most alcoholic, etc tech person you ever had to work with or follow after they got fired?

When I was first starting in consumer electronics repair, (my first job in the business, actually) I was hired to replace "Karl". He had a resume - in the '70s a well respected shop paid for him to come over from Germany.

By my time however - apparently a broken down alcoholic.

He would "repair" tons of stuff, bill out huge amounts (paid on commision) then abscond when the re-do's became too much.

I was charged with fixing his re-do's and generally cleaning up the chaos he had left behind.

Next job - another shop. They had just fired the SAME GUY. Same situation. Re-do's coming in one after another. Angry customers. Piles of screws and small hardware in a pile on one corner of the bench. Dis-assembled units all over the place, and I mean ALL OVER. A Teac A-4010 in about four different parts of the shop. No pressure... I'd never even seen one before.

Next job - SAME DEAL. By now I was getting pretty good at reverse-engineering other peoples screw-ups, but - really?

A couple examples:

Auto-reverse car cassette deck. He didn't have the correct drive belt, so he had SUPER-GLUED the ends of the old belt together. Played about 2 mnutes, if that.

A Marantz 1060 integrated amp (re-do) with a blown channel. He had substuted a driver transistors with a similar package item. Unfortunately, the part he used was a VOLTAGE REGULATOR IC and not even a transistor.

He had his "groupies" though. Some customers followed him from one job to the next.

About 1987 Bang & Olufsen in Chicago contacted our shop for a reference on this guy.

We were rolling on the floor!

Gave him an absolutely GLOWING reference. We could think of nothing funnier than the prospect of this guy working for B&O.

(no he didn't get hired)

Good times.

Reply to
Mark Zacharias

Mark,

In the early 70s there was a company that sold strips of rubber of various sizes with a razor blade, jig and a tube of super glue that was supposed to be used to make belts for consumer electronics equipment. I had never seen super glue before so I tried it. Once.

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
Chuck

If you enjoy reading such stories, there are quite a few in the "Made by Monkeys" section of various trade magazines. These highlight quality control and design failures: If you're planning on designing the next big thing in consumer products, or are wondering why some things just can't be repaired, these columns (blogs) should be mandatory reading.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

s he

Reply to
jurb6006

Those are still sold and work reasonably well:

The trick is to cut the o-ring or whatever at an angle. That does three things:

- It increases the surface contact area so that the glue has a better grip.

- It converts some of the stresses from tension to shear, where cyanoacrylate adhesives are stronger.

- When used as a compression seal, crushing the glue joint does NOT crack the rather brittle glue joint.

The only gotcha I've run into is dealing with tight turns such as very small diameter drive pulleys. Cutting the o-ring at a large angle causes the glue joint to be longer. Too long, and it will crack if wrapped around a small pully. Just size the angle for covering no more than about 60 degrees around the pully, and I think it should be ok.

Note: Super glue doesn't work if there's little contact area, so splicing thin and flat belts doesn't work. I've had some luck using contact cement with these, but not reliably.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

I was teaching one such lab, with bar magnet/coil experiments, and had an inspiration. I got some iron filings and sheets of paper, and had the students lay the paper over their bar magnet and sprinkle the filings over it.

There were a dozen bar magnets in the 'materials' box, and half of 'em had odd fields. One had five identifiable poles. Using only the dipole-type bar magnets, the class got better compliance than usual with the expected behavior of poles and coils in motion.

Reply to
whit3rd

I was called in unofficially to have a look at an X-ray machine in a university crystallography lab, there was an intermittent fault which shut it down after a few seconds of operation. The running time was getting shorter and shorter and the manufacturers had given up on finding the fault.

On the way there, I mentally ran through what I could remember about X-ray machines (apart from the obvious dangers) and realised that most of what I knew had come from reading my father's hand-written course notes in the 1950s; they dated from when he was trained as an army radiographerat the outbreak of WWII. I knew what an X-ray tube looked like as a symbol, but what did one look like in reality?

On being introduced to the faulty machine, I glanced around the room and saw a number of copper-and-glass objects on a shelf - and concluded that they must be spare tubes. Luckily, the manufacturers had furnished a full set of circuit diagrams and the lab had managed not to loose them, so I knew what I was dealing with, even if I didn't initially know how most of it worked. The circuits were all discrete components with intermixed transistor, diode and relay logic.

By the end of the first day, I had gained a fair idea of how the power supplies and safety circuits worked and had been instructed in the necessary safety drill by the technician, so I was able to fire the machine up and watch what happened. The fault showed up, but it all happened so quickly that I wan't able to spot what was going on.

Luckily, on the morning of the second day, I happened to spot the tube current meter flicker downwards and the voltmeter kick upwards just as the fault occurred. Careful monitoring of the primary of the mains transformer showed unstable mains voltage, which the control loop had been over-compensating and then tripping out on over-voltage.

The cause was a burnt contact in the main contactor, so I stripped it down and sandpapered the contacts, much to the amusement of the staff.

Fault cured - machine saved from the scrapheap.

--
~ Adrian Tuddenham ~ 
(Remove the ".invalid"s and add ".co.uk" to reply) 
 Click to see the full signature
Reply to
Adrian Tuddenham

Been there done that. Good site. The name seems to imply it is a comedy site but it is not. I hope people are not too disappointed.

Reply to
jurb6006

How to operate a tube caddy.

During the early 1970's, I ran a 2way radio service shop. Running it was an accident because no sooner was I hired, everyone else either quit or promoted themselves sideways, leaving me as the sole employee and later as part owner. This condition lasted for a few months of serious overwork, until I was finally able to hire an employee. The shop:

A major part of the operation was maintaining a mountain top radio repeater site: When something went wrong, the ritual was to load up the company pickup truck with everything imaginable, drive 12 surface miles to the base of the mountain, drive 16 miles up a windy dirt road to the top, fix something, and repeat the ritual in reverse. 6 to 8 hrs was the typical round trip time.

Replacement tubes were carried in a tube caddy. For those who have never seen a tube caddy, the individual tubes were stored in small personalized cardboard boxes, inside a wooden carrying case like these[1]: My new employee would identify the dead or dying tube, extract a new tube from the tube caddy, insert the new tube into the radio, insert the dead tube into the cardboard box, and replace the box into its place in the tube caddy. Do you see a problem here?

That worked for about 3 months when I discovered that my new employee couldn't seem to fix anything. Various theories were offered, but nothing worked. One day, my employee had a cold, so I had to do the drive up the mountain. Oddly, I also couldn't seem to fix anything either. Then, I noticed that some of the tubes I was using as replacements were obviously ancient and really didn't belong in service. Hmmm...

Upon returning to civilization, I tested almost all the tubes in the tube caddy, and found that well over half were dead[2]. When I mentioned it to my new employee, it took him a while to understand what had happened. I suspected that the thought the tube caddy somehow rejuvenated dead tubes. I resisted the temptation to thrash him about the head, because he was bigger than me. I later found him a job with a competitor.

[1] I still have a tube caddy full of tubes awaiting the demise of semiconductors. [2] Tube testing algorithm. If the tube tester says it's bad, it's bad. If the tube tester says it's good, it still might be bad.
--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

tion and things like that pertaining to linearity. It is of course bus addr essed. The one line, data or clock, doesn't matter, was clamping the signal down to like 2 volts. Leaky. That meant the data from the EPROM was not re ad when the unit got initialized so it did not know which sound system it h ad and never upped that.

I had a similar fault in an Universum CRT TV set. It came with an all white screen sympthom that I traced to the RGB decoder matrix IC holding its RGB outputs to 0V. Replaced the IC, no joy. It was I2C controlled and I even b uilt a simple interface to send commands from the PC, it seemed to ACK fine and all looked right. If not initialized at turn on the raster remained bl ack but as soon as the outputs were activated in any way it would go full w hite. Finally found the problem: it received a sandcastle signal from the sync pr ocessor IC that had the top pulses too low so the RGB IC was not locking on them. The picture improvement IC also connected to this signal had a leaky input and was eating half the signal.

Reply to
Jeroni Paul

I have a SABA music system (radio + cassette + turntable + audio in/out plu gs + remote control), time ago I had it connected to a desktop PC to play m usic from the PC. One day the printer attached to the same PC was taken out for repair due to clogged heads.

The next day I found the SABA turned on with the MUTE activated (the radio was selected so the FM display etc was all lit, but no sound). Since I neve r used to use the MUTE button and I was the only one at home to use that th ing I was quite surprised. I unmuted it and turned it off, all appeared to work correctly. The same day in the evening the same again, that made it ob vious it was not me. In the next few days the same kept happening at random times but never when I was there, and because it would turn on with the mu te set I could not hear when it happened.

Finally one day it was off, I went to the kitchen and when I came back it w as on and muted again, so I guessed a relation had to exist. Turned it off and went to the kitchen again - no joy. Repeated a few times and surprise - again on and muted. Some more experiments revealed that switching off the kitchen light sometimes would cause the SABA to turn on and activate the mu te at the same time.

The kitchen light consists of two 36W fluorescent tubes, apparently the ind uctive kick at turn off found its way into the SABA digital controls. They were two rooms apart, so not exactly next to the kitchen switch or lights. The issue did not reoccur after I plugged the printer back.

Reply to
Jeroni Paul

se

he

t

I was in a block of flats to look at a curious problem in their terrestrial TV reception. Whenever the communal stairs/hall lights were on, all TV set s in the block lost signal. The stairs lights were controlled by a timer relay that kept the lights on for a few minutes after any push button was hit, so every time someone ente red and hit the light the neighbors TV signal went out for a few minutes.

I started checking the terrestrial antenna head amplifier and found it lost mains power whenever stairs lights were on. I also observed that four or f ive lights in the stairs did not illuminate and some push buttons didn't ac tivate the lights. That one had me thinking for a while and I drew this dia gram to understand what could possibly be going on there:

Head amplifier wall plug N L | | | | |--light bulb--| | |--light bulb--| | |--light bulb--| | X | | |--light bulb--| | |--light bulb--| | |--light bulb--| | | | | | relay | | | | | +-----+ | | N L Mains supply

That turned out an accurate representation of the problem, I found "X" was a badly burned electrical terminal inside a connection box. With relay open , the bulbs happened to be in series in the neutral going to the head ampli fier and because its small current draw it had enough voltage to work. With relay closed, only light bulbs before the break illuminated and the head a mplifier got the L pole in the N wire through the non-working bulbs, so no voltage to work.

Reply to
Jeroni Paul

Sometimes a small victory makes you feel just as good as a big one.

Picked up a somewhat non-functional Micronta 22-220A multimeter. A little rough but the FET meter circuit worked - voltage readings weren't too far off and the zero control did it's job so I knew all that stuff was OK.

But the resistance function acted as though there was a 4 ohm or so resistor across the leads all the time, and the battery was draining at about 100 mA in Ohms function even with no leads attached.

Of course the 9.1 ohm Rx1 resistor was bad, but replacing it did NOT change the symptom.

After finding a schematic (not many out there...) I did find a component labelled "SA1" shorted at 4 ohms or so. The item resembled an MOV and I can only assume SA stood for spark or surge arrestor.

Removing it mostly fixed the ohms function, and I decided a couple of back-to-back 25 volt zeners would offer enough protection to satisfy my needs.

Still the ohms zeroing was erratic. Cleaning the function / range switch and ohms pot til I was blue in the face did not resolve the problem. It was kinda usable but it kept bugging me.

I tried putting a current meter in series with the test leads but couldn't really get a usable correlation between pushing, poking wiggling the function switch etc and the action of the meter which might zero fine, then show up to several ohms even seconds later with probes shorted.

It occurred to me that I could put a resistor (say 4.7 ohms on this range) across the probes and put a 'scope across that resistor to better see what the DC voltage there was doing.

Oh, yeah. the voltage as viewed on the 'scope varied wildly and looked "noisy" as the funtion switch was wiggled or tapped.

But I had cleaned that switch umpteen times.

Well, there was another switch - a leaf switch, going to the negative battery terminal hiding under the front face and also actuated by the function knob.

A quick cleaning of those contacts and the meter works like new.

A small victory to be sure, but made me feel as good as a big one.

Mark Z.

Reply to
Mark Zacharias

Ha!

Was noise filtering on the always on printer was somehow supressing the interference?

Reply to
Cydrome Leader

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.