Consumer electronics "war stories"

About 3 weeks ago, I was blessed by the addition of a Samsung Syncmaster 243T 24" 1920x1200 LCD monitor to my repair backlog. It had been sitting around the donors office for a year or two, so nobody could recall why it was retired. I plug it in and it appears that everything is working. I have two similar LCD monitors at home for running my flight simulator. A third monitor would make a start on a wrap around cockpit window view. (actually 4 is about right).

So, I take home the monitor, being careful not to bash in the screen like I did the last monitor I took home by planting the groceries dead center in the middle of the panel. It arrive safely, I plug it in, and nothing works. No power, no pilot light, no messages, no nothing.

I'm not exactly equipped at home to fix monitors, so I drag it back to the office where it sat around for a few days. I plug it, and everything works normally. I check for intermittents by beating on the monitor, but nothing happens.

At this point, a sane and rational person would tear the monitor apart, look for problems, probe around with a volts-guesser, determine the culprit, and fix it. Nope. I'm out of bench space and have no room to work on a big monitor. So, I drag the monitor home again, and once again, it's dead on arrival. So, I drag it back to the office for the 3rd time, where it once again works perfectly.

This would be a good time to guess the cause (although I haven't really revealed enough info to make a proper deduction).

I still haven't ripped it apart to see what's going on, but I do have a good guess what's wrong. It probably has the usual bulging capacitor problem in the power supply. I keep the office at 72F (22C) to keep the customers happy. At home, I prefer something around 65F (18C). The workbench, where I do my testing is not very well heated, and is probably colder. Outside temperature is now about 43F (6C).

Bulging electrolytics are detected by measuring the ESR, which increases as they leak. Heating the caps lowers the ESR back down. Cooling the caps raises the ESR back up. Incidentally, this is why some devices run merrily when warm, but won't turn on when allowed to cool off. The Samsung monitor is likely teetering between working when warm, and not running when cold.

I'll disclose what was really wrong after I fix it, probably next year.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann
Loading thread data ...

One of the advantages of my office location is that it's very centrally located. Within about 500ft is the intersection of 3 freeways, the main drag into Santa Cruz city, and smaller roads leading directly to nearby cities. All roads lead to my office, which is both a benefit and a problem. Besides making it easier for my customers to drop in, it also attracts a motley assortment of people that just happen to be driving by my dead end street, and just happen to in the mood for trashing my day with inane conversation.

One memorable day, I had 4 of these visitors perched on benches and chairs (I only have two chairs in the office to make sure they're not very comfortable). I was working on replacing some caps in an ATX power supply. Of course, I wasn't paying attention and accidentally soldered the caps in backwards. With the cover off, I plugged in the power supply, and continued the discussion with my visitors. Suddenly, several of the caps decided this would be a excellent time to explode and launch oily confetti all over the office. Everyone, except me, dived for cover under or behind tables and boxes. I just continued talking as if everything was perfectly normal and nothing unusual had happened. The visitors soon made a rather hasty exit. Oddly, they must have told their friends, because my office was free of unwanted visitors for at least a week or two.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

Going to look bad when it is a bad power cable or socket.

Could be the capacitors. A number of years ago when the bad capacitors were in many computers a friend had a computer in his basement that sometimes came on and sometimes not. He left the cover off of it and would put a light bulb next to the computer to heat it up. The computer wold come on and work fine unless he shut it off , then he had to heat it up again with the light bulb.

Reply to
Ralph Mowery

We had a couple of old Fujitsu "Eagle" disk drives - 500MB or so - inherited when we started a company in 1990. These were about 25-30kg, and 500x350x700mm in size - so you needed two strong sets of hands to get them on and off the rack mountings (they were on slides).

Anyhow, as they got older, the spindle bearings became sticky, so they wouldn't spin up after a power fail that was long enough for them to cool down. We used to get them off the racks, with one bloke at each end, then power them up and give a sudden lateral rotation to break the stiction of the bearings. Quite a risky business, coordinating two blokes to do that suddenly enough without dropping the drive, but it worked a number of times before we made enough money to afford to replace them.

We started a software company with 25 initial employees (staff buy-out) and a grand total of 2.2GB of storage - in the entire company. Imagine that!

Clifford Heath.

Reply to
Clifford Heath

Nope. Both passed the wiggle test.

I'm getting that now on my home desktop computah. I don't bother to heat the house much at night. When I wake up, it's about 45F (7.2C) inside the house. When I turn on my desktop computah, the fan makes some odd noises but eventually quiets down. The hard disk seems to boot normally, but usually some programs add oddly or crash. I reboot and they then act normally. It's probably read errors on the hard disk, but SMART shows nothing interesting. At this time, I boot to the BIOS screen, wait about 10-15 mins, and then boot normally.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

Take the fan out and refit it, rotated 90 degrees.

The bushings wear the holes elliptical, and the rotation changes the vibration modes.

Reply to
Clifford Heath

If I got out of bed and it was 45 F, I would be looking into the heating system first. I don't function when it is that cold.

Sounds like you may want to look into a solid state drive for the computer so it will start up cold.

Reply to
Ralph Mowery

If I got out of bed and it was 45 F, I would be looking into the heating system first. I don't function when it is that cold.

Sounds like you may want to look into a solid state drive for the computer so it will start up cold. "

If I got out of bed and it was that cold, I would be looking into some INSULATION first! Something 'Murricans seem to be averse to, even after decades of evidence in favor of it.

Reply to
thekmanrocks

I've woken up to -40F, during survival training.

I've installed more fiberglass batts than I care to think about, starting in the early '60s.

Reply to
Michael A. Terrell
8:29 PMMichael Terrell wrote: "- show quoted text - I've woken up to -40F, during survival training.

I've installed more fiberglass batts than I care to think about, starting in the early '60s. "

Then you must almost never turn your heat on.

'Murricans, thinking they all tough by never using heat in the winter...

Reply to
thekmanrocks

Below about 45F, I have problems working, especially when I'm walking around the house in pajamas and no shoes. I do it often and it doesn't bother me much. However, that requires that I keep moving, which is often difficult in front to the computah. So, I fire up the wood burner and run the temperature up to 60 to 70F.

An SSD is in the works. The problem is that every time I buy one for myself, some customer arrives and needs it more. Since prices are dropping, I don't mind delaying my SSD upgrade. However, for the present situation, I'm looking into an adjustable cat or dog warming electric blanket. I had a mysterious data corruption problem about a month ago, when it began to become cold, that was probably due to the startup problem. I recovered fairly easily, but it burned too much time.

This is not a conventional house: Insulation is possible, but there are other priorities that come first. Eventually, I'll need to build up the roof, which is where insulation will do the most good. One of my neighbors did that to their house that was similarly built, and it worked very well.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

No, but then I live in Florida. :)

I have a 1000W electric heater, for the rare times that it is in the

30s for several days in a row.
Reply to
Michael A. Terrell

Greetings Jeff, Your house makes me nostalgic. I grew up on the other side and at the base of the Santa Cruz Mountains in Saratoga. Hitchhiked over to Santa Cruz and then either north or south, depending on my mood. North for the beaches, south for Capitola. Or right into Santa Cruz for the Boardwalk. I hiked all over the Santa Cruz Mountains, in the summer I hiked and camped as much as possible. Dug for sharks teeth in Scott's Valley. The longest one I found was over three inches long but was broken into two parts. It was in a dry rivulet in the sand parking lot where we, and everybody else, dug for sharks teeth. I pretty much stopped hitchhiking when it became nearly impossible to get a ride because of that crazy guy who was killing hitchhiking girls. His mom and her friend too. Buried their heads under stepping stones in the yard. I think he was named Edmund Kemper. I remember buying chilled dungeness crab, butter, sourdough bread, and beer and sitting with friends on a dock eating it. Cheers, Eric

Reply to
etpm

Cat's aren't particularly adjustable, warmed or otherwise. They're notoriously stubborn.

Reply to
Dave Platt

Cat warmer: Actually, that's the neighbors cat. She typed "don't you dare try to move me" on the keyboard as a warning. I found a different computer to work on.

Incidentally, the noisy fan on cold start turned out to the small fan on the video card. Cleaning and switching to a multi-viscosity oil change seems to have solved the problem.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

Once upon a time, I was temporarily put in charge of marine radio manufactories production test department while they searched for someone to replace my predicessor, who had apparently gone missing, insane, or both.

Among the test techs was one who followed instructions dilligently, but usually failed to understand what was expected. The techs were told that in the transmitter RF chain, if they couldn't identify which device lacked sufficient gain, they could just replace each device in turn until the culprit was found. This was possible because there were only 7 devices involved, and the most likely culprits were the cheap TO-92 devices found at the oscillator end of the chain. Unfortunately, that was not communicated to this individual, who proceded to replace devices starting with the rather expensive 25 watt VHF RF output transistor. Nobody said noticed until I went to the parts room to replace a 25 watt device that I had blown up in engineering, and was informed that production test had grabbed all the available stock. I also noticed that someone had prefixed some of the printed test procedures to include turning the power on to the test equipment. I survived for about a month and was very happy when the company finally hired someone to replace me.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

Always start with the most expensive part. If it is not bad, it will be when it is ripped out.

I had a Toyota that started running vrey bad with about 130,000 miles on it. I repalaced the simple things like the plugs, wires, coil and fuel filter. Nothing helped. From there a Autozone trouble shooting chart indicated a sensor that was about $ 500. I took it to the local Toyota dealer and let them look at the problem as I did not want to spend $ 500 and not need that part. They kept it about 3 weeks and changed a few other parts including the spark plug wires as I did not use Toyota wire. Finally they changed that $ 500 sensor and it ran like it should. I would have thought a dealer could run some tests but seems like they were just parts changers. Really ticked me off as they had put on about $ 150 worth of parts that were not needed and still charged me for them not counting on the 3 weeks it took them. It may have taken longer but I emailed the Toyota company about how bad the service was at he end of the second week and got back a very nice email from them .

Reply to
Ralph Mowery

One of my customers in the computah repair business own a large house that had been converted to college student housing. My guess is about

15 students crammed into every possible available space. I was hired by the owner to maintain the Comcast internet, NAS server, printers, phone system, wiring, and wi-fi, which are all part of the package. The prime directive was to do everything necessary to prevent the students from burning down the house, which I interpreted to mean not let them make any changes or additions to the wiring or equipment. That has worked fairly well by the honor system for about 8 years.

This morning, I get a phone call that the internet and some of the phones are down. Upon interrogation, I determine that some of the "makers[1]" in the house had "optimized" the performance of the Comcast cable modem and VoIP systems yesterday. My initial guess was that they had managed to scramble the ethernet cables on the 24 port managed switch running several VLAN's. Unlike ordinary ethernet switches, a managed switch with traffic logging, required that the correct cable be connected to the correct port.

I had previously suggested that someone should take photographs of the wiring for this exact purpose. Amazingly, the students found the flash drive in the envolope that I had place in plain sight, and were able to display how things were wired before the system was "optimized". They were able to bring things back online in only a few additional minutes (I could see the switch come up from home via SNMP) and only waited a full hour before someone bothered to phone me that it was fixed and I need not waste a service call to the house.

With such a wonderful example of initiative and organized troubleshooting, I think there may be hope for our next generation of our country's leaders.

[1] Make trouble, make a mess, make love, make things break, make a quick exit, etc...
--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

Never do anything that can't be undone.

I had almost the same situation, except it was the rear oxygen sensor. The dealer wanted to replace the very expensive catalytic converter. The dealers computer diagnostics all pointed to the catalytic converter. They even called the factory service support number, which confirmed that the diagnostics conclusively pointed to the catalytic converter. However, I was able to listen to this conversation and detected considerable uncertainty and lack of competence on everyones part. Something seemed wrong, so I decided to do some checking. The internet was full of the usual mix of wisdom and useless garbage, but I did manage to find an article suggesting that the rear oxygen sensor tends to get contaminated by various exhaust products, and that since it's two orders of magnitude cheaper than a catalytic converter, it should be replaced first. I replaced it, and all as well.

I was curious as to what had just happened, so I went back to the dealer, told them the story, and asked a few questions. Basically, there is nobody in the system that knows how to troubleshoot a vehicle without totally relying on the computerized hardware. The programming is not intended to nail down the exact cause of a problem, only the most probable causes, expecting the dealer to replace things in a rational and logical order based on their experience. In other words, the computer supplies a shopping list that usually includes a large number of irrelevant replacement parts and procedures. When calling factory service support, they don't have experienced techs on the phones. They just read the computer screen. I was also told that it was amazing that the diagnostics programs worked because the factory was constantly making changes and improvements that affected recommendations.

In my distant past, I had the displeasure of integrating BITE (built in test equipment) in some radios. The test system caused more failures than it detected. The diagnostics could be easily mislead, producing insane results. Even worse, it could not diagnose itself. All I can say is that such diagnostics are NOT easy to do, especially with a moving target, such as an automobile.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

At work there is a pipe that has about 150 psi of air on it and about 2 inches in diameter. This goes to a regulator that cuts it to about 75 psi and then to a a control valve that is ran by a computer. This regular has a habit of getting water in it and in the winter time turning to ice. I got a call that the system was not working correctly so I cleared the ice from the regulator. Then an engineer wanted me to rebuild or change out the control valve because it was not working. I told her that I was not going to do it unless there were more problems. She got very mad at me and told me that her data from the computer said the control valve was not working. I told her I did not care what her old data said, it was not the control valve.

The control valve can not control anyting if no air is getting to it for it to control.

Had a lot more trouble out of some of the production engineers because of the computer. The process probably monitored over 1000 points and stored that data for a month. That data could be called up in a graphical form. Only problem is some of the data points may be 1 to 5 moments apart. Something could hapen and would not be caught during that time frame. Then it would look like something other than the origional problem was the cause.

Computers are good,but you have to undestand what they are trying to tell you and sometimes they are just wrong.

Reply to
Ralph Mowery

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.