Stupid malfunctions, quick fixes

[Alana]

I just fixed an EEPC by re-soldering components inside its PSU. This stupid malfunction gave me an idea - how about posting such stuff here. I mean - description of what went wrong and how you dealt with it.

But there is a catch - it has to be funny, unexpected or plain stupid.

[Psi]

My Panasonic PT-AE900 data projector developed a fault with the color green.
The green went to around 10% on over the whole image. Red and Blue were unaffected.

I downloaded a service manual and probed around the green lcd driver and sample/hold IC's.
After a while i found a pin on the green sample and hold IC was at 0.2V when it should be 5.5V.
The datasheet said the pin was the green bias output voltage. It didn't go anywhere except directly onto the green LCD header.

It seemed pretty obvious the IC had a faulty output. The time was ready for a dirty stupid hack...
I wired up a pot and a few resistors from a nearby 15V source and tried to just force the IC output pin back to 5.5V.
It took ~8mA and the green came back right away. The hack is still working in the projector today (4 years later)

I posted it on the AVS forum in case anyone else encountered the problem and googled for it.
If you want to read more about the hack its here
http://www.avsforum.com/t/1029165/pt-ae900-no-green-fix-mod-warning-technical

[neotesla]

Old batteries leaked in my guitar tuner. The existing contact (shown in circle) was totally corroded, and no matter how much scraping and filing I did, the power did not get to the device. Then I discovered under my desk a piece of tin that fell off the other day when I was fixing and old laptop ... I put that where the battery contact is (it fit perfectly), and... voila, the power was back on. It's not much of a story, but it did make my day.

[SeanB]

Done the tin spacer a lot, using whatever can was to hand as the donor. Best was a corned meat can, ready cut strip of tinnes steel from the band sealing it.

When fixing a LED illuminated LCD panel to use as a panel on my workbench I did not buy a mounting kit. I measured the holes and drilled the back panel in 2 places to match, then used 2 long M4 screws that I cut the heads off, and slightly mangled the thread just under the head to lock it in the brass bosses in the panel back. Then I placed it on the panel and used 4 washers and nuts to hold it from the rear flush. Works perfectly, and cheaper than the mount kit, and no added depth as well. Cost was zero as well, the screws, nuts and washers were old ones recovered from dead electronics or were in my oddments bin.

[AndyC_772]

It seemed pretty obvious the IC had a faulty output. The time was ready for a dirty stupid hack...

I had a similar experience with the timer in an exercise machine. Its clock was running slow, which meant that a 20 minute workout lasted over twice as long - and although I could probably use the exercise, faulty hardware annoys me.

I determined that the problem was with the 32 kHz oscillator. The machine was built around a single, unidentified chip, but as the crystal wasn't oscillating even after replacement, it was pretty clear that one of the two connections to the crystal from the IC wasn't working. (Quite how something managed to break within the confines of the resin blob I've no idea).

I removed the crystal and hacked in a 32 kHz oscillator module instead, and the machine has been working fine ever since.

[KD0CAC John]

How about stupid mistakes that have not been fixed yet , and looking for repair ideas
Have an old Heart inverter , and A-- Holes do not want to let anybody have access to there wiring diagrams / schematics ?
I guess I'm use to ham radio gear freely giving that info with there gear . 

[G7PSK]

How about stupid mistakes you have made yourself. I once had a job that required cutting a channel in a floor laying some cables and then cementing the channel in again and all had to be done over night and ready fo the forklifts to be able to run over it. Well I decided that quick set cement was the thing to use at first all was well with the first batch but then I realized it would set faster mixing it with hot water so the next mix I used hot water from the tap and the cement set in ten minuets so I thought to myself if hot water is good boiling water is better. I looked around and found a five gallon tea urn, so I boiled it up poured the boiling water into the barrow of cement and started mixing, 30 seconds later the shovel was stuck in the barrow and the cement was rock hard.

[SeanB]

Been there, but a better bet was mixing tile cement with bonding liquid and filling the hole. Cured in around 5 hours to rock hard. I take it you still have the barrow, but the shovel is still a piece of modern art.

[G7PSK]

The cement set solid but was very brittle and a slight tap with a 28 pound sledge hammer shattered it freeing the shovel and barrow, I did not think about keeping it, just getting the job done on time.

[grumpydoc]

Quick fixes:

Years ago when I was doing computer repairs a bloke walked into the shop with a 5 1/4" floppy drive** which wasn't working. He begged me to have a look ASAP as he was in the middle of his tax return and couldn't read any files.

Took the drive into the workshop - this was in the days when belt drive floppies were the norm. Turned out the belt had fallen off. It was probably a bit stretched but I didn't have a replacement - anyway fitted it back on and the tension seemed OKish so gave it back to him. Easiest 25 quid I ever made.

** I did say years ago

Mistakes: Oh <deity> too many to mention 

[miceuz]

I was building a linear power supply out of opamps/transistors, basic stuff, just to get some understanding in analog. Voltage regulation worked OK, then I've added current sensing / regulation and the circuit would start to oscillate like crazy when in over current mode. After a month of frustration it came to me that the current sensing opamp was powered by the same rail it was sensing - it's just that kind of specially designed current sense amps. So, current regulation was kicking in, it was lowering the voltage untill current sense opamp went bananas giving high readings, regulation would shut everythig down, then get 0 amps from sensing opamp and relax and the cycle continues.

Fixed by getting other current sense opamp that was powered by separate rail. Just dumb stupid me

[Psi]

Back in the days of 10MegaByte hard drives we had one of them die so dad and I took the top off to have a look.
One of the two heads had completely detached from the arm and was only held on by its two wires.

A small drop of superglue and a lowlevel format and it was working again

[lrai]

Yesterday I spent 2 hours designing a custom connector in eagle (what a pain). Made my PCB, drilled all the holes mounted everything and about 1 am all was soldered, tested and time for my connector when I realized that I mirrored the connector.
Solution … mount the connector on the bottom and lots of hot glue for mechanical support.

[JoannaK]

Latest..  Starting battery (AGM type Lead-acid, 'sealed') of my scooter died..  It didn't hold the charge anymore so I desided to do agains the instructions and open it up (hey.. if they leave those rubber gasgets visible, what's I'm supposed to do?). Yep, it was allmost 100% dry. I was not sure if it's safe to fill em up with distiled water (like regular Lead-acids), but decided it's easy and cheap to test. Plenty of water, full charge and it works just fine.

Saved me over 20euros + need to trash this old one. Makes me wonder how much these batteries will be trashed unnecessarely? I kinds get this feeling that these 'maintenance-free' batteries are just an excuse to sell more trash to people.

[Paul Moir]

Psi's comment reminded me of this.   Way back when I had one of the very early IDE drives, 40MB Western Digital iirc, a derivative of the only slightly older 30mb RLL WD drives if you remember those.  This particular one would throw the occasional read error when it warmed up.  It was pretty obvious that it was a tracking error and there was no way that I knew to low level format this drive to fix the tracking.  Since this one of those ones with the external square stepper moving the head, I simply drove a (happened o be) very large wood screw between the stepper and the case, which would rotate the stepper very slightly.

Of course in those days you would actually hand a drive like that down to say your brother who needed to write papers.  Much to my surprise, the wood screw was a big hit.  Any time he'd see an Abort/Retry/Fail message, he would put a bit of a turn on the screw.  You have to understand, even regular users encountered drive errors very regularly back then, and for once this gave them something to do about it rather than simply hit (R) and pray.
 

[SeanB]

Joannak, the "Sealed for life" lead acid car/bike battery is the same as the regular one with a filler that can be removed, they just have a cover or a label over it. I can go to my local battery supplier and buy a battery from 4 makes, all the same battery, either sealed for life or not, which are the same battery just with a different label. Putting water in to the cells will not kill it, but it will not last long now that it has run dry, but will last you till the cold months of winter kick in and the scooter will not start one morning when it is very cold. If you drive it every day long enough to keep it fully charged, and have clean tight terminals without corrosion it can last a while before it will suddenly die.

[Psi]

Psi's comment reminded me of this.   Way back when I had one of the very early IDE drives, 40MB Western Digital iirc, a derivative of the only slightly older 30mb RLL WD drives if you remember those.  This particular one would throw the occasional read error when it warmed up.  It was pretty obvious that it was a tracking error and there was no way that I knew to low level format this drive to fix the tracking.  Since this one of those ones with the external square stepper moving the head, I simply drove a (happened o be) very large wood screw between the stepper and the case, which would rotate the stepper very slightly.

Of course in those days you would actually hand a drive like that down to say your brother who needed to write papers.  Much to my surprise, the wood screw was a big hit.  Any time he'd see an Abort/Retry/Fail message, he would put a bit of a turn on the screw.  You have to understand, even regular users encountered drive errors very regularly back then, and for once this gave them something to do about it rather than simply hit (R) and pray.

haha, classic.

[SeanB]

REminds me of the best one repairing Avionics. Bombing computer came in with the fault "First bomb fell on target". Checked it and sent it out without opening it, after a good laugh with the pilot ( OC of the squadron as well) and a quick suggestion to my erstwhile boss there to do a self test on a certain other part. Never got a reply from him on that..............

[JoannaK]

Putting water in to the cells will not kill it, but it will not last long now that it has run dry, but will last you till the cold months of winter kick in and the scooter will not start one morning when it is very cold. If you drive it every day long enough to keep it fully charged, and have clean tight terminals without corrosion it can last a while before it will suddenly die.

Yep.. It'll soon become the end of the season anyhow, so I'll need to store the bike away for winter. I have no interest on using that machine when there are ice/snow on roads and temperature drops bellow 0C, and it would be kinda foolish to purchase new battery just before the winter brake.

[ElektroQuark]

Today's one:

Someone has given me a NAVIGON GPS unit.
It didn't power on, battery is dead and the USB charger does nothing.
I took it apart and the problem was with the USB connector. It's a SMD type and it was broken of the PCB. Someone has tried to repair it with some drops of Loctite
A little resoldering work plus two tiny pieces of cooper to fix some damaged tracks and it's working again.

[T4P]

Sorry no quick fix here, in fact no fix with my knowledge
So someone sent in a 42-incher Phillips LCD TV from 2009
Guess what was the problem? The TV came from US! The front end was only spec'd to run at 110V ... and clearly somebody put 220V through that thing
BUT i couldn't find any problem at least. Though, i found it.
wrong voltages ... 400V from a supposedly 345V area ... Oops, something's blown and i can't tell nor test

[G7PSK]

Joannak, the "Sealed for life" lead acid car/bike battery is the same as the regular one with a filler that can be removed, they just have a cover or a label over it. I can go to my local battery supplier and buy a battery from 4 makes, all the same battery, either sealed for life or not, which are the same battery just with a different label. Putting water in to the cells will not kill it, but it will not last long now that it has run dry, but will last you till the cold months of winter kick in and the scooter will not start one morning when it is very cold. If you drive it every day long enough to keep it fully charged, and have clean tight terminals without corrosion it can last a while before it will suddenly die.

Some of the sealed for life batteries have a small amount of catalyst covered gauze in the end of the cell cap to recombine the gasses.

[SeanB]

There will be a catalyst in the expensive UPS and telecoms backup batteries, but not in the price sensitive automotive start market.

[ablacon64]

A dead laptop charger. Opened it and a coil fell to the floor. Two solder points, problem solved. Client guarantees it didn't fell off (right!).