Capacitors
Ecce Homo! Loquitur...
Thread Starter
Capacitors
Hushed-Up Disaster Dept.: If you've been paying attention to online forums, you've discovered a scandal that few companies are talking about. The only major media coverage has been in the Toronto Star. And it's a weird story.
Apparently because of industrial espionage—the facts are murky—a defective formulation for capacitor electrolyte, possibly stolen from a Japanese capacitor maker, was used over the past year or two to manufacture millions of cheap capacitors, mostly by Taiwan-based component makers. The flawed electrolyte forms hydrogen, then the capacitors leak, bulge, or pop like firecrackers. These caps are now blowing up motherboards left and right.
Only Abit Computer Corp. and IBM have had the guts to admit the problem. No computer vendor is immune to this situation as far as I can tell, but most are playing dumb, threatening to sue if their names are mentioned or hiding behind NDAs. They are hoping the problem will blow over.
Tech-repair firms like to tell people that this problem is caused by power surges—blaming the victim. Do a Google search on bad capacitor electrolyte or a similar combination of terms to see what is going on. It seems these bad caps are also found in some camcorders, VCRs, and other electronic gear. When they fail, the entire board fails. There is evidence that the first batch of bad capacitors appeared in 2001, and one repair person in Utah says he's replaced 40,000 bad caps already. Geez.
What I haven't seen fully discussed is that many of these capacitors may have found their way into aircraft electronics. I hope that some reader with influence will get the message to the right people. This is a disaster waiting to happen on a lot of fronts. Can you say "fly-by-wire"?
PC Magazine
Apparently because of industrial espionage—the facts are murky—a defective formulation for capacitor electrolyte, possibly stolen from a Japanese capacitor maker, was used over the past year or two to manufacture millions of cheap capacitors, mostly by Taiwan-based component makers. The flawed electrolyte forms hydrogen, then the capacitors leak, bulge, or pop like firecrackers. These caps are now blowing up motherboards left and right.
Only Abit Computer Corp. and IBM have had the guts to admit the problem. No computer vendor is immune to this situation as far as I can tell, but most are playing dumb, threatening to sue if their names are mentioned or hiding behind NDAs. They are hoping the problem will blow over.
Tech-repair firms like to tell people that this problem is caused by power surges—blaming the victim. Do a Google search on bad capacitor electrolyte or a similar combination of terms to see what is going on. It seems these bad caps are also found in some camcorders, VCRs, and other electronic gear. When they fail, the entire board fails. There is evidence that the first batch of bad capacitors appeared in 2001, and one repair person in Utah says he's replaced 40,000 bad caps already. Geez.
What I haven't seen fully discussed is that many of these capacitors may have found their way into aircraft electronics. I hope that some reader with influence will get the message to the right people. This is a disaster waiting to happen on a lot of fronts. Can you say "fly-by-wire"?
PC Magazine
Join Date: Nov 2002
Location: 39N 77W
Posts: 1,630
Likes: 0
Received 0 Likes
on
0 Posts
Quote:
---------------------------------------------------
Tech-repair firms like to tell people that this problem is caused by power surges?blaming the victim. Do a Google search on bad capacitor electrolyte or a similar combination of terms to see what is going on. It seems these bad caps are also found in some camcorders, VCRs, and other electronic gear. When they fail, the entire board fails. There is evidence that the first batch of bad capacitors appeared in 2001, and one repair person in Utah says he's replaced 40,000 bad caps already. Geez.
---------------------------------------------------
Let's see, most 9-to-5 people work about 2000 hours a year. Say 1.5 years have elapsed since capacitors started failing - that would be about 3000 working hours.
To have replaced 40,000 caps in that period, he would be doing roughly one each 5 minutes. Diagnose and repair in just 5 minutes with surface-mounted components????
That's unbelievable unless there were a substantial team of people working.
Hyperbole seems to have set in.
Does sound like a worrisome situation, nevertheless.
seacue
---------------------------------------------------
Tech-repair firms like to tell people that this problem is caused by power surges?blaming the victim. Do a Google search on bad capacitor electrolyte or a similar combination of terms to see what is going on. It seems these bad caps are also found in some camcorders, VCRs, and other electronic gear. When they fail, the entire board fails. There is evidence that the first batch of bad capacitors appeared in 2001, and one repair person in Utah says he's replaced 40,000 bad caps already. Geez.
---------------------------------------------------
Let's see, most 9-to-5 people work about 2000 hours a year. Say 1.5 years have elapsed since capacitors started failing - that would be about 3000 working hours.
To have replaced 40,000 caps in that period, he would be doing roughly one each 5 minutes. Diagnose and repair in just 5 minutes with surface-mounted components????
That's unbelievable unless there were a substantial team of people working.
Hyperbole seems to have set in.
Does sound like a worrisome situation, nevertheless.
seacue
Join Date: Nov 2000
Location: Melbourne,Vic,Australia
Posts: 423
Likes: 0
Received 2 Likes
on
2 Posts
Diagnosis of the fault would be quite quick especially as the faulty capacitor is usually visually obvious.
The capacitors in question are typically 10mm diam x 15mm long and are used in the switchmode buck regulators (the ones near the little toroidal coils) and for general power filtering at regulator frequency (surrounding the cpu and scatteded over the board. They are usually throughhole components in the hundreds and thousands uF range. The other capacitors on the board are mostly high and very high frequency power supply filter ceramic surface mount capacitors in the nF range.
As failure of one would lead to the suspicion they were all dubious any proper repair would involve replacement of them all - typically ~ 30.
The capacitors in question are typically 10mm diam x 15mm long and are used in the switchmode buck regulators (the ones near the little toroidal coils) and for general power filtering at regulator frequency (surrounding the cpu and scatteded over the board. They are usually throughhole components in the hundreds and thousands uF range. The other capacitors on the board are mostly high and very high frequency power supply filter ceramic surface mount capacitors in the nF range.
As failure of one would lead to the suspicion they were all dubious any proper repair would involve replacement of them all - typically ~ 30.
Join Date: Sep 2001
Location: 38N
Posts: 356
Likes: 0
Received 0 Likes
on
0 Posts
Like air carriage, electronics is a business that is relatively easy to get into and relatively hard to stay in.
There's a familiar ring about the reported problems with faulty components, manufactured into faulty systems, now leading to faulty companies.
From all reports, a bit of simple electrical testing at the component mfr or the systems mfr would surely have revealed the fabrication flaw. This is what separates beginners from finishers in technology related business.
Low-cost aluminum-foil electrolytic capacitors are typically used in power suppies and power circuits in consumer products. They depend on a chemical oxide layer for the capacitive effect. The correct electrical polarity (plus to plus, minus to minus) maintains this essential oxide layer. If electrolytics are fed a net reverse polarity for any duration, the oxide layer breaks down and the parts function very like low-ohm resistors as they generate heat and converte the internal electrolyte into various gasses. This mechanism leads to progressive failure. Eventually the gasses pressurize the case enough to explode - quite dramatically for big ones. Less dramatic failures result when occasional reverse-polarity pulses gradually convert the necessary electrolyte to gas - a 1-way process. The phenomena are so well recognized that this type of capacitor is normally manufactured with a burst diaphragm scribed into the package so as to control the kaboom contingency somewhat.
(Interestingly, the oxide layer in these components also deteriorates over time if unused. Powering up electronics with them after years of shelf time can lead to unpredictable behavior for a while until the electrolytic capacitors "heal" themsleves and begin to filter properly. )
While faulty electrolyte chemistry certainly is possible, it is a pretty dumb and obvious mistake for volume production. Similar foulups happen more often than many would care to believe, which is why Quality Control is such a popular phrase in the technology field. Other common causes for failure of these components - in the general case - are circuit design errors and failures of other components, such as reverse flow-preventing diodes.
As for avionics - the use of cheap but inherently fallible aluminum electrolytics is frowned upon in critical applications for the reasons cited above. Avionics, mil electronics, and hardened computing things typically use more exotic (and exponentially more expensive) technologies to accomplish equivalent results.
There's a familiar ring about the reported problems with faulty components, manufactured into faulty systems, now leading to faulty companies.
From all reports, a bit of simple electrical testing at the component mfr or the systems mfr would surely have revealed the fabrication flaw. This is what separates beginners from finishers in technology related business.
Low-cost aluminum-foil electrolytic capacitors are typically used in power suppies and power circuits in consumer products. They depend on a chemical oxide layer for the capacitive effect. The correct electrical polarity (plus to plus, minus to minus) maintains this essential oxide layer. If electrolytics are fed a net reverse polarity for any duration, the oxide layer breaks down and the parts function very like low-ohm resistors as they generate heat and converte the internal electrolyte into various gasses. This mechanism leads to progressive failure. Eventually the gasses pressurize the case enough to explode - quite dramatically for big ones. Less dramatic failures result when occasional reverse-polarity pulses gradually convert the necessary electrolyte to gas - a 1-way process. The phenomena are so well recognized that this type of capacitor is normally manufactured with a burst diaphragm scribed into the package so as to control the kaboom contingency somewhat.
(Interestingly, the oxide layer in these components also deteriorates over time if unused. Powering up electronics with them after years of shelf time can lead to unpredictable behavior for a while until the electrolytic capacitors "heal" themsleves and begin to filter properly. )
While faulty electrolyte chemistry certainly is possible, it is a pretty dumb and obvious mistake for volume production. Similar foulups happen more often than many would care to believe, which is why Quality Control is such a popular phrase in the technology field. Other common causes for failure of these components - in the general case - are circuit design errors and failures of other components, such as reverse flow-preventing diodes.
As for avionics - the use of cheap but inherently fallible aluminum electrolytics is frowned upon in critical applications for the reasons cited above. Avionics, mil electronics, and hardened computing things typically use more exotic (and exponentially more expensive) technologies to accomplish equivalent results.
Join Date: Dec 2001
Location: europe
Posts: 546
Likes: 0
Received 0 Likes
on
0 Posts
Capacitors are to electronics what bolts are to mechanics, and as little excuse for getting them wrong.
Used to be a common failure item due to relying on spidery terminal wires to support the vibrating mass of the capacitor.
Build reliability is what you pay for when buying established brands, hopefully.
Used to be a common failure item due to relying on spidery terminal wires to support the vibrating mass of the capacitor.
Build reliability is what you pay for when buying established brands, hopefully.
