My FPCU has over 350K mi. have a question ?
-
jamesmost
- Posts: 1963
- Joined: Tue Jul 08, 2003 10:16 am
- Car: "95Qmodded, Benz wagon 4matic , 1986 MB 560sec
My FPCU has over 350K mi. have a question ?
I have a spare w/ low low milege should i swap it ? any benefit ?? I see no signs of defficiency w/ the origional.
-
Q45tech
- Moderator
- Posts: 14296
- Joined: Tue Apr 30, 2002 3:19 am
- Car: 1990 Q45 342,400 miles 22 years ownership with original engine
1995 G20t 5 speed 334,000 miles 16" 2002 wheels - 205/50/16 Sr20ve vvl
Re: My FPCU has over 350K mi. have a question ? (jamesmost)
The only component inside electronics that have a timed life are capacitors which fail based on hours and temperature............higher temp fewer hours.
In theory the FPCU should last 20+ years [depending on number cosmic/gamma ray hits from exploding stars]. Position is important as the rays must hit semiconductor junctions precisely to cause failure.
As we know the only thing that takes them out is defective fuel pumps caused by wear from the fuel you purchase.
350,000 miles ~= 10,000-20,000 hours nothing like a 24 hour computer that lasts 3 years
A generally accepted formula for estimating film capacitor life is:
L=LR*(ER/Eo)7*2 (deltaT/10)
Where:L = operating life under stated temperature and voltageLR = life at rated temperature and voltageER = rated voltage limitEo = operating voltage deltaT = difference between rated operating temperature and capacitor core temperature in C.
Aluminum electrolytics have a reputation for being troublesome in some electronic equipment. NASA, for example, does not allow them to be used in flight hardware because of the risk of failure and outgassing. Tantalum capacitors can be used instead. There has been some work in designing aluminums that could be used in space applications however.
In general, close attention has to be paid to their application if you expect good long-term reliability. The higher the operating temperature the shorter their life, and running aluminums too hot seems to be a common design mistake. This can be caused by things like excessive ripple current, poor ventilation, too high a system ambient, and/or locating them too close to a hot power supply component. Over time the oxide film tends to dissolve into the electrolyte and every 10C rise in temperature doubles this effect. The rate of deterioration is much faster in storage than if the capacitor is kept voltage biased in normal use. Actually, the most common failure in aluminum electrolytics (at least for through-hole aluminums) is not loss of capacitance or leakage, but increase in ESR, due to loss of water from the electrolyte. This is also temperature dependent. Operating life is less dependent on operating voltage than with some capacitors. A generally accepted formula for estimating electrolytic capacitor life is:
LT=LR*(ER/EO)*2(deltaT/10)
Where: LT = operating life under stated temperature and voltage (capacitor core temperature, not just ambient)LR = the life at rated limitsER = rated voltage limitEO = operating voltagedeltaT = difference between rated operating temperature and capacitor core temperature in C
Note the use of "core temperature" rather than ambient. Big aluminums are usually used in applications involving high ripple current, which causes self heating. Their wound paper/foil makes electrolytics poor at dissipating heat. Wattage dissipation in a capacitor is equal to I2*(ESR), where I is the ripple current, and ESR is equivalent series resistance at the frequency of interest. The manufacturer may not give you the ESR data however. Instead they might give a number for "allowed ripple current". A rough idea of temperature rise can be found from power dissipation, surface area of the capacitor, and the amount and kind of cooling available (forced-air, convection). Some manufacturers can provide specific guidelines. Nomographs exist for this sort of thing, but are hard to find. Clamp-on heat sinks have become available for some electrolytic sizes. One manufacturer , http://www.chemi-con.com/ , even makes some big aluminums with a hollow center (a pipe shape instead of a rod) for better cooling by air flow. Another makes some electrolytics in flat packages to increase surface area.
See:
"Selection and Application of Capacitors" http://www.evox-rifa.com/http://www.cor ... lators.htm. Dead?
Both tantalum and solid-electrolyte capacitor makers claim longer life than conventional aluminum capacitors. However, choosing the right aluminum electrolytic with the right deratings, will often allow you to meet whatever longevity requirement you have in most common applications without resorting to more exotic (and expensive) parts.
Parts kept in long-term storage (2-3 years or more) may have very high leakage when first used and should not be subjected to full rated voltage (and should be current limited) until the leakage falls to normal. Reforming of the dielectric can be done by applying a voltage that is slowly increased to maximum over a period of several hours, with current limited to rated leakage current. Some manufacturers have published such elaborate recommendations for reconditioning capacitors after long storage that it seems easier to just buy new ones. However, while common wisdom has it that storage is very bad for aluminum capacitors, manufacturer's literature does not always support this. One manufacturer has a 1000 hour/85C shelf-life test that specifies that leakage will not exceed that of a new capacitor. That's equivalent to about 7 years at 25C. It may be that modern capacitors are less susceptible to storage problems than older ones.
See:
MIL-HDBK-1131 Storage Shelf Life and Reforming Procedures for Aluminum Electrolytic Fixed CapacitorsIt´s on the web in PDF form.
Unless epoxy sealed, aluminum capacitors can receive long-term damage from chlorinated solvents, and from halides in general. When in doubt, use a non-chlorinated board cleaner. Some manufacturers even warn against chlorides in no-wash solder flux (including some that claim to be chloride-free), board coatings, and adhesives. Methyl bromide, a very common fumigant, is suspected of causing capacitor failures in at least one case. Some non-chlorinated solvents can cause damage to the rubber end-seals if exposure is too long. Manufacturers also warn against other chemicals like hydrogen sulfide and ammonia.
Don't subject aluminum capacitors to more than about 1 volt reverse polarity at 25C, and more than about 0.4 volt at 85C .
Large axial-leaded electrolytic capacitors have problems with leads breaking from vibration and must be held down with glue or a cable tie.
Manufacturers usually warn against using general-purpose aluminum capacitors in applications involving high surge currents. Although aluminum capacitors are not as vulnerable to surge currents as tantalums are (see below), possibly because higher ESRs help limit the current, they may not be reliable in such applications. Use capacitors with specified high-current and/or dV/dT ratings such as photoflash aluminum electrolytics.
Counterfeit Capacitors Just like any other electronic component, capacitors are the target of counterfeiters. This includes low quality parts mislabeled as to manufacturer and parts that are totally nonfunctional, such as tantalum capacitor cases with no actual capacitor inside. Today such parts are typically traceable to state-owned Chinese companies with go-betweens in Taiwan. The best defense against counterfeit parts is to buy from the factory, or from authorized distributors. This not a perfect defense however, some people are sneaking counterfeit parts between the factory and the distributor.
XXXXXXXXXXXXXXX
Recent news There are many reports of high failure rates/short life of electrolytics made by a number of Taiwan companies. It seems they all purchased a novel water based, low-ESR electrolyte from the same source. The formula was defective (industrial espionage gone wrong) and the electrolyte evolves hydrogen, which causes the cases to leak or even burst over time. However, the caps may be unusable even if they look normal. Time to failure is said to be from a few hundred to a few thousand hours. The number of companies involved is estimated at from 5-12, few of which admit to any problem with their parts, but blame the design of the final products. Names most often mentioned include Luxon, Lelon, Licon, Tayeh, Jackcon, JPCON, Teapo, and Rulycon, a shamless ripoff of the reputable Rubycon brand . Some brand names are actual companies, but others are fictional brands made by companies who didn't want to put their names on these products.
The problem seems to have started around 2001 and peaked in 2002, but a number of motherboard repairmen say they are still seeing defective capacitors on current production products, including major brands. In fact, defective capacitors may still be in production as of 2006. To make things worse, there are, as always, many counterfeit parts coming out of the Far East. Many products can be repaired by just replacing the capacitors, but in some cases the capacitors cause the voltage regulators to fail as well. It may be safest to stick to Japanese brands from authorized distributors.
The parts have been widely used on white-box PC motherboards and in many other products. Products sold (but perhaps not made) by major companies are involved as well, IBM, HP, and others. Why? The difference in cost between a brand name part with a history and a no-name part is pennies. The ASUS motherboard just I bought uses all Japanese electrolytics. A few companies, IBM, the PC board maker ABIT, and capacitor maker Jackon, for example, have "come clean". Other companies deny everything to the point of threatening people with lawyers.
For more information see below. Other links I had have gone 404:
http://www.niccomp.com/taiwanlowesr.htm http://www.e-insite.net/electronicnews/ obsolete?http://www.reed-electronics.co...cfd=1 http://www.burtonsys.com/bad_BP6/story4.html http://www.ec-central.org/maga...p.pdf http://www.badcaps.net/ The motherload of "bad cap" stuff. Sells caps and cap kits for motherboard repair. Includes a list of know bad brands (about 30) in the Forums section. http://www.burtonsys.com/bad_BP6/story1.html http://www.motherboardrepair.c...=home Sells caps and cap kits for motherboard repair.http://www.pcstats.com/article...D=195 http://listas.rcp.net.pe/piper....html http://www.spectrum.ieee.org/W....html http://en.wikipedia.org/wiki/C...rands
XXXXXXXXXXXXXXX
OS-CON and other polymer-aluminum capacitors TCNQ aluminum electrolytics are said to have a longer life than most traditional aluminums, but they don´t last forever. See http://www.vishay.com for a pdf file for estimating their life.
Tantalum Electrolytic As with other capacitors, operating temperature and operating voltage have an affect on service life and reliability.
Most tantalum capacitor types tolerate a limited amount of reverse polarity voltage. Manufacturers recommend no more than about 1 volt at 25C and 0.1 volt at 125C, others say no more than 10% of rated voltage at 25C and 1% at 125C. Experience shows that solid tantalums inserted backwards in low-voltage systems (a 5 volt buss for example) are unpredictable and may take from seconds to years to fail. The time to failure is much less at elevated temperature however, and the parts should failure quickly in routine quality control temperature testing.
Solid tantalum capacitors tend to have a problem with surge currents due to impurities and thin spots in the dielectric. Manufacturers typically recommend at least 1 ohm of series resistance per operating volt if the inrush current is not otherwise limited. Tantalum capacitors should not generally be used for power supply filtering unless specifically made and tested for the application (as some are). If you do use a general-purpose tantalum in a power supply application, derate the operating voltage by at least 3:1 (some say 4:1) if possible and/or use soft-start circuitry. Tantalums are often used as output filters for IC voltage regulators in small systems and the regulator will probably limit the turn-on current to a reasonable level. A tantalum on the input side of the regulator may be more vulnerable however. The progressive miniaturization of modern electronics has brought tantalums into more applications where surge is a problem, where a larger aluminum capacitor would previously have been used. A 3 or 4:1 voltage derating has a drastic impact on volumetric efficiency, so manufacturers have had to improve their manufacturing and testing methods. Leakage testing helps catch tantalums with dielectric defects. They sometimes do 100% surge screening to weed out early failures as well. Complaints of tantalums catching fire are common, SMD more often than through-hole.
Some ceramic capacitor makers have specifically targeted SMD tantalums for replacement by large ceramics. This is partly in response to the dissatisfaction of some equipment makers with the reliability of SMD tantalum bypass capacitors (through-hole tantalums seem to be much less troublesome). Problems with tantalums may be vendor specific, but ceramics do have some advantages over tantalums in general. Ceramics are not polarized so there is no fear of putting them on your board backwards. Other advantages are lower ESL and much lower ESR for a given capacitance. In many applications, ESR is more important than bulk capacitance, so a large tantalum can often be replaced by a relatively small ceramic. Ceramics generally fail by shorting however, and can burn your board, just like a tantalum.
Unlike aluminum capacitors, solid tantalums have no significant shelf-life restrictions, but I have heard that long-term moisture absorption can impair self-healing.
Small-value tantalum capacitors are sometimes used in analog circuits, such as for a cheap (and small) way to get very long filter time-constants. Some manufacturers even recommend this. This is risky. Most samples may show acceptably low leakage but a few percent will be much leakier than average and manufacturers will not usually guarantee leakage to a low value. Large-value ceramic capacitors (Class 2 and higher) will also have this problem. It will be OK in some applications, but when in doubt, use a film cap.
Wet-slug (and foil) tantalums use a sulfuric acid electrolyte and have their own problems. Cheaper wet tantalums have silver cases and are sealed with Teflon. The reliability of silver-case tantalums has always been problematical. They have almost zero tolerance of reverse bias because the silver grows dendrites that cause rapid damage. Also, the Teflon seal can allow the slow loss of electrolyte, or even outright leakage. Hobbyists should beware of old military-surplus parts. Better quality wet tantalums have tantalum cases and tantalum-glass seals. Leakage is much less of a problem, and they can tolerate several volts of reverse bias. Wet tantalums will not fail by going up in flames.
The expected failure rate of a solid tantalum capacitor is dependent on both operating voltage and temperature. One formula for estimating solid tantalum capacitor life is:
LT=LR*(ER/EO)3*2(deltaT/10)
Where: LT = failure rate under stated temperature and voltageLR = the failure rate at rated limitsER = rated voltage limitE O = operating voltagedeltaT = difference between rated operating temperature and actual capacitor temperature in C
Note the relatively high dependence of reliability on operating voltage, compared to aluminum electrolytics.
Using the formulas: Numbers for LR, ER , and TR are obtained from the manufacturers data sheets. LR is almost always given for aluminum electrolytic capacitors, and sometimes for tantalum. LR is less commonly found in manufacturer literature for other types unless they are designed for special usage. Note that the formulas do not allow for high surge-current operation. A further complication is that a "failure" might be defined as specific loss in performance that might or might not be of interest in your application, rather than a catastrophic loss of function. This especially true for electrolytics. Manufacturers are not in complete agreement as to these formulas. For example, some say that different shaped of electrolytics should have different lifetimes, all other things being equal.
LR is typically given as 1000 to 5000 hours. This does not sound like much, but remember, it is for worst-case conditions and goes up rapidly for reduced temperatures and voltages. The magic word here is "derating". The further below its maximum rated voltage and temperature you use a capacitor, the longer you can expect it to last. For film and electrolytics, the expected life will roughly double for every 10C below rated maximum operating temperature (a rough but useful approximation). For many applications, with conservative deratings, many designers will not find it necessary to run the numbers. However, some instrument makers take the opposite approach, documenting the stresses on every component and calculating expected reliability.
In theory the FPCU should last 20+ years [depending on number cosmic/gamma ray hits from exploding stars]. Position is important as the rays must hit semiconductor junctions precisely to cause failure.
As we know the only thing that takes them out is defective fuel pumps caused by wear from the fuel you purchase.
350,000 miles ~= 10,000-20,000 hours nothing like a 24 hour computer that lasts 3 years
A generally accepted formula for estimating film capacitor life is:
L=LR*(ER/Eo)7*2 (deltaT/10)
Where:L = operating life under stated temperature and voltageLR = life at rated temperature and voltageER = rated voltage limitEo = operating voltage deltaT = difference between rated operating temperature and capacitor core temperature in C.
Aluminum electrolytics have a reputation for being troublesome in some electronic equipment. NASA, for example, does not allow them to be used in flight hardware because of the risk of failure and outgassing. Tantalum capacitors can be used instead. There has been some work in designing aluminums that could be used in space applications however.
In general, close attention has to be paid to their application if you expect good long-term reliability. The higher the operating temperature the shorter their life, and running aluminums too hot seems to be a common design mistake. This can be caused by things like excessive ripple current, poor ventilation, too high a system ambient, and/or locating them too close to a hot power supply component. Over time the oxide film tends to dissolve into the electrolyte and every 10C rise in temperature doubles this effect. The rate of deterioration is much faster in storage than if the capacitor is kept voltage biased in normal use. Actually, the most common failure in aluminum electrolytics (at least for through-hole aluminums) is not loss of capacitance or leakage, but increase in ESR, due to loss of water from the electrolyte. This is also temperature dependent. Operating life is less dependent on operating voltage than with some capacitors. A generally accepted formula for estimating electrolytic capacitor life is:
LT=LR*(ER/EO)*2(deltaT/10)
Where: LT = operating life under stated temperature and voltage (capacitor core temperature, not just ambient)LR = the life at rated limitsER = rated voltage limitEO = operating voltagedeltaT = difference between rated operating temperature and capacitor core temperature in C
Note the use of "core temperature" rather than ambient. Big aluminums are usually used in applications involving high ripple current, which causes self heating. Their wound paper/foil makes electrolytics poor at dissipating heat. Wattage dissipation in a capacitor is equal to I2*(ESR), where I is the ripple current, and ESR is equivalent series resistance at the frequency of interest. The manufacturer may not give you the ESR data however. Instead they might give a number for "allowed ripple current". A rough idea of temperature rise can be found from power dissipation, surface area of the capacitor, and the amount and kind of cooling available (forced-air, convection). Some manufacturers can provide specific guidelines. Nomographs exist for this sort of thing, but are hard to find. Clamp-on heat sinks have become available for some electrolytic sizes. One manufacturer , http://www.chemi-con.com/ , even makes some big aluminums with a hollow center (a pipe shape instead of a rod) for better cooling by air flow. Another makes some electrolytics in flat packages to increase surface area.
See:
"Selection and Application of Capacitors" http://www.evox-rifa.com/http://www.cor ... lators.htm. Dead?
Both tantalum and solid-electrolyte capacitor makers claim longer life than conventional aluminum capacitors. However, choosing the right aluminum electrolytic with the right deratings, will often allow you to meet whatever longevity requirement you have in most common applications without resorting to more exotic (and expensive) parts.
Parts kept in long-term storage (2-3 years or more) may have very high leakage when first used and should not be subjected to full rated voltage (and should be current limited) until the leakage falls to normal. Reforming of the dielectric can be done by applying a voltage that is slowly increased to maximum over a period of several hours, with current limited to rated leakage current. Some manufacturers have published such elaborate recommendations for reconditioning capacitors after long storage that it seems easier to just buy new ones. However, while common wisdom has it that storage is very bad for aluminum capacitors, manufacturer's literature does not always support this. One manufacturer has a 1000 hour/85C shelf-life test that specifies that leakage will not exceed that of a new capacitor. That's equivalent to about 7 years at 25C. It may be that modern capacitors are less susceptible to storage problems than older ones.
See:
MIL-HDBK-1131 Storage Shelf Life and Reforming Procedures for Aluminum Electrolytic Fixed CapacitorsIt´s on the web in PDF form.
Unless epoxy sealed, aluminum capacitors can receive long-term damage from chlorinated solvents, and from halides in general. When in doubt, use a non-chlorinated board cleaner. Some manufacturers even warn against chlorides in no-wash solder flux (including some that claim to be chloride-free), board coatings, and adhesives. Methyl bromide, a very common fumigant, is suspected of causing capacitor failures in at least one case. Some non-chlorinated solvents can cause damage to the rubber end-seals if exposure is too long. Manufacturers also warn against other chemicals like hydrogen sulfide and ammonia.
Don't subject aluminum capacitors to more than about 1 volt reverse polarity at 25C, and more than about 0.4 volt at 85C .
Large axial-leaded electrolytic capacitors have problems with leads breaking from vibration and must be held down with glue or a cable tie.
Manufacturers usually warn against using general-purpose aluminum capacitors in applications involving high surge currents. Although aluminum capacitors are not as vulnerable to surge currents as tantalums are (see below), possibly because higher ESRs help limit the current, they may not be reliable in such applications. Use capacitors with specified high-current and/or dV/dT ratings such as photoflash aluminum electrolytics.
Counterfeit Capacitors Just like any other electronic component, capacitors are the target of counterfeiters. This includes low quality parts mislabeled as to manufacturer and parts that are totally nonfunctional, such as tantalum capacitor cases with no actual capacitor inside. Today such parts are typically traceable to state-owned Chinese companies with go-betweens in Taiwan. The best defense against counterfeit parts is to buy from the factory, or from authorized distributors. This not a perfect defense however, some people are sneaking counterfeit parts between the factory and the distributor.
XXXXXXXXXXXXXXX
Recent news There are many reports of high failure rates/short life of electrolytics made by a number of Taiwan companies. It seems they all purchased a novel water based, low-ESR electrolyte from the same source. The formula was defective (industrial espionage gone wrong) and the electrolyte evolves hydrogen, which causes the cases to leak or even burst over time. However, the caps may be unusable even if they look normal. Time to failure is said to be from a few hundred to a few thousand hours. The number of companies involved is estimated at from 5-12, few of which admit to any problem with their parts, but blame the design of the final products. Names most often mentioned include Luxon, Lelon, Licon, Tayeh, Jackcon, JPCON, Teapo, and Rulycon, a shamless ripoff of the reputable Rubycon brand . Some brand names are actual companies, but others are fictional brands made by companies who didn't want to put their names on these products.
The problem seems to have started around 2001 and peaked in 2002, but a number of motherboard repairmen say they are still seeing defective capacitors on current production products, including major brands. In fact, defective capacitors may still be in production as of 2006. To make things worse, there are, as always, many counterfeit parts coming out of the Far East. Many products can be repaired by just replacing the capacitors, but in some cases the capacitors cause the voltage regulators to fail as well. It may be safest to stick to Japanese brands from authorized distributors.
The parts have been widely used on white-box PC motherboards and in many other products. Products sold (but perhaps not made) by major companies are involved as well, IBM, HP, and others. Why? The difference in cost between a brand name part with a history and a no-name part is pennies. The ASUS motherboard just I bought uses all Japanese electrolytics. A few companies, IBM, the PC board maker ABIT, and capacitor maker Jackon, for example, have "come clean". Other companies deny everything to the point of threatening people with lawyers.
For more information see below. Other links I had have gone 404:
http://www.niccomp.com/taiwanlowesr.htm http://www.e-insite.net/electronicnews/ obsolete?http://www.reed-electronics.co...cfd=1 http://www.burtonsys.com/bad_BP6/story4.html http://www.ec-central.org/maga...p.pdf http://www.badcaps.net/ The motherload of "bad cap" stuff. Sells caps and cap kits for motherboard repair. Includes a list of know bad brands (about 30) in the Forums section. http://www.burtonsys.com/bad_BP6/story1.html http://www.motherboardrepair.c...=home Sells caps and cap kits for motherboard repair.http://www.pcstats.com/article...D=195 http://listas.rcp.net.pe/piper....html http://www.spectrum.ieee.org/W....html http://en.wikipedia.org/wiki/C...rands
XXXXXXXXXXXXXXX
OS-CON and other polymer-aluminum capacitors TCNQ aluminum electrolytics are said to have a longer life than most traditional aluminums, but they don´t last forever. See http://www.vishay.com for a pdf file for estimating their life.
Tantalum Electrolytic As with other capacitors, operating temperature and operating voltage have an affect on service life and reliability.
Most tantalum capacitor types tolerate a limited amount of reverse polarity voltage. Manufacturers recommend no more than about 1 volt at 25C and 0.1 volt at 125C, others say no more than 10% of rated voltage at 25C and 1% at 125C. Experience shows that solid tantalums inserted backwards in low-voltage systems (a 5 volt buss for example) are unpredictable and may take from seconds to years to fail. The time to failure is much less at elevated temperature however, and the parts should failure quickly in routine quality control temperature testing.
Solid tantalum capacitors tend to have a problem with surge currents due to impurities and thin spots in the dielectric. Manufacturers typically recommend at least 1 ohm of series resistance per operating volt if the inrush current is not otherwise limited. Tantalum capacitors should not generally be used for power supply filtering unless specifically made and tested for the application (as some are). If you do use a general-purpose tantalum in a power supply application, derate the operating voltage by at least 3:1 (some say 4:1) if possible and/or use soft-start circuitry. Tantalums are often used as output filters for IC voltage regulators in small systems and the regulator will probably limit the turn-on current to a reasonable level. A tantalum on the input side of the regulator may be more vulnerable however. The progressive miniaturization of modern electronics has brought tantalums into more applications where surge is a problem, where a larger aluminum capacitor would previously have been used. A 3 or 4:1 voltage derating has a drastic impact on volumetric efficiency, so manufacturers have had to improve their manufacturing and testing methods. Leakage testing helps catch tantalums with dielectric defects. They sometimes do 100% surge screening to weed out early failures as well. Complaints of tantalums catching fire are common, SMD more often than through-hole.
Some ceramic capacitor makers have specifically targeted SMD tantalums for replacement by large ceramics. This is partly in response to the dissatisfaction of some equipment makers with the reliability of SMD tantalum bypass capacitors (through-hole tantalums seem to be much less troublesome). Problems with tantalums may be vendor specific, but ceramics do have some advantages over tantalums in general. Ceramics are not polarized so there is no fear of putting them on your board backwards. Other advantages are lower ESL and much lower ESR for a given capacitance. In many applications, ESR is more important than bulk capacitance, so a large tantalum can often be replaced by a relatively small ceramic. Ceramics generally fail by shorting however, and can burn your board, just like a tantalum.
Unlike aluminum capacitors, solid tantalums have no significant shelf-life restrictions, but I have heard that long-term moisture absorption can impair self-healing.
Small-value tantalum capacitors are sometimes used in analog circuits, such as for a cheap (and small) way to get very long filter time-constants. Some manufacturers even recommend this. This is risky. Most samples may show acceptably low leakage but a few percent will be much leakier than average and manufacturers will not usually guarantee leakage to a low value. Large-value ceramic capacitors (Class 2 and higher) will also have this problem. It will be OK in some applications, but when in doubt, use a film cap.
Wet-slug (and foil) tantalums use a sulfuric acid electrolyte and have their own problems. Cheaper wet tantalums have silver cases and are sealed with Teflon. The reliability of silver-case tantalums has always been problematical. They have almost zero tolerance of reverse bias because the silver grows dendrites that cause rapid damage. Also, the Teflon seal can allow the slow loss of electrolyte, or even outright leakage. Hobbyists should beware of old military-surplus parts. Better quality wet tantalums have tantalum cases and tantalum-glass seals. Leakage is much less of a problem, and they can tolerate several volts of reverse bias. Wet tantalums will not fail by going up in flames.
The expected failure rate of a solid tantalum capacitor is dependent on both operating voltage and temperature. One formula for estimating solid tantalum capacitor life is:
LT=LR*(ER/EO)3*2(deltaT/10)
Where: LT = failure rate under stated temperature and voltageLR = the failure rate at rated limitsER = rated voltage limitE O = operating voltagedeltaT = difference between rated operating temperature and actual capacitor temperature in C
Note the relatively high dependence of reliability on operating voltage, compared to aluminum electrolytics.
Using the formulas: Numbers for LR, ER , and TR are obtained from the manufacturers data sheets. LR is almost always given for aluminum electrolytic capacitors, and sometimes for tantalum. LR is less commonly found in manufacturer literature for other types unless they are designed for special usage. Note that the formulas do not allow for high surge-current operation. A further complication is that a "failure" might be defined as specific loss in performance that might or might not be of interest in your application, rather than a catastrophic loss of function. This especially true for electrolytics. Manufacturers are not in complete agreement as to these formulas. For example, some say that different shaped of electrolytics should have different lifetimes, all other things being equal.
LR is typically given as 1000 to 5000 hours. This does not sound like much, but remember, it is for worst-case conditions and goes up rapidly for reduced temperatures and voltages. The magic word here is "derating". The further below its maximum rated voltage and temperature you use a capacitor, the longer you can expect it to last. For film and electrolytics, the expected life will roughly double for every 10C below rated maximum operating temperature (a rough but useful approximation). For many applications, with conservative deratings, many designers will not find it necessary to run the numbers. However, some instrument makers take the opposite approach, documenting the stresses on every component and calculating expected reliability.
-
Q45tech
- Moderator
- Posts: 14296
- Joined: Tue Apr 30, 2002 3:19 am
- Car: 1990 Q45 342,400 miles 22 years ownership with original engine
1995 G20t 5 speed 334,000 miles 16" 2002 wheels - 205/50/16 Sr20ve vvl
Re: My FPCU has over 350K mi. have a question ? (Q45tech)
All that was so members might understand the failure mode of the BOSE AMPLIFIERS.............the caps eventually start to fail and cause feedback noises and intermittents in gain.
Everything with capacitors will fail eventually unless the caps are routine replaced with high quality units.
NASA Sat Flight grade units are 100 times more expensive than used in automobiles.
Most don't want a $2,500 FPCU.
Since LEGAL LIFE of an automobile is 10 years or 100,000 miles SAY 5,000 hours of power on operation, no engineer gets faulted for designing things that last that long.
Why things can last 300,000 miles at 60 mph or 100,000 miles at 20 mph.
As to Bose besides ambient temperature it depends on how loud [voltage] and the type of music one listens to [duty cycle].
Hitting max loudness Bass notes immediately in a cold door can stress the OLD caps until things warm up..........gradually.
Everything with capacitors will fail eventually unless the caps are routine replaced with high quality units.
NASA Sat Flight grade units are 100 times more expensive than used in automobiles.
Most don't want a $2,500 FPCU.
Since LEGAL LIFE of an automobile is 10 years or 100,000 miles SAY 5,000 hours of power on operation, no engineer gets faulted for designing things that last that long.
Why things can last 300,000 miles at 60 mph or 100,000 miles at 20 mph.
As to Bose besides ambient temperature it depends on how loud [voltage] and the type of music one listens to [duty cycle].
Hitting max loudness Bass notes immediately in a cold door can stress the OLD caps until things warm up..........gradually.
Re: My FPCU has over 350K mi. have a question ? (Q45tech)
Now that the word Bose has been invoked, I'm curious if anyone still repairs these units?
I have been on the fence as of late whether to rip out these units with extreme prejudice and build a new system or try to replace the geriatric capacitors.
Thoughts?
BTW - most excellent response Tech!
I have been on the fence as of late whether to rip out these units with extreme prejudice and build a new system or try to replace the geriatric capacitors.
Thoughts?
BTW - most excellent response Tech!
-
Q45tech
- Moderator
- Posts: 14296
- Joined: Tue Apr 30, 2002 3:19 am
- Car: 1990 Q45 342,400 miles 22 years ownership with original engine
1995 G20t 5 speed 334,000 miles 16" 2002 wheels - 205/50/16 Sr20ve vvl
Re: My FPCU has over 350K mi. have a question ? (66mgb)
Somewhere on Web are pictures and specs [bill of materials] for replacing the caps...............don't buy from radioshacko.............get some quality high temp units in a 2x oem voltage rating [50v for 25v..........100v for 50v].
http://www.cdplayerrepair.com/fee_&_warranty.htm
Most places charge $75-$100 per speaker/amp
NOTE THE 16 volt caps not enough margin for a system that can reach 14.4 volts [13.8] from alternator same with 105C rated need higher temp units.
high temp. (105 degrees) caps:
Black Nichicon 3X 16v 10µf
Black Nichicon 3X 50v 1µf
Green Nichicon N1322X 16v 47µf
Brown Nichicon H91282X 16v 820µf
Ten Caps total.
"The 820uf caps are hard to find, I always replace them with 1000uf" some web guy
http://www.cdplayerrepair.com/fee_&_warranty.htm
Most places charge $75-$100 per speaker/amp
NOTE THE 16 volt caps not enough margin for a system that can reach 14.4 volts [13.8] from alternator same with 105C rated need higher temp units.
high temp. (105 degrees) caps:
Black Nichicon 3X 16v 10µf
Black Nichicon 3X 50v 1µf
Green Nichicon N1322X 16v 47µf
Brown Nichicon H91282X 16v 820µf
Ten Caps total.
"The 820uf caps are hard to find, I always replace them with 1000uf" some web guy
-
maxnix
- Posts: 22627
- Joined: Mon Jul 22, 2002 8:11 pm
- Car: 1995 Infiniti Q45
1995 Infiniti Q45t
2000 Infiniti Q45
Re: My FPCU has over 350K mi. have a question ? (66mgb)
http://carstereohelp.com/bose.htm66mgb wrote:Now that the word Bose has been invoked, I'm curious if anyone still repairs these units? !
Re: (CrimsonQ)
Thats not an answer, that's a book!CrimsonQ wrote:god damn, now thats an answer!
-
jamesmost
- Posts: 1963
- Joined: Tue Jul 08, 2003 10:16 am
- Car: "95Qmodded, Benz wagon 4matic , 1986 MB 560sec
Re: (superuber)
sure hope that was a cut and paste.So should i swap the FPCU ot keep it in there ??
Re: (Q_SHIP)
Interesting thought. So, james, you were looking for a yes or no. Good luck!We have been using diagnostic dice for years! Maybe my answer wasn't what you were looking for but atleast you didn't waste time readin all them words.Q_SHIP wrote:Q45tech is smart.......I just don't understand a thing he is saying.
http://www.creativedice.com/productinfo.html
