My nad c372 integrated has two bulging capacitors (1000uf 100v). I’m planning to replace them with Nichicon LKG series caps. And four power capacitors(10000uf 80v) looks fine visually. Should I replace them as well or let it be. If needed it’ll be replaced with elna lao series ones, will it be a possible upgrade in terms of audio quality ?. Also the ones that comes from nad are rated at 105c and the ones that I can find are rated at 85c. Is it okay to go with lower temp ones ?
Thankyou
Thankyou
Any electro more than about 10 years old should be replaced. They dry out. Leakage goes up, ESR may go up. Basically, they become not a very good cap.
Replacing restores it to original condition. No magic. No improvement over stock.
But why the initial failures? I would be concerned with that.
85C should be fine.
Replacing restores it to original condition. No magic. No improvement over stock.
But why the initial failures? I would be concerned with that.
85C should be fine.
Oxygen attacks elastomer seals in e-caps, working or not.
High temps can also shorten seal life.
Ecaps can be bought in different grades, from 500 hours service life or unspecified, up to 10000 hours @ 105 C on small ones. I can only find ~3000 hours service life on caps over 2200 uf. If your vendor won't tell you the service life of the replacement caps, shop somewhere else. I find farnell and &digikey have service life available in the selector table.
I've used 85 deg C long life caps in power amps and organs without failures over the 10 years since I started upgrading electronics.
You can measure rail cap performance by measuring max wattage on speaker versus that specified. An analog voltmeter with AC scale is required, or scope or RMS dvm. $12 pointer VOM from hardware usually suffices. P=(V^2)/Z where Z is speaker nominal impedance. Impedance is usually ~4/3 speaker resistance if you don't know it.
I'm not surprised by e-cap failure in any electronic product over 15 years old. 5 years for consumer grade **** like TV accessories or computer supplies. The only vendor known for using epoxy sealed e-caps was fender in the seventies. Green CDE's.
High temps can also shorten seal life.
Ecaps can be bought in different grades, from 500 hours service life or unspecified, up to 10000 hours @ 105 C on small ones. I can only find ~3000 hours service life on caps over 2200 uf. If your vendor won't tell you the service life of the replacement caps, shop somewhere else. I find farnell and &digikey have service life available in the selector table.
I've used 85 deg C long life caps in power amps and organs without failures over the 10 years since I started upgrading electronics.
You can measure rail cap performance by measuring max wattage on speaker versus that specified. An analog voltmeter with AC scale is required, or scope or RMS dvm. $12 pointer VOM from hardware usually suffices. P=(V^2)/Z where Z is speaker nominal impedance. Impedance is usually ~4/3 speaker resistance if you don't know it.
I'm not surprised by e-cap failure in any electronic product over 15 years old. 5 years for consumer grade **** like TV accessories or computer supplies. The only vendor known for using epoxy sealed e-caps was fender in the seventies. Green CDE's.
Last edited:
That is not entirely accurate. I've repaired and restored several old amps made any where from late 70's to recent years, and found out that a lot of caps after 10-30 years of service fall into datasheet specs or better. So, if you can measure capacity, esr and leakage which any of decent LCR meters will be able to do, measure and decide if caps require replacement. It worth to measure before spend money on particularly large ones.
That could be a problem if it is only few years, let it on for a few hours then measure. Sometimes (rarely) caps can bulge on its own once I had brand name caps (sourced from digikey) bulged while in storage. I was trying 3 different series of caps in the small amp, so one in question had like 50-100 hours in the amp then got replaced by another series of the same brand. Pulls were put in storage for future use bu when I come back to the it year later I noticed that caps bulged (badly).
In any case bulged replace bulged caps with same and observe or chose in that place something with higher ripple current rating. if bulged ones where 105C see if it justified by external heat source (they are placed near the heatsink, etc) and if id does replace with 105C caps otherwise 85 is OK. Main caps if you replacing them, look if you can increase capacity in the same footprint. Same series, same manufacturer preferred.
year of life for caps specified at max ripple and max temp: so if cap specified to have 2000 hours @105 and ripple of 2.5A, then @35C and same ripple it life will extends for tens of thousands of hours 🙂 I wouldn't even bother to look a this param at all for consumer grade electronics because of conditions consumer electronics used, practically guaranties that caps will work for tens of thousands of hours within specs without any problems.
:facepalm:
:facepalm: Impedance for a speaker is a set of values over the frequency range there are known speakers rated as 8Ohm with impedance drop somewhere on range of few hundreds Hz to 3Ohms.
Nichicon on some series is another 🙂. And there is long life (10k+ hours) series caps with rubber and or plastic seals from variety of manufacturers
I had to replace ecaps 4 times in 14 years for low power or idle bias wildly off in a ST70 amp. ~2000 hours a year use, 2.5 hours on weekdays, 12 hours on weekends. That is way below 10000 hours use, probably 8000 hours. TV parts store caps available then were unspecified for life. The cooler running PAS2 preamp needed ecaps twice in that time.
I doubt if any of these organs that need 100% cap replacement got 10000 hours use. Put 71 in a Hammond H182 and it needs 115 more to even out the decay time on the key sustain. 10000 hours of focussed practice is usually what piano/organ websites say it takes to make an expert. Another H182 (I destroyed case on first one moving it when helper ran away on moving day) got about 1.5 hours per week use in a church near Cinci, 75 hours per year, 25 years, 1875 hours? Nothing works right on it, wattage about 2 instead of 40, no attack, no sustain, pro service tech replaced output tubes (easy) instead of replacing ecaps (hard). An Allen 300 organ bought 1980 went silent in a service 2017, measured 2 watts/channel the next Monday. 1.5 hours per week, 3000 hours use? Allen used really good caps, the allen service people say they never need replacing (WONG). Puts out 70 w/ch now on 8 ohms speaker (100 W specified on 4 ohm speaker).
My farnell HDTV converter got about 6 years before losing function, maybe 6000 hours?I don't watch TV as much as listen to the radio. Coupla new ecaps in the switcher supply brought it back. buy the long life stuff, life is too short to do a job over 4 times.
Last edited:
Nominal impedance is a bogus param only good for very rough approximations. To understand it all you need to do is to look at the impedance curves of almost any commercial or well known diy speaker, for example ported 3 way. Too often in the range where more than 50% of power for real musical program distributed (20Hz to approx 1-2kHz) you will see impedance swing up to 10 times
Impedance is the second to last thing speaker designers looking when designing crossover. In infinite battle of compromises between desired SPL linearity, phase linearity, off-axis responses etc. impedance almost always failing as a victim, kind of don't let it fail to low, don't care about the rest.
Big can-type capacitors don't bulge. The plastic on top is just cosmetic, you should be able to dent it in. They have a vent on the bottom end, so they could leak onto the board. Do not confuse electrolyte with the mounting glue used to hold them down during production.
This sort of cap, if of good quality, tends to last for decades in domestic use. NAD has occasionally used types that were frankly garbage, the C270 had issues with this if memory serves. Normally though, these tend to be among the most long-lived caps for sheer size alone. These main filter caps also tend to lead a fairly boring life more often than not.
This sort of cap, if of good quality, tends to last for decades in domestic use. NAD has occasionally used types that were frankly garbage, the C270 had issues with this if memory serves. Normally though, these tend to be among the most long-lived caps for sheer size alone. These main filter caps also tend to lead a fairly boring life more often than not.
Last edited:
Switching power supply, high frequency, small footprint, cheap caps, all known issues. Same reason it failing in some of the consumer grade routers/switches, in two of mine were replaced by caps with higher rated ripple (not long life) and running 24/7 for the past 8 years. Same with computer motherboard caps mostly on processor supply I had repaired a number of mobo's failed do to the bad caps, replaced with caps with higher ripple rating some of them still in service more than 10 years later.
In solid state amps even from mid 70's I rarely seen failed main caps, almost always in-spec if datasheet can be found 🙂. Another question that some of them they weren't good to begin with 🙂 So it get replaced anyway if authenticity is not a requirement. Smaller caps particularly in hot places near heat-sink or near hot resistor failing way more often.
They do but on extremely rare occasions. I seen it few times. Once it was some brand labeled caps AFAIR Nippon Chemi oem, judging from the indents on the can, and once it was Elna but in than amp I needed to replace a lot of caps in a pre-amp (lost capacitance) it was very weird as it was amp made in 00's
Last edited:
Exactly as sgrossklass points out. The bulging has often been attributed to outgassing but that's not possible if there's no top vent. When hot, the plastic cover disk could possibly bulge with the expansion of trapped air but I think it's likely aided by the force applied to the rim of the plastic cover disc by shrinkage of the PVC sleeving, which would occur fractionally with each cooling/warming cycle. It seems to be a consistent occurrence with most brands/types of large can electrolytics from the past 25 or so years.