• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Low-side capacitor production: has it changed how you choose recap values?

Various thoughts; none proven, and in no particular order.

1) Then and now, a designer has to have minimum and maximum limit, not typical values. If you don't provide those you won't sell any caps to manufacturers.
2) Back then, tolerances were likely set based on what the manufacturing process could produce a high percentage of the time. +/- 20% is good enough that you'd never notice it in any power supply or coupling application. Same with -20 to +80%, though I think that's from the bad old FP can cap days.
3) No designer in his/her right mind sets any time constant that matters with an aluminum electrolytic. An aluminum electrolytic cap is not a precision component.
4) Now, very unlike then, a manufacturer might ask capacitor company to show the statistical distribution of various parameters. This became more common with ICs back when Linear Technology started doing it.
5) The capacitor production process has certainly improved. How much? Not a clue and they probably won't tell you.
6) If you have a process with a narrow statistical distribution and you can save money on foil and paper, why would you not build parts close to the minimum tolerance? How would you explain to management that you weren't going to do that? Especially if the caps met all their specs, including lifetime under worst case conditions?
7) Measuring a bunch of parts today, it would probably be possible to estimate the answer to #5.
8) There are many ways to measure electrolytic caps but getting the official pass/fail numbers requires specific procedures and methods. Most common test equipment isn't ideal for this. The stuff that is tends to be expensive! OTOH, unless you're taking the company to court, most of our more common equipment is plenty good for our needs.
 
EchoForge said:
That makes sense for checking the RIAA curve. But have you ever tested whether the actual capacitance value makes any measurable difference to the noise performance?
For example, comparing two electrolytics of the same nominal value and series, one measuring near the low end of tolerance and the other near the high end, in the exact same circuit, then comparing the noise floor or SNR ?
Click to expand...
electrolytics are not used in the riaa curve, 'lytics are used at the output of the phono stage and usually on the input unless the input has jfets .

if you used a higher rated 'lytic in those positions then you are going to be lowering the freq. cutoff and can potentially allow more low freq. to pass. unless i have a phono stage that rolls off the bass too high, i stay with the value. lowering the value will raise the cutoff freq.

so actually, if your ears were sensitive enough and you installed a cap with a lower value, it would slightly raise the bass roll off and could be perceived as less noise.(if you had a noisy turntable or listen at higher volumes and were picking up feedback) in theory...in my head and my understandings of how the circuits work.

i do not think that many people have golden ears that can tell the difference between a 2.2uf cap or a 2.0uf cap on the input of a phono stage.

the better units tend to use low leakage 'lytics on the in/out of the phono stage. its best to stick with those to keep noise down.

i've done a bit of tweaking of phono stages in different units in the past few months and my ears can not hear a difference in noise or snr and while my test gear can measure it, with the nature of the cabling and an inverse riaa pcb and my flourescent lighting in the shop, i gave up trying to measure noise.


my thoughts, however wrong they may be are this..and this is totally overthinking the situation and a totally worse case possibly even fantasy situation.

a manufacturer designed a circuit and gave it a tolerance of +-20% for a capacitor

modern manufacturers are shipping out caps at -15% of their rated value(say 85uf for a cap thats rated at 100uf)....they cover their asses saying that the capacitor will remain within +-20% of the value that was first measured. woopdee ****in do. THEY are covered

but what about the circuit that it's going in? it's designed for 100uf but it can tolerate an 80uf cap....here you are putting in a brand new 100uf cap that is only actually 85uf. technically, that should not affect the circuit but what happens when the cap starts to age and the value drifts low? if that cap drifts 6% low with age, isn't it now out of spec of the circuit?

i think this will affect perceived lifespan of a cap. while it may be within a reasonable lifespan according to the manufacturer. that technically 'good" cap is no longer good for that circuit. meaning, they will sell us another cap with a depleted value...for as long as they are making thru hole caps anyway.

it totally depends on the location of the cap in a circuit and IF the circuit really even cares.

i'm no e.e., i just wonder alot and think about useless things too much.

if said capacitor even lasts 1/2 as long as they did from the 70s, i'll be dead and won't ever see if my theory is correct.

either way, i think it's crappy that manufacturers are doing this. they care nothing about the customer or the unit the caps are going in. they only care about how many tenths of a cent profit they can squeeze out of every capacitor and billions and billions of capacitors made equals a new jet for the ceo.
 
ConradH said:
..
5) The capacitor production process has certainly improved. How much? Not a clue and they probably won't tell you.
...
Click to expand...
Based on the consistency of reports it seems like they have it down pretty well. Exactly how well certainly is an open question outside the mfgs. but "pretty well" seems apt. Else, as mentioned before, simple mfg tolerance around a nominal I think it would seem one should find some on the high side but at least anecdotally that seems rare for new caps.
 
The noise topic is interesting. Some Japanese service manuals referred to "low noise" capacitors. I've never quite grasped that, thinking a capacitor is its own filter for noise. It shouldn't be an issue. OTOH, where does noise come from? First, a pure reactance can't generate noise unless it's from external excitation (I think). The loss term, the ESR, however can/will generate thermal noise. One could argue that low ESR capacitors might have lower noise and that seems to be what the Japanese designers specified in some phono applications. I always wondered if there was a translation issue that just kept getting repeated. Similar applies to EPR, the DC leakage term. My solution is just don't use leaky high dissipation factor caps!
 
IMO, those are the things we should pay more attention to. After all, we make sound and vibrate things! It would be interesting to attach a transducer (speaker or shaker) to amps and preamps, excite them and see what shows up in the output. My guess is, more than we might hope.
 
EchoForge said:
Is the goal simply to bring the unit back within its original specifications, or is there also value in improving measurable performance where possible, while preserving the original character and intent of the design as much as possible?
Click to expand...
There are different ways to ‘restore’ this gear. Some use many newer, better parts to try to make the unit better than when new. Others just change like for like and mostly just the electrolytic caps. Lots of variation between these two almost extremes.
 
So, if I understand what is being posited here, modern manufacturing methods have made it possible to produce capacitors that are more consistently at the lower end of traditional tolerances, thus saving on materials.

Sort of like capacitor shrinkflation. And did all the manufacturers collude on this or did it just sort of happen?
 
Blue Shadow said:
There are different ways to ‘restore’ this gear. Some use many newer, better parts to try to make the unit better than when new. Others just change like for like and mostly just the electrolytic caps. Lots of variation between these two almost extremes.
Click to expand...
Absolutely. I’m not looking for a simple good/bad or acceptable/unacceptable answer, and I’m certainly not trying to judge how anyone chooses to restore equipment.

When I described one approach as pragmatic, I meant the reasoning behind it, not as a value judgment. I don’t think there is a one-size-fits-all restoration philosophy anyway. The appropriate approach depends on the class of equipment, available resources, the skill and standards of the technician or shop, and also commercial realities such as the owner’s budget and expectations.

What interests me more is whether the apparent changes in modern capacitor production have actually been incorporated into restoration practice. What, if anything, has been tested, adopted, or changed in response, and what measurable impact does that have?

One point raised here that I found particularly interesting concerns aging margin. Suppose a circuit was originally designed to tolerate a 100 µF capacitor down to 80 µF. If a modern replacement marked 100 µF starts life around 85 µF, it is perfectly within tolerance. But if its capacitance then drifts downward with age, it has much less margin before crossing the circuit’s lower acceptable value than a part that started near nominal.

That does not necessarily mean the modern capacitor itself has poorer reliability. But from the circuit’s point of view, its useful margin may be smaller simply because it started much closer to one boundary.

For one capacitor that may be insignificant. What I find interesting is what happens after a full recap if this low-side starting point is common across much of the unit. Could that have any measurable effect on the long-term maintenance interval or effective service life of the equipment, even if every replacement part was fully within specification when installed?
 
actuator said:
So, if I understand what is being posited here, modern manufacturing methods have made it possible to produce capacitors that are more consistently at the lower end of traditional tolerances, thus saving on materials.

Sort of like capacitor shrinkflation. And did all the manufacturers collude on this or did it just sort of happen?
Click to expand...
I wouldn’t go as far as calling it capacitor shrinkflation, at least not based on what we have so far.

What I’m seeing is simply that many modern electrolytics I’ve measured tend to cluster fairly tightly below nominal rather than appearing randomly distributed around it. ConradH’s point about tighter manufacturing control and the economic incentive to produce closer to the lower tolerance limit is certainly plausible, but at this stage it is still an explanation, not something we have demonstrated.

You may also want to read Blue Shadow’s thread linked below. That discussion, and particularly the measurements reported there, is actually part of what prompted me to start this thread in the first place.

Blue Shadow

Thread 'New large caps are being manufactured below specified value. Request for measurements'

After talking to a Nichicon Sales Engineer and finding out that Nichicon is producing higher capacitance caps at 10% below the rating (well within the ±20% tolerance) I asked if folks could measure some of the new caps they are using. Many can't measure caps over a hundred or so µF and most will not have the ability to measure at rated voltage. I'm not interested in peer reviewed scientific journal test results, just something to get an idea of who is doing what. The Nichicon guy indicated it is industry standard now to short us on capacitance because the manufacturing process is so...
  • Like

That is also one of the reasons I think measuring a larger sample of current-production capacitors could be interesting.
 
EchoForge said:
Absolutely. I’m not looking for a simple good/bad or acceptable/unacceptable answer, and I’m certainly not trying to judge how anyone chooses to restore equipment.

When I described one approach as pragmatic, I meant the reasoning behind it, not as a value judgment. I don’t think there is a one-size-fits-all restoration philosophy anyway. The appropriate approach depends on the class of equipment, available resources,
Click to expand...

I resemble that remark. LoL. I have a diverse range of gear from stuff like Levinson and Audio Research to really cheap TPA3116 amp boards, and between.

I'm not a restorer, per se. If something breaks I fix it. Sometimes I build stuff. Occasionally I've done a recap type thing.

What I don't do is get my knickers in a twist over values of components spec'd generally the same. If something specifically calls for matching like the JFETs did for my PassDIY F6 and B1 buildS I do it. I don't idolize Pass or others but if it's good enough for them it's probably gonna pass muster for me.

Precisely to your point, pragmatic.
 
You must log in or register to reply here.
Back
Top Bottom