Seems to me you measured all, could you share what were bad electrolytics you found in M05 ?
to get upset
you asked if this is normal or not I would say that batch you got has been stored in a hot place for years?? hence why they measure lower in value(leak) ..it is a possibility..
now if you want perfect buy elna all the elna's that were bought measured dead on with the 4k sencore cap/esr tester...IMO elna's do sound better
Elna was/is an oem for many company's been in business forever! they know how to make good long lasting caps
Lawrence
you asked if this is normal or not I would say that batch you got has been stored in a hot place for years?? hence why they measure lower in value(leak) ..it is a possibility..
now if you want perfect buy elna all the elna's that were bought measured dead on with the 4k sencore cap/esr tester...IMO elna's do sound better
Elna was/is an oem for many company's been in business forever! they know how to make good long lasting caps
Lawrence
Jmlyd, consider that you are a design engineer and choose a part with the tolerance +-20%. A good engineer calculate/simulate the consequences to have part at the outer ends. If this is acceptable, then +-20% will do. If your amps requires +-20% and your are getting parts with the range, then you don't have any problems. I'll bet as mentioned before here that some of the positions in your amps had +100% to -50%.
Notice also that the capacitance will probably change a bit with you apply DC voltage. The oxide layer will grow so the capacitance should decrease but this is nothing I have tested.
Notice also that the capacitance will probably change a bit with you apply DC voltage. The oxide layer will grow so the capacitance should decrease but this is nothing I have tested.
Thread title edited at OP request. The thread does not meet forum criteria for locking.As a test, I took a brand new Nichicon 1000uF/35V cap and measured it. Then I charged it up to 30V, let it sit for a minute, then discharged through a 220 ohm resistor.
Before: 885uF, 580mOhm ESR
After: 892uF, 118mOhm ESR
I never knew they changed so much.
Before: 885uF, 580mOhm ESR
After: 892uF, 118mOhm ESR
I never knew they changed so much.
Wow... never thought the ESR could change so much.. 😕
Can somebody explain this phenomenon?
I hardly doubt it! I have many electronice 40+ years old that there capacitors measure just fine with excellent ESR!
IMO everybody doing mods are actually giving there equipment a downgrade!!!
unless there dead leave them alone 😎
Yes you doubt it, buy a 7 year old Marantz, or 10 year old or 15 year old or 20 year old and do report back.
Different to a "mechanical" capacitor, if you allow me that word, where dielectric is a laminated film of known thickness and composition, surface is a very definite parameter too, in electrolytics, specially modern ones:
1) actual dielectric is an aluminum oxide layer, not "rolled" , not "applied" over an aluminum substrate, but *grown* because of chemical action on the aluminum surface.
That alone introduces a ton of variables in the process (time/temperature/PH/concentration/ etc.) which makes end result quite unpredictable, and to boot, chemical layer thickness varies during storage, with applied tension, the works.
Hence the old rating +100%/-40% or +60%/-30% or equally large numbers.
That this (justly respected) Manufacturer could reduce it to +/-20% is amazing, we should count our blessings instead of dissing them.
That the whole batch is kept within 5% or so from a central value, even if average is 10% below *nominal* value is, to me, being an Industrial Engineer involved in Real World production for over 45 years nothing short of amazing, kudos to them.
2) even worse: here surface is not just that of the geometrically calculated surface of the metal strips, but the one *chemically etched* which increases the plain "geometrical" one (width*length) by a large factor, one of the reasons behind the huge Electrolytic miniaturization in the last 20 years or so.
Guess what?: it also introduces even *more* variables , this being a separate and different process than oxide layer creation.
Even more praise to capacitor makers (all of them).
1) actual dielectric is an aluminum oxide layer, not "rolled" , not "applied" over an aluminum substrate, but *grown* because of chemical action on the aluminum surface.
That alone introduces a ton of variables in the process (time/temperature/PH/concentration/ etc.) which makes end result quite unpredictable, and to boot, chemical layer thickness varies during storage, with applied tension, the works.
Hence the old rating +100%/-40% or +60%/-30% or equally large numbers.
That this (justly respected) Manufacturer could reduce it to +/-20% is amazing, we should count our blessings instead of dissing them.
That the whole batch is kept within 5% or so from a central value, even if average is 10% below *nominal* value is, to me, being an Industrial Engineer involved in Real World production for over 45 years nothing short of amazing, kudos to them.
2) even worse: here surface is not just that of the geometrically calculated surface of the metal strips, but the one *chemically etched* which increases the plain "geometrical" one (width*length) by a large factor, one of the reasons behind the huge Electrolytic miniaturization in the last 20 years or so.
Guess what?: it also introduces even *more* variables , this being a separate and different process than oxide layer creation.
Even more praise to capacitor makers (all of them).
About 15 years ago, National Semiconductor was touring none other than Bob Pease around to give talks and promote NS products and such. I attended one of these on Long Island, and perhaps as expected, I was highly amused at some of his slides, which included cute maxims such as "In theory, theory and practice are the same. In practice, they're not." He probably stole that from someone else, but I won't complain, as I never heard it from whatever other source may have originated it, and I never would have heard it if it weren't for Bob repeating it.
But I was shocked at something he said regarding opamp spec sheets, and really any component that any manufacturer produced and provided data on: "DO NOT design using 'Typical' values!" I initially thought why would he say that, but then it immediately came to mind that he (and others at NS) have surely fielded many calls from engineers regarding circuits that didn't work because they relied on the typical values to be the correct ones. "Make sure your circuit works with WORST CASE values [and throughout the range between minimum and maximum]. It's only the minimum and maximum values that a manufacturer guarantees, NOT the typical."
I was in my 40s and had "been an engineer" since college, and I knew these things very well. I suppose I was shocked that other engineers didn't know this, or never noticed. Even on data sheets, there's usually asterisks or footnotes telling which columns of figures are guaranteed [usually min and/or max values] and which are not [typical values!].
What? Is there a spec for the batch? You'd be hard pressed to make a good claim there is even an implied spec for any batch of any size. Without a doubt the legalese on data sheets and/or on manufacturers' websites specifically disclaim any implied specifications, whether regarding average value of a batch or the phase of the Moon.
The data sheet specifications are for each part and ONLY for each part, not for an amalgam or any combination of parts. The engineers and attorneys at the manufacturer know this, have spelled it out in data sheets, and will have no problem defending their specs in court.
Another thing Pease used to stress was that you cannot extrapolate from the data sheet specs. The specs only include whatever is within their borders.
no I'd call it a guy with a screw cutting machine set up on a inch scale....
why so angry? you miss the joke AND the preceding useful idea I gave you on a silver plate.
Gentlemen, its me again. I expected to be be bashed, but not this hard.
Somebody wrote:
Will you agree to teach me?
So in order to learn I wrote to some of the few remaining elcap manufacturers in
Germany asking about some points that came up here.
Manufacturer A replied:
"The starting capacitance at the time of delivery is in the range of 85 ... 95% of the
nominal value in the currently normal tolerance range of +/-20%."
Manufacturer A is a big traditional manufacturer of caps of all varieties.
In a telephone conversation he confirmed that "starting capacitance" is not the minimum
that you can expect on delivery, but that indeed the caps are manufactured in a way so
that all of them have 85 to 95% of the nominal value and this for economical reasons.
He said that there is some agreement among manufacturers in this point, but it is not
explicitly published. He confirmed that the capacitance will not rise in operation (due to
"forming" as claimed somewhere here).
I must confess the fact with vendor A that caps are intentionally manufactured below
nominal value was new to me. It does not help my appreciation of today's business rules.
It was confirmed in this conversation that company A does not manufacture any
electrolytic caps in Germany any more, but somewhere else.
I tried to translate all this as good as I can. Additional complication is that you will
actually exchange letters and have phone calls with business people, not technicians.
So use of the same words does not always mean the same if you don't insist.
A representative of manufacturer B replied: "Our process generally allows for a drift of
up to minus 8% in the tolerance range of +/-20%. " He confirmed that "we offer series
with reduced tolerance spreads also. But these will not be taken out of a current
production."
Manufacturer B is among the smaller companies, a traditional vendor and a known
supplier to the "highend" business. I understand the answer in a way that they aim the
process to be at around the nominal value.
I received a short reply also from manufacturer C simply stating that "our current
production always aims at the nominal value in all electrolytic caps. Biggest spread is
in the production of the raw foil. If yo ask for a reduced spread this can be done by
additional selection".
Manufacturer C is a big traditional vendor of good reputation, now part of an inter-
national conglomerate.
I did not receive a reply from vendor D for holidays reasons, sorry for that. D is also
around for a long time and still producing in Germany. I would love to see this answer
later on.
So it seems my original statements are not justified as far as vendor A is concerned.
I am sorry for this and also for delayed post, the answers above did not arrive earlier.
Somebody wrote:
and somebody else
"Anatech" wrote:
This is true, but how can you know?
I admit I feel privileged to get attention from a most respected forum member like this.
Will you agree to teach me?
So in order to learn I wrote to some of the few remaining elcap manufacturers in
Germany asking about some points that came up here.
Manufacturer A replied:
"The starting capacitance at the time of delivery is in the range of 85 ... 95% of the
nominal value in the currently normal tolerance range of +/-20%."
Manufacturer A is a big traditional manufacturer of caps of all varieties.
In a telephone conversation he confirmed that "starting capacitance" is not the minimum
that you can expect on delivery, but that indeed the caps are manufactured in a way so
that all of them have 85 to 95% of the nominal value and this for economical reasons.
He said that there is some agreement among manufacturers in this point, but it is not
explicitly published. He confirmed that the capacitance will not rise in operation (due to
"forming" as claimed somewhere here).
I must confess the fact with vendor A that caps are intentionally manufactured below
nominal value was new to me. It does not help my appreciation of today's business rules.
It was confirmed in this conversation that company A does not manufacture any
electrolytic caps in Germany any more, but somewhere else.
I tried to translate all this as good as I can. Additional complication is that you will
actually exchange letters and have phone calls with business people, not technicians.
So use of the same words does not always mean the same if you don't insist.
A representative of manufacturer B replied: "Our process generally allows for a drift of
up to minus 8% in the tolerance range of +/-20%. " He confirmed that "we offer series
with reduced tolerance spreads also. But these will not be taken out of a current
production."
Manufacturer B is among the smaller companies, a traditional vendor and a known
supplier to the "highend" business. I understand the answer in a way that they aim the
process to be at around the nominal value.
I received a short reply also from manufacturer C simply stating that "our current
production always aims at the nominal value in all electrolytic caps. Biggest spread is
in the production of the raw foil. If yo ask for a reduced spread this can be done by
additional selection".
Manufacturer C is a big traditional vendor of good reputation, now part of an inter-
national conglomerate.
I did not receive a reply from vendor D for holidays reasons, sorry for that. D is also
around for a long time and still producing in Germany. I would love to see this answer
later on.
So it seems my original statements are not justified as far as vendor A is concerned.
I am sorry for this and also for delayed post, the answers above did not arrive earlier.
Last edited:
Hello as_audio,
Your attitudes are very common for a person in their early years in the electronics business. Idealized viewpoints really only set you up for disappointment. Relax and accept that all companies are in business to make money. Most companies are sensitive to their market share, some more than others. There is a desire to push for profit as well, again some more than others. The way manufacturing is set up is a sterile setting without concern for people new in the field that are new.
I am surprised that manufacturer A gave you a straight answer with minimal dancing around the subject. He is right as I have observed that today's electrolytic capacitors are a little low in actual capacitance, but still within tolerance. Would I rather them aim at the standard value - yes, of course. This is more important with film and foil caps that are often used for filters and timing applications. However, I'll take values closer to each other given a choice.
In practical applications for electrolytic capacitors, the absolute value is usually unimportant. What does make a difference is improved DA and high frequency performance. I'll choose those over electrolytic capacitors close to the published value - again within reason.
-Chris
Your attitudes are very common for a person in their early years in the electronics business. Idealized viewpoints really only set you up for disappointment. Relax and accept that all companies are in business to make money. Most companies are sensitive to their market share, some more than others. There is a desire to push for profit as well, again some more than others. The way manufacturing is set up is a sterile setting without concern for people new in the field that are new.
I am surprised that manufacturer A gave you a straight answer with minimal dancing around the subject. He is right as I have observed that today's electrolytic capacitors are a little low in actual capacitance, but still within tolerance. Would I rather them aim at the standard value - yes, of course. This is more important with film and foil caps that are often used for filters and timing applications. However, I'll take values closer to each other given a choice.
In practical applications for electrolytic capacitors, the absolute value is usually unimportant. What does make a difference is improved DA and high frequency performance. I'll choose those over electrolytic capacitors close to the published value - again within reason.
Thanks you for the kind words, but I am surrounded by more capable people. Just read and absorb all the great knowledge flowing around here between members of our great membership. No one can tutor a member due to time constraints we all face. But you can obtain an excellent education simply by asking questions at the right time. Most people will take the time to give you a brief explanation. Between reading (listening) and asking questions, you will learn a great deal. There is the big difference in how to interact. Ask questions rather than making a declarative statement about things you aren't sure of.
Actually, you weren't bashed. You were corrected and made aware that your views were not in sync with the industry. This is hard to take as it can feel like an attack on your person. Roll with it and come up wiser. When someone is bashed, there will be no question as to what just happened.
-Chris
Thank you for this reply. I feel honored to be estimated as "a person in their early years in
the electronics business". But there is no need for me to relax.
This particular answer from manufacturer A was on the telephone. I can not prove it. If it
conforms to today's common practice this may be one reason for the reputation of this
companies heritage is down today compared to the past.
made statements about things I can not be sure of. Most people do that most of the
time (in this thread also) and it seems to be quite common in forums. When I asked
for support here from time to time the results were low. This does not mean to belittle
the known abilities of the members, but a useful answer was probably not possible
due to unpublished diagrams etc.
I even recommended somebody to use a resistor 2% down off nominal value!
Considering the answers that I received from the elcap manufacturers and re-reading
my post (http://www.diyaudio.com/forums/parts/288738-new-capacitor-value-3.html#post4656036) I do not feel that my "views were not in sync with the industry".
But of course the small survey was limited to "german" companies and it may be
different in a global perspective.
the electronics business". But there is no need for me to relax.
This particular answer from manufacturer A was on the telephone. I can not prove it. If it
conforms to today's common practice this may be one reason for the reputation of this
companies heritage is down today compared to the past.
As seen in my criminal record here I avoided asking questions most of the time but
made statements about things I can not be sure of. Most people do that most of the
time (in this thread also) and it seems to be quite common in forums. When I asked
for support here from time to time the results were low. This does not mean to belittle
the known abilities of the members, but a useful answer was probably not possible
due to unpublished diagrams etc.
I even recommended somebody to use a resistor 2% down off nominal value!
Considering the answers that I received from the elcap manufacturers and re-reading
my post (http://www.diyaudio.com/forums/parts/288738-new-capacitor-value-3.html#post4656036) I do not feel that my "views were not in sync with the industry".
But of course the small survey was limited to "german" companies and it may be
different in a global perspective.
I think the replies you received from the manu's are rather interesting. Thanks for taking the time to get them.
Jan
Jan
- Status
- Not open for further replies.