Haven't noticed this on the cheap caps I buy.
I needed 100v caps for my slewmaster and ended up buying the cheapest I could find Samwah 22 000uf.
All 8 measured 24 000 uF on my esr meter.
... and you have not complained that they are way out of nominal value I guess?
Enzo:
Agreed! Im going to make a breadboard test setup which charges and decharges some off the capacitors, to see if the messured value alters.
JMfahey:
Mouser has a fine 30days return policy. Haveno problem in paying for return shipment. Time (money) spent to repair/reviwe the 2 amplifiers has been quite a bit.
The design of the amplifiers is not up to alteration, all parts replaced yet, has been original diodes/transistors. And yes, main amp boards got 2 banks of 10*220uF. Of cource is the lack of 300uF no big issue, but why accept an "downgrade" capacity wise when using new parts? Im trying not to "overthinking", im just wandering, is this the general issue for new elektrolytic capacitors.
Jensen567:
I prefer to think that Mouser isnt calculating on delivering 2 or 3 sort of new parts. why should they? My first post showed my messurements of different capacitor values, with the same "lack" of capacity. ok, perhaps i should eksersise some of the capacitors and evaluate the result. They are still going to be inside spec, but my guess is, with the same "to low" value. My 2 cent is - why accept, as an consumer, that parts perhaps knowingly is being "downgraded" due to an wide value definition. The best el capacitor value estimate is ( what iwe seen yet) -10/+30%.
Elvee:
Have messured both Cs and Cp, with no major differences ( again isnt nearby an english dictionary - sorry for my bad english)
Agreed! Im going to make a breadboard test setup which charges and decharges some off the capacitors, to see if the messured value alters.
JMfahey:
Mouser has a fine 30days return policy. Haveno problem in paying for return shipment. Time (money) spent to repair/reviwe the 2 amplifiers has been quite a bit.
The design of the amplifiers is not up to alteration, all parts replaced yet, has been original diodes/transistors. And yes, main amp boards got 2 banks of 10*220uF. Of cource is the lack of 300uF no big issue, but why accept an "downgrade" capacity wise when using new parts? Im trying not to "overthinking", im just wandering, is this the general issue for new elektrolytic capacitors.
Jensen567:
I prefer to think that Mouser isnt calculating on delivering 2 or 3 sort of new parts. why should they? My first post showed my messurements of different capacitor values, with the same "lack" of capacity. ok, perhaps i should eksersise some of the capacitors and evaluate the result. They are still going to be inside spec, but my guess is, with the same "to low" value. My 2 cent is - why accept, as an consumer, that parts perhaps knowingly is being "downgraded" due to an wide value definition. The best el capacitor value estimate is ( what iwe seen yet) -10/+30%.
Elvee:
Have messured both Cs and Cp, with no major differences ( again isnt nearby an english dictionary - sorry for my bad english)
I give up..
I ship the capacitors back to Mouser minnesota ware house, because online ordering gives an 30days return policy on unused parts (Reason makes no difference)
Then just feel free to order panasonic Fc series capasitors, from Mouser, they have a much closer range of -15 to-10% (5% uncertanty) below the promised value. Just plain better than the -+20% (40% range). Then you know what the stake is!!
If that deal seems ok, then go for it!
This has nothing to do with Mouser or even Panasonic. Its more the general perspektive. El capacitors comes with a wide definition, but buying parts which obviously is all manufactured at least 10% lover value than promised, cant be ok from my point of wiew. Its just not one value...
Sorry for my thoughts, perhabs im mistanken in knowledge, but never the less!
None intent to provocate, just wondering!
Last post from my side!
I ship the capacitors back to Mouser minnesota ware house, because online ordering gives an 30days return policy on unused parts (Reason makes no difference)
Then just feel free to order panasonic Fc series capasitors, from Mouser, they have a much closer range of -15 to-10% (5% uncertanty) below the promised value. Just plain better than the -+20% (40% range). Then you know what the stake is!!
If that deal seems ok, then go for it!
This has nothing to do with Mouser or even Panasonic. Its more the general perspektive. El capacitors comes with a wide definition, but buying parts which obviously is all manufactured at least 10% lover value than promised, cant be ok from my point of wiew. Its just not one value...
Sorry for my thoughts, perhabs im mistanken in knowledge, but never the less!
None intent to provocate, just wondering!
Last post from my side!
Last edited:
I don't understand what the fuss is about. If a design specifies 220uF then that means the circuit will probably be happy with anything from about 150uF to 300uF. A '220uF' cap measuring at 200uF is within tolerance and fine for the circuit. The only complaint could be if the maker claims that his caps have a broad spread around the nominal value (e.g. a rectangular or normal distribution) but no such claim has been made.
Gentlemen, I can not believe what people, some of them probably respected engineers,
write in this thread.
First let us assume the specification of the caps delivered is really +/-20% of nominal
value, second that the measurement was accurate. But I don't know.
Jmlyd told us that he ordered up to 40 of each. We don't know how many of each type
and one may argue that the number is too low for statistical evidence.
But the observation that all these caps were under nominal value is remarkable.
True that every single capacitor is within spec, but the batch is not.
If you check a batch with large enough number you can expect the same number below
and above nominal value and the mean value of a large number should be the nominal
value. If all samples are below nominal value this can be an indication of a problem
even if the batch is small.
If you gamble you can expect to get hits of more than three half of the time. If you
always get only one, two or three points there is surely a problem with the dice.
And it is not true that you should expect lower ratings or capacitance from a 105
degrees part compared to 85 deg.
So the customer may argue that the delivery was not up to specification. If all sausages
are too short, but still within, the customer will feel cheated. Will somebody show up
here and say it is better for your health this way? That is not the point in selling caps
and sausages.
It may be true that an aberration like this is not audible. But audio and diy is not only
about this but also about precision, trust, reliability and doing things as good as you can.
I ordered 20 Roederstein caps 470uF 105deg last week - it was necessary, usually I
prefer my old stock. Will not tell the source here. I checked these with my Philips
PM6303 LCR and the highest capacitance was 430uF. Many parts showed under 400uF
and two or three were under spec, it was +/-20% here. The batch is too small, but the
assumption that something is wrong is surely justified. I have not seen deviations like
this before, only with electrolytic caps of a certain origin after 25+ years of storage.
I did not have a good feeling about this. True -nearly- every single cap was inside the
range, but if the batch is consistently low there may be a problem in production or
storage.
And no, it is not more economical to build the caps to slightly lower capacitance. If
somebody argues this way I want to see a publication and some agreement in the
standards committees.
No again, I predict that the capacitance will not go up when they are in use.
write in this thread.
First let us assume the specification of the caps delivered is really +/-20% of nominal
value, second that the measurement was accurate. But I don't know.
Jmlyd told us that he ordered up to 40 of each. We don't know how many of each type
and one may argue that the number is too low for statistical evidence.
But the observation that all these caps were under nominal value is remarkable.
True that every single capacitor is within spec, but the batch is not.
If you check a batch with large enough number you can expect the same number below
and above nominal value and the mean value of a large number should be the nominal
value. If all samples are below nominal value this can be an indication of a problem
even if the batch is small.
If you gamble you can expect to get hits of more than three half of the time. If you
always get only one, two or three points there is surely a problem with the dice.
And it is not true that you should expect lower ratings or capacitance from a 105
degrees part compared to 85 deg.
So the customer may argue that the delivery was not up to specification. If all sausages
are too short, but still within, the customer will feel cheated. Will somebody show up
here and say it is better for your health this way? That is not the point in selling caps
and sausages.
It may be true that an aberration like this is not audible. But audio and diy is not only
about this but also about precision, trust, reliability and doing things as good as you can.
I ordered 20 Roederstein caps 470uF 105deg last week - it was necessary, usually I
prefer my old stock. Will not tell the source here. I checked these with my Philips
PM6303 LCR and the highest capacitance was 430uF. Many parts showed under 400uF
and two or three were under spec, it was +/-20% here. The batch is too small, but the
assumption that something is wrong is surely justified. I have not seen deviations like
this before, only with electrolytic caps of a certain origin after 25+ years of storage.
I did not have a good feeling about this. True -nearly- every single cap was inside the
range, but if the batch is consistently low there may be a problem in production or
storage.
And no, it is not more economical to build the caps to slightly lower capacitance. If
somebody argues this way I want to see a publication and some agreement in the
standards committees.
No again, I predict that the capacitance will not go up when they are in use.
Big assumption there and very wrong. Most manufacturing processes are run to a very skinny sigma, so all units from one batch are likely to be clustered together. Over time the production might drift and when it gets close to one limit will we adjusted. But getting a proper bell curve distribution within limits from a batch would suggest a very poorly optimised production line these days.
Absolutely agreed. Besides that, this whole thread is much ado about nothing other than unwarranted insecurity on the part of the OP. Even if he's correct on the values being off ..and the jury is still a long way out on that one..., those capacitors will work perfectly fine in his amplifiers.
Indeed, the whole idea of a batch is that all member parts are pretty much identical. Batch-to-batch differences are much greater.
The title of this thread is a shame. Calling something a scam out of ignorance is plain wrong. If I were Mouser I'd sue him!
Jan
billshurv and jan didden are right, of course.
I run a small MI amplifier and speaker factory for 46 years now and never ever had a problem because of lack of tolerance parts.
Lots of problems with fake transistors but that's something else.
Leo Fender even wrote in his old schematics roughly that "everything is +/- 20%" and they are still the reference in MI amplification.
Now and then I see some new builder agonizing because : "schematic calls for 250k pots and all he gets is 220k" or "need .05uF caps and only get .047 ones " or "am repairing a Peavey amplifier with 5000uF caps and all I find is 4700uF ones" while in this case there even is NO real difference, just that old parts roughly followed a "2-5-1" series (.02 - .05 - .1) (we are talking WW2 era specs) while modern follows an exponential scale, with fixed multiplier steps, so we get .022 - .047 - .1 - etc.
Certainly nobody *expects* that a .047uF cap is *exactly* .047 and not "somewhere close".
Of course, in the very very limited case where it does matter (generally circuits which must be tuned to a certain frequency or Lab/Test equipment) , feel free to get 1% parts (or better) or add small value caps in parallel as a final adjustment.
In either case, be ready to accept higher manufacturing costs, either on precision parts or in Tech time adjusting things ..... but ....... for standard series production stuff where it's not critical?
It's one nice way to (commercially) shoot oneself in the foot.
I run a small MI amplifier and speaker factory for 46 years now and never ever had a problem because of lack of tolerance parts.
Lots of problems with fake transistors but that's something else.
Leo Fender even wrote in his old schematics roughly that "everything is +/- 20%" and they are still the reference in MI amplification.
Now and then I see some new builder agonizing because : "schematic calls for 250k pots and all he gets is 220k" or "need .05uF caps and only get .047 ones " or "am repairing a Peavey amplifier with 5000uF caps and all I find is 4700uF ones" while in this case there even is NO real difference, just that old parts roughly followed a "2-5-1" series (.02 - .05 - .1) (we are talking WW2 era specs) while modern follows an exponential scale, with fixed multiplier steps, so we get .022 - .047 - .1 - etc.
Certainly nobody *expects* that a .047uF cap is *exactly* .047 and not "somewhere close".
Of course, in the very very limited case where it does matter (generally circuits which must be tuned to a certain frequency or Lab/Test equipment) , feel free to get 1% parts (or better) or add small value caps in parallel as a final adjustment.
In either case, be ready to accept higher manufacturing costs, either on precision parts or in Tech time adjusting things ..... but ....... for standard series production stuff where it's not critical?
It's one nice way to (commercially) shoot oneself in the foot.
+1 for the above four posts.
In addition, as_audio has got it wrong!
If the cap manufacturer had a "premium" grade, say +-5%, these would all be selected out at component test so would not appear in the 20% bin!
So even if the measured values showed a normal distribution, and they usually don't, there would be a great hole in the bell curve.
In addition, as_audio has got it wrong!
If the cap manufacturer had a "premium" grade, say +-5%, these would all be selected out at component test so would not appear in the 20% bin!
So even if the measured values showed a normal distribution, and they usually don't, there would be a great hole in the bell curve.
+1 for Cliff's comment. I encountered this issue often in particular with film capacitors when designing EQ at a former employer, because of selection the statistical distribution of parts we could get was bi-modal with all of the close to nominal value parts having been selected out by the manufacturer for specific (and probably better regarded) clients.
Where did the maker claim to provide a full spread of values?as_audio said:
No.
There are specific EU rules about food weight, which differ from the rules which the UK used to use. I believe the EU rules say that the average (over sufficient number of samples) must equal the stated weight, with no huge discrepancies; UK rules used to say that each sample must equal or exceed the stated weight. I am not aware of any rules about capacitor value. The two issues are not comparable, because a short sausage provides less food value but a smaller cap value is still adequate for the purpose intended.
Hi Jmlyd,
You really have to accept the advice from all the professionals who posted in your thread. I'll side with them, and I'll add that I was warranty service for Luxman, and have worked on those very amplifiers. Your new capacitors are fine, use them and smile. What would you do if the next brand measures low as well, send them back too?
Any design that is critical for the value of electrolytic capacitors and uses common brands has a faulty designer. That's a fact. You can get closer tolerance parts, and they generally cost more. I think that part of this is you trying to save face a bit.
Relax. Learn your trade, accept knowledge from those who know what they are talking about. We are passing knowledge of the trade on to you. Another thing I'll say ... this isn't the right kind of amplifier for you to work on while you are young in your career. You haven't the experience yet. I wish there was an older audio technician there to help you. The most common destructive errors young techs make? Please, please do not over-tighten any screws, and power transistors have torque values for proper mounting. They are less than what you might think they are. Tighter is definitely not better. There is so much to learn, and you are just beginning.
Hi as_audio,
You also need to learn a lot. I'll give you the same advice. Seek knowledge from others more experienced. You can't listen when you are talking.
-Chris
You really have to accept the advice from all the professionals who posted in your thread. I'll side with them, and I'll add that I was warranty service for Luxman, and have worked on those very amplifiers. Your new capacitors are fine, use them and smile. What would you do if the next brand measures low as well, send them back too?
Any design that is critical for the value of electrolytic capacitors and uses common brands has a faulty designer. That's a fact. You can get closer tolerance parts, and they generally cost more. I think that part of this is you trying to save face a bit.
Relax. Learn your trade, accept knowledge from those who know what they are talking about. We are passing knowledge of the trade on to you. Another thing I'll say ... this isn't the right kind of amplifier for you to work on while you are young in your career. You haven't the experience yet. I wish there was an older audio technician there to help you. The most common destructive errors young techs make? Please, please do not over-tighten any screws, and power transistors have torque values for proper mounting. They are less than what you might think they are. Tighter is definitely not better. There is so much to learn, and you are just beginning.
Hi as_audio,
You also need to learn a lot. I'll give you the same advice. Seek knowledge from others more experienced. You can't listen when you are talking.
-Chris
I am just surprised that you're even giving it this much thought to begin with Jmlyd what a waste of your personal time. When you originally bought the caps surely you noticed the tolerance spec +-20% or whatever. So surely, as soon as you measured them and found them to be within spec, you should have thought, well these measure within the boundaries of what was to be expected, lets just use them...like any normal person would have done.
I have measured hundreds of components over the years and yeah some show up short, some are the opposite. The real truth here is that your FC caps could have measured even lower than you encountered and would have still been within factory specification. There would be nothing wrong with them and if you were a manufacturer buying in bulk directly from Panasonic, Panasonic would laugh at you for wanting to return them and advise that in the future you buy from a range that specified +-5% if it's that important to you.
I have measured hundreds of components over the years and yeah some show up short, some are the opposite. The real truth here is that your FC caps could have measured even lower than you encountered and would have still been within factory specification. There would be nothing wrong with them and if you were a manufacturer buying in bulk directly from Panasonic, Panasonic would laugh at you for wanting to return them and advise that in the future you buy from a range that specified +-5% if it's that important to you.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.