> one with a ohmmeter. This is a fairly expensive instrument, and so far
> as I know it is very accurate.
snap again
> So here is the interesting part -- every single resister, with no
> exceptions, came in below the nominal value. This is not any kind of
> problem, of course, since they were all within tolerance. But one would
> think (wouldn't one?) that some would be over, some would be under, and
> once in a great while one would be right on the money.
Probably because is OT, you forgot the most important things:
1. which is the tolerance class of the resistors as the seller said ?
2. which is the precision of your "fairly expensive" ohmeter ?
Then will find the truth easily. There is no passive componente
without plus or minus error, just wrong measuring methodes or bad
measuring tools.
But if you want to say that Olin is great, I'm agree.
:)
>>1. which is the tolerance class of the resistors as the seller said ?
>2. which is the precision of your "fairly expensive" ohmeter ?
>
>Then will find the truth easily. There is no passive componente
>without plus or minus error, just wrong measuring methodes or bad
>measuring tools.
>
>
I didn't say there was not plus or minus error - I said that it is all
minus **if** you assume the mean is the nominal value. If you assume
the mean is something below the nominal value, which I think it is, then
everything falls into place. I don't see as the precision of the
measuring instrument would even come into play, so long as it is
consistent. The same is true of the tolerance class. The degree of
tolerance should only affect the value of the standard deviation, but
it should still be a normal curve with the apex of the curve at the
nominal value.
>> But if you want to say that Olin is great, I'm agree.
>:)
>
>
Did I say that??? Don't think so. None of the data that I saw seemed
to have any bearing one way or the other with regard to Olin's
greatness. :-)
> I don't see as the precision of the measuring instrument would even
> come into play, so long as it is consistent.
After thinking about that for a bit, I withdraw the statement. :-)
Obviously it **would** come into play. However, in this particular
instance, I tried it with three different instruments and all gave the
same results. So while it is conceivable that all three could be
reading too low, I tend to doubt it.
>Are all three instruments calibrated
>to the same (local) standard?
hah, story you may like. many years ago, when radio was still young, well,
somewhere not long post-war anyway, there was this radio factory which had a
problem. people would measure voltages when fault finding, and find them
different to the design values, but then someone would come along with
another meter, measure the voltage, and get a different reading.
Eventually the technical management decided to do something about this. All
the meters in the facility were collected up, for a "calibration", and in
the absence of a standard source, one meter which was about the average
reading of them all was selected to be the standard. All the others were
adjusted to read the same by adjusting the magnetic shunt on the meter
movement. Problem solved, and all went quiet with measurement problems.
One day a newly hired graduate turns up at the facility, complete with an
Avo Model 8 - which back then was regarded as the "bees knees" when it came
to meters. However there was a problem. The readings of the Avo 8 did not
match the readings from another meter from the facility. Cocky young new
graduate claims his Avo is correct, and the facility meter must be wrong. As
more meters from the facility were brought, and all showed the same
discrepancy to the Avo, new graduates cockiness steadily went quieter and
quieter - nobody was letting on what the secret was.
>> . . . As
>more meters from the facility were brought, and all showed the same
>discrepancy to the Avo, new graduates cockiness steadily went quieter and
>quieter - nobody was letting on what the secret was.
>
>
On 10/8/05, John Nall <.....jwnallKILLspam@spam@gmail.com> wrote:
> Vasile Surducan wrote:
>
> >>1. which is the tolerance class of the resistors as the seller said ?
> >2. which is the precision of your "fairly expensive" ohmeter ?
> >
> >Then will find the truth easily. There is no passive componente
> >without plus or minus error, just wrong measuring methodes or bad
> >measuring tools.
> >
> >
> I didn't say there was not plus or minus error - I said that it is all
> minus **if** you assume the mean is the nominal value.
I'm asking about the tolerance because if it's greater than 1% the
behaviour discovered by yourself is quite common. Only very large
tolerance (greater than 5%) have plus and minus equal dispersion from
the nominal value.
> If you assume
> the mean is something below the nominal value, which I think it is, then
> everything falls into place. I don't see as the precision of the
> measuring instrument would even come into play, so long as it is
> consistent.
Imagine you have a fairly expensive instrument with an error of -1%.
(not +/- 1%) and all measured resistors are 0.2%. Now do you
understand why I'm asking ?
> The same is true of the tolerance class. The degree of
> tolerance should only affect the value of the standard deviation, but
> it should still be a normal curve with the apex of the curve at the
> nominal value.
Only if you have a large numbers of measured samples. Which I'm
affraid you don't.
cheers,
Vasile
Btw, how do you feel as retired (if I don't mistake).
>> Imagine you have a fairly expensive instrument with an error of -1%.
>(not +/- 1%) and all measured resistors are 0.2%. Now do you
>understand why I'm asking ?
>
>
>
Yeah, I think that I already agreed with you on that one. But just in
case I didn't already say so, I do.
>Only if you have a large numbers of measured samples. Which I'm
>affraid you don't.
>
>
While I agree with you in general, the number of samples necessary to
approximate the behavior of the population is a lot smaller than most
people think it is. The greater the number the greater the accuracy, of
course, but a sample greater than 20 is amazingly accurate.
>> Btw, how do you feel as retired (if I don't mistake).
>
>
While I was getting my undergraduate degree (Florida State University) I
discovered computers (big iron, back then) and got addicted. Spent the
next 30 years working for the Computing Center there, and it always
amazed me that they were willing to pay me for something which I would
gladly have done without pay! So "retirement" is just a continuation of
work for me, except that I have greater freedom to pick and choose what
I work on and when I do it. (Although my wife and I did take a 4 year
vacation right after retirement and lived on a sailboat in the northwest
Carribean).
Very possible if all three are digital and using the same
method of measuring ( or even same chip or clones inside )
John Nall wrote:
>
> Obviously it **would** come into play. However, in this particular
> instance, I tried it with three different instruments and all gave the
> same results. So while it is conceivable that all three could be
> reading too low, I tend to doubt it.
>
> John
gmail.com> wrote:
