Accurately Measuring Precision Resistors

[Henry Kolesnik]

"Man with one watch knows exact time, man with two, not sure." Confucius

--

73
Hank WD5JFR

"Barry Jones" <bjones01@acm.org> wrote in message
news:426173b4$0$24300$8f2e0ebb@news.shared-secrets.com...

Henry Kolesnik wrote:

I too like to have accurate things to confirm something to the nth
degree.
I'm still hoping that someone will come up with a trick to verify a
precison resistor with stuff you already may have as you initially asked.
But by the looks of it, it can't be done.

You really have to have *something* you trust to start with. Two
somethings makes it even easier.

I was thinking about finding *something*, and it seems to me, at least
theoretically, that you could make a large set of resistors (or other
things) of a given precision into a smaller set of resistors (or other
things) with higher precision.

For example, suppose you have 100 1 M resistors with a precision of 1%. If
you connect them all in parallel, you'd have an equivalent 10K resistor,
but it's standard deviation will have decreased by a factor of sqrt(100).
This is from the definition of Sample Normal Distribution. Assuming the 1
M resistors had a mean of 1 M, and a somewhat normal error distribution
(it doesn't even have to be very close to normal), this should increase
the precision by a factor of 10. What say you to the analysis?

Of course there may be other errors introduced in trying to connect 100
resistors in parallel. Details, details.

Now if I could just invent perpetual motion . . .

--
Barry

Heisenberg may have slept here.

 

[Dave]

Watson A.Name - "Watt Sun, the Dark Remover" wrote:

For example, suppose you have 100 1 M resistors with a precision of

1%.

If you connect them all in parallel, you'd have an equivalent 10K
resistor, but it's standard deviation will have decreased by a factor

of

sqrt(100).

As someone said, that only works if the originals are normally
distributed. If they were different makes then it might be valid, but
otherwise no chance.

Maybe it's better to just scrounge a known good 0.1% resistor from a
piece of equipment,

I don't think you will find 0.1% resistors are that expensive.

About $5 last time I checked. So buying a few is not exatly going to
break the bank.

 

[John Fields]

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
<kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure." Confucius

---
There were watches around 2500 years ago?^)

--
John Fields
Professional Circuit Designer

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 13:51:54 +0100, the renowned Dave
<nospam@nowhere.com> wrote:

Watson A.Name - "Watt Sun, the Dark Remover" wrote:

For example, suppose you have 100 1 M resistors with a precision of

1%.

If you connect them all in parallel, you'd have an equivalent 10K
resistor, but it's standard deviation will have decreased by a factor

of

sqrt(100).

As someone said, that only works if the originals are normally
distributed. If they were different makes then it might be valid, but
otherwise no chance.


Maybe it's better to just scrounge a known good 0.1% resistor from a
piece of equipment,

I don't think you will find 0.1% resistors are that expensive.

About $5 last time I checked. So buying a few is not exatly going to
break the bank.

Less than a dollar, quantity 1, for 402R +/-0.1% 25ppm from stock
(leaded, which I'd prefer for this purpose, 1206 or 0805).


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
<jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure." Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg



Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Lord Garth]

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> wrote in message
news:gtr4619obea8ecedv1bt5arpukcspv0bkf@4ax.com...

On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure."
Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg

Someone scratched graffiti the step near him.

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 14:39:47 GMT, the renowned "Lord Garth"
<LGarth@Tantalus.net> wrote:

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> wrote in message
news:gtr4619obea8ecedv1bt5arpukcspv0bkf@4ax.com...
On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure."
Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg


Someone scratched graffiti the step near him.

It's just chalk, and not very ancient at that. ;-)


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Stewart Pinkerton]

On Sun, 17 Apr 2005 10:17:00 -0400, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure." Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg

Yes, but the original quote is rubbish, as we all know.

Confucius, he accused of many things he never say..........
--

Stewart Pinkerton | Music is Art - Audio is Engineering

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 15:01:25 +0000 (UTC), the renowned Stewart
Pinkerton <patent3@dircon.co.uk> wrote:

On Sun, 17 Apr 2005 10:17:00 -0400, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure." Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg

Yes, but the original quote is rubbish, as we all know.

Confucius, he accused of many things he never say..........

Yes, I have my own doubts about attributing "man who have hole in
pocket feel cocky all day" to the great philosopher Kong fu zi.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Watson A.Name - \"Watt Su]

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:b55461d9b3f7g8bbr4kv8v7f0vgt3rl6rh@4ax.com...

On Sat, 16 Apr 2005 17:20:01 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:


I too like to have accurate things to confirm something to the nth
degree.
I'm still hoping that someone will come up with a trick to verify a
precison
resistor with stuff you already may have as you initially asked. But
by the
looks of it, it can't be done.

Unfortunately, there *are* no tricks. In order to verify any given
electrical quantity to a high degree of precision, you must already be
in possession of a highly precise electrical quantity, be it voltage,
current or resistance.

That's not true. All electrical constants are based on physical
constants, and can be derived from them.

The conversion is straightforward, but you need
some kind of absolute measure at some point in the procedure.

--

Stewart Pinkerton | Music is Art - Audio is Engineering

 

[Watson A.Name - \"Watt Su]

"Henry Kolesnik" <kolesnik@sbcglobal.net> wrote in message
news:jys8e.10$yd7.0@newssvr11.news.prodigy.com...

"Man with one watch knows exact time, man with two, not sure."
Confucius

Well, if what Confucius said is true, then NIST must not be certain of
the exact time. But NIST claims accuracy of something like 10^14 or so
over their numerous standards..

Perhaps it should be "Man with one watch is deceived into thinking he
knows the exact time.."

--

73
Hank WD5JFR

"Barry Jones" <bjones01@acm.org> wrote in message
news:426173b4$0$24300$8f2e0ebb@news.shared-secrets.com...
Henry Kolesnik wrote:

I too like to have accurate things to confirm something to the nth
degree.
I'm still hoping that someone will come up with a trick to verify a
precison resistor with stuff you already may have as you initially
asked.
But by the looks of it, it can't be done.

You really have to have *something* you trust to start with. Two
somethings makes it even easier.

I was thinking about finding *something*, and it seems to me, at
least
theoretically, that you could make a large set of resistors (or
other
things) of a given precision into a smaller set of resistors (or
other
things) with higher precision.

For example, suppose you have 100 1 M resistors with a precision of
1%. If
you connect them all in parallel, you'd have an equivalent 10K
resistor,
but it's standard deviation will have decreased by a factor of
sqrt(100).
This is from the definition of Sample Normal Distribution. Assuming
the 1
M resistors had a mean of 1 M, and a somewhat normal error
distribution
(it doesn't even have to be very close to normal), this should
increase
the precision by a factor of 10. What say you to the analysis?

Of course there may be other errors introduced in trying to connect
100
resistors in parallel. Details, details.

Now if I could just invent perpetual motion . . .

--
Barry

Heisenberg may have slept here.

 

[Watson A.Name - \"Watt Su]

"Dave" <nospam@nowhere.com> wrote in message
news:42625bea@212.67.96.135...

Watson A.Name - "Watt Sun, the Dark Remover" wrote:

For example, suppose you have 100 1 M resistors with a precision of

1%.

If you connect them all in parallel, you'd have an equivalent 10K
resistor, but it's standard deviation will have decreased by a
factor

of

sqrt(100).

As someone said, that only works if the originals are normally
distributed. If they were different makes then it might be valid, but
otherwise no chance.


Maybe it's better to just scrounge a known good 0.1% resistor from a
piece of equipment,

I don't think you will find 0.1% resistors are that expensive.

About $5 last time I checked. So buying a few is not exatly going to
break the bank.

Oh, okay. I had no idea of how expensive they are, only that they
usually require some amount of manual labor, either in assembly or
testing. The ones I have are wirewound, and the Micro-Ohm website says
they are wound out ot wire that's as fine as .6 mil, which is 1/4 the
thickness of a human hair. Must be a very delicate and tedious
procedure.

 

[Watson A.Name - \"Watt Su]

"John Fields" <jfields@austininstruments.com> wrote in message
news:nip4615a9on9cbpqrlb1s3pj8l2nlb4pij@4ax.com...

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure."
Confucius

---
There were watches around 2500 years ago?^)

Of course. Haven't you seen those cartoons with the Ancient Greeks with
the sundials on their wrists? ;-)

--
John Fields
Professional Circuit Designer

 

[Watson A.Name - \"Watt Su]

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> wrote in message
news:gtr4619obea8ecedv1bt5arpukcspv0bkf@4ax.com...

On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure."
Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg

Shame on you for not warning us that it would take 5 minutes to load
that image, even with the fuzzed out face.

Speaking of fuzzed out faces, I saw (not by choice, btw) the sketches of
Jacko's trial and they blurred out the faces of the witnesses. Pretty
sad when they have to go _that_ far.

Best regards,
Spehro Pefhany
--

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 08:43:17 -0700, the renowned "Watson A.Name -
\"Watt Sun, the Dark Remover\"" <NOSPAM@dslextreme.com> wrote:

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:b55461d9b3f7g8bbr4kv8v7f0vgt3rl6rh@4ax.com...
On Sat, 16 Apr 2005 17:20:01 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:


I too like to have accurate things to confirm something to the nth
degree.
I'm still hoping that someone will come up with a trick to verify a
precison
resistor with stuff you already may have as you initially asked. But
by the
looks of it, it can't be done.

Unfortunately, there *are* no tricks. In order to verify any given
electrical quantity to a high degree of precision, you must already be
in possession of a highly precise electrical quantity, be it voltage,
current or resistance.

That's not true. All electrical constants are based on physical
constants, and can be derived from them.

Not very practical. ;-) Also, one of them is *defined* by a world
standard kept in Paris.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Watson A.Name - \"Watt Su]

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> wrote in message
news:g2v4611cn66voqisrhrmvtb9gad1579agn@4ax.com...

On Sun, 17 Apr 2005 15:01:25 +0000 (UTC), the renowned Stewart
Pinkerton <patent3@dircon.co.uk> wrote:

On Sun, 17 Apr 2005 10:17:00 -0400, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Sun, 17 Apr 2005 08:42:40 -0500, the renowned John Fields
jfields@austininstruments.com> wrote:

On Sun, 17 Apr 2005 12:22:39 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:

"Man with one watch knows exact time, man with two, not sure."
Confucius

---
There were watches around 2500 years ago?^)

No, but there were watch towers.

http://www.speff.com/gwall_mar13.jpg

Yes, but the original quote is rubbish, as we all know.

Confucius, he accused of many things he never say..........

Yes, I have my own doubts about attributing "man who have hole in
pocket feel cocky all day" to the great philosopher Kong fu zi.

As an aside from that, from what I've seen on the likes of Antiques
Roadshow, the ancient asian dress was just a simple pocketless pullover,
and they used a bag hung from their belt.

Best regards,
Spehro Pefhany
--

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 09:00:16 -0700, the renowned "Watson A.Name -
\"Watt Sun, the Dark Remover\"" <NOSPAM@dslextreme.com> wrote:

"Dave" <nospam@nowhere.com> wrote in message
news:42625bea@212.67.96.135...
Watson A.Name - "Watt Sun, the Dark Remover" wrote:

For example, suppose you have 100 1 M resistors with a precision of

1%.

If you connect them all in parallel, you'd have an equivalent 10K
resistor, but it's standard deviation will have decreased by a
factor

of

sqrt(100).

As someone said, that only works if the originals are normally
distributed. If they were different makes then it might be valid, but
otherwise no chance.


Maybe it's better to just scrounge a known good 0.1% resistor from a
piece of equipment,

I don't think you will find 0.1% resistors are that expensive.

About $5 last time I checked. So buying a few is not exatly going to
break the bank.

Oh, okay. I had no idea of how expensive they are, only that they
usually require some amount of manual labor, either in assembly or
testing. The ones I have are wirewound, and the Micro-Ohm website says
they are wound out ot wire that's as fine as .6 mil, which is 1/4 the
thickness of a human hair. Must be a very delicate and tedious
procedure.

The thin film type flow out of a resistor trimming machine in their
thousands much like 5% resistors do.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Stewart Pinkerton]

On Sun, 17 Apr 2005 08:43:17 -0700, "Watson A.Name - \"Watt Sun, the
Dark Remover\"" <NOSPAM@dslextreme.com> wrote:

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:b55461d9b3f7g8bbr4kv8v7f0vgt3rl6rh@4ax.com...
On Sat, 16 Apr 2005 17:20:01 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:


I too like to have accurate things to confirm something to the nth
degree.
I'm still hoping that someone will come up with a trick to verify a
precison
resistor with stuff you already may have as you initially asked. But
by the
looks of it, it can't be done.

Unfortunately, there *are* no tricks. In order to verify any given
electrical quantity to a high degree of precision, you must already be
in possession of a highly precise electrical quantity, be it voltage,
current or resistance.

That's not true. All electrical constants are based on physical
constants, and can be derived from them.

While theoretically true, this is practically useless information.

--

Stewart Pinkerton | Music is Art - Audio is Engineering

 

[Spehro Pefhany]

On Sun, 17 Apr 2005 09:18:18 -0700, the renowned "Watson A.Name -
\"Watt Sun, the Dark Remover\"" <NOSPAM@dslextreme.com> wrote:

As an aside from that, from what I've seen on the likes of Antiques
Roadshow, the ancient asian dress was just a simple pocketless pullover,
and they used a bag hung from their belt.

Ancient garments that I've seen depicted were similar to that.
According to one web site, in Europe (and relatively recently compared
to Chinese history) pockets developed from people hanging the bag
(containing coins etc.) inside their garments to prevent theft. They
then added a slit to access the bag without disrobing and pockets
developed from that.



Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Watson A.Name - \"Watt Su]

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:gr256191huomm8lgtspi0jasnqv12nbeqe@4ax.com...

On Sun, 17 Apr 2005 08:43:17 -0700, "Watson A.Name - \"Watt Sun, the
Dark Remover\"" <NOSPAM@dslextreme.com> wrote:


"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:b55461d9b3f7g8bbr4kv8v7f0vgt3rl6rh@4ax.com...
On Sat, 16 Apr 2005 17:20:01 GMT, "Henry Kolesnik"
kolesnik@sbcglobal.net> wrote:


I too like to have accurate things to confirm something to the
nth
degree.
I'm still hoping that someone will come up with a trick to verify
a
precison
resistor with stuff you already may have as you initially asked.
But
by the
looks of it, it can't be done.

Unfortunately, there *are* no tricks. In order to verify any given
electrical quantity to a high degree of precision, you must already
be
in possession of a highly precise electrical quantity, be it
voltage,
current or resistance.

That's not true. All electrical constants are based on physical
constants, and can be derived from them.

While theoretically true, this is practically useless information.

Absolutely not. Last week I needed a 0.1 ohm resistor, so I took out
the ref manual and looked at the wire tables, and found that 32 AWG has
162 milliohms per foot. This is a physical constant derived from Rho *
length / area. Rho for copper is 10.37 at 25 C. So I measured 7.4
inches of #32 wire and wound it around a 1W high value resistor, and it
measured very close to 0.1 ohm. Now I only have the problem that it has
..6 microhenry inductance.

The same can be done for a capacitor.

--

Stewart Pinkerton | Music is Art - Audio is Engineering