Accurately Measuring Precision Resistors

[Watson A.Name - \Watt Sun, the Dark Remover\]

I acquired a couple dozen Micro-Ohm 400 ohm, 0.1% resistors, and some
other values. They've probably been around a decade or more, and/or may
have been culled. In any case, I would like to check them to make sure
they're within tolerance. I thought about doing this a couple ways:

Measuere them with a Fluke 4.5 digit meter, which hasn't been cal'd in
decades, but it could give me an idea of how close they are to a common
value, even if that value isn't exactly 400.

Measure them with a Leeds Northrup Wheatstone bridge. It's been around
decades and hasn't been cal'd in a long time. But it's a bridge, so if
there is any long-term drift, the resistors should drift the same way,
assuming the resistors are all the same. My guess is they're wirewound,
which is fairly stable. But there's grunge on the switch contacts, etc.

Make a bridge out of four of the 400 ohm resistors. Again, use this to
compare them to one another, to see if they are all about the same
resistance. If I apply a few dozen volts to the bridge, I should be
able to measure a few millivolts, as long as I don't heat up the
resistors too much and cause them to drift.

One other thought. I have a precision power supply with terminals on
the back to allow me to hook up a resistance in series with the 1.000 mA
constant current source. If it is 400 ohms, then I should measure 0.400
VDc output.

I realize that 0.1% is one part in a thousand, so that's 400 + or - 0.4
ohms. That's about the residual meter lead resistance I see in my
meters. I don't have a standard resistance, traceable to the NIST or
whatever. And I don't have a friend who's working at a cal lab. So I'm
trying to make do with what limited resources I have to get the most
accurate measurement. Any helpful advice would be appreciated. Thanks.



--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS? Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted. *All* email sent to it
goes directly to the trash unless you add NOSPAM in the
Subject: line with other stuff. alondra101 <at> hotmail.com
Don't be ripped off by the big book dealers. Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html
@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

 

[Phil Allison]

"Watson A.Name -

I acquired a couple dozen Micro-Ohm 400 ohm, 0.1% resistors, and some
other values. They've probably been around a decade or more, and/or may
have been culled. In any case, I would like to check them to make sure
they're within tolerance. I thought about doing this a couple ways:

Measuere them with a Fluke 4.5 digit meter, which hasn't been cal'd in
decades, but it could give me an idea of how close they are to a common
value, even if that value isn't exactly 400.

Measure them with a Leeds Northrup Wheatstone bridge. It's been around
decades and hasn't been cal'd in a long time. But it's a bridge, so if
there is any long-term drift, the resistors should drift the same way,
assuming the resistors are all the same. My guess is they're wirewound,
which is fairly stable. But there's grunge on the switch contacts, etc.

Make a bridge out of four of the 400 ohm resistors. Again, use this to
compare them to one another, to see if they are all about the same
resistance. If I apply a few dozen volts to the bridge, I should be
able to measure a few millivolts, as long as I don't heat up the
resistors too much and cause them to drift.

One other thought. I have a precision power supply with terminals on
the back to allow me to hook up a resistance in series with the 1.000 mA
constant current source. If it is 400 ohms, then I should measure 0.400
VDc output.

I realize that 0.1% is one part in a thousand, so that's 400 + or - 0.4
ohms. That's about the residual meter lead resistance I see in my
meters. I don't have a standard resistance, traceable to the NIST or
whatever. And I don't have a friend who's working at a cal lab. So I'm
trying to make do with what limited resources I have to get the most
accurate measurement. Any helpful advice would be appreciated. Thanks.

** Convert resistance to frequency - since that is a dead easy to measure
accurately.

A 555 oscillator ( using a polystyrene timing cap) running at 10kHz to 20
kHz will allow you to compare the 400 ohm resistors with each other and a
new 400 ohm 0.1% precision resistor for calibration.



............ Phil

 

[John Fields]

** Convert resistance to frequency - since that is a dead easy to measure
accurately.

A 555 oscillator ( using a polystyrene timing cap) running at 10kHz to 20
kHz will allow you to compare the 400 ohm resistors with each other and a
new 400 ohm 0.1% precision resistor for calibration.

 

[Stewart Pinkerton]

Basically, without accurate voltage *and* current measuring
facilities, he's screwed. OTOH, if all four resistors measure exactly
the *same* (a much easier job than an absolute measurement), then it
boosts confidence that they are all very close to the nominal 400 ohms
absolute value.

 

[John Woodgate]

I read in alt.binaries.schematics.electronic that John Fields

AARGH! You've criticised Allison! You will regret it. Actually, I think
he's got a good solution, and if the project is serious, the cost of one
reference resistor is justified.

 

[John Fields]

I read in alt.binaries.schematics.electronic that John Fields


AARGH! You've criticised Allison! You will regret it. Actually, I think
he's got a good solution, and if the project is serious, the cost of one
reference resistor is justified.

 

[The Phantom]

I acquired a couple dozen Micro-Ohm 400 ohm, 0.1% resistors, and some
other values. They've probably been around a decade or more, and/or may
have been culled. In any case, I would like to check them to make sure
they're within tolerance. I thought about doing this a couple ways:

Post your snail mail address, and I'll send you several .01% resistors, checked with a
6 1/2 digit DVM.

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

Well, converting to a freq is about the same as what the DMM does but
the DMM uses 1/F or time. And I don't have an accurate freq meter, I
have a DMM with a freq range but it's nowhere as accurate as a frequency
meter.

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

Basically, without accurate voltage *and* current measuring
facilities, he's screwed.

Well, I checked the two dozen on the Leeds & Northrup wheatstone bridge.
And contrary to what you just said, above, the wheatstone bridge does a
*comparison* of the unknown to an accurate known value, so it doesn't
really care what voltage or current you use. I used a 30VDC PS for the
batt. Most were within a few dvisions on the meter (I think five
divisions is an ohm at that range). I had two 'outliers' that were
greater, and one that was less than that. But the rest were fairly
close. I used heavy short leads to minimize their resistance and their
effect on the readings.

OTOH, if all four resistors measure exactly
the *same* (a much easier job than an absolute measurement), then it
boosts confidence that they are all very close to the nominal 400 ohms
absolute value.

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

I read in alt.binaries.schematics.electronic that John Fields


AARGH! You've criticised Allison! You will regret it. Actually, I think
he's got a good solution, and if the project is serious, the cost of one
reference resistor is justified.

So what's it cost for a resistor that's more accurate than the 0.1%
resistors I have? In this same batch, I have a few resistors that are
0.05 percent. ;-) But they're as old as the others.

Sometimes I think I'm trying to talk myself into buying a more accurate
bench DMM. I have an old 4-1/2 digit, how many X's in X-1/2 digits do I
need to measure the 0.1% resistors accurately? How about a used HP or
FLuke DMM from Ebay?

 

[Phil Allison]

"Watson A.Name -
"Phil Allison"

to 20kHz will allow you to compare the 400 ohm resistors with each other
and a new 400 ohm 0.1% precision resistor for calibration.

Well, converting to a freq is about the same as what the DMM does but
the DMM uses 1/F or time.

** A 4.5 digit multimeter meter permits 0.1 ohm resolution with a 400 ohm
resistor.

My suggestion offers about 0.02 ohms resolution.

And I don't have an accurate freq meter,

** Boo - hoo !!

High time you got yourself one or simply built one.




............ Phil

 

[Spehro Pefhany]

Well, converting to a freq is about the same as what the DMM does but
the DMM uses 1/F or time. And I don't have an accurate freq meter, I
have a DMM with a freq range but it's nowhere as accurate as a frequency
meter.

*Most* 4-1/2 digit DVMs are pretty linear and repeatable, so all you
need is a reference resistor in the same range. You can buy or borrow
one, or compare the L&N bridge vs. the DVM and if they happen to
agree, call it a day. Chances are if you can't tell the difference, it
doesn't much matter to you, right?



Best regards,
Spehro Pefhany

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

On Mon, 11 Apr 2005 19:58:26 -0700, "Watson A.Name - \"Watt Sun, the Dark Remover\""

Post your snail mail address, and I'll send you several .01% resistors,
checked with a 6 1/2 digit DVM.

Which remindw me, I have some precision resistors made by Cinema
Engineering out in the garage. I think some of them are that low a
tolerance, but they're big ol' bobbins wound with lots of very fine
wire. I'll have to dig around and check them out.

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

*Most* 4-1/2 digit DVMs are pretty linear and repeatable, so all you
need is a reference resistor in the same range. You can buy or borrow
one, or compare the L&N bridge vs. the DVM and if they happen to
agree, call it a day. Chances are if you can't tell the difference, it
doesn't much matter to you, right?

I can compare them to each other, and to the known resistances of the
bridge and the DVM's (uncal'd) accuracy. But my concern is that the
measurements I'm making are comparing or measuring resistances that are
close relative to each other, but may be off from the absolute value.
If the DMM and the bridge are both off in the same direction by about
the same amount, then I would be fooling myself if I accepted them as
absolutely accurate because they happen to closely agree. No? (BTW,
the L&N bridge has a cal sticker that says it's due for recal in Nov.
1969. )

I measured some very old Cinema Engineering wirewound resistors with the
bridge. The label says 3390 but I got 3404 ohms. That's 0.4% high, but
since it's only a 1% tolerance, it's still within. Another one is
labeled 100 ohms 0.1% and I set the bridge for 100.0 and I can't see the
needle deflect, so I'm getting the feedback that the bridge agrees
closely with the marked resistance.

Most of these resistors are bobbins manually wound with a measured
amount of resistance wire, so I feel that they are still accurate even
if they're very old. But the two dozen Micro-Ohms look like chokes
about the size of 1W resistors, plastic cylinders with one end shiny
where they poured the potting epoxy in. I found this info on them.
They're part number 135, subminiatures, but they're so old that the
markings don't obey their current system. Can you imagine winding these
with 0.6 mil wire? That's a quarter the thickness of a human hair!
http://www.micro-ohm.com/precision/axial.html

 

[John Fields]

Sometimes I think I'm trying to talk myself into buying a more accurate
bench DMM. I have an old 4-1/2 digit, how many X's in X-1/2 digits do I
need to measure the 0.1% resistors accurately? How about a used HP or
FLuke DMM from Ebay?

---
4-1/2 digits = 1.XXXX, where the X's can go to 9, so you've got a
meter which can read to one part in 20,000. That's +/- 0.005%.

There'll be a spec on the number of counts that it can be in error,
and if you look on Fluke's web site you can get that spec.

Why don't you just spend a few bucks to get the thing calibrated and
be done with it?

 

[Vidar Løkken]

Post your snail mail address, and I'll send you several .01% resistors, checked with a
6 1/2 digit DVM.

I'm building a measure bridge, and I'd love two of these. How much do
you want? I've turned "on" this address, so you can mail me at
[email protected]

 

[Tim Shoppa]

But my concern is that the

measurements I'm making are comparing or measuring
resistances that are
close relative to each other, but
may be off from the absolute value.

An older 400 ohm 0.1% axial resistor is almost certainly a wire-wound
unit. Long-term stability is very good. If operated over the power
limit or in extreme humidity/temp there may be some irreversible shifts
in resistance, but if never used then you probably don't have to worry.

Tim.

 

[J. Yazel]

---- SNIP ----
absolutely accurate because they happen to closely agree. No? (BTW,
the L&N bridge has a cal sticker that says it's due for recal in Nov.
1969. )

Thanks. You really made my day.

Jack

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

---
4-1/2 digits = 1.XXXX, where the X's can go to 9, so you've got a
meter which can read to one part in 20,000. That's +/- 0.005%.

There'll be a spec on the number of counts that it can be in error,
and if you look on Fluke's web site you can get that spec.

Why don't you just spend a few bucks to get the thing calibrated and
be done with it?

Something doesn't seem quite right about that +/- figure. When I
meaasured the 2 dozen, I had to use the 2k range. Most of them measured
400.3 give or take a few tenths of an ohm, except for 3 or 4 outliers
which were the same as the ones the bridge IDed (I marked them with some
white paint). On the 2k range, the Fluke can only see a tenth of an
ohm, so it seems to be able to do no better than .01% not including the
+/- counts.

I still have to deal with the R in the leads (I shorted the 12 gauge
wires together and it measured zero). Mine's an 8600 and doesn't have
4-wire R measurement (I used short lengths of 12 AWG in banana plugs to
make the measurement). Here's a 5-1/2 on Ebay for probably not much
over a hundred, and it has the 4-wire R measurement. I think the Fluke
8600s like mine go for well under a hundred on Ebay, and the cost of
calibrating them would be muich more than that, so it's kind of not
worth it. It's probably wiser to buy a 5-1/2 or better meter on Ebay,
just to get the 4-wire capability. Anyway, my Fluke measurements agree
closely with the L&N wheatstone bridge measurements, which is
reassuring. Yhanks.

 

[Watson A.Name - \Watt Sun, the Dark Remover\]

Thanks. You really made my day.

Jack

???