EEVblog #26 - Multimeter Tutorial - Counts, Accuracy, Resolution & Calibration

A tutorial about digital multimeter measurement display counts, accuracy, and calibration.

Hi! Welcome to the EEVblog.
I'm your host Dave Jones

and this is episode number 26.
There was a comment on one of

my earlier blogs from
someone called Walter.

He had a question about multimeters and
the counts like a 4000 count multimeter

What are counts and how that
relates to multimeter accuracy.

I thought that's a pretty good question.

It's something a lot of people
don't understand all that well.

So I thought I'd clear it up.

A 3½ digit meter can actually
display plus or minus 1999.

The half actually means that the most
significant digit can only go to one.

It can only be zero or one.
So a 3 ½ digit meter can go up to 1999.

A 4½ digit meter can go to +/- 19999.

They've started advertising 3¾ and

three-quarter digits or

three and two thirds digits and that's
where it gets a bit confusing is very

real

standard for that but Tom typically if
it if it's not like two-thirds of three

quarters

or something like that by reeling main
that he can go a bit faded with them

once thought might be

I'm 3 know I know I know I know it might
be to

999 or even Fort 9900 something like
that

not because it started becoming quite
confusing

manufacturers started talking in terms
is our

the number of care what's and I think
this is a much better way to do it

so I i three and a half digit multimeter
at plus/minus 1990 is actually

a 2000 count me to because it counts up
to

plus/minus 2000 and likewise before and
after each meter is actually a twenty

thousand county can use these tents
interchangeably

now has this relight accuracy well

it actually doesn't rely lauter nothing
to do with accuracy so let's talk about

this

now accuracy is a term when it comes to
motivate his people get

roll all the time and mining I'm guilty
of sin at 23

she laziness or and I have it it's wrong
to side

that this may die has and accuracy

a four-and-a-half digits or an accuracy
of 10,000 carrots

that is completely wrong the

camels the number I've I am

digits is the resolution so it's

you shoots I that this may not has a ten
thousand count

resolution not a ten thousand count
accuracy

and if you unite if you say the wrong
thing

Mike use a market adults especially if
you're in a job interview or something

amass a question I can make you look
like you don't know it

talking about okay now let's take a look
at

of three typical mode he made his and
hell

the the number of digits for the
resolution

fix I'm the usability of these made its

now I've got a fluke 79 series 2

he at now this is i on 4,000 kill

or three and three-quarter digit made up

I've got made me in 37 X sake and

this one is a 10,000 count made a

the flick 87.5% he is I to
four-and-a-half digit made up but I've

got it in

it's actually got a mud that he's
actually I'm 6,000 count low

okay don't fade in the Sun Bowl teaching
two or three

mayes and you can see they all read
fields love it and I've is this slight

differences they

but let's have a look what happens when
Roy chein the only jockey

K now if I had a three and a half

digit me to heal which I do it it
wouldn't actually dropped down to two

decimal places

because it would have been over the one
know and I know I'm limit

but none of her reach that yet so let's
keep turning this up

K&C which one

actually dropped its range first

and I'm sure you can guess which one its
gonna pay

being a there it is the flu

seven or eight series 2 because this is
a four thousand count major

we r others I that 4,000 where

basically over four so it's got it go up
to the next highest

range so we lose one digit all

resolution okay not accuracy now

week a tennis up foreign

that the flick a 75 here is a 6% sales
in Kansas in as a guest to six

or just slightly II Bay its

Chinese upper range also I so I this is

up Dale these are giving the signs
resolution okay this is a much more

expensive made up

I'm at six thousand chaos I K but

its but we've got the sign resolution
know this maiden name

37 XIII 10,000 count you'll still get
him

the extra digital resolution and this
won't change ranges I'm to a guy right

up to a

and to your casa no i know i know i know
i limit and being a race

after just they're all back on an even
playing field but let's switch

the fluke 87 52 for her digit night

okay now we've got this extreme teacher
a resolution he

not accuracy resolution now you for it

that this ID 7542 hostage it may be that
is useful

until once again we get up to

game we get up to twenty adults and
bingo it's going back

so you can save that for her digit made
up only has more resolution for twenty

percent of its range

ok the 10,000 count me to

I said before that resolution the number
digits on the sly has nothing to do with

accuracy n that's true

I K accuracy of the made-up is

based on the percentage of the actual
reading

not a false Kyle so I made his accuracy

will be specified as the percentage of
the Radian

plus camps now I typical

lie a good made up might have point and
accuracy of

no 0.1 percent plus to

camps now what that means is if you're
measuring

one fault okay you have an it

it will have and uncertainty all plus

or minus one Miley vault

.1 percent a one-vote is one million
baht

but you have to add on two camps to that
plus or minus

to Celts to get displayed value

saw it it will be if it's actually
raining exactly one vote

okay 1.00 0

then it can actually be I one

point 0 0

the re or plus/minus

for lockport no I'm 9

7 so that's resolution and accuracy

now let's talk about calibration
everyone knows

that unite you should get you made a
calibrated

was actually many what it means is that

the meda has been checked it's been it

doesn't calibration does not mean
adjustment will be made when you see in

this away for calibration

its they usually don't adjusted what
they would do

is all measure it against a reference
stain did

and I will give you 80 s report for
these

I'm giving you the actual typically
giving you the actual fee got

and you know and the a rock against that

absolute reference all that transfer
stain

and what you do with that calibration
information determines

help this morning made is going to be
used

within your company one applications
it's going to use for and how often you

going to calibrate

now just because a meter has been
calibrated yesterday

does not mean it's going to be within
spec

to die its I it's all about calibrations
about the history of this made at

Racine's best traceability

in a bit rice not the calibration of
this may DNA test article

record of it overtime and then you build
up confidence

in this may tap that it's not drifting
this is why some companies in some

industries will actually I to unite you
have to actually use

to the mall he made is too tight I'm
measurement just in case one of them

has unite is is actually help and I will

unite anyone serious about electronics
will actually have to multimeters I want

rely on just

one though have to you so that you can
cross check each other

and you can actually use a good known
Mon made a as I

as a bit of a transferred stain did to
not check

and measure other instruments and if
you've got a punch him only made as you

can all cross measure

themselves and you get more confidence
that you made a is within

specification usually you follow the
manufacturer's recommendation which my

peace I every 12 months

but when you first get used to it you
might wanna calibrated

more often so that we build up more that
reference

that historical reference data so you
can track it

and one Chino this mater is not you have
to really stable it's not true 15 at all

then you might actually widened here I
period of calibration

12 months then United still has
interested or you might actually change

it to you