So what exactly do you need a 6-1/2 digit multimeter for again?

[FenderBender]

After watching the SignalPath video on the Rigol 6.5digit meter and reading another discussion...I was wondering, what field of electronics requires such a high resolution test instrument? When will it matter whether you have 5.12345V vs 5.12344V?

I understand for test equipment calibration or perhaps frequency standards (in satellites or cell towers) or maybe NASA stuff, but why would would we need 6-1/2 digit meters on our benches?

Just wondering..

[HLA-27b]

For marketing purposes of course...

if you want to outsell competitors 5.1 model you need a 6.1 model.

If the competitor comes up with a 3D  multimeter you need a 4D multimeter to outsell it...

[FenderBender]

Just as I expected.

[rr100]

When will it matter whether you have 5.12345V vs 5.12344V?

Do not confuse resolution with accuracy! If we take a look at the specs for DM3068 in three months to one year from calibration you should get, for 5V (that needs 20V scale):
+/- (0.0040% reading + 0.0005% range)
That is
0.004/100*5 + 0.0005/100*20 = 0.0003
So your:
5.00000V reading (for example, to keep things simple) can mean anywhere between 4.99970 and 5.00030.
The range can be 60 times more than what you implied. You are barely sure about the "milivolt" digit, if that.

Now run the numbers for your $50 multimeter (with 3 1/2 digits) and then see if you can properly test even a LiIon charging circuit.

[robrenz]

Personally I find I gain a deeper understanding of a topic when I attempt to learn all the nuances involved in high levels of precision.  Being a machinist my real appreciation for high precision started when I bought my first electronic indicator with 1 micro inch resolution (38 years ago and I still use it frequently).  My eyes were opened when I could breath on a piece of metal and watch it grow or press on my lathe (which I always assumed to be infinitely rigid) with one finger and watch it deform.  Did I actually need that indicator when I bought it?  No, but the educational value and gut level understanding of how materials behave was invaluable. The same applies for me with electronics. I bought a 6.5 digit 24ppm accuracy bench meter. Do I actually need it? No, but I am learning a lot by being able to see things happening that are not apparent with a regular meter.  As with all metrology, knowing what is involved in making accurate measurements at the bleeding edge of precision helps you improve your understanding and the accuracy of your every day measurements.

[olsenn]

For simple repair work and debugging (checking if you're getting  between 4.5 and 5.5 volts for example), you would not require a 6.5-digit multimeter. However, if you are testing/designing a precision voltage reference for an ADC or DAC, or testing the stability of a system for example, having a high accuracy, high precision DMM is crucial. Furthermore, the accuracy rating for 6.5-digit meters (good ones) are high higher than for say a Fluke 87 and so instead of the instrument guaranteeing that it will be within +-0.3V of the displayed value, it will guarantee between +-0.01V for instance... this can save your ass in numerous applications.

[WorldPowerLabs]

I've also seen the argument that this sort of resolution is good for keeping an eye on subtle battery behavior (self-discharge, etc).

[FenderBender]

I've also seen the argument that this sort of resolution is good for keeping an eye on subtle battery behavior (self-discharge, etc).

Not arguing with you, but hell, I think you could get away with 3-1/2 or 4-1/2 meter for things even like that. When you get out into 6+ digits, subtle changes could practically be caused by anything in the environment.

I don't know.

[FenderBender]

When will it matter whether you have 5.12345V vs 5.12344V?

Do not confuse resolution with accuracy! If we take a look at the specs for DM3068 in three months to one year from calibration you should get, for 5V (that needs 20V scale):
+/- (0.0040% reading + 0.0005% range)
That is
0.004/100*5 + 0.0005/100*20 = 0.0003
So your:
5.00000V reading (for example, to keep things simple) can mean anywhere between 4.99970 and 5.00030.
The range can be 60 times more than what you implied. You are barely sure about the "milivolt" digit, if that.

Now run the numbers for your $50 multimeter (with 3 1/2 digits) and then see if you can properly test even a LiIon charging circuit.

No I'm not confusing accuracy with resolution. I guess the proper word should be "When will it matter whether you have measured 5.12345V vs 5.12344V?"

I'm most likely just ignorant, but in real life, are all of those digits actually all that helpful? I think 4-1/2 digits is a reasonable resolution. 6-1/2. Gosh, I just wouldn't know what to do.

I guess if I don't know why a product exists, I shouldn't be questioning it.

[Bloch]

If some one gave me a 6-1/2 digit multimeter would I use it ?

Not sure.

I never had the use for one. 


Would i like a more precise 3-1/2 digit multimeter

YES

[HLA-27b]

All of what you guys say is true of course but the real issue is that we do not have confidence in the last digit (or two) of any multimeter. Perhaps we would be better off if we simply tape over the last digit and forget that it is there at all.
So the best use for a 6.1 digit multimeter is as a 5.1 digit multimeter 

[T4P]

When will it matter whether you have 5.12345V vs 5.12344V?

Do not confuse resolution with accuracy! If we take a look at the specs for DM3068 in three months to one year from calibration you should get, for 5V (that needs 20V scale):
+/- (0.0040% reading + 0.0005% range)
That is
0.004/100*5 + 0.0005/100*20 = 0.0003
So your:
5.00000V reading (for example, to keep things simple) can mean anywhere between 4.99970 and 5.00030.
The range can be 60 times more than what you implied. You are barely sure about the "milivolt" digit, if that.

Now run the numbers for your $50 multimeter (with 3 1/2 digits) and then see if you can properly test even a LiIon charging circuit.

No I'm not confusing accuracy with resolution. I guess the proper word should be "When will it matter whether you have measured 5.12345V vs 5.12344V?"

I'm most likely just ignorant, but in real life, are all of those digits actually all that helpful? I think 4-1/2 digits is a reasonable resolution. 6-1/2. Gosh, I just wouldn't know what to do.

I guess if I don't know why a product exists, I shouldn't be questioning it.

Reason? More stability and more resolution
Try doing the NPLC thing on your 4-1/2d DMM
AANNDDD get into the mV range that's where a 6-1/2" kicks in
Would 10mV look different from a 4-1/2" and a 6-1/2"? Yes it would. For reasons so 6-1/2d DMM's are much more stable and you can just go and sleep then wake up having see the same reading the next day if the source was actually stable
Mind you just so you do know, a fluke 87 can drift overnight ...

[FenderBender]

Well if you are talking about a 6-1/2 digit meter, >95% of the time, you are talking about a bench meter. Bench meters, (well atleast the ones worth considering), are intrinsically stable. Just take a look at all of their fancy precision references and Caddock resistors you see. Yes, most of them are darn stable.

The question is more on the lines of... Ex. 2.55554. For practical matters, I'm curious to know when you'd need to know the fate of that last digit. I'm not really trying to make a point. I honestly want to know what application you need such precision for

And if we really just want a 6 1/2 digit meter so we can have a better 4 1/2 digit meter...why can we not make a better 4 1/2 meter?

[T4P]

It goes historically, i always wanted more precision and counts.
I don't know how to explain it but in some cases extra counts does really matter
Speaking of which i just remembered why i want a 6-1/2d DMM
there was a case of one of my circuits not working even with the 5th digit stable with my UT61E
I borrowed a 3456 and i was able to see why, this certain circuit even with minute fluctuations wasn't working! It just wasn't!
*might be bulls*** since i can't really remember

[desirider]

Hi FenderBender,

Here's an article I came across a while ago.  It is not directly related to high-accuracy voltage measurements, but gives you some insight into what is required for resistor matching to within 0.1%:

http://tangentsoft.net/audio/resmatch.html

Regards
Desirider.

[desirider]

Following-up my previous post, here's a quote I saw in DIYAUDIO forum:

From Analog Devices' AN734 appnote:

"Difference amplifiers are commonly used in high accuracy circuits to improve the common-mode rejection ratio, typically known as CMRR.
For this type of application, CMRR depends upon how tightly matched resistors are used; poorly matched resis
tors result in a low value of CMRR.
...
For example, when R7 = R6 = 10 k, and R2 = R4 = 1 k, and error = 0.1%, CMRR improves to better than 80 dB.
...
Select resistors that have much tighter tolerance and accuracy. The more closely they are matched, the better the CMRR. For example, if a CMRR of 90 dB is needed, then match resistors to approximately 0.02%."

[free_electron]

2 words : DYNAMIC RANGE.
That's why you need so many digits.

Let's say you want to measure a signal that starts at 1mV and can go as high as 8 volts.
with 3 1/2 digit Dmm ( 0 ... 19.99) at the 10 volts range your lsb is 10mV ... so you cant resolve that
with a 4 1/2 digit dmm 0..19.999 your resolution is 1 mV you still can't resolve that
with a 5 1/2 digit your resolution is 100nV you have 10 steps to resolve the signal... 0.15% _+/1 1 lsb ... i'd like a bit more resolution
so we go for  1/2 digit.

Why not use autoranging ? well , because the range switching changes the accuracy of the meter !
let's say that signal that we are measuring is the output of a precise DAC. when the meter switches range you will see a noticable shift in the curve plot.. simply because the meter has changed its input attenuator. so you messed up. with a machine that has that many tigits you can select a fixed range and still capture a large dynamic signal.

Now , there is another reason. machines that can resolve 6 1/2 digits can do 5 1/2 digits fast .. like faster than 3 readings per second which is common for a handheld multimeter.
they can also do 4 1/2 digits very fast. like 3000 times a second ( or more )

You are tasked to plot inl/dnl of a 14 bit DAC... 16384 steps . Averaging 32x per reading to get rid of random noise... and you can read 3 times a second... so you are looking at roughly 180000 seconds runtime ... that's 48 hours or two days just to collect data...
Switch that machine in 4 1/2 digit and you are done in 2.9 minutes....
For home usage / repairs etc you don't need that. Do any kind of reasonable design work ... you want one.

[Lukas]

6½ digit DMMs may come in handy for troubleshooting too. You are able to measure the voltage drop across a pcb trace leading to a shorted cap or so, which helps a lot identifying the faulty component on a supply rail.

[StubbornGreek]

Personally, I was impressed with the range of capacitor the Rigol was able to test.

[FenderBender]

Well I think I see the point now. Just I don't have a need, so I don't see one for the rest of humanity.

[Psi]

When will it matter whether you have 5.12345V vs 5.12344V?

If you have say, a 200x gain differential amp measuring a tiny voltage change from reference.
You need to check your reference voltage is exact down to many decimal points because the high gain will amplify any errors.

Just one example of the top of my head.

[digsys]

The trouble is, when you get to those resolutions - the test device introduces unknown error !!
Often they are 1Mohm, 15pF say, even 10M + the stray capacitance you pick up will add error.
But the higher the Meter Impedance, the more RF / noise / inductance you pick up.
There is ONE really useful function that I did love, tracking a rogue current. I made a very tiny
pair of sharp pins, spaced and fixed app 2mm apart, then I'd puncture the copper resist and
actually follow the current paths, looking for a dead or faulty component. REALLY handy.

[RRobot]

Honestly, for me cause I like having a bench meter which is always in the same place, never gets knocked over etc... The fact that its a 6 1/2 digit is just a bonus.

[saturation]

Free electron said it best, but here's a simple application where the extra digits save you a lot of time,  testing self discharge or quiescent drain on very power frugal devices.

On a standard NiMH battery, after you charge it full, connect it to 6.5 digit meter, and you can see in real time, the battery discharging in the uV range ~ 1uV/10 seconds or less.

Do the same with a LSD NiMH battery, and you'll see the same uV discharge in minutes to tens of minutes; once the battery has settled in to 85% of its rated mAH, it takes hours for a 6.5 digit DMM to detect a uV change.  If you have a graphing DMM, you can actual see the rate of change, and graph the exponential decay.

The above is a very quick bench way to detect if a NiMH is actually LSD or not;  with less digits, you'd have to wait longer to detect a change, taking up a lot of time.

Because sleeping devices consume uA or pA of current when OFF, its very difficult to get a sensitive ammeter to measure OFF current draw.  But you can detect small drops of the battery supply at the uV level.  For example, measure the battery drain to 6.5 digits in 1 min to detect self discharge rate, if at all, do the same with the low power device on OFF, with the same battery.   If the drain is very small, increase the measurement time from 1 min to 10s of minutes or more. 

The future of electronics is low power, devices, so in many ways the 6.5 digit is a meter of choice because of its wide availability and low cost; the next level is already 8.5 digits with 10x the cost, and there are not many 7.5 digit DMMs.  At uV level, one has to use low level measurement techniques to avoid noise, stray fields, etc.,

Well if you are talking about a 6-1/2 digit meter, >95% of the time, you are talking about a bench meter. Bench meters, (well atleast the ones worth considering), are intrinsically stable. Just take a look at all of their fancy precision references and Caddock resistors you see. Yes, most of them are darn stable.

The question is more on the lines of... Ex. 2.55554. For practical matters, I'm curious to know when you'd need to know the fate of that last digit. I'm not really trying to make a point. I honestly want to know what application you need such precision for

And if we really just want a 6 1/2 digit meter so we can have a better 4 1/2 digit meter...why can we not make a better 4 1/2 meter?

[T4P]

Eventually we will all have to use 8-1/2d meters

[dilbert]

For marketing purposes of course...

if you want to outsell competitors 5.1 model you need a 6.1 model.

If the competitor comes up with a 3D  multimeter you need a 4D multimeter to outsell it...

Your signature reminds me of an anecdote.
We were trying to run basic flow code with matlab.To do so,we simulated a curved wall,defined by a nicely-given function.
Check with the HP calcs,ok.Run the code...something strange on the outer wall,right in the final part where troubles can arise..but that problem shouldn't have come out.
Hand check the functions,the discretizations,calc and draw some flow lines by hand,everything ok.
It looked like that part of wall was and wasn't there.Open the variable table and check them all,everythings' fine.
Then i had a lightning.
The wall function was discretized in points,and matlab was reporting them as discretized values.
But the variable editor just showed the first 4 or 5 digits,not the exact value.
So 99.99999999 was displayed as 100.0000,which is nice for a cake recipe,but not for the boundary conditions checks who red "fluid" instead of "wall" on those points because 99.99999999<100.0000
So there you go,a true case when a 1E-8 can waste you an afternoon.I still have the hand drawings somewhere.

[HLA-27b]

99.99999999<100.0000

True

 BUT
 99.99999999...=100

Those three dots make all  the difference.

[poorchava]

It makes sense when building measurement equipment.

I had a need for such meter when building device for measuring C-V characteristics of semiconductor junctions and MOS structures (my master's thesis). I wanted to measure resistances of PCB tracks and such to nullify as many arrors as possible in software. Keithley 2001 was a really nice thing to have.

Shame I can't afford to buy it for private use

[free_electron]

The wall function was discretized in points,and matlab was reporting them as discretized values.
But the variable editor just showed the first 4 or 5 digits,not the exact value.
So 99.99999999 was displayed as 100.0000,which is nice for a cake recipe,but not for the boundary conditions checks who red "fluid" instead of "wall" on those points because 99.99999999<100.0000
So there you go,a true case when a 1E-8 can waste you an afternoon.I still have the hand drawings somewhere.

Which is why real precision mathematics needs to be done using packed bcd arithmetic. There precision is absolute.

Packed bcd is a format where you use 1 byte to store two digits. .  1 byte is two nibbles
If you draw the hex table then 0 to 9 represent the numbers. A represents the sign , b represents the comma , c represents infinity , d represents the sign of the exponent, e represents where the exponent begins , f represents 'j' so you can have imaginary numbers as well.
In case no imaginary numbers are needed you can use it as the fraction sign ( f = fraction , e = exponent ) so you can store things like 1/3

For example
-1.7e-10 is stored as a string
0xa1b7ed10

12e13 would be 0x012e13

And so on

The problem is speed hit. You need a decoder and encoder. Cpu's in general can do bcd arithmetic.
As long as you have storage space you can store numbers.
If you have a decimal number with 500 digits you need a string of 257 bytes to store it. But it will be correct without rounding error.
The ieee765 floating point system is the worst thing ever invented. Because it gives you errors.

Packed bcd libraries exist for 'big iron' like Cray and others.

[nukie]

When you have 6-1/2 digit multimeter you get an additional feature its called warm up time. If you don't need it then stick to lower digits meter.

tapatalk

[FenderBender]

Very insightful guys. Starting to see the point...but no way in hell I'm getting one, unless I get it for free. 

[rr100]

Which is why real precision mathematics needs to be done using packed bcd arithmetic. There precision is absolute.

Problem is it really works only with arithmetic, that is the 4 basic operations and at most up to fractions. Once you throw in radicals, logarithms, trigonometry (and even limits/series/integrals) you have the problem that you can't compare the results as the same number can be represented by different strings (and there is no guaranteed "canonical form" or way to get from one representation to another).

[anotherlin]

Which is why real precision mathematics needs to be done using packed bcd arithmetic. There precision is absolute.

If you want absolute precision, use "symbolic computation".
There is no way you can perfectly represent 1/3 (0.333) using 2 (binary) or 10 (decimal) base system.
You have also the problem of constants like pi and irrational numbers (like square roots).

Double precision floating points have 57-bit precision. Proper floating point usage can be tricky (and it is often overlooked in computer science courses).
But I don't see any point of using BCD arithmetic. Especially if you consider the fact there's quite a few libraries for arbitrary long integers (BigNum). GNU even offers an arbitrary precision floating-point library.

For our DMMs, if you have 6.5 digits, that can easily fit into 24-bits.

[analex]

2 words : DYNAMIC RANGE.
That's why you need so many digits.

Yes, more SNR could carry more information. 6.5 DMM gives more DR abilities to get more SNR and more informations.

As 6.5 DMM has 0.1m ohms resolution, it is easier to locate the short circuit point by using 2-Wire or 4-Wire(better way) OHM.

Some of 6.5 DMM have data plot function, and DMM have variable BW(change NPLC), and uV/nA/mohm level resolution,  is very easy to find out low freq characters of a "noisy" signal. BTW, DSO can't do that, as DR/resolution reason; SA can't do that, as it can't go down to low freq range.

6.5 DMM is faster than 5.5 etc. Most of time, I set my 6.5 DMM to DCV 1000V range, so I can measure form 1mV~1000V without waiting for auto range. As this I can find out over load immediately at the power up moment, and cut off power supply before smokes.

[saturation]

That would be nice.  One could be lucky today and get a used one in $1000-3000 in perfect order.  Its the calibration costs that are high, if you want the best it can deliver.

The 1989 designed 3458a 8.5DMM sells for about $10,000, about its new price in 1989, in could be called cheaper given the depreciation of the US $. 

However, with 1nV resolution and 4ppm stability, few labs can calibrate the 3458a properly.  It runs about $600 for the 4ppm calibration, but the 0.2ppm version is about $1500.

http://lvldstdslabagilent.blogspot.com/2005/01/faq-what-3458a-calibrations-are.html

http://www.febo.com/pipermail/volt-nuts/2011-August/000840.html

Only a Josephson Junction has the resolution to calibrate an 8.5 DMM these days, and even if it were made into a turn-key standard, the labor to maintain it is far more than buying a 732a and keeping it in an airconditioned room, so calibration costs won't drop soon.

Eventually we will all have to use 8-1/2d meters

[nukie]

I used mine as a coffee warmer today, it works quite well.

[ejeffrey]

The wall function was discretized in points,and matlab was reporting them as discretized values.
But the variable editor just showed the first 4 or 5 digits,not the exact value.
So 99.99999999 was displayed as 100.0000,which is nice for a cake recipe,but not for the boundary conditions checks who red "fluid" instead of "wall" on those points because 99.99999999<100.0000
So there you go,a true case when a 1E-8 can waste you an afternoon.I still have the hand drawings somewhere.

Which is why real precision mathematics needs to be done using packed bcd arithmetic. There precision is absolute.

No, the only reason to use BCD is if you need to represent exact decimal numbers, typically with currency. This works OK because there is a finite required precision, and you only have to be correct on addition and subtraction -- multiplications are allowed to be rounded before adding to the accumulator.  Physical quantities have no special preference for base 10, so BCD is rarely if every used for scientific or engineering numerical computing (it is used in CAD for geometry definition, but converted to floating point for things like FEA).  Infinite precision arithmetic, either binary or decimal, is only used in specialized circumstances, for any sort of general algorithm the required size grows exponentially with the number of steps in the computation, and in any case operations other than +,-,* are not even possible.

The real lesson here is 'don't compare floating point values for equality'.