Why 5 digit multimeters?

[jpb]

Precision (rather than absolute accuracy) is important in some circumstances.

As part of a previous job I needed to calibrate voltages in a pulsed measurement instrument and a standard DVM was simply too coarse (I can't remember the details but I think it involved balancing two separate quantities). Though even in this case 4.5 digits was ok.

Though I agree that some meters have many more digits than they can justify from their intrinsic accuracy and the last few dance around so much as to be useless.

I do like my Keithley 2000 and 2015THD and it is nice to be able to monitor small adjustments in voltages on the OCXOs I play around with.

[ElectroNub]

I got my 6.5 digit meter for playing with thermocouples and to measure tempco of various parts. And to also help keep all my cheapo hand held meters in spec. An older model is not that much money. I paid less than $200 for a HP 3456A. A high quality modern multimeter with less digits costs a lot more.

My philosophy on all tools is... It's better to have it and not need it then need it and not have it! I just wish I could afford the expensive stuff 

[Phoenix]

We recently had a power supply to service, fairly catastrophic failure. After tracking down and replacing all the failed components to get it up and running again it passed all acceptance tests except for efficiency. It was measuring 81.88%, with a pass of 82%. The problem was traced to a completely fried relay (15x nominal contact resistance).

The efficiency test is done using volts x amps using a calibrated shunt (it's a DC output 12V 70A). I haven't done the math on how many digits/accuracy required to find this problem, but if the accuracy and calibration of the DMMs and shunts involved weren't up to spec the power supply may have been returned to the customer with a fried relay. It looks like our 34401A and a 34410A were used in this service.

[eevblogfan]

Hello

Pardon my love to Hp/agilent old stuff my reason was calibrating abuillity/cost (400 dolars for two hp34401A's is awsome deal asspecialy when you can calibrate and abjust them to be within 1ppm of each other for free )

I do have to say my words though , as part of my spare time lately , i do appreciate "brain power" rather then "hourspower" .

as I like to read free-electron's long posts , i Also enjoy analyzing the circuit myself and asppecially as the blogger from "wattsup" power blog have written(  http://powersupply.blogs.keysight.com/2015/02/when-measurements-and-mathematics-agree.html?m=1 ) I am super happy when math and practicality are going hand crossed , so my case I've learned Alot from been Able to read RMS values and understand the releashionship to the power circuit with transformers Etc ..

The main point I'm trying to rais here is the huge luck of schematics in our enviorment , as i understand ther's majority of companies who'se hiding from us the real deal .. and I'm afraid some guys having hard time raising their comfidance and actually do electronics using brain power (i am pointing to that lovely feeling you either solved or created (new)circuit .. )

So for the 5 digit race :

Comfidance is what am i going to sum it up to , IMHO

So all of that rubbish about he's capabilities ?

Yea .. he can defenetly read fast , read accurate , do math ,communicate to GPIB etc .. and measure kelvin type resistance , you've mentiond shorts ?

Well when I've had to fix for my job like 23 out of 40 prodigit modules , i did noticed like 4-7 of them had shorts with two examples , one was shorting trace (i did apply surge from our high current supply in order to vanish he's appearnce . forgotten the exact output but we have had two swichers . one was like 5V 500A .. and the other was some voltage at 1000A )  so that solved that , and the other case was rainging from one fet shorted to 4 tops ( out of like 8-9 ) .   

And I admit i've had the access to pretty good gear for A begginer , hp6632B at current mode (i've once measured he's stabuillity for given time , it was superb of other supplies after warmup periode )and keithley 2000 (at 5A output and math function of the keithley "just becouse I could" i told him to tallk to me Ohmic language so that was fun )

But otherwise ? , damm build your own 1A CC and kelvin measure with your own 10$ meter it is fucking good enough at mv range to see shorts .. so maybe at 200A switcher you'll have too lower values to measure , but how often one needs that ? )

Good article punch=> "Practical Low Res Measurements Aero Electric Connections" in google , first result

Excuse me becouse I am going to post and later fix becouse my point is valid and i am almost late to my new day
 good day everyone

[PA4TIM]

Why you want to know there are 5,5 digit multi-meters ?
Why not then also ask why there are :
- capacitance meters with aF resolution
- L meters with pH resolution
- resistance meters with uOhm resolution
- meg ohm meters that can measure pohm
- Volt meters that can measure 5kV
- scopes that can show 40GHz signals
Or is it you secretly want one yourself ? 

The answer is different per user. A year ago I used my gear for hobby only. Now I'm glad I bought good stuff because now I need it for my work.
I repair measurement and calibration gear,  my most used meter is a Keithley 2000. I have two 5,5 digit, two 6,5 digits and one of my two 7,5 digit meters on my bench. I have some 3,5 and 4,5 meters and lots of analog meters but (except for 3 special analog meters) not on the bench.

Non of my gear is cheap and chinese. I rather buy an older repairable top-product as a new cheap toy with more digits then accuracy

[Howardlong]

In repair, a higher resolution meter helps track down a short. Obviously 7 digits is overkill for this, but for some people it is a status thing (and others have genuine requirements I'm sure )

This. I, too, have been a naysayer, but I did discover very recently that tracing shorts on a board is made an awful lot easier with a 50000 count meter. I used to have to gingerly remove parts off boards to isolate a short across power rails, a very common fault. With a high resolution meter, you can trace where on the board that short is without lifting any parts. You don't need absolute accuracy though, just as long as the device's ADC is monotonic that's all that's required.

There are also cases particularly in the microcontroller world where you are trying to work out the current draw of turning on a particular piece of functionality, and having the resolution helps.

You can teach an old dog new tricks, but for absolute measurement (as opposed to measuring relative or delta data) I personally don't have any need for 5 digits.

Regarding frequency references, I do have a rubidium source, and use it for frequency measurement in the GHz for narrow band devices, where you need a reasonable amount of accuracy. In the past I have used it to perform doppler ranging for orbit analysis as well as characterising thermal and even bus voltage characteristics of satellites in orbit by measuring frequency aberrations from a prediction model, but I guess those are a bit of an edge case.

[Grapsus]

Here are my reasons for having five or more digit multimeters as a total hobbyist. I was contaminated fairly recently, mainly because of this forum 

The most impressive thing with a 6.5 digit meter is the ability to see an underlying physical reality of components and circuits. A regular 4000 counts DMM sees a battery as a super stable voltage source but with more resolution you start seeing all sorts of interesting events. Likewise you can witness the value of resistors and the voltage drop of diodes change as you touch them or as the temperature in the room increases. You can measure the offset voltage of various op amps etc. It may not be immediately applicable to building better circuits, but it sure gives you a better understanding of components and physics as it all becomes less abstract and all this data and graphs in datasheets make more sense.

There are also many practical applications as others have pointed out. You can directly measure various sensors like thermocouples or strain gauges and make sure that your circuit does a good job at reading those sensors.

Finally it's just a hobby like many others. For example, with digital electronics, a cool project might be the design of a 8*8*8 LED cube. You learn a lot and it's super cool, but it won't have any use in your everyday life. Building a 10 ppm voltage source is exactly the same. It may not have a lot of use apart from checking DMMs, but you will learn a lot and it's super cool. Well the last part is subjective, but it will become obvious if you start reading the Test Equipment forum here.

[Cliff Matthews]

Loads of people seem to lust over 5 (or more) digit multimeters. Forgetting they are not that accurate on any range except perhaps DC volts, why does anyone need that accuracy anyway? I can understand the need for accurate frequency meters for broadcast etc, but why voltmeters? OK, there are some applications like calibrating lower accuracy meters, but not many people do that. Is there anything more to this than a "mine's better than yours" desire?

3 digits reflect the early DVM's and still found in pocket junk under $10
3 1/2 digits reflected the DAC's prevalent before ~2000
4 1/2 digits reflect the DAC's after ~2000
5 digits may just be evolution (hopefully accompanied by better fusing, shunts, cases, etc.. again, maybe safer)

I thought I saw Dave expose a 4200 count DAC found in one 5-digit teardown...

[wraper]

I thought I saw Dave expose a 4200 count DAC found in one 5-digit teardown...

IIRC that was Dave's mistake. Actualy ADC was 40k or 50k count if I'm thinking about the same thing.
BTW ADC, not DAC.

[robrenz]

My thoughts from the So what exactly do you need a 6-1/2 digit multimeter for again?  thread.

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.

[zapta]

Safety?
Most 5 digit meters are built to real Class-3 specs it allows us to live another day (could be important).
If a 220v arc-plasma starts inside a cheap 3 1/2 digit DVM, who can say where it goes?

People brag about their digits. I am yet to see safety related bragging.

;-)

[switcher]

It's a 'precision gold' model in a kind of dark gloomy grey colour. I bought it on a special promotion they had and it was one of the ones that also measures capacitance and temperature. I also chose it for the 200uA current range it offers and the 'instant' continuity buzzer ( as fast as any I've used). However, it's only a 2000 count meter so it isn't anything special. I don't think I've adjusted it since the 1990s and I just keep sticking PP3 batteries in it when it runs low. It still seems to be as accurate as when I last adjusted it although it can only display 12V as 12.00V for example.

Cold war grey; very nice!
Mine has long gone, unfortunately.

The meter I always reach for is a Tek TX3; I love the large, dual display, and it has the fastest, loudest continuity buzzer....ever!