Resolution?

[andy257]

Hi Guys

Could anyone explain what is meant by resolution with regard to electronic equipment. E.g rosolution of a oscilloscope or multimeter etc.....

ive tried searching for it on the net but all i get is monitor resolution etc. (same thing?)

cheers

andy

 

[Nigel Goodwin]

Hi Guys

Could anyone explain what is meant by resolution with regard to electronic equipment. E.g rosolution of a oscilloscope or multimeter etc.....

Resolution is the smallest change that can be resolved, probably the easiest to understand is an analogue to digital converter. If it's got '8 bit resolution' this means it returns an 8 bit word, so it can give readings between 0 and 255 (2^8 is 256). Likewise a 10 bit A2D can give readings from 0 to 1023 (2^10).

 

[kinjalgp]

Reslution is the smallest reading that you can measure with your instrument.

 

[mechie]

Resolution vs Accuracy ?

Resolution isn't quite the same thing as accuracy - think about it
If a digital multimeter is 4.5 digits (can show upto 19999) then that implies a 15 bit ADC which can therefore resolve to 1 part in 32768 but you may well see a quoted accuracy of 0.1% +/-1 digit.

For an oscilloscope (and multimeter!) you need to be clear about its technology, a DSO (and the pocket LCD ones) has an ADC which therefore sets a maximum to any resolution, an LCD display may further reduce this to its screen height in pixels ('scope) or number of digits (DMM). The final resolution will be a function of both of these values, along with possibly other factors.

A resolution figure for an 'old fashioned' analogue oscilloscope or moving coil multimeter is difficult to calculate as there is no theoretical limit, it can display a line anywhere up the screen. My cheap 3.5 digit multimeter will give me far more accurate voltage readings despite having a much lower resolution, the accuracy of the analogue instrument is limited by how carefully you interpret the display, two people could read different values - this is not an issue with a digital display, everybody should see the same digits.

 

[kinjalgp]

Thanks Mechie, I have removed "accurately" from my definition.

 

[Reloadron]

I have always defined, in its simplest terms as resolution is the ability to read an instrument or of the instrument to be read. This takes the term resolution to back before a digital display or an analog to digital conversion. That said when choosing an instrument for a measurement process Resolution and Accuracy (or uncertainty) both figure into things as well as resolution.

As to defining accuracy? I like to define precision as a high measure of repeatability and accuracy as unbiased precision. The following image illustrates this fairly well.

Now as to resolution?

If the above temperature readings were taken using a Type K Thermocouple with a standard uncertainty:

MAXIMUM TEMPERATURE RANGE

Thermocouple Grade

– 328 to 2282°F

– 200 to 1250°C

Extension Grade

32 to 392°F

0 to 200°C

LIMITS OF ERROR

(Whichever is greater)

Standard: 2.2°C or 0.75% Above 0°C

2.2°C or 2.0% Below 0°C

Special: 1.1°C or 0.4%

Anything at all to the right of the decimal is not worth considering. I can have resolution out 6 places to the right of the decimal but the numbers will mean nothing. It will however, look pretty cool but that's about it.

Ron

 

[gophert]

I have always defined, in its simplest terms as resolution is the ability to read an instrument or of the instrument to be read. This takes the term resolution to back before a digital display or an analog to digital conversion. That said when choosing an instrument for a measurement process Resolution and Accuracy (or uncertainty) both figure into things as well as resolution.

As to defining accuracy? I like to define precision as a high measure of repeatability and accuracy as unbiased precision. The following image illustrates this fairly well.

View attachment 111912

Now as to resolution?

View attachment 111913

If the above temperature readings were taken using a Type K Thermocouple with a standard uncertainty:

MAXIMUM TEMPERATURE RANGE

Thermocouple Grade

– 328 to 2282°F

– 200 to 1250°C

Extension Grade

32 to 392°F

0 to 200°C

LIMITS OF ERROR

(Whichever is greater)

Standard: 2.2°C or 0.75% Above 0°C

2.2°C or 2.0% Below 0°C

Special: 1.1°C or 0.4%

Anything at all to the right of the decimal is not worth considering. I can have resolution out 6 places to the right of the decimal but the numbers will mean nothing. It will however, look pretty cool but that's about it.

Ron

What the heck is going on Ron? Did you fall into the "Way Back Time Machine"?
You responded to a 14-year-old thread!

Either
Middle-school Andy is now an electrical engineer with 8-years of experience and knows exactly what resolution is...
Or,
Middle-Aged Andy is now retired and doesn't give a crap.

 

[Reloadron]

What the heck is going on Ron? Did you fall into the "Way Back Time Machine"?
You responded to a 14-year-old thread!

Either
Middle-school Andy is now an electrical engineer with 8-years of experience and knows exactly what resolution is...
Or,
Middle-Aged Andy is now retired and doesn't give a crap.

Gopher, not sure how I managed to do that? I haven't any idea exactly what I did. With the age of the thread it could not have been showing up as current yet I posted to it. I have no clue why I posted to that thread and no clue how I even found and saw the thread. Hell, I didn't even get here till 2009. Hopefully it will go away before too many people notice my screw up.

Ron

 

[gophert]

Gopher, not sure how I managed to do that? I haven't any idea exactly what I did. With the age of the thread it could not have been showing up as current yet I posted to it. I have no clue why I posted to that thread and no clue how I even found and saw the thread. Hell, I didn't even get here till 2009. Hopefully it will go away before too many people notice my screw up.

Ron

Oh, you know better than that. We'll have at least three more posts - one that repeats what you said and two more correcting/clarifying what you or your 2004 predicessors said.

 

[Reloadron]

Oh, you know better than that. We'll have at least three more posts - one that repeats what you said and two more correcting/clarifying what you or your 2004 predicessors said.

Gopher, you're not helping.

Ron

 

[JimB]

What the heck is going on Ron? Did you fall into the "Way Back Time Machine"?
You responded to a 14-year-old thread!

No he replied to a spammer who has subsequently been deleted.
Hence he "appears" to be replying to a very old thread.

JimB

 

[gophert]

No he replied to a spammer who has subsequently been deleted.
Hence he "appears" to be replying to a very old thread.

JimB

I suspected something like that happened but it was a great opportunity to raz Ron for a while.

 

[crutschow]

No he replied to a spammer who has subsequently been deleted.
Hence he "appears" to be replying to a very old thread.

JimB

It would be nice if a notice was left in the thread that a spammer was removed so there wouldn't be this confusion.

 

[Reloadron]

No he replied to a spammer who has subsequently been deleted.
Hence he "appears" to be replying to a very old thread.

JimB

I kept looking and thinking I replied to something new? All of a sudden nothing made any sense to me. The classic what the moment. When the spam post went away the count in the thread reset and showed an unbroken chain. Thanks Jim for mentioning that.

Ron

 

[Reloadron]

I suspected something like that happened but it was a great opportunity to raz Ron for a while.

Thank you Gopher.

Ron

 

[unclejed613]

I have always defined, in its simplest terms as resolution is the ability to read an instrument or of the instrument to be read. This takes the term resolution to back before a digital display or an analog to digital conversion. That said when choosing an instrument for a measurement process Resolution and Accuracy (or uncertainty) both figure into things as well as resolution.

As to defining accuracy? I like to define precision as a high measure of repeatability and accuracy as unbiased precision. The following image illustrates this fairly well.

View attachment 111912

Now as to resolution?

View attachment 111913

If the above temperature readings were taken using a Type K Thermocouple with a standard uncertainty:

MAXIMUM TEMPERATURE RANGE

Thermocouple Grade

– 328 to 2282°F

– 200 to 1250°C

Extension Grade

32 to 392°F

0 to 200°C

LIMITS OF ERROR

(Whichever is greater)

Standard: 2.2°C or 0.75% Above 0°C

2.2°C or 2.0% Below 0°C

Special: 1.1°C or 0.4%

Anything at all to the right of the decimal is not worth considering. I can have resolution out 6 places to the right of the decimal but the numbers will mean nothing. It will however, look pretty cool but that's about it.

Ron

looks almost like the definition for precision, accuracy, and uncertainty we got in out first day of class in TMDE school.

 

[Reloadron]

looks almost like the definition for precision, accuracy, and uncertainty we got in out first day of class in TMDE school.

Maybe because during the 1975 through 1979 I taught the Navy Intermediate Level Calibration course at NAS North Island? I don't think that the basics of Test Measurement and Diagnostic Equipment has changed much over the last 40 years or so, not the terminology anyway. The target used in post #6 was my own doing. Anyway, that may explain the similarity.

Something humorous is a little training film we started each course with:

Pretty sure the cartoon was made during the mid 1960s. While obviously off topic it is a cute little training film.

Ron

 

[JimB]

While obviously off topic it is a cute little training film.

A nice high level overview of why calibration is important in the military and commercial world.
(Yes I watched it all the way through).

In the hobbyist world, not quite so important.

It also occurs to me that there have been great advances in electronic component stability since that film was made, so that even cheap equipment like £10 DMMs can be relied on to last a long time.

JimB

 

[JonSea]

I had to question the knowledge of one navy calibration lab when I needed some accelerometers weighed for a test procedure. I needed the weight in grams; the scale they used to weigh the accelerometers was accurate to 0.01 pound. The list they gave me said:

Accelerometer #1: 57.145749 grams.

Seemed to be some confusion about significant figures! If they had had a better calculator, my results would have been even more accurate!

 

[Reloadron]

A nice high level overview of why calibration is important in the military and commercial world.
(Yes I watched it all the way through).

In the hobbyist world, not quite so important.

It also occurs to me that there have been great advances in electronic component stability since that film was made, so that even cheap equipment like £10 DMMs can be relied on to last a long time.

JimB

Heck Jim, I could not agree more. For the hobbyist the term close enough more frequently applies. The commercial and military works paint a much different picture to where millions of US dollars are spent on test, measurement and diagnostic equipment all in pursuit of accurate numbers. Additionally when instruments were calibrated years ago it was not unusual to have to "tweak and peak" but today they retain their accuracy and the entire way they work is different. Never could have dreamed of buying a DMM at the local hardware or home improvement store for $10 USD. Pretty cool really.

Ron