Which is more accurate Digital or Analog Meters?

[Eddy Current]

Which is more accurate Digital or Analog Meters?

 

[kingpb]

Accuracy is determined by certification. Given the the same level of accuracy, and certified to be accurate within those limits, than it would not matter.

The main difference I see, is that analog need to be rechecked periodically, where as digital is digital. A lot of the accuracy of the measurement is going to be based on the CT's or VT inputs as well.

 

[GeorgeB]

That depends on what is being measured. I've used high quality, in-spec, DVMs to measure (specifically generated as an exhibit) AC voltage and seen over a 30% error of __real RMS value__

I've seen similar results with analog meters. In that specific case, it depends on the concept used to obtain the DC value which is displayed by the instrument.

THEN I've seen the older moving vane AC voltmeters ... inherently handle moderately distorted waveforms.

Then there are the "thermal watt converters", substantially obsolete, but which do well.

For DC values, a properly designed and calibrated digital meter can easily offer better than 18 bit resolution, 16 bit accuracy, 0.002%

What is your real question?

 

[kwired]

There are a lot of factors that can go into this as some have been mentioned.

You could have a very accurate very expensive instrument that remains accurate if left on a lab table in a relatively stable environment. Take that same instrument and let it bounce around in a service truck and be subject to different temparatures all the time and it may still be accurate, but within a different range of tolerance.

The instrument that is subject to the service truck conditions often does not need to be extremely accurate is another thing that needs consideration. Most of the time with servicing you don't care if you have 120 volts or if you have 118.34 volts, you are just concerned that it is close to 120.

 

[zog]

That depends on what is being measured. I've used high quality, in-spec, DVMs to measure (specifically generated as an exhibit) AC voltage and seen over a 30% error of __real RMS value__

If it had a 30% error how would it pass calibration?

 

[zog]

Which is more accurate Digital or Analog Meters?

As kingpb mentioned, being properly calibrated is the key, digital meters should be calibrated annually while analog meters are calibrated semi-annually, so that tells you something right there. Analog accuracies are based on full scale readings, so at lower values that error becomes much more signifigant. (3% accuracy @1000V is 997V to 1003V, but a 1000V meter reading 10V can be 7V to 13V, which is a 30% error). Digital meters accuracy goes across the range.

Plus you have the factor of misreading an analog meter, you have to look stright at it to get correct value (The term for that is on the tip of my tounge but can't recall) so I would have to say digital is more accurate.

Edit: Paralax error is the term I was looking for

 

[raberding]

i think the correct term is Parallax.

Another thing is understanding the difference between "accuracy" and "precision"...look it up in Wik

http://en.wikipedia.org/wiki/Accuracy_and_precision

 

[Open Neutral]

Most of the time with servicing you don't care if you have 120 volts or if you have 118.34 volts, you are just concerned that it is close to 120.

Or zero.....

 

[K8MHZ]

Analog meters have one big advantage over digital, that is the response time. You will be able to see rapid fluctuations on an analog meter that a digital can't respond to.

Digitals are more precise for steady voltages, all other things being the same, as it's easier to see four decimal places on a digital than on an analog meter.

 

[rcwilson]

Analog panel meters have an inherent advantage for an operator to quickly know the status. If the digital ammeter display is reading 150 amps, the operator has to read the number and relate that to the motor full load amps of 75A. With a properly ranged analog meter the operator knows the motor is in trouble with a quick glance at the pegged needle. As he is walking up to the panel, he can see the needle pegged long before he can interpret the numbers on the digital display.

 

[GeorgeB]

If it had a 30% error how would it pass calibration?

We did a show-n-tell on how peak, average, and true RMS meters displayed various signals; this one happened to be a 1% pulse train. AC cal for some meters is done with sine waves. We also used square and triangle waves from function generators.

MOST analog meters of the day assumed a sine wave, half-wave rectified and scaled the meter ASSUMING "stuff".

Even today, working as I do in electrohydraulics with PWM power to solenoids, I can take various meters of various manufacturers and get ... various ... readings<g>.

 

[GeorgeB]

Analog meters have one big advantage over digital, that is the response time. You will be able to see rapid fluctuations on an analog meter that a digital can't.

And you can see trends. As an engineering co-op student in the 60's working with analog radios in our final test, some of the guys thought they wanted one of them newfangled digital meters ... when you are adjusting to a peak or null, analog is (OPINION MODE) much better. It probably took 3 or 4 times as long with a digital meter.

 

[steve66]

Plus you have the factor of misreading an analog meter, you have to look stright at it to get correct value (The term for that is on the tip of my tounge but can't recall) so I would have to say digital is more accurate.

Edit: Paralax error is the term I was looking for

That's why some meters had a mirror strip across the dial. If you were looking straight at the meter, the reflection of the needle would be hidden behind the needle.

 

[zog]

Analog panel meters have an inherent advantage for an operator to quickly know the status. If the digital ammeter display is reading 150 amps, the operator has to read the number and relate that to the motor full load amps of 75A. With a properly ranged analog meter the operator knows the motor is in trouble with a quick glance at the pegged needle. As he is walking up to the panel, he can see the needle pegged long before he can interpret the numbers on the digital display.

Excellent point

 

[Strife]

Methinks you posed the question wrong
My opinion is: most of the times (MOST, not always), the $ 5,000.00 meter will be more accurate than the $ 20.00.

Which is more accurate Digital or Analog Meters?

 

[Besoeker]

Accuracy is determined by certification.

With respect, I disagree.
Certification doesn't make a meter accurate.
A meter can be accurate, certified or otherwise.
A calibration certificate shows that the meter calibration has been checked.

Given the the same level of accuracy, and certified to be accurate within those limits, than it would not matter.

The main difference I see, is that analog need to be rechecked periodically, where as digital is digital.

There a couple of other points you miss here.
Firstly, operator skill.
A digital meter gives a readout in digits - obviously.
If it displays 123.4V, that's that. Anyone can read that and see the same number.
On an analogue meter, you'd be hard pressed to determine whether it was 123.4V or 123.5V and any two users might have different views even with an anti parallax mirror scale.

And that's the second point.
Resolution. Four or five digit meters are not uncommon. There is no way you could read to that degree of resolution on an an analogue instrument.

Please understand that I am not knocking the analogue meter. It is very useful and user friendly for some applications. If, for example, you have a drive motor hunting it is much easier to look at this on an analogue meter than a bunch of digits rattling round.

 

[SG-1]

That's why some meters had a mirror strip across the dial. If you were looking straight at the meter, the reflection of the needle would be hidden behind the needle.

Here are some examples of paralax error. Notice how the pointer is made on this meter. Same thickness as the lines on the scale.

I do not care for digital meters without a bar graph. The bar graph can respond as fast as the analog.

 

[Besoeker]

I do not care for digital meters without a bar graph. The bar graph can respond as fast as the analog.

Probably faster but not usually seamlessly.

 

[weressl]

Analog panel meters have an inherent advantage for an operator to quickly know the status. If the digital ammeter display is reading 150 amps, the operator has to read the number and relate that to the motor full load amps of 75A. With a properly ranged analog meter the operator knows the motor is in trouble with a quick glance at the pegged needle. As he is walking up to the panel, he can see the needle pegged long before he can interpret the numbers on the digital display.

Precisely for this reason in addition to the digital display many digital meters incorporate an analog, segmented LCD bar readout so the observer can quickly glance to see where the reading is in the range of the full scale. Digital switchboard meters are also available where there is a circular bar about 270+ degrees around the circumference with a smaller digital readout in the middle. The circular bar can also be programmed to show alarm by the bars changing to red in the high range and yellow in the low range, and green in the normal range.

 

[gar]

111104-1437 EST

Eddy Current:

Now that you have had a number of responses can you more precisely define your question?

I won't describe why, but most of the time I use my Fluke 27 for most measurements.

.