ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [3]   Go Down

Author Topic: How do you know a volt is really a volt?  (Read 6062 times)

Mark Cadwallader

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1333
  • Helena, Montana USA
Re: How do you know a volt is really a volt?
« Reply #20 on: December 06, 2017, 01:09:19 pm »

(snip)

Sometimes, your measuring device doesn't need to be accurate, as long as the variation is consistent across the range you are measuring. If you're looking to identify voltage variations between two different circuits, and you're more concerned with the difference rather than the actual voltage, it doesn't matter if you're off by -3.2 volts, as long as you're off by -3.2 volts in each measurement you take.
[/quote]

Precision (repeatabilty of a measurement) is often more important than nominal accuracy. Or so I have found.
Logged
"Good tools are expensive, but cheap tools are damned expensive."

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16606
  • Hickory, Mississippi, USA
    • Resotune
Re: How do you know a volt is really a volt?
« Reply #21 on: December 06, 2017, 01:26:13 pm »

It was probably a fluke. But not a Fluke. ;)

Sometimes, your measuring device doesn't need to be accurate, as long as the variation is consistent across the range you are measuring. If you're looking to identify voltage variations between two different circuits, and you're more concerned with the difference rather than the actual voltage, it doesn't matter if you're off by -3.2 volts, as long as you're off by -3.2 volts in each measurement you take.
Just to be clear a -3.2V error when measuring 120VAC is unlikely to be a -3.2V error around 0V.  That is the difference between a scale error, or offset error.  Cheap meters are more likely to have a scale error (say from cheap components or flaky internal reference) that result in measurements being off by a multiplier factor. Offset errors are probably way way smaller than volts.

When dealing with AC measurements another huge source of potential (pun intended) errors is rectification... I have seen some gross errors from toy VOMs. I wouldn't trust the frequency response of AC measurements at anything other than mains frequency for inexpensive VOM.

I do not think the reasonable performance (so far) seen with my relatively cheap VOM is a fluke. The innards are likely a single large scale IC that will likely either work decently or no at all.

Oddly I paid $10 more for a feature on the Fluke (diode scale) that was included in the $15 VOM for nothing extra. 

One obvious feature difference in the Fluke is the lighter action in the main selector switch that is nice when changing ranges with only one hand, etc... The fluke also came with a fancy hanging gadget to prop it up for hands free bench measurements...

So nice ergonomic features in the more expensive VOM that are not obvious from the spec sheet.

JR
Logged
Don't tune your drums half-ass. Listen to what a properly "cleared" drum sounds like.   http://circularscience.com/

Stephen Swaffer

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2361
Re: How do you know a volt is really a volt?
« Reply #22 on: December 06, 2017, 10:41:29 pm »

The real question is, does it really matter?

Sent from my 2014817 using Tapatalk

To borrow an oft used phrase-it really depends.

For 98% of the industrial troubleshooting I do, I would be content with plus/minus 10% accuracy-as long as I have a CAT III or IV safety rating.  Practically speaking?  a quality meter verified against mains voltage gets the job done.

Honestly, for Mike's purposes calibration against a diode junction/zener diode or a battery cell might be good enough.

For lab R&D work, NIST calibration is probably worth the cost and effort.

For low voltage logic or analog circuit design-even at the hobbyist level-accuracy is more important than a CAT III/IV safety rating.

I guess I have enough meters, I'll never know for sure what the right voltage is-though I'd bank on the out-of-date calibrated Fluke I picked up at an auction of stuff from an industrial engineer.
Logged
Steve Swaffer

Brian Jojade

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1660
    • HappyMac Digital Electronics
Re: How do you know a volt is really a volt?
« Reply #23 on: December 08, 2017, 07:05:04 pm »

I would say yes. I teach an audio electronics class at University level, and have my students bring in whatever digital meter they want to class. It's amazing just how many different voltage readings you can get from the same source. Now, being 2 or 3 volts off on a 120-volt reading may not be significant in many cases. However, I do a lot of ground and power troubleshooting where a volt or two difference can send my failure theory in different directions.

If you use different meters giving different readings, then yes, this would be a problem. However, if you use the same meter, it shouldn't really matter much if the starting voltage read 120 or 125 volts. It's the difference that you're looking for.  So, accurate calibration to the exact volt doesn't matter.  Consistency in the measurement is what matters.
Logged
Brian Jojade

Jay Barracato

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2025
  • Solomons, MD
Re: How do you know a volt is really a volt?
« Reply #24 on: December 08, 2017, 07:41:02 pm »

If you use different meters giving different readings, then yes, this would be a problem. However, if you use the same meter, it shouldn't really matter much if the starting voltage read 120 or 125 volts. It's the difference that you're looking for.  So, accurate calibration to the exact volt doesn't matter.  Consistency in the measurement is what matters.
Ah... A discussion of measurement uncertainty...
One of my specialties, precision first and foremost, you can adjust for accuracy as needed.

Sent from my Moto Z (2) using Tapatalk

Logged
Jay Barracato

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9045
  • Atlanta GA
Re: How do you know a volt is really a volt?
« Reply #25 on: December 08, 2017, 07:47:55 pm »

Back in a "previous life", I was a component level repair guy for NEC working on 6 and 11GHZ microwave transmitters and receivers.

Test gear was REAL picky, and MUST be right.  We would have to clean the cables before every connection.  You could measure the difference.

We had a calibration company come in twice a year to calibrate all of our gear, back to traceable standards.

It would take them a couple of weeks.

I was always "pretty sure" that my readings were pretty accurate.

And YES, at 11Ghz, cables DO make a difference-even when 1 or 2 meters long. 
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Lyle Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1556
Re: How do you know a volt is really a volt?
« Reply #26 on: December 14, 2017, 03:15:31 am »

The "international volt" was defined in 1893 as 1/1.434 of the emf of a Clark cell. ... Prior to the development of the Josephson junction voltage standard, the volt was maintained in national laboratories using specially constructed batteries called "standard cells".

+1
Logged

ProSoundWeb Community

Re: How do you know a volt is really a volt?
« Reply #26 on: December 14, 2017, 03:15:31 am »


Pages: 1 2 [3]   Go Up
 



Page created in 0.037 seconds with 26 queries.