Brymen DM869S DCV, DCV +ACV, DC+ACV +ACV functions

[ercapoccia]

Hello there,

i was playing with a couple of cheap AD584LK voltage reference modules with my trusted BM869S. (In my job i don't need absolute accuracy so i'm doing this just as learning experience.)

I am quite surprise by the results.

Code: [Select]

34401A    BM 869S DCV         DCV+DCV+ACV       DC+ACV+ACV DCV 5-4/5
2,49955                 2,4987 2,4995               2,4996         2,49882
5,00173                 5,0000 5,0012               5,0014         5,00012
7,5005                 7,498 7,499               7,500                 7,4987
10,0024                 9,999 10,000               10,002         10,0002

2,49942                 2,4986 2,4993               2,4995         2,49873
5,00037                 4,9982 4,9994               4,9996         4,99833
7,5004                 7,495 7,496               7,497                 7,4963
10,0007                 9,995 9,996               9,997                 9,9957
Seems that reading DC+AC is more accurate than just the DC reading. The scope says 12mVpp of ac noise.
Reading the references with both my BM257s and BM235 values are spot on.

What those number means? Is the BM869s filtering the AC noise? What about the Agilent?

Marco

[edigi]

I maybe wrong here but at least I have a theory.
If you don't have a solid understanding of RMS AC vs RMS AC+DC this video is something that worth watching:


So it's pretty normal that RMS AC+DC will be a higher value than DC. (What is not normal is that relatively big noise, 12mVpp is a lot in my view.)
Because RMS AC+DC is the square root of (AC square + DC square) it will be alway bigger (at best equal) than DC (which is just an average and noise will be averaged out pretty efficiently).

This means however bad news regarding the total accuracy (which you seem to expect) of one of your meters because it's not the RMS AC+DC that should agree with the Agilent but the DC... (especially that reference is supposed to be pure DC).
This however does not tell which is the most accurate.

Normally I'd assume that it's the benchtop meter as the DMM is even at its best accuracy range (DCV) only 0.02%+2 digit accurate (one order of magnitude worse than what is expected from the bench meter)...
The good news is though that the 2 meters are not that far away, if we consider the tolerances of both meters (I did not check though one by one your table, I had just a quick look). As a DMM is typically not used for a really accurate measurement the accuracy of your DMM is pretty OK in my view.

Note: The 5-4/5 mode of the BM869s can add more resolution but not accuracy.

Should you need the best accuracy probably it's the benchmeter that can be trusted.

[CDaniel]

BM869S is showing a little bit less for DC , if the bench multimeter is for sure accurate . That's why AC + DC is closer if that reference has the "right " amount of noise ... just a coincidence in this particular case

[ercapoccia]

Thanks for your suggestions.

I'm sure it is just a coincidence that reading were matching.

I don't know how a multimeter works in details, but in my mind there was a bandwidth related measurement. So i thought that a more advance multimeter would isolate better the dc part of a signal. This is what i meant.

One of my modules has not been really tested with a bench multimeter because i saw the same module with the same values in a video on yt.
The other module who knows, maybe those number are just made up, with just one cheap noisy reference i can't say what is what. If i had to bet i would trust my DMM.

I'm more an electrical than electronic guy.