Low AC voltage measurement issues

[pwnell]

I am trying to measure an AC 1kHz signal’s RMS voltage using a Keysight 34465A bench DMM.  However it measures 000.000 0mVAC.

I have ac filter to >3Hz. Manual 100mV range. Using bnc cable. I am generating a 1kHz 17.666mV signal using my signal generator. That I measured using the Keysight. I feed that into a simple 46.64k + 463 ohm resistor voltage divider network to get out 0.175mVAC RMS. Yet when I measure there, I get 0 as per above. I know there is voltage as the output of the voltage amp this signal feeds into, reflects the amplified signal.

What am I doing wrong? Surely this DMM can measure 175 micro volt?

[alm]

Why would you expect the DMM to measure 175 uV RMS AC? The datasheet has this footnote for the ACV specifications:

Specifications are for sine wave input > 0.3% of range and > 1 mVrms

Which is much better than for its predecessor, the 34401A:

For sine wave input > 5% of range. For inputs from 1% to 5% of range and < 50 kHz, add 0.1% of range additional error

Joe Geller wrote an article about the limitations of the 34401A for measuring low level AC (I'm linking to an archived version since he took his website offline).

Any meter using an analog RMS converter will have such a limitation due to the noise/offset of the converter, which, assuming it's uncorrelated with the input signal, adds as the root of squares. They normally won't read zero with shorted inputs (though presumably in the 34401A there's a firmware hack to round small values down to 0). The ones that digitize the AC signal, like the 34465A, do better, but still have offsets and noise.

[pwnell]

Thanks for the response. That paper makes it sound that with the improved 34410A model they were able to measure down to 100uV. I am trying 175uV with a better meter.

When using the digitizer I can clearly see the sinusoidal wave, and using the stats, calculating the rms from Vpp gives a result of approx. 367uV which is wrong (input is pure sine wave).  Is there any way I can make that measurement with my meter?

[bdunham7]

What am I doing wrong? Surely this DMM can measure 175 micro volt?

I think it is not specified to measure 175uVAC.  Interestingly, an ancient, obsolete, cheap HP 403B will measure it just fine!

It is interesting that they choose to show you 000.000mVAC rather than some noise.  Not very sporting, IMO.

[Andreas]

Hello,

attached a linearity comparison between 34401A K2000 and my oscilloscope for AC voltages <1mV in 100 mV range.

with best regards

Andreas

[David Hess]

Any meter using an analog RMS converter will have such a limitation due to the noise/offset of the converter, which, assuming it's uncorrelated with the input signal, adds as the root of squares.

I have not seen an RMS converter which does it yet, but the active precision rectifier used for the first stage can have zero offset because it can be AC coupled at the input and output to remove all DC error, which gives the possibility of noise down in the microvolt range to 10s of kHz for a modern design without any offset voltage correction.

[pwnell]

I tested the accuracy of my 34465A:

I used a 1kHz signal from the MSO5104 signal generator at 600, 300, 150, 100, 75 and 50mV Vpp output.  I measured the RMS using two methods - using the Vac measurement from the Keysight, and the digitizer.  For Vac I used the graph over time to get an average value.  For the digitizer I took between 6 - 12 sine waves, measured Vpp and calculated Vrms.  I also measured the input to the resistor divider I built (using 100kΩ and 1kΩ 0.1% metal film, 5ppm/C Ohmite resistors) using the Keysight Vac measurement, and previously I calculated the average actual resistance of the two resistors over a 20 minute period.  Using ohm's law I calculated the theoretical value.To me it seems like the digitizer is pretty good down to 174µV - with an error of 1.38%.  The Vac function started diverging at approx. 350µV with an error of 7.57%. 

Only issue is the digitizer only works for pure sine waves - I cannot easily calculate RMS for mixed frequency signals.

[Conrad Hoffman]

How 'bout a preamp?

[pwnell]

How 'bout a preamp?

Yeah that is my second problem: https://www.eevblog.com/forum/beginners/low-noise-voltage-opamp-oscillating/msg3861761/#msg3861761

[srb1954]

How 'bout a preamp?

Yeah that is my second problem: https://www.eevblog.com/forum/beginners/low-noise-voltage-opamp-oscillating/msg3861761/#msg3861761

Or use a LF spectrum analyser (or a suitably configured sound card) and do some simple maths to get the RMS sum of the harmonic components.

[jonpaul]

Bonjour a tous:

As designer and consulting in audio and digital audio since 1970s, we are very familiar with low level AC measurements.
The old TEK AA501  audio analyzer has great low lev capability 200 uV FS.

 We also used the HP analog meters like 400FL, and  3400A, 1 mV ranges and good noise figure.

Fluke digital RMS wideband 8920A has 1 mV FS and 28 uV RMS with 50 Ohm term.

Using these meters and the SG505 TEK with AA501, we could verify the AC capabilities of both Keysight 34465A and our old Fluke 8842A on AC at freq up to 300 kHz and levels well within specs.



We recommend use of a LP filter to limit noise to band of  of interest eg voice or audio.

J'espère que mes notes prouvées d'intérêt.


Jon