How to measure the Vpp, Vrms of a low frequency signal (<10Hz)

[Rajashekar]

Hi,

Can anyone suggest me to measure the Vpp, Vrms of low frequency signal (<10Hz i.e. 1Hz, 2Hz, 0.16Hz, 0.8Hz etc.) through DMM (PXI 4070) or Digitizer (PXI 5122).The DMM is not able to read the rms value of lower frequency and through Digitizer i am not able to get the accurate value.

Is there is any different methodologies in measuring low frequency signals.

Regards,

Rajashekar

[crossrulz]

How are you currently calculating them?  Can you show some code with some sampled data?  What rate are you digitizing?  For how long?

[johnsold]

The DMM should do fine if you use it in the digitizer or DC modes.  Sample at about 1000 samples per second and then do the peak and rms calculations in software.

Lynn

[crossrulz]

I'm in agreement with Lynn except for 1 thing, but it really depends on how accurate you really need to be.  I had a test setup for something similar (1/2Hz up to 40kHz).  I found that in order to get the accuracy we needed, I had to truncate the waveform to have an integer number of cycles.  In order to do this, you need to know your sample rate and your test frequency.  If you think about the units, you can figure out the calculations needed.  I don't have them with me nor am I allowed to post them.

[johnsold]

Crossrulz has pointed out something important: When doing rms calculations on digitized data you must work with data sets containing integer numbers of cycles.  The mathematical definition of the Mean part of rms applies to one complete cycle of the periodic signal.  Hardware instruments use some kind of time-based weighting to minimize the effects of making continuous measurements. (And some of them work equally poorly at low frequencies).

If your signals have relatively low noise, zero-crossing techniques can be used to find the beginning and ending of cycles.  If your signals are noisy and the frequency is not known independently, it can get quite difficult.

Lynn

[Rajashekar]

For lower frequencies i am configuring my digitizer with sample rate and record length to capture atleast one full cycle. With the captured waveform i am extracting the Vm value using the Extract Single Tone Information.vi function and then calculating the Vpp or Vrms using the Vm. Please refer the attache files for Digitizer Configuration.

For frequency 1Hz onwards the sample rate: 2000 and record length: 6000 is configured.

For frequency <1Hz the sample rate: 2000 and record length: 18000 is configured.

In some cases i use the niScope Read Measurement.vi function for reading the Vpp or Vrms value. The auto measurement of the scope will consider some noise peaks in the signal. There is a difference between the auto and manual measurement reading (taking the measurement by using the cusror in the voltage domain) reading the of signal in oscilloscope which i checked in the external tektronics oscilloscope.

Is there any way to configure the lower frequency (filter) settings in the DMM for measuring the Vrms of low frequency signals.

Regards,

Rajashekar

[crossrulz]

---

@Rajashekar wrote:

Is there any way to configure the lower frequency (filter) settings in the DMM for measuring the Vrms of low frequency signals.

---

Unfortunately not.  The AC measurements for a DMM typically AC couple the input using a capacitor.  And who knows what frequency that will cut off at.  You are pretty much stuck digitizing and analyzing.

[AnalogKid2DigitalMan]

From the specs:

AC Specifications

Note

All AC speed specifications apply with Auto Zero disabled.

AC System Speeds

Range or function change ...................... 10/s

Auto Range time, AC V and AC I ......... 250 ms

Trigger latency ....................................... 2

μs

Maximum trigger rate ............................ 1 kHz

Readings/s NMRR Conditions

10 >100 dB

* All noise sources >46 Hz

50 (60) > 60 dB

† 50 (60) Hz ±0.1%

*

With high-order DC noise rejection; 100 ms aperture.

†

With normal DC noise rejection; 20 ms (16.67 ms) aperture.

Digits Reading Rate Bandwidth

6½ 0.25 S/s 1 Hz to 300 kHz

6½ 2.5 S/s 10 Hz to 300 kHz

6½ 25 S/s 100 Hz to 300 kHz

6½ 100.0 S/s 400 Hz to 300 kHz

5½ 1.0 kS/s 20 kHz to 300 kHz

-AK2DM

[Kevin_Price]

Just repeating something again for emphasis.  You said:

For lower frequencies i am configuring my digitizer with sample rate and record length to capture at least one full cycle

What has been recommended is to capture for exact integer multiples of one full cycle.

-Kevin P

[Rajashekar]

How to capture the signal for exact integer multiples of one full cycle? Can you provide some details please.