# Frequency Measurement - Oscilloscope

#### Mazaag

Joined Oct 23, 2004
255
Hey guys,

I'm working with a Trektronics digitl oscilloscope to measure the frequency content of some signal.. I was wondering if someone can give me a DSP explanation as to what settings I require in order to avoid aliasing..

lets say that i am interested in measuring the frequency content at 100Hz.. what do I have to set the time base to be (or the sampling rate) ? and how does it relate to the samples per second of the oscilloscope...

I hope my question is clear...

Thanks guys

#### bloguetronica

Joined Apr 27, 2007
1,521
Given that example, the time base would be 1/100s. The period of a signal, measured in seconds, is the inverse of it's frequency.

#### Mazaag

Joined Oct 23, 2004
255
..but doesn't the sampling rate have to be atleast twice the frequency of interest? Thats as far as I know..but my question relates to the connection of the time base to the sampling rate of the actual oscilloscope...how can I determine the actual sampling rate used by the oscilloscope (which ii'm assuming is related to the time base setting)

#### bloguetronica

Joined Apr 27, 2007
1,521
..but doesn't the sampling rate have to be atleast twice the frequency of interest? Thats as far as I know..but my question relates to the connection of the time base to the sampling rate of the actual oscilloscope...how can I determine the actual sampling rate used by the oscilloscope (which ii'm assuming is related to the time base setting)
Indeed, the sampling rate would be at least twice the frequency of interest, so you could make a better visual analysis. I don't know your model, but the models I knew would auto sync with the signal. The only thing I did was adjusting the coarse frequency.

#### Mazaag

Joined Oct 23, 2004
255
the idea is that I have low frequency content that I would like to detect, so I wanted to know what to set the time base setting to in order to detect such components....

the model i'm using is a Tektronix TDS2004... it says that it has a Sampling Rate of 1.0 GS/s.. (with a bandwidth of 60Mhz.) now i'm assuming that is the maximum rate that it could sample at... is that correct?

It also says that the Record Length is 2.5K points on all time-bases... I have no idea what that means...

There is also a possibility to use different types of windows.. ( Rectangular, Hanning , Flat-top)

How do all these (Sampling Rate, Record Length, Window type) relate to each other and to measuring a specific frequency of interest...?

Thanks guys

#### beenthere

Joined Apr 20, 2004
15,819
Ah, the joys of DSO's. You might want to look on the web to see if you can scare up a manual for the beast, if you haven't got one.

I have not used that beast, but they are prone to aliasing. You might practice on a known waveform, like the output of a function generator to see if the reproduction on the screen resembles the signal.

#### nanovate

Joined May 7, 2007
666
It also says that the Record Length is 2.5K points on all time-bases... I have no idea what that means...
It means that it will only give you 2500 data points per file-- limits your "recording" time.