5203 Bandwidth measuring?

Post any questions you may have about our current range of oscilloscopes
Post Reply
Samy
Newbie
Posts: 0
Joined: Thu Nov 12, 2009 7:10 pm

5203 Bandwidth measuring?

Post by Samy »

Hello,

I bought a 5203 because I want to use it to measure frequency, power, bandwidth.
With the bandwidth I have a problem. I tried it with spectrum mode at:
Spectrum Range 125Mhz
Spectrum Bins 16384
Window Function Blackman
Display Mode Average
Scale Logarithmic
Logarithmic Unit dBm 50 Ohm


First I searched a radio station and found it at 98.5 MHz with peak at -57,19dBm.
From the peak I go down with the second level ruler to -3dB of the peak.
With the two frequency rulers I go to the crossingpoints left and right of the signal
and I get a bandwidth of 17,23kHz. I think a stereo broadcast should be wider than
17,23kHz. If I'm right the chanal bandwidth is a max. of 150kHz.

Now I'm very confused because I tried to change the spectrum bins up to 65536, this should
be 1,9kHz Bin width, before it was 7kHz, so I hoped to get a better result. What I get
is something complete other then in the first try. Now I measure a peak of -65,3dBm that
is a really to big different. At -3dB under the peak I measure a bandwidth of 6kHz.

Of course there will be different between measuring with 16384 or 65536 Bins but I never
will know what I measure because on every changing the Spectrum Bins I will get an other
result.

With the radio station it is only an example, I tried also with other signals like the
output of amateur radios on different frequencys.

How can I get a result of measuring which is a exakt measuring which I can trust?

Samy

PeterF
Advanced User
Advanced User
Posts: 435
Joined: Thu Jun 07, 2007 10:53 am
Location: Cambridgeshire

Re: 5203 Bandwidth measuring?

Post by PeterF »

Hi Samy,
You are set to "averaging" which will reduce the side-bands visible on the signal unless the station is transmitting at maximum modulation continuously. You are, at present, in effect, just looking at the carrier frequency. You will need to switch to "Peak Hold" and wait a while for the spectrum to develop. The signal may have a maximum bandwidth of 150kHz but the station will normally use less than that because of less than 100% modulation. Selecting different settings and numbers of bins will, of course, give different results.
Regards,
PeterF.

Samy
Newbie
Posts: 0
Joined: Thu Nov 12, 2009 7:10 pm

Re: 5203 Bandwidth measuring?

Post by Samy »

Hello PeterF.

many thanks for your answer.
Selecting different settings and numbers of bins will, of course, give different results.
With this I have a little problem :wink:

If I use the scope mode and I measure 150mV with an Input Range 200mV I will see this 150mV also with an Input Range of 500mV / 1V etc. The result will not differ at factor 6.
So if I don't missunderstand you, it is not possible to use the spectrum mode for measuring signals because the result will differ in a wide range from the real signal.

If I want to measure a power output of a small radio with an output of 2W it is not possible to say something about the output power because it could be 2W or 12W (8dBm) because I use 16k bins or 65k bins. The spectrum is the bigger reason for me to buy the 5203 so now I'm not sure if I bought the right product for my need.

Samy
Ps: Is there a German support avaible?

PeterF
Advanced User
Advanced User
Posts: 435
Joined: Thu Jun 07, 2007 10:53 am
Location: Cambridgeshire

Re: 5203 Bandwidth measuring?

Post by PeterF »

Hi Samy,
I am afraid we do not have German support except via "Google Translate" on the Help-desk.
Switching between the different number of bins and types of spectrum plot should not have nearly the effect you find. Measuring power in spectrum mode is different from normal voltage measurements in scope mode in that a peak reading at the fundamental frequency is just the power at that frequency and it ignores power in the side-bands and harmonics. Spot readings are more accurate with a high number of bins but the refresh rate is much slower. To get a better total power figure in spectrum mode the carrier power must be added to the power in the major side- bands and harmonics. You must be doing something wrong if you get a variation of 8dB with different settings. Are you sure you have the right voltage range set when in spectrum mode? If the waveform is being "clipped" that will give to major errors.
Regards,
PeterF.

Post Reply