Low signal level in Spectrum option

[STMN]

Hi,
I’ve got a question about the spectrum option. In the attached screenshot you see an 455kHz IF filter measurement on channel A. AWG output is generating a sweep signal 400-500kHz for the filter input. My output signal on channel A is 80mVpp, but in the spectrum view it is shown as a -78dBµ level.
Could you please tell me, why the represented output level is so low? -78dBµ would actually mean 275µVpp.

Many thanks in advance.
STMN

[Gerry]

Hi STMN,

The answer to your question is because the time domain signal that you are looking at has frequency components spread out across the spectrum (which you can see in the Zoom overview). If you were to add up all of the frequency components, in the correct phase, then they would total to 170mV p-p (which is includes the +ve and -ve peaks).

Regards,

Gerry

[STMN]

Hello Gerry, and thanks for that information.
What would be the correct settings to sweep an 455kHz IF filter? Is it possible to do it in the spectrum mode, or is the X/Y mode more recommended for that purpose?

Best Regards.

STMN

[Gerry]

Hi STMN,

The answer to your question depends upon what you are trying to achieve with your measurement. So, what exactly are you hoping to see from your measurement?

Regards,

Gerry

[STMN]

Hi Gerry,
thanks for that, my intention is to measure the filter curve, and to determine the center frequency of that IF filter. What would you recommend for this case?

Best Regards.

[Gerry]

Hi STMN

Sorry for the delay in answering.

I'm assuming that by 'filter curve' you mean response curve for the filter, in which case you can do what you ask by using some software created by another one of our users that performs a frequency response analysis, see here: https://bitbucket.org/hexamer/fra4picoscope/wiki/Home.

Regards,

Gerry

[STMN]

Hello Gerry,
sorry for late answer and thank you for your help,
yes, it’s response curve what I mean. I used spectrum analyzer mode for that measurement in attached screenshot. For this purpose I need linear frequency axis, I guess It’s not possible to change the X axis to linear in RA4PicoScop, am I right?

Best Regards.

[Gerry]

Hi STMN,

That would be a question to pose to the users and designer of the software here: https://www.picotech.com/support/topic14311.html?&p=143080&hilit=hexamer#p143080.

Regards,

Gerry

[STMN]

Hello Gerry,

Many thanks for that. I’ve got a question regarding X/Y mode. I learned that the signal generator is internally synchronized to oscilloscope. The sweep generator is beginning at 400kHz and stopping at 500kHz. Is that possible in X/Y mode to measure exact time I have marked in the attached screenshot? For this purpose I have to trigger with the “starting impulse” of sweep generator (sync impulse), but not by the voltage level on channel A, as it shown in the picture. In some user cases it’s not possible to measure in spectrum analyzer mode (signal mixer in the signal path etc.). As an alternative method I need the X/Y measurement to determine center frequency of IF in that way. I couldn't find a way to use a FRA tool for it. Or do you have better idea for that?

Many thanks in advance.
Best Regards.

[Gerry]

Hi STMN,

For most PicoScopes in our range the DAC in the Signal Generator and ADC in the Acquisition channel run from the same clock, so whether or not yours does depends upon the model of PicoScope that you have. Which one do you have?

However, for the PicoScopes that do use the single clock, the Signal Generated frequency is not tied to the Acquisition rate in any way other than the fact that it is ultimately derived from the same clock (i.e. it can be a seemingly unrelated frequency). The Signal Generated frequency comes from a Direct Digital Synthesizer, which can create very precise ultra high resolution increments in frequency.

Unfortunately, if you are using X/Y Mode then you are giving up the Time axis, so you would need something like a 3 dimensional plot to measure Time as well as the 2 relative channels, which we don't have.

Could you explain, in detail, what your measurement goal is (what are you trying to do with the Intermediate Frequency of a mixer and why are you trying to do it?).

Regards,

Gerry

[STMN]

Hello Gerry,
thank you for that, my Picoscope is 2206B. My goal was to see the response curve of the intermediate filter (in a receiver device) and to see the resulting center frequency of that filter, which can be little different. This works typically with a sweep generator and a some scope in X/Y-Mode (both devices synchronized with each other), in this way I measure the time and can get the center frequency of the filter.
That task can be done well with Picoscope in spectrum analyzer mode, but I noticed that I cannot reproduce that measurement result in X/Y-Mode, the reason is probably giving up the time axis, as you explained it before.
My intention was to compare both measurements in getting the same center frequency as a result.

Best Regards

[Gerry]

Hi STMN,

O.K. now that you have explained what your goal is, you can visualize the time measurement with a little more work, i.e. by capturing the X axis data on one channel, the Y axis data on a second channel, and then exporting the data file as CSV to import into Excel, or exporting a CSV or mat file to import into Matlab. In Excel you can create 3d plots as explained here: http://www.graytechnical.com/blog/xyz-data-inside-excel/, or more easily in Matlab by using its graphics functions. You may then be able to, for instance, see where the centre of the x/y co-ordinates are and guage where that point is on the z-axis.

Regards,

Gerry