Test and Measurement Forum

Trouble FFT and distortion mesure on Pico 2203

Post any questions you may have about our current range of oscilloscopes

Trouble FFT and distortion mesure on Pico 2203

Postby Glubbein » Mon Jul 02, 2012 3:07 pm

Hello,

I bought a PICO 2203 about one year ago, but starded to use it juste a few weeks ago. Everything works fine except distortion analysis

To show you the problem I've done two tests :
a) with a very low distortion sinus generator (Marconi TF 2005R), an HP8903 distortion analyseur (witch cane give me TRS and TDH (in %)) in one hand, and the Pico2203 (using same cable than with HP8903) one the other hand.
- the THD value given for a 200Hz signal is : HP : 0,07 PICO : 0,78
- the THD value given for a 633Hz signal is : HP : 0,038 PICO : 0,157
- the THD value given for a 6kHz signal is : HP : 0,033 PICO : 0,229
RMS and frequency values are the same with HP and PICO.
The Pico FFT analyzer gives me the folowing screens (Hanning window) :
200Hz signal :
FFT Hann Average Marconi 2.jpg

633Hz signal :
FFT Hann Average Marconi.jpg


b) with the Pico AWG (at 600Hz) with signal out shorted with channel A (using the same cable as before and Hanning window) :
Sinus 600Hz AWG.jpg



As ou can see something is wrong... Is this due to sofware or hardware???

Thanks a lot for helping.

Best regards.

Glubbein
Glubbein
Newbie
 
Posts: 0
Joined: Mon Jul 02, 2012 2:14 pm

Re: Trouble FFT and distortion mesure on Pico 2203

Postby ziko » Fri Jul 06, 2012 8:45 am

HI and thank you for your post.

From the post you sent I can tell that you have put averaging on (picture says it!), but I require more information about your setup. For example you say Hanning Window do you mean Hamming or do you mean Hann window? Also what is the amplitude of the signal that you are feeding in.

Also you seem to have a big frequency span across your signal, 5MHz, I would drop the frequency range down to say something like 9.766 kHz or lower.

Could you send a settings file?

Bear in mind this is an 8 bit oscilloscope and not a dedicated spectrum analyser.

Kind regards
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Trouble FFT and distortion mesure on Pico 2203

Postby Glubbein » Sun Jul 08, 2012 12:56 pm

Hello Ziko,

thanks a lot for your help.

1°) I've done a mistake : the amplitude setting in spectrum analyzer was wrong (too low ---> big distortion!!) :oops:

2°) But I still have many problems :
a) mesurement of THD (in %) :
signal : 1,7kHz sinus 3V peak to peak
spectrum settings : 39,06kHz Hann Window average amplitude : +/-20V to +/-2V (see below)
+/- 2V ---> 0,07%
+/- 5V ---> 0,03%
+/- 10V ---> 0,05%
+/- 20V ---> 0,67%
With my HP8903E (setting with 30kHz filter to reduce bandwidth) : 0,01%
b) worth :cry:
I put a 250Hz, 1V peak to peak in both channel (at the same time) with my Marconi generator :
results are completely different on channel A and on channel B...
A and B +-2V 250Hz 1V.jpg

I thought it could be my cables, so I invert the two cables : still the same!
A and B (inverted cables) +-2V 250Hz 1V.jpg

Why do I don't have the same specrum view in both channels??? Why do I have so many harmonics?

Best regards

Glubb
Glubbein
Newbie
 
Posts: 0
Joined: Mon Jul 02, 2012 2:14 pm

Re: Trouble FFT and distortion mesure on Pico 2203

Postby Glubbein » Mon Jul 09, 2012 6:03 pm

Hello,

I found a video here in the topic named : "Picoscope 4262 distortion mesurement floor". So I dicided to do the same operation than in the video.

1°)Using exactely same settings (spectrum frequency range=5MHz) :
- signal : AWG sinus 1kHz 900mV
- acquisition : 100ms/div 32 100ks +/-1V
- spectrum : Blackman window, dBu, frequency range=5MHz (only point different), average bins : 16384
See the spectrum view : :cry:
AWG 1kHz 900mV Black dBu 1V.jpg

Two mesurements :
- SNR : 11,6dB
- THD : 168%


2°)Using exactely same settings but with spectrum frequency range=39,06kHz :
AWG 1kHz 900mV Black dBu 1V 39,06kHz.jpg

Two mesurements :
- SNR : -48,81dB
- THD : 0,141%
OK it's better, but even with an 8bit scope, is it normal? Could it be the noise of the Picoscope?

Thanks a lot for answering.

Best regards.

Glubb
Glubbein
Newbie
 
Posts: 0
Joined: Mon Jul 02, 2012 2:14 pm

Re: Trouble FFT and distortion mesure on Pico 2203

Postby ziko » Thu Jul 12, 2012 11:31 am

Hi Glubb,

I will answer the first forum post you sent on Sunday first :

Try using a repeat trigger instead of an auto trigger as you may be getting skewed results due to hysteresis, particularly on the larger voltage ranges, +/-20V.

In terms of your two signals try using the same voltage range, it looks to me like you are using channel A on +/-1V and channel 2 on +/-2V.

In terms of your second post sent on Monday, the signal generator on the 2203 is NOT a low distortion sine oscillator and therefore it will have noise on it, its NOT meant to replace a dedicated signal generator, particularly a low distortion one.

An 8 bit scope has a dynamic range of -48dB which is 256 steps and therefore 20*log(1/256) = 20* log (0.00390625) = 20*-2.4082 = -48.165dB.


Hope this helps.
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Trouble FFT and distortion mesure on Pico 2203

Postby Glubbein » Thu Jul 12, 2012 2:44 pm

Hell Ziko,

1°) repeat trigger instead of auto trigger : it doesn't change anything :
signal : 1,7kHz, sinus, 3v p to p (from my Marconi generator)
acquisition with repeat trigger, both channels at +-10V
Spectrum : Hann, average,... both channels at +-10V
Marconi 1,7kHz 3V Ch A B +-10V.jpg


2°) OK for the information about the AWG. Is it the same for every Picoscope's series?

3°) What I don't understand it's why I have so much harmonics when the signal comes from my low distortion Marconi generator :
Marconi 1kHz 900mV Black dBu repeat trigg2.jpg

Is it due only to the 8 bits resolution DAC of the 2203 Picoscope, or could't it be other thing (hardware disfunction)?

Best regards.

Glubb
Glubbein
Newbie
 
Posts: 0
Joined: Mon Jul 02, 2012 2:14 pm

Re: Trouble FFT and distortion mesure on Pico 2203

Postby ziko » Fri Jul 13, 2012 11:07 am

Hi Glubb,

Some of the higher end products will have better signal generators but none of them will be as good as a dedicated signal generator.

The harmonics you see are because of the fact that the scope is an 8bits that when sold was less than £200. The other thing I notice is that you are using +/-10V signal on a +/-1.5V signal why are you not using +/-2V?

The scope itself is an entry level scope you really need to use something like the 4262 oscilloscope This is a 16 bit oscilloscope and has a significantly better sine wave generator.

Please see attached image of a 4262 with same settings, 1.7kHz +/-3V signal, Averaging and Han Window with a spectrum range of 100kHz.

Regards
Attachments
4262.jpg
1.7kHz Sine Wave from Black Star LDO Sine/Square Oscillator
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Trouble FFT and distortion mesure on Pico 2203

Postby Glubbein » Mon Jul 16, 2012 11:31 am

Hello Ziko,

thanks a lot (especially for the last picture)... Now everything is clear : my problem is the fact that the 2203 is an 8bits, and an entry level scope (as you say), and nothing else!
A 4262 would be the solution, but since my use is not a professional one, the cost of the 4262 is a problem :roll: I must find another solution...

Regards.

Glubb
Glubbein
Newbie
 
Posts: 0
Joined: Mon Jul 02, 2012 2:14 pm


Return to USB PC Oscilloscopes

Who is online

Users browsing this forum: No registered users and 0 guests