Test and Measurement Forum

Excessive noise.

Having problems ? let us know the details here

Excessive noise.

Postby musinteg » Tue Jan 23, 2018 5:07 am

I have had my 2206b for a few months now and seem to be at a dead end.

The main issues are noise and triggering.

Noise: I'm not sure where the noise is coming from, but the scope detects excessive amounts of it, where it does not exist. I am a 30 + year audio professional repair technician and own some good gear. My main piece of gear is my Amber 3501 noise and distortion measuring system.

When I hook up the picoscope to the amber, so I can look at an output wave, there are excessive amounts of noise present on the signal, according to the picoscope.

My Amber is certified that the output never exceeds the -99dB threshold for noise on an unbalanced signal, yet on my picoscope spectrum analyzer, the noise floor is at roughly -62 dB ( basically useless ). That is plugging directly bnc to bnc, using shielded cable with 99% braid and 100% foil shielding.

This "noise" also plays havoc with the triggering. No matter what I do, I cannot get a nice stable trigger unless the signal is above 3v pk.

This would not be as intolerable as it is, but the noise causes nothing but triggering issues, causing even simple audio measurements to be frustrating.

How can I get rid of this noise ?

I have tried using the filtering on the input channel menus and all that does is take it off the screen, but it is still messing up the triggering. I have also tried the resolution enhancement up to 12 bits.....same thing. It looks better but the functionality is still compromised.

I have tried different grounding schemes, suspecting noise difference issues between my gear and nothing ever improves. If I plug in a good old analog scope, everything is fine.

Musinteg.
musinteg
User
User
 
Posts: 3
Joined: Mon Nov 06, 2017 3:38 am

Re: Excessive noise.

Postby Gerry » Wed Jan 24, 2018 10:28 am

Hi Musinteg,

Just to give you the correct expectation, the PicoScope 2000 series are 8-bit Scopes, which means that, on the smallest range of +/-50mV, the smallest step size that you can get for a digitized signal is just over 390uV. The noise specification for the PicoScope is <220uV on the +/-50mV range. Now, even if you have a minimal noise environment, realistically, as signals have multiple levels, the noise will be spending a fair amount of time straddling between 2 digital values, which means that the actual minimum noise level will be somewhere between 390uV and 780uV (up to 2 bits). So, the best signal to noise ratio you can hope to get at the input to the converter is between 256 to 2, and 256 to 1, which is a ratio of between -42dB and -48dB.

You can usually see better SNR values in the Spectrum analyzer because the noise floor is artificially pushed down by the process of performing the FFT (called Process Gain) according to the number of bins used for the calculation. However, the bandwidth of the Spectrum plot has significant impact on the SNR, so if you have been restricting everything the audible spectrum for analysis with the Amber analyzer, then you should do the same with the settings in PicoScope 6 Spectrum Mode.

So, with all that in mind, we need to identify how much of the problem is incorrect settings before looking at how to fix what you are seeing. Could you post psdata files of both a Scope Mode capture, and a Spectrum Mode capture, so that we can see your trigger setup, and captured data. Could you also include your SNR measurement.

Incidentally, the PicoScope 2206B is not really a good choice for audio. Ideally you need a low noise high resolution PicoScope such as the 16-bit PicoScope 4262.

Regards,

Gerry
Gerry
Technical Specialist
Gerry
PICO STAFF
PICO STAFF
 
Posts: 226
Joined: Mon Aug 11, 2014 11:14 am

Re: Excessive noise.

Postby musinteg » Wed Jan 24, 2018 9:00 pm

I posted one file but the scope file is too big and the forum will not let me post it.

The test was from my amber analyzer doing a self test for IMD using a standard SMPTE 60Hz - 8kHz mixed waveform.

I also took a picture of the unit during self test and it read 0.0012% IMD, ( around -98dB), but that is not uploadable either due to the size.

I am very disheartened at the explanation of the issue and the fact that I now need an almost $2000 dollar canadian ( with taxes ) product from pico, just to accurately look at audio waveforms is totally impractical. Also , with a bandwidth of only 5MHz the suggested product will not be able to examine the eye pattern of either a CD or DVD player, as that required a minimum bandwidth of 40MHz.

Musinteg
Attachments
IMD test analyzer view.psdata
(42.03 KiB) Downloaded 121 times
musinteg
User
User
 
Posts: 3
Joined: Mon Nov 06, 2017 3:38 am

Re: Excessive noise.

Postby Gerry » Mon Jan 29, 2018 12:28 pm

Hi Musinteg,

First of all, I'm sorry that you're disheartened about the explanation, that wasn't my intention. I had to give you an explanation and, unfortunately, there is no workaround for the laws of physics (an 8-bit digitization process has quantization noise that fixes how low the noise floor can go).

However, regarding the PicoScope 4262 comment, perhaps I should have said the PicoScope 2206B is not really the best choice for audio, because it CAN be used for a number of audio applications where you don't necessarily need high accuracy, (for example where you are investigating a filtering effect on your signal, and just need the general outline of a Bode plot to look at). Also, we take pride in the fact that we do have quality instruments that can be used for high accuracy work, and the PicoScope 4262 is one of them which is priced accordingly, as are all higher accuracy instruments (as I'm sure you also found with the Amber 3501 noise and distortion measuring system).

That said, we do understand that choosing the correct PicoScope for an application is not always the easiest of tasks (which is why we have a very knowledgeable Technical Specialists on our help desk to guide customers through the process) so, if after we resolve your triggering problem, you still feel that you need higher accuracy, there may be something that we can do to help rectify your situation (in which case you should contact our help desk at support@picotech.com).

So, to resolve the triggering we do need to look at your data files. Could you upload them to a cloud storage drive (such as Google Drive, or Drop Box) and give us the link to download them.

Regards,

Gerry
Gerry
Technical Specialist
Gerry
PICO STAFF
PICO STAFF
 
Posts: 226
Joined: Mon Aug 11, 2014 11:14 am


Return to Getting Started

Who is online

Users browsing this forum: No registered users and 0 guests