crosstalk between multi channel of PS2407B

Post any questions you may have about our current range of oscilloscopes
Post Reply
minjie
Newbie
Posts: 0
Joined: Sun Jul 07, 2019 11:44 am

crosstalk between multi channel of PS2407B

Post by minjie » Sun Jul 07, 2019 2:40 pm

Hello,

We just get started with the PS2407B with a laptop. We noticed that when more than one channel are switched on in the PicoScope software, some noise is induced on the scope.
PS2407B was connected with a laptop(later also checked with PC and same behavior).
A 1kHz pulse signal with 2V Max voltage was sent to channel C. The statistic of Max value shows sigma of ~10mV.
As channel D was switched in the software, the sigma raised to ~60mV, whenever the physical port was open, connected with a 50ohm resistor, or any other input.
The screenshot and data file of both cases are attached.

Would you please identify how does this noise come from and how can we prevent it?
Thank you very much.

Best regards,
Minjie
Attachments
chCchD.png
chC.png
20190705-chCchD.psdata
(16.41 KiB) Downloaded 22 times
20190705-chC.psdata
(13.29 KiB) Downloaded 19 times

Martyn
Site Admin
Site Admin
Posts: 3710
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: crosstalk between multi channel of PS2407B

Post by Martyn » Mon Jul 08, 2019 8:49 am

I am not sure that this is noise, but is instead linked to the difference in sampling rate between the two captures and the application of Sin(x)/x Interpolation.

If you right click on the display and select "View Properties" you will see that you have 1nsec sampling interval for one active channel and 2nsec interval for two active channels. If the value beside the "No. samples" has an additional number in brackets this means that interpolation is being applied to fill in the gaps between samples. I would suggest going to Tools->Preferences->Sampling and turn this off to see what the raw data is like. You will then see the individual samples around the peak you have and how this differs between each buffer that is collected.
Martyn
Technical Support Manager

minjie
Newbie
Posts: 0
Joined: Sun Jul 07, 2019 11:44 am

Re: crosstalk between multi channel of PS2407B

Post by minjie » Mon Jul 08, 2019 5:17 pm

Hello,
Thanks for the reply. We checked as you suggested but I don't see different behavior.
If I look at the property when switching on more channel, the no. samples decrease as more channels are activated. Statistics also get worse with more channels, reached the largest number with two idle channels(A:signal;B,C:activated but not connected).
Best regards,
Minjie
Martyn wrote:
Mon Jul 08, 2019 8:49 am
I am not sure that this is noise, but is instead linked to the difference in sampling rate between the two captures and the application of Sin(x)/x Interpolation.

If you right click on the display and select "View Properties" you will see that you have 1nsec sampling interval for one active channel and 2nsec interval for two active channels. If the value beside the "No. samples" has an additional number in brackets this means that interpolation is being applied to fill in the gaps between samples. I would suggest going to Tools->Preferences->Sampling and turn this off to see what the raw data is like. You will then see the individual samples around the peak you have and how this differs between each buffer that is collected.
Attachments
20190708-chABCD.psdata
(16.89 KiB) Downloaded 20 times
20190708-chABC.psdata
(14.7 KiB) Downloaded 19 times
20190708-chAB.psdata
(14.77 KiB) Downloaded 23 times
20190708-chA.psdata
(13.78 KiB) Downloaded 17 times

minjie
Newbie
Posts: 0
Joined: Sun Jul 07, 2019 11:44 am

Re: crosstalk between multi channel of PS2407B

Post by minjie » Fri Jul 19, 2019 12:44 pm

Hallo Martyn,
Would you please help with this issue? Thank you very much.

Martyn
Site Admin
Site Admin
Posts: 3710
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: crosstalk between multi channel of PS2407B

Post by Martyn » Fri Jul 19, 2019 1:39 pm

When you add in more channels, the sampling rate drops, so with your narrow pulse you can't always guarantee to take a sample at the highest part of the pulse. Therefore any peak measurements you take will be compromised.
Martyn
Technical Support Manager

Post Reply