[Solved] PS4824: First samples are wrong

Post any questions you may have about our current range of oscilloscopes
Post Reply
LouisRast
Newbie
Posts: 0
Joined: Mon Feb 02, 2015 1:30 pm

[Solved] PS4824: First samples are wrong

Post by LouisRast »

Hy,
I'm using the 4824 with a custom C# application.
I have the following issue which I was not able to resolve yet:

When starting an AC coupled measurement in Block mode, the first thousand samples are wrong.

Test setup
Input Signal: 600mV DC off a Lecroy waveStation2012
Picoscope channel configuration: Channel A, Range 1V, Coupling AC
Sample count: 1300, Timebase: 8100

Test procedure
OpenDevice
SetChannel
GetTimebase
SetDataBuffers
Collect sample in blockMode
wait for ps4000aBlockReady
getValues
calculate AdcToMv

Expected result: 1300 samples around 0mV.
Measured result: Declining values from Sample(1): 300mV to Sample(1300): 79mV. If i collect a second block of data the values are correct.

Observations
There is no problem when setting the channel to DC coupling.
Waiting >10s between channel setup and measuring does not help.
The input voltage is correct (Measured with a benchtop scope).

Which leads me to the question: Ist there something I have to do before I can start measuring AC only? Am I missing something obvious?
Last edited by LouisRast on Mon Oct 19, 2015 8:53 am, edited 1 time in total.

Martyn
Site Admin
Site Admin
Posts: 4491
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: PS4824: First samples are wrong

Post by Martyn »

What you are seeing is the signal slowly decaying away to zero as the DC blocking capacitor charges up. It is only switched in by your call to collect the sample.
Martyn
Technical Support Manager

LouisRast
Newbie
Posts: 0
Joined: Mon Feb 02, 2015 1:30 pm

Re: PS4824: First samples are wrong

Post by LouisRast »

Martyn wrote:What you are seeing is the signal slowly decaying away to zero as the DC blocking capacitor charges up. It is only switched in by your call to collect the sample.
Ok, so how can I set the PS4824 up so that the first measurement is correct?
Is collecting dummy samples until the capacitor is charged the only way to work around this?

Martyn
Site Admin
Site Admin
Posts: 4491
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: PS4824: First samples are wrong

Post by Martyn »

Collecting dummy samples would provide a solution, waiting for the values to settle would be an alternative approach.

This effect will also be present if the DC component of the signal changes quickly during the data collection, each change will cause the capacitor to charge/discharge and this will be visible in the readings.
Martyn
Technical Support Manager

LouisRast
Newbie
Posts: 0
Joined: Mon Feb 02, 2015 1:30 pm

Re: PS4824: First samples are wrong

Post by LouisRast »

Ok. Many thanks for the quick reply and clarification.

Post Reply