ps6 / sdk : offset differs

Forum for discussing PicoScope version 6 (non-automotive version)
Post Reply
Posts: 0
Joined: Thu Jul 16, 2015 9:50 am

ps6 / sdk : offset differs

Post by bonelli »


Using a 5444D with
- Channel A physically shorted with a clean RF short.
- 16 bits mode on channel A, 5V range, no offset, 20MHz BW
- 12500 samples (16ns sampling time)

I'm doing 2 things :
- Export a .csv from the picoscope 6 software
- Generate a .csv in C++ with the SDK and ps5000aGetValues()

Below are the calculation of min/max/mean/stedev for both set of data (all values in mV):

mean 7 / 27
min -8 / 9
max 24 / 44
sdtdev 4 / 4

Repeating the measures numerous time always give the same feeling: Noise is the same but offset is always 2-3 time more accurate using PS6. Is there a auto-calibration using PS6 that is not performed using the SDK?

Thank you

Posts: 734
Joined: Mon Aug 11, 2014 11:14 am

Re: ps6 / sdk : offset differs

Post by Gerry »

Hi bonelli,

There is no auto-calibration performed by ps6 (not possible without shorting the input), however you can perform a manual one using the 'zero all ranges' function in the Channel Options for an input channel, which requires you to physically short the channel input. In fact, when running the Scope using PicoScope 6, you're using the same API functions as those that are available in the SDK.

I would revisit your method. The results that I get for PicoScope 6 using the available stats for the Measurement functions (Minimum, Maximum, and DC Average) when performing multiple groups of multiple captures (because the stats only have a capture count of 20) suggest a larger deviation between DC averages. So, it is possible that you're inferring a behaviour that isn't based upon statistical significance. What statistical sample size are you using for both sets of data?


Technical Specialist

Post Reply