Probe compensation

Having problems ? let us know the details here
Post Reply
skyflyer
Newbie
Posts: 0
Joined: Fri Jul 10, 2015 9:08 pm

Probe compensation

Post by skyflyer »

Hi!

Just received my Picoscope 2204A with probes today. I did a probe compensation and I'm wondering if this is OK?

I mean, there is 19 mV spikes present -- is this normal? 19mV seems a lot to me... or are the supplied probes that inaccurate?

To see what I mean, have a look at the image, where I tried to compesante the supplied probe.
Image

Kind regards!

bennog
Advanced User
Advanced User
Posts: 206
Joined: Mon Nov 26, 2012 9:16 am
Location: Netherlands

Re: Probe compensation

Post by bennog »

According to https://www.picotech.com/oscilloscope/2 ... ifications the 2204A s a 8 bit scope.
In your image the scope is set to +/- 2V
So 4V / 256 is about 15 mV per bit.
And it is normal voor digital scopes to wobble the LSB.

In PS6 you can do resolution enhangement to 12 for your scope, what this actualy do is increas sampling speed and the average out the samples (so you get a virtual resolution of 12 bits)

Regards,
Benno

skyflyer
Newbie
Posts: 0
Joined: Fri Jul 10, 2015 9:08 pm

Re: Probe compensation

Post by skyflyer »

bennog wrote:According to https://www.picotech.com/oscilloscope/2 ... ifications the 2204A s a 8 bit scope.
In your image the scope is set to +/- 2V
So 4V / 256 is about 15 mV per bit.
Thanks! So, if I understand this correctly, if I were to define the range from 0 do 1.5V, for example, I would get ~6mV accuracy? Whereas by oversampling to 12bit, I could get to 1mV (theoretically).

I didn't know that the vertical scale affects the probe accuracy (resolution).

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

Re: Probe compensation

Post by Martyn »

Nearly, but it is related to the vertical ranges you can select, 0 to 1.5V is not a settable vertical range for the device.

+/- 2V which gives 4V peak to peak divided by 256 so approximately 15 mV per bit.
+/- 1V which gives 2V peak to peak divided by 256 so approximately 7.5 mV per bit.

Now if you add resolution enhancement at 12bit, instead of dividing by 256 you divide by 4096

+/- 2V which gives 4V peak to peak divided by 4096 so approximately 976 uV per bit.
+/- 1V which gives 2V peak to peak divided by 4096 so approximately 488 uV per bit.

Resolution enhancement uses a moving average filter so it will affect the rise times on the vertical edges of your signals.
Martyn
Technical Support Manager

skyflyer
Newbie
Posts: 0
Joined: Fri Jul 10, 2015 9:08 pm

Re: Probe compensation

Post by skyflyer »

Martyn wrote:Nearly, but it is related to the vertical ranges you can select, 0 to 1.5V is not a settable vertical range for the device.
Yes, I found that out yesterday, while trying to figure out why I can't set the range from 0 to 50V :)

It makes sense now. I also saw that oversampling (resolution enhancement) in this case (square signal) does not give desired results, especially because it affects the rise times so badly.

Post Reply