Math channel-alarm-save to .txt file problem

Forum for discussing PicoScope version 6 (non-automotive version)
Post Reply
vanAsten
Newbie
Posts: 0
Joined: Thu Aug 08, 2013 2:47 pm

Math channel-alarm-save to .txt file problem

Post by vanAsten »

I use PicoScope 5203 with PicoScope software 6.7.28.7 and Windows XP. I use an external trigger and the data I want to save is in channels A, B and math channel A*B. I use an alarm to save the current buffer to a .txt file on capture. I use Scope Mode and 'repeat' for the trigger with the intent to save hundreds of .txt files. I encounter two problems:

1. The data of the three channels is not completely saved, the .txt file is only 256 lines long (it goes from about -103ns to about -40ns, where it should go to about +403ns).
2. If the saved data from channels A and B is correct, then the saved math channel A*B is not A multiplied by B. Up to -40ns the signals are just noise in channels A and B, say -0.00027682 V and -0.00036909 V.

If I try the same without math channel A*B the .txt file seems to contain everything.

How come?
Thanks in advance.

Hitesh

Re: Math channel-alarm-save to .txt file problem

Post by Hitesh »

Hi vanAsten,

Could you please post a pssettings or psdata file for your setup or e-mail it to support@picotech.com

Thanks,

vanAsten
Newbie
Posts: 0
Joined: Thu Aug 08, 2013 2:47 pm

Re: Math channel-alarm-save to .txt file problem

Post by vanAsten »

Hi Hitesh,

Here are the psdata files for the two different measurements:
20130813-0001 (2).psdata
This is the psdata file when I measure with math channel A*B turned on.
(10.84 KiB) Downloaded 383 times
20130813-0001 (5).psdata
This is the psdata file when I measure with math channel A*B turned off.
(10.02 KiB) Downloaded 404 times
Thanks in advance.

Hitesh

Re: Math channel-alarm-save to .txt file problem

Post by Hitesh »

Hi vanAsten,

I've checked both data files using PicoScope Beta 6.8.2 and it's only displaying 250 samples.

How many samples were you expecting in the buffer?

Thanks,

vanAsten
Newbie
Posts: 0
Joined: Thu Aug 08, 2013 2:47 pm

Re: Math channel-alarm-save to .txt file problem

Post by vanAsten »

Hi Hitesh,

These are the text files resulting from the same measurements the psdata files are from:
20130813-0001 (2).txt
This is the text file with math channel A*B turned on.
(11.86 KiB) Downloaded 370 times
20130813-0001 (5).txt
This is the text file with math channel A*B turned off.
(72.34 KiB) Downloaded 366 times
As you can see, the file with math channel A*B turned off contains the complete buffer as was shown on my screen initially, from -104ns to +403.75ns, while with math channel A*B turned on I only get -104ns until -41ns: 3 header lines and then 253 samples for a total of 256 lines in the file.
What I want is the result from 20130813-0001 (5) with a correct math channel included.

Thanks in advance.

Hitesh

Re: Math channel-alarm-save to .txt file problem

Post by Hitesh »

Hi vanAsten,

Could you please post the 20130813-0001 (5).psdata file again please? The one that I downloaded only seems to have 250 samples which doesn't match the text file you sent.

Could you also confirm the timebase used and the number of samples requested so I can try to replicate this.

Thanks,

vanAsten
Newbie
Posts: 0
Joined: Thu Aug 08, 2013 2:47 pm

Re: Math channel-alarm-save to .txt file problem

Post by vanAsten »

Hi Hitesh,

Here it is again, although I did not notice anything strange while uploading the psdata file the first time:
20130813-0001 (5).psdata
(10.02 KiB) Downloaded 348 times
In PicoScope 6 I asked for 50 nanoseconds per division (so -100 nanoseconds to + 400 nanoseconds is expected, pre-trigger is 20%) and I just left the default value of 1 MS. The text file is precise enough though, 1 sample per 0.25 nanoseconds will do.

Kind regards,
vanAsten.

Hitesh

Re: Math channel-alarm-save to .txt file problem

Post by Hitesh »

Hi vanAsten,

The maximum sampling rate with 1 channel enabled is 1GS/s corresponding to a 1ns sampling interval. With 2 channels enabled, the sampling rate drops to 2ns, and over 500ns this will give you 250 samples.

It appears that the data in your files may have been interpolated.

Are you using any other oscilloscopes?

Please could you download and try PicoScope 6.8.2 Beta from http://labs.picotech.com/ and see if this makes a difference.

Thanks,

vanAsten
Newbie
Posts: 0
Joined: Thu Aug 08, 2013 2:47 pm

Re: Math channel-alarm-save to .txt file problem

Post by vanAsten »

Hi Hitesh,

I will try the Beta, it might take a few weeks before I have the time to do so though. I am not using any other oscilloscopes at the moment.
Meanwhile, do I understand correctly that if the data in my files is interpolated, the reason 20130813-0001 (2).txt only contains 253 samples is because the math channel only calculated A*B for the initial samples, not the interpolated ones creating a difference in available number of samples to put in the .txt file for channel A and B on one hand and math channel A*B on the other, therefore the .txt file was cut short at -41ns when PicoScope ran out of A*B results to put in the .txt file? Meaning that except for the first sample, all math channel results are listed at the incorrect time because there should be 2ns between every math channel result and not 0.25ns?

Thanks in advance,
vanAsten.

Hitesh

Re: Math channel-alarm-save to .txt file problem

Post by Hitesh »

Hi vanAsten,

It's possible that there was bug in that version meaning somehow the interpolated results were somehow incorporated.

I've also run some tests with the recently released PicoScope 6.7.35 and the files appear to be produced correctly.

I hope this helps.

Post Reply