ADC212/100 on VB6 - Jagged signal at 100MHz

mike_ims · Post by **mike_ims** » Tue Aug 24, 2004 11:32 pm

I have run into another problem using the adc212/100 in VB6. In the system I am developing, I have set up a virtual oscilloscope using the adc212.

I am looking at repetitive ultrasonic signals and notice a big change in signal shape when I move from digitizing at 50MHz to 100MHz. At 100MHz the signal appears "jagged". The general shape of the signal is the same, but, instead of a smooth line it is jagged. At first, I thought it was high frequency noise of some sort. After attempting to average, I found that this occurrence was repetitive and could not be averaged out. At any lower digitization rate this affect does not show up - only at 100MHz. Note that when I load the Picoscope software and view the signal there, the signal is as expected - smooth. Also, the affect does not show up on a Tektronix scope.

Is situation symptomatic of any common programming errors with the adc212?

To further investigate this, I modified the visual basic example program to work from an external trigger, digitize at 100 MHZ (timebase = 0), and collect 4096 samples. These are settings I have been working with previous to now. The same jagged affect occurs.

I am nearing my deadline so any expedited help you can give me would be invaluable.

Many thanks,
-Mike

markspencer · Post by **markspencer** » Thu Aug 26, 2004 7:58 am

Dear Mike,

I am sorry to hear that you are experiencing this problem.

I have looked at the pictures you have sent displaying the traces and would like you to repeat the following experiement, but first download the latest software release from our website at:

http://www.picotech.com/software.html

Once you have done this and installed the software, open Picoscope up and select the fastest timebase, by increasing the number of values in the Settings | Options | Advanced and only having one channel active you should be able to force the unit to sample at 100Msps. Start dropping the timebases, and watch the signal. Now switch on channel B, when the unit is running at 100Msps does it show similar results to your traces at 100Msps

You can confirm that it is sampling at 100Msps by saving the results as a .txt file.

When switching on both channels, this will automatically reduce the timebase to 50Msps for each channel. Does the trace then smooth out again you can check the sample rate by saving the data as a .txt file.

If the above does not reproduce your problem and the sample rates in the .txt files are as stated 10nanoseconds time difference for 100Msps and 20nanoseconds time difference for 50Msps. Then us the adc20032.dll in the Picoscope installation with your program.

If the problem is displayed in Picoscope then the unit will need to be reclaibrated. Please let me know when the unit was purchased and its batch number and your contact details. I will then issue you a returns number.

Best regards,