• Resolved

ADC34J45EVM: Please help me choose the right devices...

Part Number: ADC34J45EVM

Hi everyone!

My name is Fernando Sanz. I'm a Cosmic ray researcher.

I am sorry to take your time but I have to explain my problem a little bit.

For a few years I was studying the theoretical aspects of radio detection of cosmic rays.

Now, I'm eager to setup a simple but effective experiment just to see how it works in real life.

I know that there are some very professional researchers in this field in the world but my goal is not that complicated.

To achieve my goal, I need few devices including an ADC and a data capture card.

Since my selection list of a band pass filters is limited, I have to choose the ADC wisely to satisfy my needs as well as the Nyquist criteria.

My band pass filter has a passband of 55-67 and stopbands at 44-79 MHz and my 4 antennas are going to work in 30-80 MHz frequency range.

that means for an oversampling I should go beyond 158 MSPS or stay at 80 MSPS to be in the second Nyquist zone.

Now, all I want is to be able to capture the signals from the antennas and to be able to do few things:

1- to Save and store signals for later (offline) analyzes.

2- To be able to remove signals which are weaker than an specific strength.

3- To be able to store data only if they've come from all channels in an specific time windows.

4- if there be an option for an external triggering it would be really good because I may be able to add a particle detector as a trigger to this setup in the future

I searched Ti's products and I'm wondering whether ADC34J45EVM with TSW14j50EVM or any other product can do the job for me.

Please Help me.

Thank you and sorry for my bad English.

  • Genius 12785 points
    Hi Fernando,

    The ADC34J45 and TSW14J50 can do the following:
    1. TSW14J50 +HSDC Pro is used to capture data from the ADC34J45 to be stored for later analysis - HSDC Pro only does FFT and time plots by default.
    2. TSW14J50+HSDC Pro can be triggered externally to capture data when needed.

    The ADC34J45 EVM is AC coupled via a transformer and expects a 50 ohm input - please take this into account when designing your sensor interface. The TSW14J50EVM is limited to 256Msamples of 16bits. This will be split across the number of channels.

    Ken.

    HSDC

    High Speed Products

    HPA

     

  • In reply to Ken C:

    Hi Ken,
    Thank you so much for your quick reply.

    "1. TSW14J50 +HSDC Pro is used to capture data from the ADC34J45 to be stored for later analysis "

    let me ask this: Can I achieve what I want (2,3) either by direct programming or using other software like MATLAB or LabVIEW?

    "The TSW14J50EVM is limited to 256Msamples of 16bits. This will be split across the number of channels."

    So, Despite having 160 MSPS (per channel?) on the ADC, we won't be able to actually use that, right?
    What about any other option like TSW14J56? The user guide indicate that It can save up to 2G of 16 bit.
  • Genius 12785 points

    In reply to Fernando Sanz:

    Fernando,

    Yes you can automate HSDC pro with labview or matlab - this will allow you to control HSDC Pro with Matlab or Labview.  Once you can control HSDC pro then you can control any of our capture cards (TSW14xxxx).  Look in this installation directory for examples.

    C:\Program Files (x86)\Texas Instruments\High Speed Data Converter Pro\HSDCPro Automation DLL\Manual and Examples

    The memory depth does not limit the sampling speed.  You can still use the ADC at 160Msps, it will store each of those samples from each channel at speed.  However you will only be able to capture enough time to fill the memory.  In this case you will only be able to capture 64Msamples per channel (4 channels) and at 160Msps that will be about .4s.  The TSW14J56 will be able to capture more time if needed.

    Ken

    HSDC

    High Speed Products

    HPA

     

  • In reply to Ken C:

    Thanks again.

    I'm sorry but I can't quite understand the second part of your reply.
    May I ask for a more clear explanation?

    Honestly, The signals that I'm going to capture from the antennas have a very short life (10-20 ns) so I want to make sure I don't misunderstand something.

    Thanks.
  • Genius 12785 points

    In reply to Fernando Sanz:

    Fernando,

    How fast the ADC samples and how much memory you have are 2 independent things.  You could have a 1 sample depth memory - and you would only be able to save 1 sample from the ADC regardless if the ADC was running 1sps or 160Msps.

    In this case you have access to 256Msamples of memory to be used for 4 channels.  This means you can store up to 64Msamples for each channel.  If you run the ADC at 160Msps, then you can store up to 64Msamples/160Msamples/sec=0.4s of time.

    At 160MSps ADC rate you are getting time resolution of 1/160M=6.25ns time slice per sample.

    Does this make sense or am I not understanding your intended use case?

    Ken

    HSDC

    High Speed Products

    HPA

     

  • In reply to Ken C:

    I'm sorry if I'm asking simple questions.

    Actually I was eager to know if I'm loosing any part of the real signal because of the size of the memory on the data capture card.

    What I understand from your reply is two different scenarios, Which one is correct?

    We have a container which does have 64M slots on it, Now the ADC samples the data at the rate of 160 MSPS...
    These samples goes to the data capture card until it fills and because it does have 64M slots, it takes something about 0.4s to fill and then the next wave of samples comes in.

    Or,

    Every second the ADC samples 160 MSPS, now because the data capture card only have 64M slots, it captures only 40 percent of those samples every second.

    Thanks again.
  • Genius 12785 points

    In reply to Fernando Sanz:

    Fernando,

    Both cases are correct. You are talking about the same thing. Let me illustrate with an example. You have 3 variables in this case
    Xsps=Nsamples/timeS

    Choosing 2 will result in the 3rd.

    In both cases: you have 64Msamples, and the ADC is sampling at 160Msps, this results in Time=64Ms/160Ms/s=0.4s

    The ADC can be running for any length of time, but the capture card can only capture 0.4s of that time frame. So it will depend on when you start the capture. If you can trigger it closely with your event then it will capture 0.4s from the trigger point.

    Ken

    HSDC

    High Speed Products

    HPA

     

  • In reply to Ken C:

    Thank you so much!

    Just a quick question:

    Does ADC3443 have the capability to be connected to TSW1400EVM directly?

    Can I control HSDC Pro and TSW1400EVM using MATLAB or LabVIEW too and do what I motioned earlier?

    Thanks.
  • Genius 12785 points

    In reply to Fernando Sanz:

    Yes - the ADC3443EVM will connection to the TSW1400 directly. You will need to provide an external clock as there is no clock chip on this EVM. Have a look at the users guide section 3.2+
    www.ti.com/.../slau579c.pdf

    Matlab/labview can be used to control HSDC Pro which in turn controls TSW1400.

    Ken.

    HSDC

    High Speed Products

    HPA