sbRIO9637 the corruption of data to loop speeds

Hello, I send a waveform AO0 and extracted AO1 of loop using a FIFO, is attached.  If the timer loop is 1 msec then the output is clean.  However when I take the timer loop up to 50 usec (20 kHz) can I get occasional glitches in the output, a record of the scope of such an event is attached.  If I take the loop up to 20 usec timer, then the total duration of the production is not much shorter, but the waveform is severely damaged.  I would like to understand how corruption occurs and how to get reliable results with a 50 kHz update rate.  I'm guessing that the fracture of fixed point might use too much CPU, but I do not see how to put a value I16 in the 16 bit DAC, it expects a voltage of + /-10.

Thank you

JEB

Hi John,.

Good job governs me execution of writing - the big problem was adding a loop FOR and placing the FIFO_Write inside the loop.  This means that I said to do the tail 3, a thousand times in a row.  The idea used FOR autoindexing change table 2D in a 1 d accepted by the FIFO entry.  But as I learned periodically in LabVIEW, never do something with a loop FOR if there is a VI somewhere who will do the same thing in one bite!  In this case Array reshape transforms the data directly into a table 1 d containing all points (see extract), and I ask all points at a time.

With this change it on at 320 analog net Ksamples/s on.  It is close to the maximum side, so the results are now considered to be very good!

(problem solved, but I do not see the button for this)

thanx

JEB

Tags: NI Hardware

Similar Questions

  • Best practices for the reader to 'Data' between VM?

    Hello

    So on my box ESXI, I have a 250 GB drive. I was wondering what the best practice is to have a 'data' drive shared between VM? I'm pretty new to virtualization so would like to view

    I would basically following drive configuration...

    Win 2008 R2 - 60 gb

    Win 2008 R2 - 60 gb

    Ubuntu 10.10 - 20 GB

    (Shared between the two areas of 2008) DATA - 100 GB

    Thank you.

    The only way to do this is to assign the drive to a virtual machine and create a network share. Unless you use a file system that supports concurrent access to files, an attempt to present the disk to several systems would probably end by the corruption of data.

    André

  • the speed of the analysis of data VI

    Hello

    I wrote a labview vi to help me analyze some rather large data sets. The vi works however its very laggy and uses a lot of memory. Hoped someone could take a look at my code and give me some ideas on how I can modify it to improve the speed in which it runs. I have attached my labview vi. I hope someone can help.

    Best regards, R

    Things I see wrong.

    1. you are constantly re-read the same files in a loop.  Don't expect the files to edit?  Files must be read before the loop and data manipulated inside the loop.  I think that this is your primary problem.

    2. the abuse of local variables.  You have many copies of AllRows for example.  You should have a single wire coming out of reading the file in spreadsheet on your terminal and direction that, whenever you need it.  Many other controls are duplicated unnecessarily with local variables.

    3. a lot of code dupliated.  You do the same operations on many charts.  You can probably simplified by putting that into a Subvi and replicate the Subvi.  You can probably all this to wrap in a loop For.

    4. your cursor property nodes appear to be back.  You're the X readling current position before you set the active cursor.  That and a race condition with the property of other parallel means node not sure which X value read you corresponds to what slider.  Read the two sliders in a single node and in the order appropriate like this:

    5. other things as well as become more evident once the big objects above are supported.

    I recommend you watch the LabVIEW tutorials online
    LabVIEW Introduction course - 3 hours
    LabVIEW Introduction course - 6 hours

  • What is the speed of data transmission when using the dashboard of data?

    Hello, everyone, I've seen it doesn't matter if an iphone or a Chinese tablet is used, the speed of transmission of data with the shared Variables and dashboard of data is almost the same and is at reduced speed. Does anyone know how I could measure this baud rate? Thank you in advance.

    Hello there;

    The rate of updating the dashboard data for 1.0 is set to 1 second of webservices and 0.2 second for variable network shared, of course depends on the amount of information on the network continuously. On the second version of the dashboard of data, you can specify your update frequency within a range

    You can find this info in this KB

    Hope this info helps.

    See you soon

  • How to make the exchange of data between 2 whole loop real-time

    Hello
    I have 2 while loop
    the 1st loop includes data acquisition program
    the 2nd loop includes control program
    --------------------------------------------------------------------------------------------------------------
    My question is how do the exchange of data between 2 whole loop real-time

    --------------------------------------------------------------------------------------------------------------
    I tried with the variable and direct wiring between the 2 local while loop
    It does not work (there is a delay)


  • Restarting a task for the acquisition of data inside a For loop

    Hello

    I need iterate through my acquisition of data. Currently, I'm doing this through the creation, implementation and tasks for the acquisition of data inside a loop For which is iterated according to the needs of compensation. Unfortunately, the creation of these DAQ tasks slow down my code.

    I would like to be able to create the tasks outside the loop, pass them in and revive the tasks at the beginning of each iteration. Is there an easy way to do this?

    Otherwise, is there a way to make the standard DAQmx digital startup trigger trigger several times (so that it starts each pulse data acquisition in a long pulse rather than just the first pulse train)?

    Thank you!

    -Evan

    I whent before and created this example for you (and many others.)

  • In the attached VI why looping force the type of data while the other does not?

    In the attached VI why looping force the type of data while the other does not?

    I'm guessing that you have created the Enum on the front panel.  If you right click and create an indicator, it corresponds to the type and be an enum.  LabVIEW represents enums as U16, but because the types are not the same (for example, your enum has 3 values, but a U16's 65 536 values), LabVIEW automatically converts value (or converts) the smallest representation (enum) in the largest value (U16).

  • My Cookie file was corrupted in data and my programs continually pop up with the 'corrupt file' error when a program opens, pointing to \Users\name\Appdata\Roaming\Microsoft\Windows\Cookies.

    My Cookie file was corrupted in data and my programs continually pop up with the 'corrupt file' error when a program opens, pointing to \Users\name\Appdata\Roaming\Microsoft\Windows\Cookies.

    The problem is, Chkdsk will not fix this, and I can't access the file. This opens up a world of problems when I need to fix this error, so I can upgrade to Windows 7. Help, please!

    I just ran the scan again. He said of the errors found, but could not be repaired.

    This problem actually has nothing to do with internet explorer. It's the cookies folder that is corrupt in the Windows folder in roaming. The error file is in conjunction with the programs, that I try to run (AIM, Skype, so forth). Trying to enter the folder it says it's corrupt. I don't know what is the root of the problems or how to solve this problem, but these measures have not helped so far. Regarding the repair, I will try that next.

    http://Tinypic.com/view.php?pic=14mw4ly&s=4

    Here is a picture. This is the error of the lower right

    It's what the tool SFC to see system files corrupted and try to repair the system files.
    SFC could not fix it.

    Looks like you will need to do this:

    http://www.Vistax64.com/tutorials/88236-repair-install-Vista.html

    Read the info on the link above.

    Good luck with it.

    Let us know how you go with it.

    See you soon.

    Mick Murphy - Microsoft partner

  • RMAN backup on the database with data file with corrupted block

    Hi friends,

    I have a problem with a database of oracle 11g XE on windows server 2003, the database is in production, it is open, is operating normally.

    The database has no backup configured before, so I put the database archiving mode and tried to backup using rman utility, but it show me some errors.

    First of all, it showed me:

    ORA-19501: error in reading file... SYSAUX. DBF

    ORA-27070: async read/write failed

    OSD-04016: error queuing an asynchronous i/o request.

    And when I passed the dbv on the SYSAUX file utility. DBF showed the fatal error.

    So, I switched CHKDSK utility on windows to repair the disc of the error. After that, run utility dbv and he passed and showed me the corrupt on SYSAUX blocks. DBF

    Then, came to rman and ran:

    RMAN > validate the logical database checking;

    State of the file marked corrupt blocks examined empty blocks high YVERT

    ---- ------ -------------- ------------ --------------- ----------

    3 0 1111 115840 78859075 FAILED

    File name: C:\ORACLEXE\APP\ORACLE\ORADATA\XE\SYSAUX. DBF

    Type block for lack of treated blocks blocks

    ---------- -------------- ----------------

    Data 0 0

    Index 0 0

    Other 38 114729

    After that: watching the content of v$ database_block_corruption;

    Show me a lot of lines like this:

    SQL > select * from v$ database_block_corruption;

    FOLDER # BLOCK # BLOCKS CORRUPTION_CHANGE # CORRUPTIO

    ---------- ---------- ---------- ------------------ ---------

    3 4440 1 FRACTURED 0

    3 4439 1 0 ZERO ALL THE

    3 4386 2 FRACTURED 0

    3 4384 1 0 ZERO ALL THE

    3 4334 1 0 ZERO ALL THE

    ...

    33 selected lines.

    Finally I try to rerun rman and he shows me the error:

    ORA-19566: exceeded the limit of 0 blocks corrupt file C:\ORACLEXE\APP\ORACLE\ORADATA\XE\SYSAUX. DBF

    How can I do backup with rman? Maybe I can by pass the corrupt block?

    I'll be grateful for any help you can provide.


    =)

    It would be better if you try to find the reason of corruption and try to recover the corrupt blocks. A block corrupted in the backup would really not must use. But if you're ready to take backup, you can ask RMAN to ignore corrupt blocks and move forward.

    Use the pseudo-code for the backup

    Run {set maxcorrupte for the database to 10;

    backup tablespace sysaux ;}

    HTH

    Aman...

  • A loop in a refcursor for the list of dates as a string

    Hello
    I have a simple procedure that gives me the list of dates and dates
    I need to get the dates dates separated by commas
    I don't know if this can be done directly by making a loop on the refcursor or I have to go pick her up in a file/table and then concatenate with comma, or is there anything else that can be done.
    I tried some stuff like below

    PS help out me
    the procedure is that returns the list of dates
    CREATE OR REPLACE procedure SALUSER.prm_sp_rpt_payslip_lop_dates(p_empid in int,p_tran_year in int,p_tran_month in integer,o_dates out sys_refcursor)
    as
    begin
     open o_dates for select  to_char(PHL_LOP_FROM,'DD-Mon-YYYY'),to_char(PHL_LOP_TO,'DD-Mon-yyyy') 
                     from prm_h_lop
                     where phl_emp_id=p_empid
                       and phl_tran_year=p_tran_year
                       and phl_Tran_month=p_tran_month;
    
                   
     end;
    /
    I need my o/p as
    dates :<date1>,<date2>...etc
    Kind regards

    Maybe sth. as

    SQL>  var cur refcursor
    
    SQL>  declare
     cr sys_refcursor;
     procedure prm_sp_rpt_payslip_lop_dates (cr in out sys_refcursor)
     as
     begin
       open cr for select hiredate from emp;
     end prm_sp_rpt_payslip_lop_dates;
    begin
     prm_sp_rpt_payslip_lop_dates(cr);
     open :cur for select 'Dates: ' || column_value dates from xmltable('string-join(//text(), ", ")' passing xmltype(cr));
    end;
    /
    PL/SQL procedure successfully completed.
    
    SQL>  print cur
    
    DATES
    --------------------------------------------------------------------------------
    Dates: 17-Dec-1980, 20-Feb-1981, 22-Feb-1981, 02-Apr-1981, 28-Sep-1981, 01-May-1
    981, 09-Jun-1981, 19-Apr-1987, 17-Nov-1981, 08-Sep-1981, 23-May-1987, 03-Dec-198
    1, 03-Dec-1981, 23-Jan-1982                                                     
    
    1 row selected.
    
  • accelerate the backup of data received through series

    Hello

    I hope you can give me some advice. So far, I have not found a solution.

    I have a sensor connected by USB-serial converter to the computer (COM 2). After the beginning of the measure, the sensor sends continuously data as a string. The sensor is able to operate at different speeds and it is controlled through strings sent to the device.

    Now, I have programmed a small piece of code to read and evaluate the data from the sensor. There is no problem while looking at live data in several locations. But as soon as I try to save the data, I have speed issues. The loop is not fast enough to receive the data, because it is provided by the sensor. Temporarily remove the plots and so on, don't help not, therefore the problem is data backup. As far as I know, I should use the tables that I did and I don't think that editing bays should be the bottleneck, because they should be able to manage data in general much faster.

    In the Hyperterminal with windows, the unit works fine.

    I use VISA series with a timeout of 2 s and a buffer of 128. Baud rate is set to 460800.
    The real VISA read command is set to read 50 bytes max or stop a tank of termination (line break).
    After initialization, and sending the command start for the sensor, the device starts to send its data...
    The different speeds are:
    1 - 4 strings per second
    2 - 10 per second
    3 - 100 per second
    4 - 500 per second

    Each line gets cut, evaluated and plotted. Each value is also stored in a table (which could be saved after the program stops).

    In the photo, I deleted several commands sent to the probe, because they are not relevant and it doesn't matter (of failure) if I remove them in the program or not. It goes the same for plots. After setting the device at the higher speed, the program fails, because there is more data than labview bed. (I have already said, that without the tables it works very well, I do not have it?).

    So, what can I do? I'm doing something completely stupid? Is there a faster way to collect data with labview? I mean, 500 HZ, which must be nothing.

    Thanks for your help.

    Kind regards

    HD

    PS I'm continues LabVIEW 8.6

    Computer: Intel Core2 Duo E8500 3.16 GHz with 3 GB of RAM

    SW: Windows XP SP3

    You use insert into table with each iteration and forcing LabVIEW to keep reallocation memory for the berries. It is not very effective. Instead, I advise you to use the architecture of producer/consumer and transmit data back to the consumer loop where you write the data. Shift registers and who would skip all the functions to insert it in the table. You also use local variables and property nodes. Neither one of those who are very fast either. Can't tell from the picture why those are used. Also, consider using a structure of the event for these Booleans wired for the structures of the case. Do you really need those who questioned all the time and run the code with each iteration, or do you just need them to run once when the Boolean value changes?

  • Data acquisition high speed (200 ksps / s) with pci-6224 or usb-6211 is possible?

    Hello

    I need assistance to complete a task.

    The task:

    Record a switch contacts twists every 5 microseconds (200 ksps / s).  DAQ cards I have at my disposal at the present time are PCI-6224 and the USB-6211. The two cards are rated up to 250 kech. / s. I only watch 1 channel that will monitor the voltage.

    Here are the settings I use:

    CSR

    Continuous samples

    Sampling rate of 200 kech. / s

    Analog DBL1 1 sam chan

    timeout of 10 seconds (Read DAQmx)

    The problem:

    (1) every time I try to taste 200 kech. / s I get an error that reads

    Measurements: Tried to read samples that are no longer available. The requested sample was already available, but has since been replaced.

    Increase in the size of buffer, most frequently the reading of data or by specifying a fixed number of samples to read instead of reading all available samples would correct the problem.

    (2) when you try to taste at a much slower pace of 50 kech. / s, the speed of the loop iteration is not realizing this speed setting.

    My question is this: are these 2 cards (PCI-6224 or USB-6211) DAQ fast enough at my request and I do something wrong? OR

    I use the wrong hardware to accomplish my task?

    Thank you in advance,

    Gerardo

    The number of iterations has nothing to do with the number of samples if you do it right. You specify the rate and the number of samples you want and number of samples returned. You DO NOT loop 200,000 times. If you ask for 200 000 s/s and 200,000 samples, you will get this number of samples every second. It's really as simple as that.

  • B1 - 710 stuck on the screen 'formatting data"after trying to do a factory reset.

    After occassionly gel to start the past two weeks, causing necessary reboots, the tablet in a loop start.  The last time this happened, about a year, a factory reset solved the problem. This time, however, the Tablet is stuck on the screen layout of the data. Now, even holding the volume up button and the power button on the system goes back to the screen set data formatting and nothing else happens.

    Is it possible to retrieve the tablet or is - this one probably a hardware problem?

    It seems a hardware problem, to be honest, but may also be a corruption of OS/recovery if you are still under warranty, support Acer a call.

  • What are the causes of data loss during an update?

    Just what are really the causes of data loss or corruption when you update. Most sites will tell you to back up what I'm doing, but none of them actually appoint a cause to save.

    -Thanks in advance

    Minor updates in an OSX * may * break the file system.  Fully upgrades to other OSX replaces almost all file systems.

    Some backup of people even before you install some software packages, because they know how this package changes directories OSX.

    Backups are never a waste of time.

  • Hello. I buy the Macbook Air Mid 2015... After the upgrade to El Cap internet speed slowed painfully. What should do? Help, please

    Hello. I buy the Macbook Air Mid 2015... After the upgrade to El Cap internet speed slowed painfully. What should do? Help, please

    When you see a beachball cursor or the slowness is particularly bad, note the exact time: hour, minute, second.

    These must be run as administrator. If you have only one user account, you are the administrator.

    Run the Console application in one of the following ways:

    ☞ Enter the first letters of his name in a Spotlight search. Select from the results (it should be at the top).

    ☞ In the Finder, select go utilities ▹ of menu bar or press the combination of keys shift-command-U. The application is in the folder that opens.

    ☞ Open LaunchPad and start typing the name.

    The title of the Console window should be all Messages. If it isn't, select

    SYSTEM LOG QUERIES ▹ all Messages

    in the list of logs on the left. If you don't see this list, select

    List of newspapers seen ▹ display

    in the menu at the top of the screen bar.

    Each message in the journal begins with the date and time when it was entered. Highlight date back to the time you noted above.

    Select the messages entered since then until the end of the episode, or until they begin to repeat, whichever comes first.

    Copy messages to the Clipboard by pressing Control-C key combination. Paste into a reply to this message by pressing command + V.

    The journal contains a large amount of information, almost all this is useless to solve any particular problem. When you post a journal excerpt, be selective. A few dozen lines are almost always more than enough.

    Please don't dump blindly thousands of lines in the journal in this discussion.

    Please do not post screenshots of log messages - text poster.

    Some private information, such as your name, may appear in the log. Anonymize before posting.

    When you post the journal excerpt, an error message may appear on the web page: "you include content in your post that is not allowed", or "the message contains invalid characters." It's a bug in the forum software. Thanks for posting the text on Pastebin, then post here a link to the page you created.

    If you have an account on Pastebin, please do not select private in exposure menu to paste on the page, because no one else that you will be able to see it.

Maybe you are looking for