Reading USB-6210 tension slowly decreases with switch

Similar Questions

• Clock and hw external trigger with USB-6210 on Linux with NOR-DAQmx Base?

  I have two devices USB-6210 I need to synchronize so that they both collect data exactly at the same time. I was told by support OR I can send the clock off Dev1/PFI4 and have the two USB-6210 s read the clock in through their own PFI0. I also want to trigger data collected for each device by sending a trigger off Dev1/PFI6 and have two devices to receive the signal on PFI2.

  All my attempts to try this are filled with error messages and my research online seem to say that's not possible with USB devices on NOR-DAQmx Base 3.4.0f2 on Linux.

  I "ve tried using example AI programs and those who do not seem to work either for external clocks. Here is the code I tried:

  #include "NIDAQmxBase.h"#include 
  
  #define DAQmxErrChk(functionCall) { if( DAQmxFailed(error=(functionCall)) ) { goto Error; } }
  
  int main(void){    // Task parameters    int32       error = 0;    TaskHandle  taskHandle = 0;    char        errBuff[2048]={'\0'};    int32       i;
  
      // Channel parameters    char        chan[] = "Dev1/ai0";    float64     min = -10.0;    float64     max = 10.0;
  
      // Timing parameters    char        clockSource[] = "/Dev1/PFI7";    uInt64      samplesPerChan = 1000;    float64     sampleRate = 10000.0;
  
      // Data read parameters    #define     bufferSize (uInt32)1000    float64     data[bufferSize];    int32       pointsToRead = bufferSize;    int32       pointsRead;    float64     timeout = 10.0;
  
  printf("Calling CreateTask...\n");    DAQmxErrChk (DAQmxBaseCreateTask("",&taskHandle));printf("Calling CreateAIVoltageChan...\n");    DAQmxErrChk (DAQmxBaseCreateAIVoltageChan(taskHandle,chan,"",DAQmx_Val_Cfg_Default,min,max,DAQmx_Val_Volts,NULL));printf("Calling CfgSampleClkTiming...\n");    DAQmxErrChk (DAQmxBaseCfgSampClkTiming(taskHandle,clockSource,sampleRate,DAQmx_Val_Rising,DAQmx_Val_FiniteSamps,samplesPerChan));printf("Calling StartTask...\n");    DAQmxErrChk (DAQmxBaseStartTask(taskHandle));printf("Calling ReadAnalogF64\n");    DAQmxErrChk (DAQmxBaseReadAnalogF64(taskHandle,pointsToRead,timeout,DAQmx_Val_GroupByChannel,data,bufferSize,&pointsRead,NULL));
  
      printf ("Acquired %d samples\n", pointsRead);
  
      // Just print out the first 10 points    for (i = 0; i < 10; ++i)        printf ("data[%d] = %f\n", i, data[i]);
  
  Error:    if( DAQmxFailed(error) )        DAQmxBaseGetExtendedErrorInfo(errBuff,2048);    if(taskHandle != 0) {        DAQmxBaseStopTask (taskHandle);        DAQmxBaseClearTask (taskHandle);    }    if( DAQmxFailed(error) )       printf ("DAQmxBase Error %d: %s\n", error, errBuff);    return 0;}
  

  When I run the resulting program, I see this:

  $. / acquireNScans-ExtClk
  The CreateTask call...
  Call for CreateAIVoltageChan...
  Call for CfgSampleClkTiming...
  Error-89136 DAQmxBase: route specified cannot be satisfied, because the hardware does not support it.

  For example, a clock and a trigger can be imported via one of the PFI lines by using a USB-6210 on Linux with NOR-DAQmx Base? A clock and a trigger exportable via one of the PFI lines?

  If so, does anyone have the code example illustrating how to do this, or can you at least tell me the names of the lines ("PFI0/Dev1" or other) so I can try again?

  Clues or suggestions would be helpful.

  Thank you

  -Tom

  The clockSource in the example specifies an output rather than an input channel channel. Change source "/ Dev1 / PFI0" solved the problem.

  Please close this post.

• -error message 200284 when calling DAQmxReadAnalogF64() with a USB-6210

  I have tried the code a CVI application to communicate with a USB-6210. I want to start the acquisition to gain 30 samples and then collect them. My problem is when I call the DAQmxReadAnalogF64() function, I get this message error of-200284. The error message means that the data has not yet been acquired in a nut shell. He said I should do to ensure that my startup trigger is configured correctly. I tried to use the internal clock to start the acquisition. I tell myself that I have not configured something correctly.

  In the DAQmxCfgSampClkTiming() function, I set the sampleMode to DAQmx_Val_FiniteSamps I thought I would collect 30 samples (number_of_scans = 30) when I called DAQmxStartTask(). It does not work. When I put sampleMode to DAQmx_Val_ContSamps I can collect data. Here is a stripped down most of my code:

  returnAcq = DAQmxCreateTask ("myAcqTask", & acq_taskHandle);
  returnAcq = DAQmxCreateAIVoltageChan (acq_taskHandle, physicalChannel, "acq_channel", DAQmx_Val_Diff, minVal, maxVal, DAQmx_Val_Volts, "");

  returnAcq is DAQmxCfgSampClkTiming (acq_taskHandle, "OnboardClock", sweepsInterval, DAQmx_Val_Rising, DAQmx_Val_FiniteSamps, number_of_scans);.
  returnAcq = DAQmxStartTask (acq_taskHandle);
  returnAcq = DAQmxReadAnalogF64 (acq_taskHandle,-1, 5, DAQmx_Val_GroupByChannel, Data_readings, number_of_scans, & read, 0);
  Disable the task
  DAQmxClearTask (acq_taskHandle);

  I guess my question is: am I set up so that the internal clock will begin the acquisition when I call DAQmxStartTask()? I thought when I put the second parameter in DAQmxCfgSampClkTiming() to 'OnboardClock', this would use the internal clock.

  Thanks in advance for your help,

  Don Pearce

  Don,

  You're right, your task is configured to use a sample clock internal with a trigger for starting the software.  The description of the error pointed out several reasons why people might see this error.  What is the value of 'sweepsInterval '?  I think that as configured DAQmx should be pointing at the speed (in Hz) specified by sweepsInterval.  By the way-1 in the entrance to numSampsPerChan, I think DAQmx will try to read all samples for a finite acquisition (30 in your case).  If you run below 6 Hz, the lira could be the time-out period before these data were available.

  When it is configured as a continuous, what are the values is passed in the sampsPerChanRead parameter?

  Dan

• 15-in-1 reader (USB 2.0 with 1 USB port) 2 USB 1 1394 Audio (headphone Sockets and microphone)

  I have a pavilion d4996t on which the 15-in-1 card reader is a failure.  (USB 2.0 is no longer detected by the system and cannot write on cards SD.)  I want to buy a replacement but can't find documents in saying what the product number is for the 15-1card Reader / USB 2.0 unit is.  Anyone happen to know?  I've been looking at different websites hoping that I could find a match with a picture, but the few who have photos aren't mine.

  Thank you.

  Real components are separated.   http://PartSurfer.HP.com/search.aspx?SearchText=KC787AV

  The 15 in 1 player's 5070-2040, USB/Audio Unit depends on the connectors, IE 5070-2055 or 5070-4708.

• Amnesty International and counter sync + USB signal stream (USB-6210 vs USB-6341)

  Hi all

  I'm at a stage of identification of a material suitable for the following tasks:

  • 5 analog inputs (AIs) of reading at the same time, tensions at a rate of kSps (at least) 10,
  • application captures 2 inputs using timers (detection of contours with timestamps), square wave entry with duty ratio of 50 percent and about 1.5 kHz frequency and variable pulse width / frequency (from 2 sensors hall, representative of the DC motor rotation speed and direction, quadrature signals), resolution of timestamps should be (at least) 50 ns,
  • AIs and counters should behave in a deterministic way, and must be synchronized in a way,
  • data to be transferred via the USB port of a host computer with Matlab Data Acquisition Toolbox (unfortunately not LabVIEW).

  I've identified the long USB-6210 USB-6341 and potential candidates of material to accomplish the above tasks, but after reviewing several documentation and the topics of the forum, I'm still a bit confused, if both are fully working and my approach described below is not working properly.

  Counters: I intend to use the internal time base available 20 MHz as being the source of meter to get into account the resolution of timestamp 50 ns. External impulses hall are used as sample clock (about 1.5 kHz, see above). As the pulse width varies, the sample clock is not constant.

  AIs: Using a 10 kHz internal clock signal derived from the time base of 20 MHz for timing and analog inputs (trigger) start-up and counters simultaneously material should translate into the required synchronization and deterministic behavior.

  It work? Other recommendations?

  Next is the USB data transfer: all HAVE 5 and 2 data entry of the meter must be correctly transferred to the host computer (the corresponding rates are shown above). USB-6210 is capable of 4 USB signal flow, device USB X range (6341) offers 8 of them. Unfortunately, I could not understand the exact meaning of the expression "signal flow" still. Do I need 1 flow of input signals (would be 7 for my application described) or 1 stream for all analog inputs and 1 for counter inputs (lead 2 streams for my request). Is there no further details on this approach (more than Streaming of signals of NOR) USB signal flow?

  Any challenge to the described application that I might have forgotten? 6210 USB seems to a very limited number of entry PFI, maybe even too low for my meter participate application?

  Looking forward to your comments and advice.

  Concerning

  jAwA

  1. I recommend the X-6341 series on the M-series 6210 sake of counters/timers.  It is more of them, and each of them is more capable.   It can also have a great FIFO embarked for meters that may be important in certain tasks, although I don't think that you currently deal with one of them.

  2. your general concepts on timing & sync are satisfactory.  You will be able to share and to route signals that help ensure synchronization and determinism between the timestamps for your various tasks.  Note that for meter entry tasks, you need set up the trigger 'Arm Start' rather than the regular start trigger.

  3 is not authoritarian, but I believe that the flow of signal # will correspond to the tasks #.  For you, it would be 1 task of HAVE and tasks CI 1 or 2.  (Not clear if you have 1 Encoder with 2-channel quad that would require 1 task of CI, or if you have 2 encoders with 4-way quad).

  4. pay attention to the hall effect signals that are not virgins.  Digital filtering is available and probably better on the X-series, the series M.

  5. strictly speaking, edge detection is a type of digital input task that produces samples but no timestamps.  Ideally, I would like to parallel wires on the two digital inputs for the entries of detection and counter change to position quadrature decoding.   Then I would sample the counters Encoder 1 or 2 using the internal pulse 'event of detection of change '.  I would create another counter timestamp change detects pulses as well.

  -Kevin P

• sensor pressure and USB-6210

  I am installing a USB-6210 to read pressure sensors (specifically: http://www.omega.com/pptst/PX309-100mv.html), with labview2013.

  Its one task relatively simple and should not be this hard...

  I used a similar system to take similar measures in the past, but its been awhile.

  I don't know if it's just a wiring problem or incorrect installation of labview (program), but I can't measure the tensions that are remotely close to the sheet of calibration of the probe.

  Thoughts?

  It turns out that my diet was not come out correctly. Passed to a diet different and everything works as it should.

• USB-6210 voltage divider Calibration questions

  I'm trying to calibrate 14 tensions that I read from a voltage divider in 14 channels of a device USB-6210. Attached is an Excel spreadsheet of data I'm trying to understand. On the bottom of the sheet, I applied the same voltage to all channels and draw the raw tension for all channels and the differences between the expected voltage and voltage of raw. I find that these differences are not constant but linear I do not understand. I then use the basic linear function to find the factors in the top of the worksheet. Therefore, in the upper part, I read the input voltage by using a probe high voltage and voltage of Labview which has been adapted by the linear function of base in the background.  Most of the multiplying factors are ~ 1000 as expected, but the tensions for the last 4 channels are significantly different and are not linear for the last two channels. I mounted the two worst voltage channels with polynomial which is ok but not make any sense. Does make sense to anyone. I'm using Labview 9.0.1 and Windows 7. I ran Diagnostic Utility for the acquisition of data on the device and it passed all tests. I looked into the issue of ghosting, but the criteria seem not valid here because of tension in the neighbouring country of the channels is small. I also enclose the VI that supports voltages raw to show how these functions are used. According to me, the results are stable and precise, but I can't explain the above mentioned discrepancies.

  The problem was solved by changing the sampling frequency in the DAQ assistant. I think the default frequency is 1 kHz and I had given up to 750 Hz. By lowering the frequency 150 Hz, the capacitor is allowed time to discharge so that no ghost image occurs. In summary, the question always was ghosting and solve the I: question

  -Removed in the wizard DAQ, channels that have been disconnected and therefore not used.

  -Change the wiring to the DAQ assistant by sending the low voltage in channel 1 and the second lowest voltage on channel 2, etc.

  -Change the sampling frequency of 1 kHz to 150 Hz to let the capacitors discharge time.

• USB-6210

  Hello!

  We just bought a USB-6210 hardware DAQ, which we could successfully install and it appears as well in the device OR monitor.

  So, we come the 6210 connected via USB to our Win7 Pro 64 bit machine and wanted to see if his works.

  We opened NI Max (ver. 14.5), our device is recognized and we have opened a "Panel Test". We don't see anything, but the noise that we didn t connect anyhing to any unit. But we see a lot of crazy things. Our analog inputs are beginning to increase and saturate all about 11V. Some other entries are reduced to 11V slowly to 0...

  However, we connected a photoreceptor (thorlabs) to acquire certain signals... nothing is showing just all the same strange signals.

  We went on our device USB-6000 and everything works. We don't get anything with USB-6210.

  Any ideas?

  Thank you

  The readings that you see with nothing plugged are expected. When you do not connect a signal, what is your configuration (diff, CSR, etc.) and what is your wiring? Looks like you selected a configuration but your wiring does not match then you still have a nice.

• I need help to change a VI of the NI PXI-1042 q to USB-6210

  Hi all.  This is my first post, and I apologize in advance if it is in the wrong place or is not suitable... I was instructed to come here to help.

  I'm working on the evolution of a VI that works perfectly with an NI PXI-1042 q on a much smaller unit, the USB-6210.  Supposedly the USB-6210 can handle input/output same I need (a read analog at 30 kHz, 2 outputs digital input, written at 10 kHz), however I am having a little trouble to make it work properly.  The exits seem to go where they are supposed to, but the new system is orders of magnitude slower than the original, and for the system, I use this is unacceptable.

  I'm sure I'm doing something wrong, but I'm about as new to the LabView you can get, so I have no idea where to start.  Any help would be appreciated.

  More info you might need:

  LabVIEW 2010

  The VI is attached.

  Respectfully,.

  Emrys Maier

  University of Texas at Arlington Research Institute

  Research Assistant

  The device is unable to produce the frequency required by the output digital, but the outings of the counter do it perfectly.

  Emrys-

• USB-6210 on Linux - is DAQmxBaseCfgAnlgEdgeStartTrig supposed to work for this unit?

  I have two devices USB-6210 I need to synchronize so that they both collect data exactly at the same time. I want to trigger the data collected for each device by sending a trigger off Dev1/PFI6 and have two devices to receive the signal on PFI0.

  I "ve tried using the example AI program that receives a left analog trigger, and it doesn't seem to work. Here is the code I tried:

  /*********************************************************************** ANSI C Example program:*    acquireNScans-AnlgStart.c** Example Category:*    AI** Description:*    This example demonstrates how to acquire a finite amount of data*    using the DAQ device's internal clock, started by an analog edge*    condition.** Instructions for Running:*    1. Select the physical channel to correspond to where your*       signal is input on the DAQ device.*    2. Enter the minimum and maximum voltage range.*    Note: For better accuracy try to match the input range to the*          expected voltage level of the measured signal.*    3. Set the number of samples to acquire per channel.*    4. Set the rate of the acquisiton.*    Note: The rate should be AT LEAST twice as fast as the maximum*          frequency component of the signal being acquired.*    5. Set the source of the start trigger. By default this is*       analog input channel 0.*    6. Set the slope and level of desired analog edge condition.** Steps:*    1. Create a task.*    2. Create an analog input voltage channel.*    3. Set the rate for the sample clock. Additionally, define the*       sample mode to be finite.*    4. Define the parameters for an Analog Slope Start Trigger.*    5. Call the Start function to start the acquistion.*    6. Read all of the waveform data.*    7. Call the Clear Task function to stop the acquistion.*    8. Display an error if any.** I/O Connections Overview:*    Make sure your signal input terminal matches the Physical*    Channel I/O Control. In this case wire your signal to the ai0*    pin on your DAQ Device. The analog start trigger must be connected*    to PFI0.** Recommended Use:*    1. Call Configure and Start functions.*    2. Call Read function.*    3. Call Stop function at the end.**********************************************************************/
  
  #include "NIDAQmxBase.h"#include 
  
  #define DAQmxErrChk(functionCall) { if( DAQmxFailed(error=(functionCall)) ) { goto Error; } }
  
  int main(int argc, char *argv[]){    // Task parameters    int32       error = 0;    TaskHandle  taskHandle = 0;    char        errBuff[2048]={'\0'};    int32       i;
  
      // Channel parameters    char        chan[] = "Dev1/ai0";    float64     min = -10.0;    float64     max = 10.0;
  
      // Timing parameters    char        clockSource[] = "OnboardClock";    uInt64      samplesPerChan = 1000;    float64     sampleRate = 10000.0;
  
      // Triggering parameters    //orig    char        triggerSource[] = "PFI0";    char        triggerSource[] = "/Dev1/PFI0";    uInt32      triggerSlope = DAQmx_Val_RisingSlope;    float64     triggerLevel = 1.0;
  
      // Data read parameters    #define     bufferSize (uInt32)1000    float64     data[bufferSize];    int32       pointsToRead = bufferSize;    int32       pointsRead;    float64     timeout = 10.0;
  
      DAQmxErrChk (DAQmxBaseCreateTask("",&taskHandle));    DAQmxErrChk (DAQmxBaseCreateAIVoltageChan(taskHandle,chan,"",DAQmx_Val_Cfg_Default,min,max,DAQmx_Val_Volts,NULL));    DAQmxErrChk (DAQmxBaseCfgSampClkTiming(taskHandle,clockSource,sampleRate,DAQmx_Val_Rising,DAQmx_Val_FiniteSamps,samplesPerChan));        printf("Calling CfgAnlgEdgeStartTrig with %s...\n", triggerSource);    DAQmxErrChk (DAQmxBaseCfgAnlgEdgeStartTrig(taskHandle,triggerSource,triggerSlope,triggerLevel));    printf("Done with StartTrig.\n");    DAQmxErrChk (DAQmxBaseStartTask(taskHandle));    DAQmxErrChk (DAQmxBaseReadAnalogF64(taskHandle,pointsToRead,timeout,DAQmx_Val_GroupByChannel,data,bufferSize,&pointsRead,NULL));
  
      printf ("Acquired %d samples\n", pointsRead);
  
      // Just print out the first 10 points    for (i = 0; i < 10; ++i)        printf ("data[%d] = %f\n", i, data[i]);
  
  Error:    if( DAQmxFailed(error) )        DAQmxBaseGetExtendedErrorInfo(errBuff,2048);    if(taskHandle != 0) {        DAQmxBaseStopTask (taskHandle);        DAQmxBaseClearTask (taskHandle);    }    if( DAQmxFailed(error) )        printf ("DAQmxBase Error %d: %s\n", error, errBuff);    return 0;}
  

  and here is the result of the execution of the program:

  $. / acquireNScans-AnlgStart
  Calling CfgAnlgEdgeStartTrig with/Dev1/PFI0...
  DAQmxBase error-200428: value passed to the control task/channels is not valid.

  The file README.txt NIDAQ-mx says this is taken in charge for USB-621 X. Can someone tell me why it does not work?

  Tom,

  Analog start and trigger reference are supported only on the PXI, PCI, and the NOR-6210 PCIe form factors.  The Readme file has a list of the functions that are available for USB.

  Kind regards

  Justin

• voltage measurement usb-6210

  Hello

  I'm looking for to accomplish three parallel en measures of tension by using a USB-6210 card. Regardless of the means of control used (vi in LabVIEW + acquisition of data SignalExpress or measurement and Automation Explorer), the read signal is not one I send on the map. Attached an example from the control panel. Has someone it already faced this type of results? I suspect a gross error on my part the signal car didn't Exchange read regardless of the voltage that I send on the map (using a generator). Yet it seems to have respected the wirings and my drivers are up to date.

  Small additional difficulty: I'm working on German envorinnements.

  Thanks in advance for the help

  Hello Saiko,

  Thank you for having posted the Forum of National Instruments.

  In your screenshot, you perform a differential measurement. Well connected the negative potential (mass) at the AI9 entrance? Maybe you connected the mass to the GND pin of GOT it, that she only works when a CSR measure.

  I invite you to download the manual of your card: here and here.

  Kind regards

• The streaming of multiple channels directly to CSV USB-6210?

  I would like to stream up to 16 channels of data to a CSV file. My sample rate can vary from a file in a file, but I can reach 1 kHz. I understand the disadvantage of using text rather than PDM or binary files, but I have to be able to read these files later with programs such as Excel (Excel 2007 can make 1 million points).

  I have no trouble writing to TDMS files as seems very straighforward. I wonder though, it would be more efficient to use the TDMS files as the data are acquired and then after I stop, read the PDM and convert to CSV or just directly save to CSV? I only say this because I don't know if Labview can broadcast on a text or CSV file at 1 kHz.

  The scenario. The user will begin acquisition (using the USB-6210) to start playback 16 channels, applying multipliers for some data, and then send to a drive. Don't worry about the file size, the operator will be told that if it samples fast it must limit its total sample time. He has the ability to run slower to achieve longer sampling frequencies.
  Bottom line is I want good data for all 16 channels timestamped. No header information, nothing other than data raw beginning to end.

  I've seen several posts where the notice has been using TDMS files. Yet once again, I am not opposed to that, but I need to end up with text files.

  Jeff

  LabVIEW can handle this, if you do it right.

  1. If you are not already, use the design data of producer/consumer model.  It should be a template provided with LabVIEW.  This will allow you to take continuous data because your data treatment/backup of the routines take place in a separate thread.  If the hard data recording longer than reading the data, LabVIEW will keep only the size of the queue more and more with new data and I hope that you will not run out of memory system.

  2 write. you data in "chunks".  If you expect to write each data point to the disk as soon as it is available so that will probably be very tax on performance due to the excessive amount of 'read/write' header to your disk controller information and other things.  It is better to save 1000 samples at once, for example.  This way you are 'hit"your disc 10 times per second instead of 1000 times per second.

  3 ASCII files are very large, but I too tend to use them due to their convienance.  If you think about it, a 32-bit value can be "12332.0123" which in the ascii code is rather 80-bit (1 byte per character, 8 bits per byte, 10 characters is 80 bits or more than twice the amount if stored as a number 32-bit floating point).  It is therefore logical that if you have problems with performance (even after the application of 1 and 2) you write to a binary file.  Then, you can easily write a program that reads the binary file and writes a text file.

  4. If your computer has enough memory, you could put everything in memory and save it in a file at the end of the data collection.  You will need to add more controls and other things if you were writing this LabVIEW application for someone else.

• How to out high-level digital signal (5V from 1V) on usb-6210?

  Hi, I work with a OR usb-6210 and so I did not all channels of analog output that I can use. I hope that I can use the digital output for my application because I am creating a switch I can synchronize my editing. I need 5V output for my device run and currently, I am only able to get 1V.

  -Teo

  Outputs digital are dip lent TTL of the stent, since, your load is attractive more current that can provide the device and drop the voltage. Please provide information on what you are connected to or look at the other posts where users have had exactly the same problem. You probably just a conductor connected to the digital output.

• A USB-6210 will work on USB 1.1?

  Quick question...

  A USB-6210 will work OK only with USB 2.0 or USB 1.1 ports?

  I am considering buying one of these, and there is a small chance I might need to use it on an old computer that only has USB 1.1 ports.

  I only need 16 channels of analog input with a sampling rate around 1 kHz, without input/output digital or counter.

  Hello, MIG,.

  I hope you are well today!  In fact, you are able to use our devices for acquisition of data USB 2.0, USB 1.1, but with the decrease in speed.  Please see this link here for more information on styling.

  Thanks MIG!

  Brandon G

• slowly decreasing oscillation for PID

  In my project, now I am facing a very interesting problem, and I would like to ask the opinion of others.

  I control the temperature of a phyiscal object using the LabView PID toolkit, with good results, I set my controller manually using the Ziegler-Nichols closed loop adjustment rule. The process value (PV) reaches the set point (SP) quick (about a year and a half period time and swing left), and I've reached the ultimate accuracy (deviation around SP) fast enough.

  However, I realized that in fact, in my experience, a swing slowly decline would be beneficial at the beginning of the process because of certain physical phenomena.

  I guess I should use some kind of hourly? For example, I would that my control system having an oscillation decreases with only 5% of amplitude in a period time.

  All tips are welcome!

  Kind regards

  I think that the right adjustment of the values, you could get the controller for you to do. Try to increase the proportional gain - but do not forget that with the LabVIEW PID, which will also affect the gain full and derived, then you may need to set those too. You know, Ziegler-Nichols tuning approach, that there is a proportional gain during which you get stable sustained oscillations, so looking for a proportional gain below the ultimate gain which allows you the degrowth in oscillations that you want.

  For more info on math, try a search on "rate of decay PID. Conventional tuning methods aim ratio of disintegration of a quarter - each successive swing has an amplitude of 1/4 of the previous swing. Looks that you want your controller tuned for a greater ratio decomposition, where each successive swing has an amplitude of only slightly lower than the previous.

  Alternatively you could overlay a swing decaying about to set. This might be easier.

  Gain scheduling is used when you want the different PID gains according to the set point. From your description, I think not hourly is a solution to your problem.