Daniel Rech

  • Content Count

  • Joined

  • Last visited

Everything posted by Daniel Rech

  1. Is there a document that explains the changes somewhere?
  2. Last time I checked was with Waveforms 3.3.7. Nowadays I read that Version 3.5.4 is the current version. --> Are there any changelogs available? I wasted a lot of time debugging this problem last time and I don't want to get into this again... @neoclassic I don't know if the problem is solved
  3. I didn't test your code but from my experience with high speed python variable declaration is critical: results1=[0.0]*len(rgc) ^-- this could happen before the loop --> then within the iteration you would not need to wait for the declaration. Did this solve your problem? Or did I misunderstand you question?
  4. Hi ravi_butani, I experienced similar issues with Digilent Adept 2 and a RaspberryPi 3. I think these issues might be correlated to the 4.4 kernel that is coming with it. On Beaglebone Black with a 3.8 kernel, the Digilent Adept works fine.
  5. Ok, then I'll stop trying to get it working Thank you for you help. I am using the Digilent Adept 2 with an older Beaglebone Black image (kernel 3.8) and it works perfectly.
  6. Ok, I'll look into it. The example you uplaoded uses DigitalIn. Which pins should I connect to the trigger signal? Is it wrong to use the AnalogIn to generate a trigger signal? The core of my problem is a weird behavior with Linux and kernels > 3.8. If I run my program on a Beaglebone Black with a 3.8 kernel, the triggering works fine, but if I run the same program on a Raspberry Pi 3 with a kernel 4.4 the scope seems to crash without a message. Only the dwfcmd seems to work with the kernel 4.4 system.
  7. Thank you for your response. I ended up with a solution where I don't need to compile dwfcmd. Thank you for the hint with the repository, I'll look into it!
  8. I found the dwfcmd in /usr/share/digilent/waveforms/samples/dwfcmd/dwfcmd.cpp. How do I compile this file? g++ throws many errors starting with "ISO C++ forbids forward references to 'enum' types" and "use of undeclared identifier".
  9. I have a weird issue with my Analog Discovery 2 and I was wondering if there are ideas out there how to track it: I use a Digilent Adept with the Waveforms 3.3.7 software on linux. It seems to work fine but does not detect my triggers. If I use the provided pyhton examples or C examples trigger detection does not work meaning the program just stays in "Waiting for trigger" state all the time. But if I use the dwfcmd: cd /usr/share/digilent/waveforms/samples/dwfcmd && dwfcmd analogintrigger.txt I get one trigger without a problem. It took me so much time to detect this behavior and now I don't have any idea how to solve this. The dwfcmd is not powerful enough for my application but maybe I'll end up modifying the code to fit my needs. I'd really like to keep using python, any ideas what I am doing wrong or any known bugs? Thanks Daniel Rech
  10. I have the same issue but no idea how to solve it. Did you solve it already?
  11. Am I the only one who's having problems on linux systems? I still haven't figured out how to solve this.
  12. Are you using Kernel 4 or 3? My linux driver keeps crashing with Kernel 4.4.24. I'm still looking into it. Did you solve your problem in the meantime?
  13. I get weird triggering behavior with the same code and 3.3 version. Not it seems that there are 400 triggers but my normal scope only shows 70 Is there an API change in 3.3 that I missed?
  14. I know, that I could use record mode but I need the triggers to clarify when the signal started and the application requires the 50Hz (== 20ms) rate. Thank you for your response. Right now we use 3.2.6 but I will try to get 3.3 running.
  15. Thank you for the reply. I started looking into the C code after some segmentation fault using the python libraries
  16. Is anyone using a kernel 4.1 with this board? Or which board are all the linux users out there using?
  17. Hi, I wrote a small programm to test the speed limitations of triggering and I have a problem: After collecting data the setup for the next trigger takes 15ms. The application I'm building requires to capture 10ms of data with 100000 Hz, so I want to capture 1000 samples. After that I only have 10ms to setup the next trigger. The programm collects all the triggers with not too many samples. But if I capture more than 500 samples (5ms) then I start missing triggers. What can I do to do this high speed data acquisition with the Analog Discovery 2? This is the cpp-Code I use: #include <signal.h> #include <stdlib.h> #include <stdio.h> #include <unistd.h> #include "sample.h" int bufferSize = 500; // int bufferSize = 600; // int bufferSize = 1000; int terminate = 0; void my_handler(int s) { terminate = 1; } int main(int carg, char **szarg) { signal (SIGINT, my_handler); HDWF hdwf; STS sts; printf("Open automatically the first available device\n"); if(!FDwfDeviceOpen(-1, &hdwf)) { printf("Device open failed"); return 0; } FDwfAnalogInFrequencySet(hdwf, 100000.0); FDwfAnalogInBufferSizeSet(hdwf, bufferSize); FDwfAnalogInChannelEnableSet(hdwf, 0, true); FDwfAnalogInChannelRangeSet(hdwf, 0, 4.99); // disable auto trigger FDwfAnalogInTriggerAutoTimeoutSet(hdwf, 0); // one of the analog in channels FDwfAnalogInTriggerSourceSet(hdwf, trigsrcExternal1); FDwfAnalogInTriggerTypeSet(hdwf, trigtypeEdge); FDwfAnalogInTriggerChannelSet(hdwf, 1); FDwfAnalogInTriggerLevelSet(hdwf, 1.5); FDwfAnalogInTriggerConditionSet(hdwf, trigcondRisingPositive); // wait for the offset to stabilize Wait(2); printf("Starting repeated acquisitions:\n"); int iTrigger; for(iTrigger = 0; iTrigger < 200; iTrigger++){ // begin acquisition FDwfAnalogInConfigure(hdwf, false, true); while(true){ FDwfAnalogInStatus(hdwf, true, &sts); if(sts == DwfStateDone) { break; } if (terminate == 1) { break; } } if (terminate == 1) { break; } } printf("#%i\n", iTrigger+1); FDwfDeviceCloseAll(); } Compile it with gcc -ldwf trigger.c && ./a.out It waits for triggers on external trigger source 1. The only way I see to solve this is to not care about triggers and capture the whole signal but I would really like to solve it in a nice way. Thank you for your replies Daniel
  18. I have a similar error with 3.3.3 on arm linux: dpkg-deb (Unterprozess): Archiv-Element wird dekomprimiert: interner gzip-Lesefehler: »<fd:4>: invalid literal/lengths set« What am I doing wrong?
  19. The same setup works with the old kernel and the device is already built into the system, I cannot attach a USB Hub to it. Do you have more details on where I can find more information about that "USB OTG driver/firmware issue"? Thanks for the quick reply!
  20. With the original linux installed on my Beaglebone Black the scope worked but then I installed a newer debian version with a 4.1.22 kernel and now I keep getting an error with scope power: dwfcmd analoginout.txt FAILED: FDwfDeviceOpen The device needs at least 2W/400mA from USB cable. To satisfy this: - Try to use different USB cable and computer plug. - Use the USB plug on the back of the PC, not the one on the front panel. - Use short cable, do not use long cable or cable extender. - Use powered USB-hub, avoid un-powered hub. - Use 5V auxiliary power supply. USB: 0.01V 777mA AUX: 3.58V 580mA The Analog Discovery 2 has its own 5V power supply and I use the original cable I don't see a reason why this shouldn't work. Anyone with the same problem or a solution out there?
  21. Right now I'm just installing it without xdg-utils and keep getting a dpkg error of missing dependencies, I can live with that but I was wondering if there is a better solution.
  22. Digilent Waveforms depends on xdg-utils which depends on x11 and many things. Can I install waveforms without all of this and how? Is anyone using this on an embedded device like the Beaglebone Black or Rasberry Pi?
  23. Thank you, --> gcc analogin_trigger.c -ldwf worked. Had to adjust the samples to get them running because the C syntax was dort of broken. Anyone interested in a github repository with fixed code? Am I even allowed to share the samples, what's the license?