Dear Jack, Thank you so much. Sensor scripts are working well in RFSOC devices. Recently, I used *pfb_fir_generic* block instead of *pfb_fir_real*. I faced this issue after using that block. I used 4 channels ADC and plotted 4 spectra simultaneously, even though the 4x2 board is working well. Whenever I use a specific block pfb_fir_generic, even 2 channels ADC, it gets a runtime error. Is there something with that block? Did someone face a similar issue while using the pfb_generic block? pfb_generic block bit code significantly increases the current in 2 channels VCC_0V85: 0.88 Volt, 8.83 Amp more than 4_channel spectrometer VCC_0V85: 0.88 Volt, 5.83 Amp This is my observation so far. As of now, I continuously run the 4-channel spectrometer bit code 5 times, this FPGA doesn't show a runtime error. Any suggestions are welcome. Thank you.
Best Regards Bishnu Sharma Institute of Astronomy and Astrophysics Academia Sinica On Wed, Jan 22, 2025 at 11:16 PM Jack Hickish <jackhick...@gmail.com> wrote: > > > On Wed, 22 Jan 2025, 01:28 Bishnu Kumar Sharma, < > bksha...@asiaa.sinica.edu.tw> wrote: > >> Dear Jack, >> Thank you for your valuable suggestions. The model is bigger than the >> spectrometer design of the RFSOC4x2 tutorial design. When I continuously >> run that bit code a few times 4x2 board, I can feel the board getting hot. >> - I will think about putting bigger heat sink. >> - As of now, I tested the smaller design multiple times and the board is >> working very well. I can feel that the board is relatively very cooler than >> running that big design. >> - Is a 4x2 board not that reliable? I checked the top utilization report >> from Vivado compiler, the design doesn't occupy 100% resources. >> > > I don't think it is a reliability problem, but rather the core voltage > power supply is relatively weak compared to what the FPGA is capable of > drawing. Also, the heat sinking of the FPGA isn't really suitable for big > designs. I suspect both of these decisions are somewhat deliberate on > xilinx's part, to avoid the board eating into the markets of more expensive > rfsoc boards provided by other vendors. > > For the ZCU this script will report temperature and voltage rail levels > https://github.com/realtimeradio/souk-firmware/blob/main/software/rfsoc_scripts/get_sensors.py > . It uses the RFSoC on chip SYSMON core so should work on all RFSoC > platforms. > > The current measurements of the other script unfortunately only work on > the 4x2, which has dedicated current monitoring hardware (I love > RealDigital for that). I don't know if the ZCU216 has any such monitoring, > but if it does I unfortunately don't have a script for it. > > Cheers > > > Thank you very much for the Python script. >> I am using the ZCU216 board also, and I haven't faced this issue so far. >> Do you have a Python script for the zcu216 board to check power and >> temperature? >> I ran this script and got the following results. >> casper@localhost:/lib/firmware/rfsoc4x2$ python3 get_power_sensors.py >> VCC_0V85: 0.88 Volt, 3.24 Amp (8% max) >> VCC_3V3: 3.33 Volt, 0.44 Amp (7% max) >> VCC_1V8: 1.82 Volt, 0.61 Amp (10% max) >> SYZYGY_VIO_BUS: 3.38 Volt, 0.01 Amp (0% max) >> VDAC_AVCC_0V925: 0.96 Volt, 0.00 Amp (0% max) >> VDAC_AVCC_AUX_1V8: 1.82 Volt, 0.00 Amp (0% max) >> VDAC_AVTT_2V5: 2.53 Volt, 0.01 Amp (0% max) >> VADC_AVCC_0V925: 0.95 Volt, 1.52 Amp (50% max) >> VADC_AVCC_AUX_1V8: 1.82 Volt, 1.12 Amp (55% max) >> Thank you so much. >> Best Regards >> Bishnu Sharma >> Institute of Astronomy and Astrophysics >> Academia Sinica >> >> >> >> On Tue, Jan 21, 2025 at 7:17 PM Jack Hickish <jackhick...@gmail.com> >> wrote: >> >>> Hi Bishnu, >>> >>> Is your design "big"? I have found it easy to overwhelm the power >>> supplies of the 4x2 with a large design, at which point they automatically >>> shut down, and the board becomes unresponsive to both ping and casperfpga >>> commands. >>> If I remember correctly, if this happens there are some error LEDs on >>> the board which will light up. >>> I've also found that putting a bigger heatsink on the board can buy a >>> little more margin. >>> >>> Assuming your board's OS has the appropriate I2C drivers, if your issue >>> is power related, you may find this script useful - >>> https://github.com/realtimeradio/souk-firmware/blob/main/software/rfsoc_scripts/get_power_sensors.py >>> If you run it on the board, it will report the current drawn on each of >>> the power rails.in my experience, the board will shut down once 0V85 >>> current draw gets near 100%. >>> >>> If you are only running small firmware, then I suspect your issue is >>> something else. >>> >>> Cheers >>> Jack >>> >>> On Tue, 21 Jan 2025, 06:41 Bishnu Kumar Sharma, < >>> bksha...@asiaa.sinica.edu.tw> wrote: >>> >>>> Dear Casperites, >>>> I hope you are doing well. I have faced a little bit strange situation >>>> when I use an FPGA 4x2 board to run bitstream. Sometimes when I run >>>> bitstream in a virtual environment, programming gets stuck and reports >>>> errors. like: >>>> Traceback (most recent call last): >>>> File "rfsoc_4x2_corr.py", line 123, in <module> >>>> fpga.upload_to_ram_and_program(bitstream) >>>> File >>>> "/home/bishnu/Desktop/m_work/casper_venv/lib/python3.8/site-packages/casperfpga/casperfpga.py", >>>> line 348, in upload_to_ram_and_program >>>> rv = self.transport.upload_to_ram_and_program( >>>> File >>>> "/home/bishnu/Desktop/m_work/casper_venv/lib/python3.8/site-packages/casperfpga/transport_katcp.py", >>>> line 654, in upload_to_ram_and_program >>>> raise RuntimeError('%s: no programming informs yet. ' >>>> RuntimeError: 192.168.50.198: no programming informs yet. Odd? >>>> I already upload the bitstream with the proper virtual environment. >>>> When I power on the FPGA, it works for a while and after running some >>>> bitstream file, it >>>> happens. I even checked the ping for the network, it doesn't show the >>>> flow of the network. >>>> I have also attached some of the screenshots for reference. Thank you >>>> for your help. >>>> Best Regards >>>> Bishnu Sharma >>>> Institute of Astronomy and Astrophysics >>>> Academia Sinica >>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "casper@lists.berkeley.edu" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to casper+unsubscr...@lists.berkeley.edu. >>>> To view this discussion visit >>>> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CACqpYn5VXe%3D-fgQALUJ9Cqk2tugA%2BH7f_ZYfGj2wDAbP%3D1NH%3Dg%40mail.gmail.com >>>> <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CACqpYn5VXe%3D-fgQALUJ9Cqk2tugA%2BH7f_ZYfGj2wDAbP%3D1NH%3Dg%40mail.gmail.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "casper@lists.berkeley.edu" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to casper+unsubscr...@lists.berkeley.edu. >>> To view this discussion visit >>> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSnkiTTpYGP3k%3DP_kaadU1L9kG2ug9KDW3a9qSc7quxgWA%40mail.gmail.com >>> <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSnkiTTpYGP3k%3DP_kaadU1L9kG2ug9KDW3a9qSc7quxgWA%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- >> You received this message because you are subscribed to the Google Groups >> "casper@lists.berkeley.edu" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to casper+unsubscr...@lists.berkeley.edu. >> To view this discussion visit >> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CACqpYn61NTa6wsmd9T4TwRV7bBq07XRPv5jbLP8jwOQRSRgA-A%40mail.gmail.com >> <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CACqpYn61NTa6wsmd9T4TwRV7bBq07XRPv5jbLP8jwOQRSRgA-A%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> > -- > You received this message because you are subscribed to the Google Groups " > casper@lists.berkeley.edu" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to casper+unsubscr...@lists.berkeley.edu. > To view this discussion visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSnz3H%3Dct6-pQC9aWemvum5%2B3U7LrRH6aOePa1O18r3cdQ%40mail.gmail.com > <https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSnz3H%3Dct6-pQC9aWemvum5%2B3U7LrRH6aOePa1O18r3cdQ%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "casper@lists.berkeley.edu" group. To unsubscribe from this group and stop receiving emails from it, send an email to casper+unsubscr...@lists.berkeley.edu. To view this discussion visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CACqpYn55LxUXnfRUqAJwtNoN-%3DY0QtoCjsnoacxg8DX2P%2BfdJQ%40mail.gmail.com.
