Post Go back to editing

problem with CN0399 SDP-S and SDP-Z

Category: Software
Product Number: CN0399 SDP-S AND SDP-Z

Dear reader,

 I can't get the combination of SDP-S (controller) and SDP-Z (RF unit) to work.

My PC has Windows 7

When I connect the SDP-S/SDP-Z to a USB port, the correct PC System messages come up:

Analog Devices System Demonstration platform SDP-S

 At Details it says:

Creation date 02-12-2020

Version 2.14.0.0

After starting GUI ADI_SDP_Writer1 ( version April 12, 2017,9.39.52) and pressing button “CONNECT”

comes after a few seconds: SDP connected.

After this, according to the procedure, some calibration value entered with the GUI.

 

If I now measure the power, a very wrong value appears on the GUI display.

As an example I take the measurement at 1 GHz with a known connected level of 0 dB

A positive number of +39.5 dB appears on the GUI, so 39.5 dB is too high.

 If you offer a level of -10 dB at the input, the GUI will indicate +29.5 dB, so 39.5 dB is too high.

At an input of -20 dB, the GUI indicates + 20.0 dB, so 40 dB is too high.

With no input signal, the GUI shows +5.0 dB?

 In the table 3PointCal-LUT, the frequency of 1 GHz has the following value:

1000.60.5-34.0330578512397.58.2,-35.3780068728522,,895,2059,2664.

 I think the last 3 values ​​represent the counts of the AD converter. And if this assumption is correct

then the combination SDP-S and SDP-Z would work well. But the GUI makes the calculation mistake.

 I can't figure out what and where it goes wrong can you help me?

I am an amateur (72 years) and have tried a lot but I can't get it to work properly.

 Thank you in advance for your response. Best regards,

 LLLRRR

 

 

 

Parents
  • Hi,

    It's not clear if the calibration file is being written correctly. Can you delete the contents of the file are start again?

    Regards,

    Kieran 

  • Hi Kieran,

     

    I still get the combination CN0399 SDP-S and SDP-Z not working. What strikes me is that there is nowhere to find an example of a working copy on the internet. A demonstration was given by you on YouTube.

    https://www.youtube.com/watch?v=kaeRGOqxwE4

     

    In the last part of this movie, the monitor of the PC also shows also a very high dB value (34.44 dB). And if you look closely, there is also an alarm message on the screen that the input level is too high. The power meter can only measure a maximum of 15dB.

    What is going wrong is there an error in the program or am I doing something wrong?

    This combination of SDP-S and SDP-Z cost me 300 Euros and it's no use to me now.

     

    I hope you can help me. Thanks in advance and with kind regards,

     

    Loek Riemens

     

  • Hi,

    In that video the device was not sampling. It was a static with no sampling so this is not an accurate representation.

    I'm out of office until Monday. So I will work through this with you next week. I have one of these boards and will try to reproduce your issue. I'll update you on early next week. 

    Regards,

    Kieran 

  • Hi Kieran, Thanks for your comment.

    My knowledge of the English language is limited so I communicate via GOOGLE translate. That is why communicating via a direct chat or Phone call is very difficult for me. I live in the Netherlands and am retired . I am 72 years old and for the hobby I started experimenting with CN0399. In your email of 7/7/2022 you wrote that you would try to reproduce my issue. Did this work or does it work for you?  

    Thank you in advance for your response, best regards,

     Loek Riemens

  • Hi Loek,
    I was able to reproduce a similar response. However I only have a version of the board from prerelease that I'm not 100% in it working. I'll need to spend time on this, this week and get back to you on it.

    Thanks,

    Kieran 

  • Hi Loek,

    I've been doing some testing in the lab today. I haven't identified an exact root cause yet but it is intermittently working for me. As an experiment would you be able to try measure with a higher frequency e.g. 1GHz?

    Thanks,

    Kieran 

  • Hi Kieran,

    I performed the following steps:

    1st Completely empty the file “3pointCal-LUT.cvs” with Excel.

    2nd Started the GUI with the application “ADL5904 low power RFpower meter .

    3rd Press the “Connect” button via the GUI, after which the message “SPD connected” appears.

    4th Press the “Calibration” button through the GUI and select “3-Point Calibration”

    5th via the GUI the measurements performed at 1GHz LowCal (-20dBm)-MidCal (0dBm) HighCal(+10dBm). With every measurement the send level on the input is adjusted and the GUI comes per message with “Code saved for -20/0/+10dBm power.

    6th With the GUI close the measurement for 1GHz by pressing the button “Calibrate” the message comes “Calibration done for 1GHz”.

    7th repeat steps 5 to 6 for the frequency of 2GHz.

    8e In the Excel file “3pointCal-LUT.cvs” the following data is now stated:

       1000,60.3,-33.9137645107794,57,55,-35,5343179843614,,894,2045,2648

       2000,61.6,-32.4512987012987,61.3,-32.6101141924959,,773,1999,2615

    9th With the GUI selected the button “Power Measurement”

    10th Chosen with the GUI for 1GHz and there will be the correct measurement result

    11th Chosen with the GUI for 2GHz and even then the good measurement result comes

    12th After this closed the GUI

    13th After this, restarted the GUI and chose step 10 or 11 . Now comes a wrong value like 2617dBm or 46dBm.

    14th If you then carry out the calibration steps 2 to 7 again, errors will continue to be measured.

    So it seems that correct measurements are only obtained if the file “3pointCal-LUT.cvs” is empty at the start of the calibration. And that there are only correct values ​​immediately after the calibration and that when starting the GUI with existing calibration values ​​in the file “3pointCal-LUT.cvs” no correct values ​​are obtained.

     

    Sincerely,

     loek

  • Hi Loek,

    I'm not able seeing the same issue as you are on my setup. I've tried emailing you with the GUI version I'm using but the attachment gets blocked. So I've uploaded to our FTP location:  ftp://pllev_ftp:3U46DK5w@ftp.analog.com 

    Can you try download 'CN0399 Software (Rev 1).zip' from there?

    Thanks,

    Kieran

Reply
  • Hi Loek,

    I'm not able seeing the same issue as you are on my setup. I've tried emailing you with the GUI version I'm using but the attachment gets blocked. So I've uploaded to our FTP location:  ftp://pllev_ftp:3U46DK5w@ftp.analog.com 

    Can you try download 'CN0399 Software (Rev 1).zip' from there?

    Thanks,

    Kieran

Children
  • Hi Kieran,

    First of all I would like to thank you for the effort you are making to help me.

    I tried to download the File you specified via the link pllev_ftp:3U46DK5w@ftp.analog.com. This was not successful. I see some other files but none of them are downloadable.

     Just to be sure, I reloaded the file “CN0399 Software(Rev1) via the link provided by Analog Devices ftp.analog.com/.../ and then installed it again.

    Could there be a difference in the GUI you use or is the same program on this link?

     Also now the connection goes well via the GUI. Then I performed all Calibration steps (10MHz to 6GHz). If I perform a power measurement immediately afterwards, wrong values ​​are displayed such as +50 dB. Even when restarting the GUI, the power measurement value remains incorrect.

    This is again a different behavior than I described last time. Then a power measurement immediately after Calibration was good. And it went wrong after restarting the GUI.

    In the EXCEL_file "3PointCal_LUT " are all Calibration values ​​per frequency on one row with a comma as separator. The last three values ​​per frequency are I think the counter value of the A/D converter of the three levels that are entered during the calibration. This value corresponds well with what is stated in the IC datasheet.

     It still seems that the power measurement does not calculate or calculates incorrectly with the data in the file “3PointCal_LUT”.

    If you perform a new Calibration after the Calibration, the ode lines will remain and the new ones will be added at the bottom of the list. The manual does not describe how to perform a new Calibration if necessary.

    Do you first have to empty the file“3PointCal_LUT” via EXCEL or does the GUI always take the last entered data?. 

    So I use a Windows 7 (professional 64 bits) PC with EXCEL from Microsoft Office 2007.The SDPSDriver comes from the same folder found at the ftp.analog.com/.../ link and is dated November 8, 2013.

    Could it be that the version of EXCEL that is on my PC influences the functioning of the GUI. So I mean before viewing the contents of the file “3PointCal_LUT “ but also when emptying it? 

    Is the source code of the GUI possibly available then I could do some more research myself? 

    I am still working daily to solve my problem. I have set all permissions of the applied folders to the highest level. I am now also working on a USB sniffer to see what data is going over the USB. Maybe that way I can figure out where it's going wrong. In this way I hope to be able to answer the following questions:

     What value comes from the A/D converter during a Calibration measurement?

    What values ​​come from the A/D converter during a power measurement? 

    Thank you in advance for your response, best regards,

     Loek

     

  • Hi Loek,

    You're welcome. I'm happy to help.

    I've since checked with the version of software that you have and I'm getting the same results as I get with my version. So there should not be any difference. 

    Here is a copy of the table that I am using: 

    0 -0.05 210 -0.05 210 11 11 10
    100 0.6 -28.3333 0.6 -28.3333 11 11 23
    1000 51 -4.45098 10.85 -20.9217 10 227 737
    1500 62.3 -7.07865 21.5 -20.5116 11 441 1064
    2000 63.9 -11.4554 35.7 -20.5042 18 732 1371
    2500 56.6 -19.682 50.55 -22.0376 103 1114 1680
    3000 63.6 -15.6918 46.65 -21.3934 65 998 1634
    3500 58.4 -13.6986 42 -19.0476 44 800 1384
    4000 62.7 -10.1595 48.3 -13.1884 154 637 1264
    4500 62.5 -8.896 43.5 -12.7816 121 556 1181
    5000 65 -8.2 42.2 -12.6303 111 533 1183
    5500 60.3 -9.22056 43.6 -12.7523 120 556 1159
    6000 59.9 -7.62938 38.7 -11.8088 70 457 1056

    As a test, could you replace the contents of your csv file with this and then take measurements?

    I don't expect that you'll see as accurate results as I'm seeing with my board due to part to part variation. But values should be a lot more sensible than you are seeing currently. 

    I pasted the source code into the same FTP location: ftp://pllev_ftp:3U46DK5w@ftp.analog.com 

    You should be able to download if download 'WinSCP': WinSCP :: Official Site :: Download

    The file is called 'ADL5904 Low Power RF Power Meter.zip'.

    Regards,

    Kieran 

  • Hi Kieran,

     

    I succeeded in downloading the FTP files via WinSCP.

    The BIN/Debug/ folder contains the following files:

    3PointCal_LUT (1KB)

    ADL5904 low power RF power Meter (282 KB) : Application File

    ADL5904 low power RF power Meter (50KB) : PDB File sdpApril1 (1.142KB) : DLL File

    When I start the program with “ADL5904 low power RF power Meter (282 KB) : Application File” Then the GUI starts but when pressing the button “connect” the error message appears:

    “No matching system found. Press Rescan to retry or Cancel to abort. If your SDP is recently connected, it may be in the process of booting. Wait ~40secs and Rescan.” This message continues to appear even after several attempts with “Rescan”.  After that I did some experimenting:

    The program “ADL5904 low power RF power Meter (284 KB) : Application File” that I had used so far, I have replaced with the program that I downloaded via WinSCP. So all the other 3 parts (PDB/3PointCal/DLL) that I have always used are still there.

    Now a “connection” is established. The GUI layout looks slightly different. There is an extra window with I think the value of the A/D counter.

    Also now all wrong measured values ​​appear in the dB window. But what is striking is that the value in the dB window is equal to the value in the extra window.

    Furthermore, with every measurement you perform, there is always an error message in my case in Dutch with the message: "The format of the input string is incorrect".

    To be clear, this message also appears in the old situation, so with the files I used until now. I'm really sure that the correct order was always used when calibrating.

    What is the reason that the program loaded via WinSCP does not work?

    Which driver did you use?

    I still think the failure for me has something to do with reading the file 3PointCal_LUT . When I look at the data file “3PointCal_LUT” I received from you, all values ​​are in a separate cell. 

    0 -0.05 210 -0.05 210 11 11 10
    100 0.6 -28.3333 0.6 -28.3333 11 11 23
    1000 51 -4.45098 10.85 -20.9217 10 227 737
    1500 62.3 -7.07865 21.5 -20.5116 11 441 1064
    2000 63.9 -11.4554 35.7 -20.5042 18 732 1371
    2500 56.6 -19.682 50.55 -22.0376 103 1114 1680
    3000 63.6 -15.6918 46.65 -21.3934 65 998 1634
    3500 58.4 -13.6986 42 -19.0476 44 800 1384
    4000 62.7 -10.1595 48.3 -13.1884 154 637 1264
    4500 62.5 -8.896 43.5 -12.7816 121 556 1181
    5000 65 -8.2 42.2 -12.6303 111 533 1183
    5500 60.3 -9.22056 43.6 -12.7523 120 556 1159
    6000 59.9 -7.62938 38.7 -11.8088 70 457

    If I now enter new Calibration value and leave the old one, these will come under the existing data cells but no longer per cell data. But per calibration frequency, all values ​​are in one cell separated by commas. See example below.

     0.60.45,-35,6906534325889,61,26666666666667,-35,348204570185,,634,1553,2762100,59,35,-35,6781802864364,58,93333333333333,-35,8597285067873,,640,1524,27111000,59,9,-33,9398998330551,58,9,-34,5161290322581,,855,2033,26321500.0,60.5,-32,2479338842975,59.55,-32.7623845507976,,760,1951,25562000,61.1,-30,9819967266776,60,7,-31,1861614497529,,679,1893,25042500.0,62.6,-29,8402555910543,61.1,-30,5728314238953,,646,1868,2494.

     Could this be the reason and how can this be solved? 

     Looking forward to your response with kind regards, 

    Loek

     

  • Hi Loek,

    When I start the program with “ADL5904 low power RF power Meter (282 KB) : Application File” Then the GUI starts but when pressing the button “connect” the error message appears:

    - The only time I see this appear is when a connection has already been established with another instance. In this case, is your original GUI already connected?
    - What I would suggest here is to close down the already connected version and physically remove the cable and reconnect before trying again to connect on the 'ADL5904 low power RF power Meter' application.

    The program “ADL5904 low power RF power Meter (284 KB) : Application File” that I had used so far, I have replaced with the program that I downloaded via WinSCP. So all the other 3 parts (PDB/3PointCal/DLL) that I have always used are still there.

    I would suggest using the existing 3PointCal_LUT file that comes with the download of the source code. This actually contains the LUT content that I pasted in my previous message. 

    Furthermore, with every measurement you perform, there is always an error message in my case in Dutch with the message: "The format of the input string is incorrect".

    I've seen this happen before and it's due to the program not reading correctly. I have usually started the program again and it fixes it.

    Which driver did you use?

    I'm not sure what you mean?

    If I now enter new Calibration value and leave the old one, these will come under the existing data cells but no longer per cell data. But per calibration frequency, all values ​​are in one cell separated by commas. See example below.

    CSV format separates cells by commas. I'm not sure what's happening. But I suggest using the 3 point LUT file that comes with the download as a first step to seeing if measurements are reading somewhat correctly. 

    Regards,

    Kieran 

  •  Hi Kieran,

     

    Another addition to the experiments I've performed:

    Of the files that I downloaded via WinSCP, I first emptied the file “3PointCal_LUT”. Then the DLL file has been replaced by the DLL file that I have used so far. The GUI boots up and “connect” works fine. After that Then a full Calibration was performed.

    After this the power meter works well. I can read values ​​from -25 dB to +13 dBm. But when I close the GUI and start it up again, “connect” works fine, but the dB values ​​are now totally wrong and very high. This same behavior was also present with the files I used so far.

    So it only works when measuring directly, so without leaving the GUI.

    Looking forward to your response with kind regards,

     Loek

     

  • Hi Kieran, When I apply the file obtained with WinSCP see screen dump 

     then the error message comes with connect”

    All possibilities you mentioned such as “ close down the already connected version and physically remove the cable and reconnect before trying again to connect on the 'ADL5904 low power RF power Meter' application.” have been performed several times and all have been unsuccessful. 

    So the only way to use the new GUI and “3PointCal_LUT -file“ is to replace the DLL-file with the DLL-file I had used so far. Or to place the GUI download via WinSCP in the configuration that was used until now. This is of course not really the usual way, but I was curious if the power meter would work. 

    Because everything downloaded with WinSCP does not work, the cause could also be the installed driver. The driver I installed was listed in the manual CN0399-SDP_Z.PDF and could be downloaded from the ftp site: anolog.com/pub/cftl/CN0399 .

    The driver details are:

    Supplier: Analog Devices,inc

    Date: 24-10-2013

    Driver: 2.1.2.0.

    The error message “The format of the input string is incorrect" appears with every manual measurement and in the "repeat" position the message appears the first time 

    Looking forward to your response with kind regards, 

    Loek