Cleverscope and Bode plots

I have modified, slightly, a Cleverscope example. I'm using it to sweep an output voltage from 100 Hz to 20 kHz in 100 Hz increments into a low-pass filter with corner frequency of about 750 Hz. The asked for sampling frequency is 10,000 samples per second. This generates 200 frames of data for each of channels A and B which I read back using 'replay' and then calculate the magnitude and phase changes using 'Extract Single Tone Information.vi' to measure frequency magnitude and phase. This seems to work well for magnitude, but not for phase. Phase is expected to rise asymptotically from 0 degrees to 45 degrees and then decline asymptotically to 90 degrees. I sort of get this but at higher frequencies some phase data points dip to -300 or rise towards +300. I am using Matlab to get a better picture of of each individual frame. What I find baffling, and I'm not sure it's related, is that as I look at the frames in succession, the data seems to become corrupted towards the end of a frame, veering from sinusoidal to (what?) spurious?. Certainly not sinusoidal.

Can anyone give me an explanation of what is going wrong? Thanks.

I have modified, slightly, a Cleverscope example. I'm using it to sweep an output voltage from 100 Hz to 20 kHz in 100 Hz increments into a low-pass filter with corner frequency of about 750 Hz. The asked for sampling frequency is 10,000 samples per second. This generates 200 frames of data for each of channels A and B which I read back using 'replay' and then calculate the magnitude and phase changes using 'Extract Single Tone Information.vi' to measure frequency magnitude and phase. This seems to work well for magnitude, but not for phase. Phase is expected to rise asymptotically from 0 degrees to 45 degrees and then decline asymptotically to 90 degrees. I sort of get this but at higher frequencies some phase data points dip to -300 or rise towards +300. I am using Matlab to get a better picture of of each individual frame. What I find baffling, and I'm not sure it's related, is that as I look at the frames in succession, the data seems to become corrupted towards the end of a frame, veering from sinusoidal to (what?) spurious?. Certainly not sinusoidal. Can anyone give me an explanation of what is going wrong? Thanks.

Hello,
Yes there is a bug in earlier versions of the driver which meant that if you are using the multi-frame capture method (as Bandpass Response does), then the last sample is spurious, and accumulates. So after 200 frames you will have 200 incorrect samples at the end of the frame. This has been fixed in driver 2.1. We also suggest you make sure your firmware is up to date. You can use the Bandpass example to look at the data in the frame by over-typing the x scale values on the Time graph.
Bandpass will automatically try and get 10 cycles - so the 100 Hz minimum frequency yields a capture time of 100ms. By setting the number of samples to 10,000 I got a resolution of 0.1/10,000 = 10 us, which is the same as yours. This will yield only 5 samples per waveform at 20 kHz.

I'll update after having modified bandpass response to show phase.

Hello, Yes there is a bug in earlier versions of the driver which meant that if you are using the multi-frame capture method (as Bandpass Response does), then the last sample is spurious, and accumulates. So after 200 frames you will have 200 incorrect samples at the end of the frame. This has been fixed in driver 2.1. We also suggest you make sure your firmware is up to date. You can use the Bandpass example to look at the data in the frame by over-typing the x scale values on the Time graph. Bandpass will automatically try and get 10 cycles - so the 100 Hz minimum frequency yields a capture time of 100ms. By setting the number of samples to 10,000 I got a resolution of 0.1/10,000 = 10 us, which is the same as yours. This will yield only 5 samples per waveform at 20 kHz. I'll update after having modified bandpass response to show phase.

Hi Bart,
My driver, according to Windows Vista Computer Manager, seems to be version 1.6.6, so I guess I should update. On the back of my Cleverscope it says serial number IF3472 . Your 'Cleverscope Update Instructions.pdf' says that if my serial number is less than 4000 (what happened to the IF?) then I should use 'cscope_fw53xx.csf'. I have downloaded 'Cscope Update 4649' and extracted it, but can see no reference to cscope_fw53xx. When I run setup.exe I see 4 checkboxes, only one of which is unchecked, viz., 'Install Classic Cleverscope CS328 Driver' (Legacy?). I would have thought that this would be the one to install. I'm confused. Can you give me clear and detailed instructions, please.

Thanks.

By the way. I have to disagree with you about the 10 us. According to the output file I had Cleverscope generate, the selected sampling period (DT) was 0.015 or 15 us.

Hi Bart, My driver, according to Windows Vista Computer Manager, seems to be version 1.6.6, so I guess I should update. On the back of my Cleverscope it says serial number IF3472 . Your 'Cleverscope Update Instructions.pdf' says that if my serial number is less than 4000 (what happened to the IF?) then I should use 'cscope_fw53xx.csf'. I have downloaded 'Cscope Update 4649' and extracted it, but can see no reference to cscope_fw53xx. When I run setup.exe I see 4 checkboxes, only one of which is unchecked, viz., 'Install Classic Cleverscope CS328 Driver' (Legacy?). I would have thought that this would be the one to install. I'm confused. Can you give me clear and detailed instructions, please. Thanks. By the way. I have to disagree with you about the 10 us. According to the output file I had Cleverscope generate, the selected sampling period (DT) was 0.015 or 15 us.

Hello Harry,
We have some confusion reigning! We will try and make the update instructions clearer. However not reading the update instruction is no excuse :-)
We have four bundles of firmare/software/logicware to worry about here.
1. We have the Cleverscope application itself. The current version is 4649. When you run update setup.exe, the update manager checks the registry to see what has already been installed. If the application in your PC is older than in the update, we check the Update Cleverscope Appliocation option for you.
2. Second we have the USB driver. We assume when setup.exe is run that the Cleverscope is NOT connected (because we may need to update drivers, without them being locked). So we don't know what is connected. This means you have to know what type of cleverscope you have. Hence the serial number question. For the classic CS328 we use a FTDI driver (USB1.1 12 Mbit/sec). For the newer CS328A (serial number >4000, USB 2.0 480 Mbit/sec) we use the National Instruments Visa driver. The registry is used to tell us if these devices have already been installed, and the check box are set appropriately. For the CS328A, if you had a really old version, we may also install VISA for you. The USB driver is usually only installed once.
3. Next we have the Firmware/Logicware inside the Cleverscope acquisition unit. This is NOT loaded by setup.exe (there is no cleverscope attached, remember). Instead you must update the firmware, once you have a working system, by using ROM loader. This is explained on page 5 of the update instruction (but I guess no one reads this!). We have three version of firmware out there, and it is really important you load the right one. ROM Loader will help, by only allowing you to load the right one (unless you over-ride it, something we need to stop). Please use the version of ROM loader that came in the update, because then you will be using the latest. We currently have:
a. The CS328A USB firmware -'cscope_fw6445.csf' which resides in executablesUSB Firmware.
b. The CS328 (classic) firmware - 'cscope_fw5336.csf' which is also in executablesUSB Firmware
c. The CS328A Ethernet firmware - 'cscope_lan 6445.bin' which resides in ExecutablesEthernet Firmware.
Please do NOT load the wrong firmware into your Cleverscope. If you have an Ethernet cleverscope do NOT load .csf files - use the ROM loader that is in the Ethernet Firmware directory.
We will be unifying the Ethernet and USB firmware at some stage, but until then you have to be careful.
4. Finally, and this is the one that applies to you, if you are using the Labview driver vi, or the Labwindows, C++, C#, Delphi or Basic DLL, then there is a version to this as well. The current version, reported by the Function call 'version' returns 2.1, and are marked as 2.1 in our Resources page. If you are using these environments - and please confirm that you are using the Labview 'Bandpass response.vi' as the basis of your testing, using driver version 2.1 will put right the problem you were having. There is no need to update the firmware version, the USB version or the Application version, unless yours are really old (and given the 2006 date, they probably are).

And now it is my turn to be confused. You say the Cleverscope output file gives 15us. Are you using the Cleverscope application to do bandpass response, or the example 'Bandpass response.vi' which uses fast sig gen stepping inside the acquisition unit to do the response?

Does this all makes sense?

Hello Harry, We have some confusion reigning! We will try and make the update instructions clearer. However not reading the update instruction is no excuse :-) We have four bundles of firmare/software/logicware to worry about here. 1. We have the Cleverscope application itself. The current version is 4649. When you run update setup.exe, the update manager checks the registry to see what has already been installed. If the application in your PC is older than in the update, we check the Update Cleverscope Appliocation option for you. 2. Second we have the USB driver. We assume when setup.exe is run that the Cleverscope is NOT connected (because we may need to update drivers, without them being locked). So we don't know what is connected. This means you have to know what type of cleverscope you have. Hence the serial number question. For the classic CS328 we use a FTDI driver (USB1.1 12 Mbit/sec). For the newer CS328A (serial number >4000, USB 2.0 480 Mbit/sec) we use the National Instruments Visa driver. The registry is used to tell us if these devices have already been installed, and the check box are set appropriately. For the CS328A, if you had a really old version, we may also install VISA for you. The USB driver is usually only installed once. 3. Next we have the Firmware/Logicware inside the Cleverscope acquisition unit. This is NOT loaded by setup.exe (there is no cleverscope attached, remember). Instead you must update the firmware, once you have a working system, by using ROM loader. This is explained on page 5 of the update instruction (but I guess no one reads this!). We have three version of firmware out there, and it is really important you load the right one. ROM Loader will help, by only allowing you to load the right one (unless you over-ride it, something we need to stop). Please use the version of ROM loader that came in the update, because then you will be using the latest. We currently have: a. The CS328A USB firmware -'cscope_fw6445.csf' which resides in executablesUSB Firmware. b. The CS328 (classic) firmware - 'cscope_fw5336.csf' which is also in executablesUSB Firmware c. The CS328A Ethernet firmware - 'cscope_lan 6445.bin' which resides in ExecutablesEthernet Firmware. Please do NOT load the wrong firmware into your Cleverscope. If you have an Ethernet cleverscope do NOT load .csf files - use the ROM loader that is in the Ethernet Firmware directory. We will be unifying the Ethernet and USB firmware at some stage, but until then you have to be careful. 4. Finally, and this is the one that applies to you, if you are using the Labview driver vi, or the Labwindows, C++, C#, Delphi or Basic DLL, then there is a version to this as well. The current version, reported by the Function call 'version' returns 2.1, and are marked as 2.1 in our Resources page. If you are using these environments - and please confirm that you are using the Labview 'Bandpass response.vi' as the basis of your testing, using driver version 2.1 will put right the problem you were having. There is no need to update the firmware version, the USB version or the Application version, unless yours are really old (and given the 2006 date, they probably are). And now it is my turn to be confused. You say the Cleverscope output file gives 15us. Are you using the Cleverscope application to do bandpass response, or the example 'Bandpass response.vi' which uses fast sig gen stepping inside the acquisition unit to do the response? Does this all makes sense?

Hi Bart,
Some of the fog is starting to lift. But first, let me revisit the serial number issue. As I mentioned in my previous post, the serial number on my CAU reads 'IF3472'. In your post, and in the pdf files I've seen, the serial number is said to be 'xxxx', never 'IFxxxx'. Has serial numbering changed so that the 'IF' part is not included anymore or is it just taken for granted?
After reading your post I went ahead and upgraded both the computer software (4649) and the CAU firmware (cscope_fw5336.csf). Everything must have worked fine because the computer still talks to the CAU and vice-versa. Also my application, which by-the-way is based on 'Bandpass response.vi', still executes. It still reports a sampling period of 0.015 and the phase still flips by 360 degrees, but I haven't got around to sorting out this 'version 2.1' part yet. You said that a call to 'version' (presumably a C++, C#, Delphi or Basic routine) returns the driver version number, but I'm still looking for a vi that does this in LabView (it's all still quite new to me).
Regards, Harry.

Hi Bart, Some of the fog is starting to lift. But first, let me revisit the serial number issue. As I mentioned in my previous post, the serial number on my CAU reads 'IF3472'. In your post, and in the pdf files I've seen, the serial number is said to be 'xxxx', never 'IFxxxx'. Has serial numbering changed so that the 'IF' part is not included anymore or is it just taken for granted? After reading your post I went ahead and upgraded both the computer software (4649) and the CAU firmware (cscope_fw5336.csf). Everything must have worked fine because the computer still talks to the CAU and vice-versa. Also my application, which by-the-way is based on 'Bandpass response.vi', still executes. It still reports a sampling period of 0.015 and the phase still flips by 360 degrees, but I haven't got around to sorting out this 'version 2.1' part yet. You said that a call to 'version' (presumably a C++, C#, Delphi or Basic routine) returns the driver version number, but I'm still looking for a vi that does this in LabView (it's all still quite new to me). Regards, Harry.

Hello Harry,
We are just referring to the numeric part. The two letters are still included in the full serial number, but for the sake of knowing which version you are using XYNNNN has NNNN as less than 4000 for the classic and NNNN greater or equal to 4000 for the A version. We still use the XY part - it actually corresponds to the Month - Year of manufacture.

If you run the 'Simplescope.exe' application included in the driver (the source is there for it as well in Cscope control driver.llb as a top level vi ), You will see a Function Command down the bottom right. Set the Function Command to Get Driver Ver and hit Do Function. This should return the value 21.

After updating the firmware from something which was presumabely quite old, I suggest you do a full calibrate. Please follow the last page in the Update Instructions.

Now if you look at the Single Tone Information.vi' you will see there is an input to set the phase to flip at 180 or 360. You might like to play with that.

Please let me know how you go!

regards, Bart

Hello Harry, We are just referring to the numeric part. The two letters are still included in the full serial number, but for the sake of knowing which version you are using XYNNNN has NNNN as less than 4000 for the classic and NNNN greater or equal to 4000 for the A version. We still use the XY part - it actually corresponds to the Month - Year of manufacture. If you run the 'Simplescope.exe' application included in the driver (the source is there for it as well in Cscope control driver.llb as a top level vi ), You will see a Function Command down the bottom right. Set the Function Command to Get Driver Ver and hit Do Function. This should return the value 21. After updating the firmware from something which was presumabely quite old, I suggest you do a full calibrate. Please follow the last page in the Update Instructions. Now if you look at the Single Tone Information.vi' you will see there is an input to set the phase to flip at 180 or 360. You might like to play with that. Please let me know how you go! regards, Bart

Hi Bart,
Thanks for that. Sorry it has taken me so long to reply, but something else has come up and I'll have to put this on the back burner for a while. I'll get back to you.
Regards, Harry.

Hi Bart, Thanks for that. Sorry it has taken me so long to reply, but something else has come up and I'll have to put this on the back burner for a while. I'll get back to you. Regards, Harry.