Rigol DS1000z/2000/4000/6000 Waveform Capture Utility

• Rigol DS1000z/2000/4000/6000 Waveform Capture Utility
  Posted by AlessandroAU on 18 Dec, 2014 11:48
• I posted this Utility in the Rigol DS1000/DS2000 jitter thread and since it seemed to attract some interest I figured I'd start a new topic.
  
  Its written in Labview so you will need the Labview RTE http://www.ni.com/download/labview-run-time-engine-2014/4887/en/
  
  Looking at the VISA commands this should also be compatible be somewhat compatible with the entire DS series from Rigol. I would expect the Real-time Capture to have no issues but there may be problems with the Long Memory Capture on other models scopes (I am using a DS1074z). If you have a DS2000/4000/6000 please test this and let me know.
  
  
  Current Version V2.2.3
  
  
  Rigol V2.2.3
  
  http://s000.tinyupload.com/index.php?file_id=90659414641703074947
  
  Rigol V2.2.1:
  
  http://s000.tinyupload.com/index.php?file_id=06194785972356290806
  
  Rigol V2.2
  
  Download Link: http://s000.tinyupload.com/index.php?file_id=16159701921368811937
  
  Source: http://s000.tinyupload.com/index.php?file_id=05957966868497599811
  
  Known Bugs:
  
  Long Memory Capture only supports one channel at a time and trigger channel must be the same as capture channel. For now it is a good idea to disable other channels when using this.
  
  Memory Depth setting is only updated when a new menu selection is made.
  
  In real-time capture mode changing channel options sometimes results in a communication error, stop cont. sampling before changing memory settings to avoid this.
  
  Waveform statistics cannot compute more than 3Mpoints of data (labview limitation) currently statistics are calculated from the first 50Kpoints to improve speed.
  
  Rigol DS1074z Bugs:
  
  Scope complains of 'Invalid Input' but processes commands anyway. (Perhaps I do not known the exact sequence it expects) commands work regardless.
  
  When using :WAV:DATA? to request samples there is an upper limit of exactly 1179644 data points that can be downloaded at any one time requesting more points using :WAV:START and :WAV:STOP will result in a no data being returned at all.
  
  Note this is when downloading the data as bytes ie :WAV:FORM BYTE, if using :WAV:FORM ASC (ascii) the limit is reduced to around 60Kpoints.
  
  However it is possible to download the entire waveform memory in blocks and this is the technique I have used to download more than 1.1Mpoints.

  ---

  

  

  

  
• #1 Reply
  Posted by 0ff on 18 Dec, 2014 21:08
• Dear AlessandroAU,
  
  your Utility seems to be quite cool!
  Unfortunately, being on a mac I can't run your tool (as it's compiled for windows).
  
  Would you be able to compile this as a mac app? (I don't know anything about labview, maybe you can just select this as output?)
  
  Thanks in advance!
• #2 Reply
  Posted by pa3bca on 18 Dec, 2014 21:42
• Hi AllesandroAU,
  Unfortunately something broke between your original version and the latest (2.2). I cannot reproduce my original nice FFT.
  I compared both versions with identical scope settings. (same as the one you re-published)
  If I do a long mem capture (1.2 Mpts) the capture(buffer) in 2.2 is corrupt (see screenshot below).
  Also the FFT is obviously incorrect (even when based on the capture buffer). See the other screenshots.
  Also I see an almost permanent message on the scope screen: Parameter Limited! which indicates an input error.
  (also did a long memory capture with 12 Mpts, with scope also on 12 Mpts settings, see below. no difference).
  
  Do you need any more info from me to troubleshoot?
  
  

  ---

  

  

  

  
• #3 Reply
  Posted by AlessandroAU on 18 Dec, 2014 22:00
• Quote from: pa3bca on 18 Dec, 2014 21:42

  Hi AllesandroAU,
  Unfortunately something broke between your original version and the latest (2.2). I cannot reproduce my original nice FFT.
  I compared both versions with identical scope settings. (same as the one you re-published)
  If I do a long mem capture (1.2 Mpts) the capture(buffer) in 2.2 is corrupt (see screenshot below).
  Also the FFT is obviously incorrect (even when based on the capture buffer). See the other screenshots.
  Also I see an almost permanent message on the scope screen: Parameter Limited! which indicates an input error.
  (also did a long memory capture with 12 Mpts, with scope also on 12 Mpts settings, see below. no difference).
  
  Do you need any more info from me to troubleshoot?
  

  
  
  Very odd, it works fine on my unit.
  
  Can you confirm the following happens:
  
  1. Make sure only one channel is enabled and that the trigger channel is the same channel as which you are capturing from.
  
  2. When changing the memory depth the number the change is reflected under the acquire menu on the rigol screen.
  
  3. Does the run/stop button turn red during the capture and back to green afterwards
  
  5. Does channel selection on the real-time screen work for you?
  
  4. Does realtime and screen shot capture work for you?
  
  Also do you have 12mpoints or 24mpoints option on your scope?
  
  
• #4 Reply
  Posted by AlessandroAU on 18 Dec, 2014 22:09
• Quote from: 0ff on 18 Dec, 2014 21:08

  Dear AlessandroAU,
  
  your Utility seems to be quite cool!
  Unfortunately, being on a mac I can't run your tool (as it's compiled for windows).
  
  Would you be able to compile this as a mac app? (I don't know anything about labview, maybe you can just select this as output?)
  
  Thanks in advance!
  

  
  Unfortunately I cannot compile it into a max executable as I don't have access to the OSX Labview Environment.
  
  Possibly it may work in VMware?
• #5 Reply
  Posted by pa3bca on 18 Dec, 2014 22:23
• Quote from: AlessandroAU on 18 Dec, 2014 22:00

  
  Very odd, it works fine on my unit.
  
  Can you confirm the following happens:
  
  1. Make sure only one channel is enabled and that the trigger channel is the same channel as which you are capturing from.
  
  2. When changing the memory depth the number the change is reflected under the acquire menu on the rigol screen.
  
  3. Does the run/stop button turn red during the capture and back to green afterwards
  
  5. Does channel selection on the real-time screen work for you?
  
  4. Does realtime and screen shot capture work for you?
  
  Also do you have 12mpoints or 24mpoints option on your scope?
  

  1: One channel enabled, trigger on this channel
  2:Yes, but only the first few times. After some buffer capturing all goes south, and nothing works any more (must restart scope and application te be able to capture again)
  3:Yes
  4:Realtime: Yes. Screenshot: No. Screen stays black. And after I try to get a screenshot I need to restart the application in order to be able to get a capture... (and realtime screen also fails).
  After trying a few time I suddenly got a better result. See screenshot. Although even here there are serious glitches in the captured waveform. The FFT here however looks ok.
  But right after that a next capture failed again.
  
  Yes I do have 12 and 24 Mpts available. But same results here

  ---

  

  
• #6 Reply
  Posted by AlessandroAU on 18 Dec, 2014 22:27
• Quote from: pa3bca on 18 Dec, 2014 22:23

  Quote from: AlessandroAU on 18 Dec, 2014 22:00

  
  Very odd, it works fine on my unit.
  
  Can you confirm the following happens:
  
  1. Make sure only one channel is enabled and that the trigger channel is the same channel as which you are capturing from.
  
  2. When changing the memory depth the number the change is reflected under the acquire menu on the rigol screen.
  
  3. Does the run/stop button turn red during the capture and back to green afterwards
  
  5. Does channel selection on the real-time screen work for you?
  
  4. Does realtime and screen shot capture work for you?
  
  Also do you have 12mpoints or 24mpoints option on your scope?
  

  1: One channel enabled, trigger on this channel
  2:Yes, but only the first few times. After some buffer capturing all goes south, and nothing works any more (must restart scope and application te be able to capture again)
  3:Yes
  4:Realtime: Yes. Screenshot: No. Screen stays black. And after I try to get a screenshot I need to restart the application in order to be able to get a capture... (and realtime screen also fails).
  After trying a few time I suddenly got a better result. See screenshot. Although even here there are serious glitches in the captured waveform. The FFT here however looks ok.
  But right after that a next capture failed again.
  
  Yes I do have 12 and 24 Mpts available. But same results here
  

  
  What firmware are you running?
• #7 Reply
  Posted by pa3bca on 18 Dec, 2014 22:30
• Quote from: AlessandroAU on 18 Dec, 2014 22:27

  What firmware are you running?
  

  Ooh old.. Software Version is 00.02.03.SP5
  (nb: your v1 app runs perfectly every time)
  
• #8 Reply
  Posted by AlessandroAU on 18 Dec, 2014 23:21
• Quote from: pa3bca on 18 Dec, 2014 22:30

  Quote from: AlessandroAU on 18 Dec, 2014 22:27

  What firmware are you running?
  

  Ooh old.. Software Version is 00.02.03.SP5
  (nb: your v1 app runs perfectly every time)
  

  
  I'm running 4.00 so rigol may had made some changes to their firmware since then.
  
  Can you try this Version?:
  
  Rigol V2.2.1:
  
  http://s000.tinyupload.com/index.php?file_id=06194785972356290806
• #9 Reply
  Posted by JDubU on 19 Dec, 2014 01:34
• Mostly works on my DS2072 (HW v2.0, FW v00.03.01).
  
  The only obvious problem that I noted on a quick test using the scope's 1KHz probe calibration square wave was that, in the Long Memory Capture, there is a large impulse-like discontinuity between each sample chunk.  It looks like the DS2000 scope wants total number of samples to be a 10x multiple of 1.4  instead of the DS1000's 10x multiple of 1.2.
  
  

  ---

  

  
• #10 Reply
  Posted by AlessandroAU on 19 Dec, 2014 02:58
• Quote from: JDubU on 19 Dec, 2014 01:34

  Mostly works on my DS2072 (HW v2.0, FW v00.03.01).
  
  The only obvious problem that I noted on a quick test using the scope's 1KHz probe calibration square wave was that, in the Long Memory Capture, there is a large impulse-like discontinuity between each sample chunk.  It looks like the DS2000 scope wants total number of samples to be a 10x multiple of 1.4  instead of the DS1000's 10x multiple of 1.2.
  

  
  Can you through the ultrasigma control center send the following: WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  
  and paste me the result?
• #11 Reply
  Posted by JDubU on 19 Dec, 2014 04:37
• Quote from: AlessandroAU on 19 Dec, 2014 02:58

  Can you through the ultrasigma control center send the following: WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  
  and paste me the result?
  

  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ†€ƒ~~‚}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  Edit:
  The result changes when command is repeated:
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ…€ƒ~‚}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100Ž‚Ž‚…€ƒ~~‚~}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ†€ƒ~‚~}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ…€ƒ~~‚}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ†ƒ~‚~}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ…€ƒ~~‚~}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100ŽƒŽƒ…ƒ~~‚}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100Ž‚Ž‚†€ƒ~‚~~}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  -> :WAV:START 1;:WAV:STOP 100;:WAV:DATA?
  <- (Return Count:112)
  #9000000100Ž‚Ž‚…€ƒ~~‚~}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}}
  
  
• #12 Reply
  Posted by pa3bca on 19 Dec, 2014 18:14
• Quote from: AlessandroAU on 18 Dec, 2014 23:21

  I'm running 4.00 so rigol may had made some changes to their firmware since then.
  
  Can you try this Version?:
  
  Rigol V2.2.1:
  
  http://s000.tinyupload.com/index.php?file_id=06194785972356290806
  

  Same issues. The first time I power up your program it works ok (1.2 Mpts). Also on 12 Mpts.  There are a few glitches in the captured buffer but probably not enough to upset the FFT calculation.
  But if I then try 24 Mpts I only get a flat line in the buffer window, and nothing in the FFT window. And then the 12 and 1.2 Mpts captures produce the same errors as can be seen on my previous screenshots (hickups in the buffer window, and FFT way off).
  Screencapture still does not work. And after I try a screencapture the realtime display acts funny. Have to enable/diable it a few times for it to start again.
  
  Also the scope's screen is flashing errors like crazy (all versions of them)
  I have to restart both the scope and the program to get them to play nice again (but then only at 1.2 and 12 Mpts)
  
• #13 Reply
  Posted by nixfu on 21 Dec, 2014 07:09
• I am just curious if you guys thank that this could replace the need to purchase an actual SA?
  
  I have a DS1074z, and I need to maybe check some filter responses, and maybe look at RFI or such stuff, or even look for harmonics on a transmitter.
  
  I have almost every piece of gear I ever have told myself I needed, except for an SA because they are soooo expensive compared to most other gear.
  
  So, for a hobbyist can this replace the need for a SA for mostly simple stuff when combined with a good signal generator, and a DS10xxZ scope?
• #14 Reply
  Posted by pa3bca on 21 Dec, 2014 13:00
• Quote from: nixfu on 21 Dec, 2014 07:09

  I am just curious if you guys thank that this could replace the need to purchase an actual SA?
  
  I have a DS1074z, and I need to maybe check some filter responses, and maybe look at RFI or such stuff, or even look for harmonics on a transmitter.
  
  I have almost every piece of gear I ever have told myself I needed, except for an SA because they are soooo expensive compared to most other gear.
  
  So, for a hobbyist can this replace the need for a SA for mostly simple stuff when combined with a good signal generator, and a DS10xxZ scope?
  

  Up to a point: yes. If you look at my comparison between the DSA815 and the FFT analysis you can see that up to say 3 MHz and -50dBm they are almost exactly the same. But further down in amplitude they begin to differ somewhat.
  So i think that is useable as a SA substitute to possibly a few tens of MHz (remember the scope's BW is only 100 MHz and de sampling rate "only"  1GSa/s). And signals below say -50 dBm are also under suspicion. The scope's AD converters are only 8 bits, and by massively oversampling (say 12 to 25 MPts) you can up the dynamic range to possibly a practical 12 bits (I'm guessing here but the -70 dBm noisefloor of the FFT's seem to suggest this).
  A real SA goes to at least 1.5 GHz (as my DSA815) and has a much larger dynamic range. the DANL is -130 dBm for the 815....
• #15 Reply
  Posted by AlessandroAU on 21 Dec, 2014 14:13
• Quote from: pa3bca on 21 Dec, 2014 13:00

  Quote from: nixfu on 21 Dec, 2014 07:09

  I am just curious if you guys thank that this could replace the need to purchase an actual SA?
  
  I have a DS1074z, and I need to maybe check some filter responses, and maybe look at RFI or such stuff, or even look for harmonics on a transmitter.
  
  I have almost every piece of gear I ever have told myself I needed, except for an SA because they are soooo expensive compared to most other gear.
  
  So, for a hobbyist can this replace the need for a SA for mostly simple stuff when combined with a good signal generator, and a DS10xxZ scope?
  

  Up to a point: yes. If you look at my comparison between the DSA815 and the FFT analysis you can see that up to say 3 MHz and -50dBm they are almost exactly the same. But further down in amplitude they begin to differ somewhat.
  So i think that is useable as a SA substitute to possibly a few tens of MHz (remember the scope's BW is only 100 MHz and de sampling rate "only"  1GSa/s). And signals below say -50 dBm are also under suspicion. The scope's AD converters are only 8 bits, and by massively oversampling (say 12 to 25 MPts) you can up the dynamic range to possibly a practical 12 bits (I'm guessing here but the -70 dBm noisefloor of the FFT's seem to suggest this).
  A real SA goes to at least 1.5 GHz (as my DSA815) and has a much larger dynamic range. the DANL is -130 dBm for the 815....
  

  
  
  You must consider fft processing gain which is equal to 10log (m/2) where m is the number of samples. So the noise floor of the fft can actually reach that low (-130dbm)!
  
  Edit: See Image
  
  

  ---

  
• #16 Reply
  Posted by pa3bca on 22 Dec, 2014 15:43
• Quote from: AlessandroAU on 21 Dec, 2014 14:13

  You must consider fft processing gain which is equal to 10log (m/2) where m is the number of samples. So the noise floor of the fft can actually reach that low (-130dbm)!
  

  Yes theoretically. But -130 dBm starting from a cheap scope with an 8 bits ADC?.
  The scope itself is 8 bits, so almost 50 dB dynamic range. To get to 130 dB you would have to get 200 Mpts (80 dB = 10 log (200x10^6  /2). Perfectly doable.
  But... Using these formulas, with 24 MPts (see screenshots below) the FFT would add 10 log ((24 x 10^6)/2) = - 71 dB!
  If all were perfect I would expect the noisefloor to be at -50-71 = -120 dB. I only see -80 (also at 200 MHz and there I would not expect to see any "real" signal out of the scope). So there seems to be 40 dB missing somewhere. I am no expert on FFT (calculations) so I do not have the answer to this puzzle.
  
  Also if you look at the screenshots below there is a significant discrepancy after say 5 MHz. The FFT (with 24 MPts) shows peaks that are not there (according to the DSA815).
  So the FFT is certainly usable as a substitute for an SA, but up to a point (obviously in frequency, but also in amplitude. Although I do not know how to discern where the useability ends...)
  Rigol DSA815:
  
  24 Mpts FFT from DS1074Z:
  
  
  

  ---

  

  
• #17 Reply
  Posted by JDubU on 22 Dec, 2014 17:51
• Is that a new version of the capture utility that has been updated for DS2000 and DS4000?
  Is there a download link for it?
  
  
• #18 Reply
  Posted by AlessandroAU on 23 Dec, 2014 05:15
• Fair enough the dynamic range is not that high, but its definitely possible to have a noise floor at -130ish with 24mpoints (by zooming in). However I concede that the best dynamic range appears to be around 105dBm, which is still very good.
  
  Also, just to show that is possible to take measurements exceeding 8 bits limit; Look at image 3, the scope is set to 5v/div or 40v max range. A 10mv pk-p signal is applied.
  
  This is 20*log10(0.01/40) = -72db below full scale, a full 22db or so below the 8 bit resolution limit. Even though the waveform looks like barely like anything, thanks to the mathematics of the Fourier transform this signal is still measurable. Very neat!
  
  Edit:
  The tektronix MDO series also only uses an 8bit adc internally but has a DANL of -130dbm from 50khz to 5mhz (http://www.tek.com/datasheet/mdo4000/mdo4000b-series-datasheet-0) so it is definitely possible approach this kinda of performance with 8 bits.
  
  also see: https://www.google.com.au/url?sa=t&rct=j&q=&esrc=s&source=web&cd=22&ved=0CC4QFjABOBQ&url=http%3A%2F%2Fwww.tek.com%2Fdl%2F48W-28882-2%252520%252520%252520MDO%252520Performance%252520WP_.pdf&ei=p_2YVOHwJcLZ8gWIsoKABg&usg=AFQjCNHaLpRkqk9Fxl4CPInDNz58hWFfLQ&sig2=vefpuXdZKo0KQWZVw9PI1g&bvm=bv.82001339,d.dGc&cad=rja
  
  Just wanted to make people aware of this fact! I would trust the scope for use as an SA up to its bandwidth of 100MHz personally.
  

  ---

  

  

  
• #19 Reply
  Posted by AlessandroAU on 23 Dec, 2014 05:39
• Quote from: JDubU on 22 Dec, 2014 17:51

  Is that a new version of the capture utility that has been updated for DS2000 and DS4000?
  Is there a download link for it?
  

  
  I am still working on support for other scopes, currently it is quite hard as I only have access to a 1074z. So far my program only detects which series scope is selected. It does not yet load configuration changes for the higher model scopes. If you have a ds2000 I could send you a WIP to try if you would like.
• #20 Reply
  Posted by JDubU on 23 Dec, 2014 06:29
• Quote from: AlessandroAU on 23 Dec, 2014 05:39

  I am still working on support for other scopes, currently it is quite hard as I only have access to a 1074z. So far my program only detects which series scope is selected. It does not yet load configuration changes for the higher model scopes. If you have a ds2000 I could send you a WIP to try if you would like.
  

  
  I do have a DS2000 and would be happy to test a WIP.
  
• #21 Reply
  Posted by AlessandroAU on 24 Dec, 2014 03:28
• Quote from: JDubU on 23 Dec, 2014 06:29

  Quote from: AlessandroAU on 23 Dec, 2014 05:39

  I am still working on support for other scopes, currently it is quite hard as I only have access to a 1074z. So far my program only detects which series scope is selected. It does not yet load configuration changes for the higher model scopes. If you have a ds2000 I could send you a WIP to try if you would like.
  

  
  I do have a DS2000 and would be happy to test a WIP.
  

  
  
  Let me know if this version works for you:
  
  V2.2.3
  
  http://s000.tinyupload.com/?file_id=90659414641703074947
  
  pa3bca, Please also try that version and let me know if it fixes any of the issues you were having.
  
  
• #22 Reply
  Posted by aroby on 24 Dec, 2014 04:39
• I tried the prior version (2.2.1) with my DS2072A / DS2302A.  I had to manually enter TCPIP::10.10.10.45::INSTR in the drop down menu, but it was able to connect.  With 2.2.3, I do the same and just get Error in the Model Detected field.
  
  Anthony
• #23 Reply
  Posted by AlessandroAU on 24 Dec, 2014 04:40
• Quote from: aroby on 24 Dec, 2014 04:39

  I tried the prior version (2.2.1) with my DS2072A / DS2302A.  I had to manually enter TCPIP::10.10.10.45::INSTR in the drop down menu, but it was able to connect.  With 2.2.3, I do the same and just get Error in the Model Detected field.
  
  Anthony
  

  
  I haven't implemented TCP/IP, can you try with usb?
  
  Edit, Hang on, were you able to connect over TCP/IP in version 2.2.1 and download a waveform?
• #24 Reply
  Posted by aroby on 24 Dec, 2014 04:52
• Quote

  I haven't implemented TCP/IP, can you try with usb?
  
  Edit, Hang on, were you able to connect over TCP/IP in version 2.2.1 and download a waveform?
  

  
  Yes.  Just tried it again.  I can do a screen capture.  In the Realtime Capture tab, I can select channels, I can see a waveform (but it says "Input is invalid!" on the scope), if I change the scales on the scope, the graphs change.  On Long Memory Capture, changing the # of points gets reflected on the scope (albeit with the issues previously described).  When I hit Capture, the scope Stops, I get Input is invalid, and nothing gets displayed.