User:Steven J. Koch/Notebook/Kochlab/2009/02/19/Getting Feedback0B Working

From OpenWetWare
< User:Steven J. Koch‎ | Notebook‎ | Kochlab‎ | 2009‎ | 02‎ | 19
Jump to navigationJump to search

Note: This page was transferred from the private wiki, and thus links and images may be broken until they are fixed.

Steven J. Koch 22:50, 18 February 2009 (EST)Maybe about Spring of last year, Richard (and a bit me) worked on putting Feedback0B information on the public wiki. Richard got the software to compile (with dummy DLLs) in version 6.1, and made a whole distribution avaible via sourceforge, and also on our server somehow (via SVN). I definitely succeeded in getting this to run (on Kochlab-01 or -02) I think, which have been scheduled for the graveyard. I guess that's OK since these machines can't use full-size PCI cards. However, I need to get it working on some machine. So, I'm using Kochlab-03 now, which should work OK as far as I know. Linh (or someone) has already installed the old E-series DAQ card in this machine. I have the machine working and now need to re-learn how to use SVN and all that. First, I'll find the communications from Richard and Caleb last year.

Possibly related / relevant pages

Installing SVN client

Following my 2008-3-10 notebook from above, I'm trying to install SVN and get everything on Kochlab-01.

File:090218 SVN Feedback0B screenshot.JPG
this looks promising
  • I think it worked. To the right, you can see in a screenshot where I think the files are.
  • Steven J. Koch 23:29, 18 February 2009 (EST): It doesn't open, though. Looks like missing driver libraries. Some LabVIEW installation that needs to be done...
  • Removing NI-DAQ 7.3 and NI-DAQmx 8.7, then I'll reboot
  • OK rebooted (it found new hardware, presumably the e-series card and I canceled). Now installing NI-DAQ 6.9.something from the pub drive LabVIEW6.1 CDs.

Feedback0B-Main compiles (with fake DLLs)

  • re-installing old DAQ worked out.
  • It asked for a project configuration file, so I chose the INI file in \\controller\webpub\files\data\Koch_Data\Popping_VC_AvaI\ and it seemed to work! Very cool. It loaded the feedback sequence I used back then, and created a new data directory.
    • I don't remember how to create a new project, other than using the "wizard" that Richard wrote,...not sure if that would work.

Feedback0B seems to run--impressive

screenshot of feedback program running
  • Feedback0B runs with the fake DLLs. I can only get it to take a couple of data points, perhaps due to AOD hard limits or something, not really sure. But the main architecture of the program seems to be working. I had to make a foot pedal out of a couple of pieces of wire, but that's not an issue, I don't think.
  • Steven J. Koch 00:33, 19 February 2009 (EST): From here, I'm not sure where to go. Port to LV8? Get version6.1 working for our own hardware? Any of these are going to take a while, I think. Hmmm
    • I could try to replace things that require fake DLLs. Such as the AOD stuff? Even that might be major, requiring making new feedback modules.
    • A good thing is that we again have an analog-controlled 1-D piezo module, so that stuff should work. In principle, actually, I could replace the AOD modules with other modules for diode current, and the software might be immediately useful.

Feedback0B compiles in LV7.1

Steven J. Koch 00:47, 19 February 2009 (EST): I tried opening Feedback0B in LV7.1 and it compiles and runs. That's interesting, but don't know if it will help me. In order to get it to do this, I had to tell it where a bunch of the DAQ vis were, which isn't surprising.

  • I saved the main VI to a new directory in my LV7.1 development directory (on the local machine). I also copied over the vi.lib/DAQ directory from LV6.1 to LV7.1 so it can find those necessary VIs.
  • I am confused though as how to go forward, because it's not very useful to be in 7.1 if we're not using DAQmx
  • I probably need to also look at "Secret Peeking Software" and other data analysis software, so I can understand how to change things without breaking data anlaysis too much.