8:00-5:30

• Prepping for LDAT demo – went pretty well
• Working on publish GUI – done with the first version. Need to make the text field associated with editing code a little more friendly (tabs should insert 4 spaces, etc)

7:30 – 3:00

• Well, I come in and see that the script runner seems to be running just fine, but the scripts aren’t executing. Upon inspection it turns out that runPeriodicScripts isn’t executing the scripts. Need to add that in. Done. Everything seems to be working. Going to let things cook while I’m at meetings
• Need to be able to delete scripts in the GUI
• Need to be able to delete outputs in the GUI
• Discovered what my time difference problem was last night. The date object were reading a 24 hour dateTime as a 12 hour dateTime without and AM/PM. This meant that all the dates were AM, which made things really odd in the afternoon.
• Had a good demo with Anne
• Created a chart for the LDAT folks for tomorrow’s meeting
• Creating a PublishDataObject for exchange between the client and server. Will use this for the basis of deleting published data.

• Shielded or shortened all signal wires
• Shield needed to be grounded to motor controllers ground.
• Seems much more predictable now.  The wires leading to strain gauges are still a bit too long and picking up interference but its hard to make them any shorter.  Remaining interference will have to be filtered out.
• Strain gauges were going haywire all day not sure if its because the wires were loose.  It does seem that the Wheatstone bridge can only be hooked up one way.  For some reason it seems if you change which SG goes to signal negative or positive it unbalances the bridge.  Does not really make sense considering there really is no difference but I ran into this symptom many times.
• Now that all the hardware problems should be solved or as good as they will get at the moment i can finally get back to software side.

• Attempted to make motor stop before stalling by setting speed to 0 whenever force reaches a certain threshhold.  This doesn’t work very well.  I think its because it is still trying to get to the position as it.
• Changing the way the motion is set up or using a get current position may fix the problems I was having.
• Reduced length of some wires that are hard to shield.  Noticed some improvement.
• Strain gauges are a bit more sensitive with the shielded wires had to turn down the gain.

8:00 – 4:30

• Discussions with John W and Mike C about the new dataset and hours until the end of the FY
• Discussions about integrating with LDAT. Michelle (?) is becoming quite our champion. I demo’s to several people and it looks like we should be able to tie directly into LDAT data
• More autorunning scripts. Could be done! Testing over night.
• 10:45 demo at Anne’s tomorrow.

7:30 – 3:30

• JW is back from vacation
• Mike C is setting up a new ingestor. Hopefully, he’ll be able to run it himself
• Demo today at noon with S14. Will add notes tomorrow
• 4:00 iPhone appointment
• Working on autorunning of scripts

Apparently I can only connect to this log from within the network.  I tried from home a few times and kept getting connection reset.

Thursday:

Steps for preparing strain gauges:

• Warm resin to room temp
• Sand surface that will have SG attached
• Clean with acetone until the Q-tip doesn’t get black anymore
• apply conditioner
• neutralize
• Apply resin and wait for it to dry
• Apply second layer of resin, attach strain gauge and allow to dry
• Test resistance with tester
• Cover with polyurethane
• Protect connections with silicone

Cleaned up code a bit from when I changed it

Friday:

• Second SG stick was ready to use.  Noted a lot of problems with calibration and consistency.
• Phil recommended averaging samples to clean up the signal inconsistency
• Showed it to Clift and he said that the signal wires needed to be shielded because the magnetic field from the motors were causing interference.

Saturday:

• Attached shielded cable to the SGs and Wheatstone bridge.
• This helped a lot with holding the calibration and consistency.
• There are signal wires going from the wheatstone bridge to the amplifiers which are still not shield which may help with lowering the inconsistencies when not in the dead zone as well as it still loses calibration sometimes.
• Averaging the samples is probably still a good idea even if it seems to be working perfectly(which it isn’t now) without it.
• Started reading OpenGL book but haven’t tried any tutorials yet.

7:30 – 4:30

• The strain gauge isn’t broken this morning – let’s see how it works!
• Need to have a ScriptRunnerServlet that actually runs scripts. The current script needs to be moved to PublisherServlet
• Need to add the mechanism for setting up periodic ingestion. Going to use a calendar and time widget to set initial time and then frequency thereafter. Most of the scheduler is working. Spent some time grocking RegExValidator.

8:00 – 4:00

• The sex change went well. I am now known as Phillese :-b
• More scripting/servlet integration. Working!
• Checked out space at Savage Mill
• Built a spare sensor rod (1/2 hr)

• Fixed code so there is only one executive.
• Made inputController the base class for anything that takes strain gauge input and turns it into motor movement.
• IFControl(input force) is a child inputController that turns force into movement speed.
• KFControl(kenetic force) is also a chile of inputController which turns force into displacement positioning.
• Should be set to go once the strain gauge is ready minus a couple tweaks.

• Fixing RCS program so that it only has one executive and many controllers  instead of many executives and few controllers.( almost complete)
• Strain gauge on left side of output motor is messed up.  It was reading 2k instead of 1.2k. I was able to fix it but then when I moved it to collide with the ball again it went off again and was not fixable.
• Probably should route the wire somewhat differently so it doesn’t get pulled on when the rod collides with the ball
• Took a while to figure out why the strain gauge wasn’t working.  I guess the moral is if the bridge is not balanced check the resistance of the strain gauges before going crazy with other things.
• Need to make a second input class to handle the output motors strain gauge events.   Instead of translating force into motion, it will have to translate force into offset from the central position.

7:30 – 5:00

• Adjusted to my moved desk and then was informed that I may be moving back
• Getting servlets to work – roughed out
• Changing the publishing mechanism so that output is published to the DB. The servlet then accesses the db using the path information to determine which output to show

How to setup the web.xml file to allow Tomcat to  access a servlet that your browser can see:

<web-app>
<display-name>Archetype Created Web Application</display-name>

<!- declare –>
<servlet>
<servlet-name>MyServlet</servlet-name>
<servlet-class>com.fgm.servletSandbox.MyServlet</servlet-class>
</servlet>

<!- map –>
<servlet-mapping>
<servlet-name>MyServlet</servlet-name>
<url-pattern>/MyServlet</url-pattern>
</servlet-mapping>

</web-app

7:30 – 5:00

• Worked at getting an LDAT account. Couldn’t find a location with account info, so sent MG a note asking for some help.
• Setting up forms demo on Monday
• Big VISIBILITY demo this afternoon
• Working on integrating autoscripting execution with the rest of the scripting functionality. Ran into a problem – the auto run server is different from the remote object server, which screws everything up. Going to have to move over to a servlet architecture

• Made it so movement was based on changing the max speed by a percentage determined by the offset when the bar is pushed or pulled.
• There are some problems with the offset not being consistent.  Going to try to lower the resistance of the strain gauges to see if that helps if not we will probably need to use those other amplifiers.

7:30 – 5:00

• Sent off Dong’s questions to Jim and his other portfolio manager
• Noticed that the LDAT folks had attempted to ingest something, but it didn’t ingest right. Waiting for a ping on that
• Going to work on getting periodic ingestion working – Got the servlet working and the accessor method. Spent freaking forever dealing with odd Date() behavior.
• Morning interview
• Noon performance review process overview – need a charge number