Today was a relatively short day.  The morning was cold, humid, and cloudy.  The sun came out at times in the afternoon.

  • Last night, I did a power budget on the pressure sites, which I should've done long ago.  We need more solar charging at all the pressure sites.  A single 145W panel at each site should be adequate.  We will likely be requesting that a total of six 145W panels be shipped from Boulder for this purpose.  We'll then swap those in and use the 65W panels to beef up the main towers.
  • We checked the sensors using Cockpit.  P5 was down due to power.
  • We cleaned all radiometers and EC150s, and checked cable tensions

    • Upper convergence tensions

      • see Guy tensions page
    • Lower convergence tensions
      • see Guy tensions page

    • Release tower tensions

      • North low 360

      • North mid 320

      • North hi 400

      • Southwest low 360

      • Southwest mid 290

      • Southwest hi 370

      • Southeast low 370

      • Southeast mid 300

      • Southeast hi 370

    • Init tower trensions

      • Northwest 300

      • Southwest 300

      • East 320

  • P5 came back online by 11:30AM, when we were done with sensor cleaning.
  • We added an extra (well used) battery to P5.  It now has two dying batteries (better than one, right?).  We plan to double up batteries at the rest of the pressure sites as soon as we receive the new batteries we ordered, and purchase three additional coolers from Target.  We'll add an extra battery back to init.b too.
  • April emailed and mentioned the next IOP is the evening of 10/15.  Additionally, U of Wisconson students will be arriving tomorrow to set up sensors.


  • No labels

3 Comments

  1. Unknown User (gilmer) AUTHOR

    Oct 13, 2018

    Pressure site batteries are dying and they don't have enough charging power.  To give the batteries enough juice for Monday night's IOP, I shut down their ubiquities for 9 hours.  They should come back around 8AM local tomorrow.  Then, I'll probably shut them down again.  The ubiquities are about 2/3 of the pressure site power consumption.

    In other news,

    • April emailed and #1 asked if we could take theodolite scans of the U-of-W students' sensors, and #2 if we she could get access to the R data.  I don't see a problem with #1, but I'm not sure how to respond to #2.  The R data hasn't been QA'd, and I'm hardly proficient in accessing it.  My current plan is to point her towards N-charts and say the data will be available later.  However, I'm open to suggestions.
    • Dan and I plan to take tomorrow off and check out Chicago.

  2. Gary Granger

    Oct 14, 2018

    We could setup a cronjob on the DSMs which turns the ubiquiti radios off and on periodically, like on for thirty minutes of every 3-6 hours.  I presume that would be easier than visiting all the sites, and the most it should do is delay the data by 3-6 hours.  Should I do that the next time the radios are on?

    Things to confirm:

    • Do the radios need to stay on all night in case there's an IOP?  I'm guessing real-time data from the pressure sites is not useful during IOPs.
    • The ubiquitis are plugged into the dcdc relay right?  So pio dcdc off turns the radio off?
    • What is the optimum schedule for the radios?  Is 1 hour on out of every 3 hours good enough?  Or should it be 30 minutes?  I think 30 minutes is a good minimum amount of time to be on for the rsync of the latest data to succeed, but an hour just gives a greater likelihood of success.  So then what should the fraction of on time be?  1/2, 1/3, 1/4, ...?


  3. Unknown User (gilmer) AUTHOR

    Oct 15, 2018

    Gary, 

    Switching the ubiquitis on for 30min every 3 hours is a great idea!  I can have the radios on by noon local if you want to try then..

    to answer your questions

    • I sent an email to April asking if the radios need to be on during an IOP.  I'm awaiting a response.  Is it possible to make this user selectable?  Three options that may be desired are
      • ON  (already available. > pio dcdc 1)
      • OFF (already available. > pio dcdc 0)
      • 3hrs on, 0.5hr off (some new script and command)
    • Yes
    • From viewing the power budget, 0.5hrs(on) / 3hrs(off) (ON fraction = 1/7) gives us a sufficiently low power that the stations will reliably stay on.  I can't speak to how long the radios need to be on to finish transmitting though.  Should we implement this and see how it works?