Died about 11pm last night.  I just restarted it with control-R, and it seems happy again.

We have no idea why this is happening – always at this site, three different TRHs tried.  I guess we'll throw together a DSM script to send control-R to it periodically.

 

  • No labels

3 Comments

  1. Gordon Maclean

    /home/daq/isfs/bin/check_trh.sh is a script that was used in CABL on the tower. It power cycles a TRH if the temp is outside a range (-50:50) or Ifan is < 10.

    It can be run from post_dsm.sh, and takes a port number as an argument.  

    It would require some significant changes to send a ctrl-R, '\x12', to the TRH instead of power cycling it. The stdout of rserial is sent to awk and not the other way around.

  2. Steve Oncley AUTHOR

    So...we took a trip to the marsh today, since we noticed erratic TRH behavior was happening:

    • verified that the fan wasn't running when current was reading 0 and that the fan speed was erratic when the current was reading less than about 30mA (in other words, the current monitor was working)
    • swapped to an emerald port (for some reason, probably jumpering, ports 9-12 weren't working, but port 7 was alive).  Fan speed was still erratic (in other words, this isn't a problem specific to a port or port type)
    • swapped cable with the spare cable from the boat.  Fan speed was still erratic (not a cable issue)
    • opened the TRH for inspection.  No obvious mechanical issues: blade spun when powered off and blown upon; blade current went up with finger pressure applied to the hub. (in other words, nothing was obviously wrong)
    • swapped with spare sensor.  Fan speed was steady, and still steady now that we are back in the trailer an hour later.  This would seem to indicate something wrong electronically, however the spare we used was the one that we had pulled from this site last week because it had similar issues (yet had run flawlessly in the trailer for several days straight)
    • back in the base, we fired up the one we just removed.  It started just fine, but the fan speed/current monotonically decreased to zero now, over a period of 15 minutes. (so something is actually wrong with this sensor)

    We'll have to see if TRH16 runs as well back on this station as it did in the trailer.

    Also, I have found that simply using: echo $'\x12' > /dev/ttyS2 does in fact reset the TRH.  I've put this in crontab every 4 hours.

  3. Steve Oncley AUTHOR

    Further base testing has isolated the failure to the fan monitor board (it still failed with swapping fans and PIC-SHT module and that fan and PIC module work on another housing).  At least we have it consistently failing in the base for testing.  It is kind of cute; mostly stopped, but gasping for breath by turning the fan on briefly every 10s to 5min.  We're going to see if we can wire up a spare fan monitor as a replacement.