- Overnight, we had the first instance of a pillow dropout (#1) since I implemented a daily restart. The restart did bring it back, but we lost data from 00-08 last night.
- Apparently, there was more snow overnight. The upward radiometers (both K&Z and NR01) act like they are covered. Also, several EC150s have died and come back.
- The 1m.uw sonic is still buried. We'll move it to 2m when we're there in 2 weeks.
- The TRHs are all back reporting normally, though there may be some small biases.
- Tsnow.0.9m.uw now appears to be buried
- Tsnow.0.6m.d is now buried as well
In addition to the csat.1m.uw sonic being buried:
- TRH.6m.c seems to have died this morning (no response to rs and pio doesn't help)
- ec150.1m.ue also died this morning (the raw signals are about 0.0, rather than 0.9, indicating a signal blockage)
Note that there was a bunch of new snow overnight, so I would hope that the ec150 will come back. I'm less sure about the TRH.
- bring out 2-person shelter
- bring out radiometer wipes
- bring out Panduit ties
- bring out more AA batteries (lithium?)
- clean radiometers
- check why Rsw.in is low on both NR01 and K&Z (presumably CVF4 is full of snow; maybe ask Dan and Eli to brush off NR01?)
- move csat.1m.uw to 2m (or 4m?)
- remove csat.1m.c? (since there already is a 2m.c – means sledding in CSAT and EC150 cases)
- check/move other 1m csats to 2m
- fix TRH.6m (it was unresponsive for most of 28 Dec, but came back – reseat cable?)
- test mobile Tsnows (might already be done by Danny and Eli)
- resecure Tsnow "base" mote (Chris was worried about cable tie breakage, so they are loose)
- log snow pillow electronics serial numbers (just because; also Alpine Hydromet wants to know which one we unsealed)
- debug TRH057 if time and/or bring back (would be done at RMBL using spare cables, DSM, and wstar. Goal is to determine whether TRH probe or housing or both are bad. Symptom was Rfan reporting erratic values and initially fan appeared not to be working – high T values during day. I bet it will appear to work fine now!)
- bring back spare soils
Of course, dsm_server died about noon on Xmas. I've restarted.
Also, pillows 4 and then 3 stopped today. Just restarted.
Interesting that this happened during a time of a lot of wind/blowing snow (as seen by the FlowCapt sensor). It may be that this sonic is actually buried by snow now. Chris reported that the snow cover was quite variable, with only 10cm near c, but over 2' at RMBL. He didn't work around uw yesterday and may have missed a sonic almost buried. The webcam seems to show that it is in the snow!
Antonio and Chris skied into Gothic (road is now closed) on Dec 20. We retrieved a TRH from the storage to swap out the 12m TRH.
On Dec 21, we go an early start and skied out to the Kettle Ponds site.
1) Antonio climbed to 12m and swapped the entire TRH (probe and housing). Old/bad (poor fan RPM control) probe was SN 057, The new probe was SN 003.
2) Chris opened the snow pillow #3 control box; plugged the firmware update cable directly into the snow pillow control PCB, and was able to update the firmware of snow pillow 3. Previous attempts use the connector on the outside of the control box, which did not allow the firmware update to occur (there must be an issue with the cable between the bulkhead and the PCB that will not allow updates).
3) We installed the Tsoil mote on the center tower just above the CL (center lower) DSM . We were able to push the USB connecter through the bulkhead opening and plugged the USB into the DSM serial PCB of the CL DSM.
4) We measured the voltage on both downwind lidar power supplies (This is the only tower which uses the AC power bricks for the lidars - the other two towers use 12 VDC power from the Victon to power the lidars). Ethan had indicated that he had seen some low voltage conditions on the lidars. Both power supplies read 12.2 V (while loaded). We did not want to change the power supplies without further communications with Ethan, and possible direct communications to determine the impacts of swapping. The two spare 12V lidar power supplies where left in the salamander room in the user sensor tub.
5) We looked at the radiometers on the downwind tower. Antonio climbed and tried to clean the downward facing radiometers. This was not very effective, and conditions were not ideal for this. The reach to the outer radiometer was The upward facing radiometers are very difficult to reach especially in the existing weather conditions. This should be done during a low wind day with higher temps.
We left the following items in the storage room:
The snow pillow programming cables and equipment.
The spare lidar AC brick power supplies (12 VDC out)
Bluetooth Mote to be used with extra Tsoil sensors.
The old TRH (with bad fan) - labeled in a bag. It is possible that the TRH or housing are OK.
Snow depth around the center tower were around 3 to 6 ". Snow depth along the walking path between the center and downwind varied between ~6 to 12". Snow depth around the downwind tower appeared to be about 9". The UW and UE towers were not visited, but we do have some pictures of them.
\
The webcam captures the crew on site:
Chris and Antonio will write a bunch more about their site visit this morning, but I thought it good to write what I've done for the (mobile) Tsnows.
These Tsnows were intended to be deployed at ue, but weren't to avoid interference with the FlowCapt sensor. The intent is to have this set (3 popsicle sticks with 4each thermistors) connected to a mote that could be manually installed into the snowpack at places of interest. I have created a mote with a set of AA batteries inside for power and a Bluetooth module to send the data. Thus, this box is self-contained. Note that the users will have to open the box, turn on the power and then press the white button twice to enable the Bluetooth module.
On the other side, Chris and Antonio today attached another white box to cl. This box just has a Bluetooth module on a small PCB with an FTDI chip to convert to USB. This USB device appears to be /dev/ttyUSB1. Thus, I've edited the config (sos.xml) so far just on cl, to read this set of Tsnows from this port. They will be called Tsnow.x.xm.c. I've verified the USB connection by typing "$$$" to enter the Bluetooth module's command mode, then typing "h" to see a list of commands, then exiting command mode with "—". All of this works.
The next step will be to bring out the Tsnow mote (presently left in the salamander room) and then attach some probes. This will wait until Jan.
Snow Pillow 3 stopped, and a "pio BANK1 0; sleep 5; pio BANK1 1" was performed to bring it back.
dsm_server crashed again at 5:30 this morning. Restarted (also statsprocs, which had died). Then found 2 pillows not reporting (2&3), so restarted them as well.
- Have done my second restart in 24 hours of pillow #3. We'll try one more attempt to upgrade its software next visit.
- EC150s (h2o/co2) all died with low signal levels (presumably due to frost) 2 days ago. Most came back later in the day. 1m.uw came back last night, 1m.ue still hasn't come back. I'm hoping that the internal heating setting is enabled on all of these.
- TRH.12m continues to have a dead fan, that will need to be replaced
- We have not seen a recurrence of the Apogee issue since the cables were swapped. Since both the original and replacement cables that Apogee sent were their new style, we suspect that the issue we saw before was our termination of the cable that possibly let water in. Our termination of the new cables was done better.
Sorry guys, got too involved with stuff the past couple of days. dsm_server died again shortly after noon yesterday. I just restarted.....
P.S. Even harder than I thought. Had to kill -9 statsprocs. Now should be running...