NCCOOS Trac Projects: Top | Web | Platforms | Processing | Viz | Sprints | Sandbox | (Wind)

Ticket #39 (task)

Opened 13 years ago

Last modified 12 years ago

Resolve Issues with YSI System Integration

Status: closed (fixed)

Reported by: haines Assigned to: haines
Priority: critical Milestone: Buoy-System-2010
Component: system integration Version:
Severity: must have Keywords:
Cc:
  • review the upgrades/retrofits to the buoys and tower package
  • see how the systems are constructed and operating, and make any changes that might be wise
  • work through the communications protocols implemented
  • collect documents and software

Change History

03/30/11 16:20:25: Modified by haines

  • status changed from new to assigned.
  • description changed.

04/26/11 16:58:19: Modified by haines

  • description changed.

Rescheduled for May 10

07/12/11 11:08:34: Modified by haines

  • description changed.

We received the buoys and all the "canisters" for both buoy systems and tower system. One canister holds CR1000, power management and the second canister for the serial MUX. A third canister holds the Iridium modem.

Let the configuring, testing, and assembly fun begin!

08/02/11 09:47:55: Modified by haines

  • priority changed from major to critical.
  • description changed.

This email summarizes the issues and testing timeline since delivery of buoys to IMS.


Subject: UNC Buoy and Tower Systems Date: Tue, 2 Aug 2011 08:55:08 -0400 From: Sara Haines <sara_haines@unc.edu> Reply-To: <sara_haines@unc.edu> To: Ted Hardman <thardman@ysi.com>, Tony DiSalvo? <tdisalvo@ysi.com>, Kevin Simpson <ksimpson@ysi.com> CC: Fred Barreiros <fbarreiros@ysi.com>, Harvey Seim <hseim@email.unc.edu>

Hi Ted, Tony, Kevin:

I had hoped that things would be better by now. At this time, without more help from YSI in getting the buoy integrated systems back to the shape they were in when tested before shipping, we cannot deploy them and cannot sign-off on the last invoice. We request your help to send someone to troubleshoot the issues we are encountering. I have detailed a timeline and list of issues that we now face.

I re-assembled one buoy (B1) by myself two weeks ago. Last week, we (Harvey and I) re-assembled the second (B2). We were hampered by a serious wiring issue in the power system causing both buoy systems to crash unexpectedly and/or crash computers connected directly to the system. With Fred and Kevin's help, we were able to identify and fix the power grounding problem. However, we are still hampered by non-functioning systems.

In summary, one buoy system (B1) has not reported sensor data since receiving the systems. The other buoy system (B2) has not reported reasonable scientific values in all fields that were outlined in design document. Also, we have yet to re-establish communications via either satellite communications since configuring for static IPs. From the data, neither system is functioning upon receipt and re-assembly the way they appeared to be in data prior to shipping.

Timeline: June 27-28 testing buoy systems in MA prior to shipping July 11 --- received buoys and systems July 13-14 bench tested both systems in NC (DC from power supply) July 20-21 buoy 1 assembled (DC from power supply --- polarity issue with wiring from battery prevented using battery)

July 27-28 buoy 1 (fixed polarity issue, but then had to fix a grounding issue) July 27-28 buoy 2 assembled (fixed grounding issue same as buoy 1)

We ran both systems overnight July 28 and found the following problems in the downloaded data tables:

B1

Baro_RawData -- all NaN for July Compass_Raw -- no entries for July CompAvg? -- no entries for July, only one set of dir and speed should be two -- (WVc(1) and WVc(2), pitch and roll) DBG_AWindsRaw -- only data for July 20 while debug level at 9 DBG_Compass -- only data for July 20 while debug level at 9 (no string, all NaNs? for Pitch Roll and Heading) GPS -- no entries for July Link10 -- no entries for July Met_6minAvg -- since July, baro=NaNs?, RH~=-0.1, TEMP=-30, raintot=-100, PSP~=0.0, PIR~=-30000 WindMax? -- Looks Right!?

B2

Baro_RawData -- data only since July 27, but look OK Compass_Raw -- OK CompAvg? -- only one set of dir and speed should be two -- WVc(1) dir too low and WVc(2) speed too constant, pitch and roll are OK GPS -- fix for Marion, MA in all records in July, no valid fix. Link10 -- Looks okay except voltage is *10 (e.g. 130 when it is 13.0) Met_6minAvg -- baro OK, RH OK, TEMP OK, raintot=-1300, PSP=-300.0 or NaN, PIR=-30000 WindMax? -- Looks OK!

SAT COMMS: Fred had tested both GlobalStar? and Iridium with direct dialup prior to shipping. All but one Iridium modem tested okay as direct dial but one would not hold settings on power cycle.

GlobalStar? modems -- they were still at ECI until *July 20* only installed last week for the first time with static IP. Probably a baud rate setting between CR1000 and modem. These need to be investigated with direct connect. Fred sent me some instructions on how to do this.

Iridium modems -- we only have one, YSI has one, one is getting fixed. I'm still troubling with Joubeh and Iridium over access to the ground station via telnet. We are still waiting for the delivery of at least one other functioning Iridium modem to go on buoys.

We cannot deploy without reading sensors correctly, without output reported correctly and without communication to the buoys. We had hoped for an August 18 deployment. I hope that we can work something out so that we can get some help troubleshooting these systems. As it is, we cannot sign off on the last invoice until we have working systems.

Respectfully,

Sara Haines

-- Sara Haines Department of Marine Sciences University of North Carolina Chapel Hill, NC 27599-3300 919-962-1253 sara_haines@unc.edu

08/02/11 09:48:48: Modified by haines

  • summary changed from Review YSI System Integration to Resolve Issues with YSI System Integration.

01/11/12 12:11:38: Modified by haines

  • status changed from assigned to closed.
  • resolution set to fixed.

After 3 days with Fred Barreios and Nuno Canha, many of the issues with power, sensors on both B1 and B2 and comms through GlobalStar? working. Other items resolved were

SC105 on B? was not configured so GlobalStar? mdeom was not functioning. Fred reset the configuration via connecting to directly -- since can't use terminal emulator on LoggerNet? to talk-thru to it. The SC105 is beyond the CS I/O port that is shared by direct connect and GlobalStar? to communicate to the datalogger. It will be hard to field service the SC105 if it becomes reset again as it is in the main buoy pod.

Outstanding Issues

  • sampling CTDs -- setup to autosample and store data internally, modify code to request last sample instead of take sample
  • Iridium modems

Since the systems are working and sensors reporting as best as we can tell, and since we have all the parts (even if Iridium modems not functioning for static IP/Rudics), we have signed off on the work with YSI as completed.

This ticket is going to be closed. Use additional tickets for further issues and troubleshooting