Data Notes/Interpretation

Most of the pumping, storage, and usage data presented on this website are posted automatically from our SCADA computer daily just after 6AM. Occasionally, power failures, equipment malfunctions, or radio communication channel interruptions will result in corrupted data or no data being posted. When this happens, daily pumping and storage data are estimated based on historical average production and actual daily run times for each of our two wells, and manually-integrated into our website data files, providing a very accurate and continuous history of system operation. Significant historical aberrations in the automated reporting system are summarized below, in reverse chronological order:

2021 - June 28-30: Well#2 fails to automatically or manually activate resulting in fallback to Well#1. Electrician found and repaired loose neutral at contactor.

2021 - April 7: Failed cellular data communications for tank level. Data file manually patched.

2020 - November: Months of sporadic Well#2 overload alarms on startup attributed to motor minder default configuration of 2% variance in phase voltage. Setting custom tolerance threshold at 10% eliminated overload alarms.

2020 - October 20: Our Badger HRE-LCD 4-20 encoder failed again and was replaced on 10/21. Errant interim pumping data was estimated based on historical production and known daily run times for each well. SCADA virtual registers reset to 0 gallons to sync with the new encoder.

2020 - July/August: Water conservation alert issued for Well #2 pump failure on 7/23/20. ProControl reconfigured to use only Well #1, Well #2 pump pulled on 7/31, and replaced on 8/12, but when brought back online 8/13, an additional problem was discovered with the #2 electronic motor starter so it was replaced with a motor minder on 8/28.

2020 - April 18: Failed cellular data communications for tank level and pumping processes. Data files manually patched.

2020 - February: Major break on Cholla Ct. (Feb 10) drained water tank. Related main rupture on Cactus Ct. (Feb 14) isolated system from water treatment plant. Emergency low-pressure water supply provided by small well during 10-day repair delay. Repairs completed on Feb 19 and CDPHE clearances issued on Feb 20. (Detailed chronology)

2019 - December: The 12/26/19 status report failed to increment the ProControl virtual register for 6.7 hours of Well#2 pumping. Volume was estimated and manually posted as described above. Subsequent pumping appears to be accurately recorded and diagnosis on 12/27/19 by our electricians and EOS tech support could not duplicate the problem.

2019 - December: Our daily status report began reporting erroneous pumping data for Well#1 on 12/5/19. Analysis revealed the 4-20 mA flow monitoring output from the totalizer meter encoder was erroneously pegged at full scale, indicating 170 GPM continuous flow from both wells (rated for 50 and 17 GPM maximum), irrespective of pump activation. Daily pumping estimates were manually calculated and posted as described above through 12/20/19 when the defective encoder was replaced.

2019 - October 13 / November 17: Corrupted cellular data communications for tank level and pumping processes.

2019 - August: The 4-20 mA flow rate encoder failed during a fill cycle on August 17, 2019 at 12:51 AM and was not repaired until 9 AM on August 27. Historical pumping data during this time frame are estimated based on interpolation of manual weekly totalizer readings and daily run times for each well. SCADA virtual registers reset to new meter on 8/30/19.

2018 - July: Failed pulse-sensing totalizer meter. Replaced with new LCD meter and 4-20mA flow measurement on 7/31/18. Daily and weekly data reporting from 7/25/18 to 8/12/18 estimated based on interpolation and projection of most recent manual totalizer readings.

2013 - September: FEMA flood disaster recovery. Emergency water via hydrant cross-connect with City of Boulder from 9/23/13 - 11/11/13.

Missing / Corrupted Data: When one or more data attachments for tank level or pumping processes are not correctly formatted as file attachments, but rather intermixed with the html body text of the SCADA computer's status report, our software can only partially recognize the daily data that is provided by the SCADA report. When this corruption occurs, the tank level and intraday data on our website are most likely to be impacted because the missing day’s 144 tank level and pumping data points cannot be synchronized with calendar-correlated daily summary data.

Intraday chart example: When tank level or pumping data attachments are missing or corrupted, the intraday chart will confuse the previous day’s tank level data, but without the corresponding pumping cycles, and the demand/supply model will display a very unusual negative water demand as shown in the intraday chart below.

Tank level example: When the daily tank level data attachment is missing or corrupted, the tank level chart simply interpolates the calendar-correlated data with a connector line bridging the missing 10-minute or hourly time points as shown in the tank level chart below.

Missing tank level data that has been graphically interpolated can also be identified by hovering the mouse over any point in the data series. This will highlight known points in the data series and skip over interpolated data points as illustrated below.

These occasional failures are most likely caused by data transmission corruption in the cellular communication link between our pumphouse and the Verizon cellular network.