-----------------------
We should note at the close of this week long pressure sensor comparison/test that ALL other features of our Meteobridge PRO unit are working beautifully with no issues to report of ANY KIND!
All Weather Nets connect and report as scehduled with no errors, Twitter strings post on schedule with no errors, as does all other enabled features. The only issue we continue to experience is with the unit's internal Barometer/Pressure Sensor (s)
CONCLUSION SUMMARY of Internal Barometer Tests: If you would like all the details of my final comparison testing, it is all listed further down the page. But here are the key points and issues I have discovered over the last 7 days of monitoring and comparing the two system's pressure sensors...
When comparing the sensitivity and overall accuracy of the MBP's internal Barometer and the VP2 Console's Barometer, I have reached the following conclusions based on our location and elevation:
1) The VP2 Console is more accurate and responsive to pressure changes than the sensor inside the MBP.
2) The MBP sensors appear to lack sensitivity within the .01 to .03 inHg range and fails to measure/report pressure changes within this range nearly 100% of the time.
3) The MBP is very slow to track pressure trends and lags behind the VP2 reported pressure readings by several hours, due in large part to #2.
4) MBP pressure reports observed during these periods of lag often results in a disparity as high as 00.30 inHg between current MBP reported pressure and current VP2 reported pressure. Once the pressure at our location stabilizes and reaches a balanced/steady state, the MBP will "eventually" catch up to the VP2s and report a pressure value within 00.03 inHg + or - of our two VP2 Consoles. But rarely has it returned the exact same reading, and usually requires an offset adjustment of between 00.02 and 00.10 inHg + or - in order to produce a pressure reading that is exactly the same value as the two VP2 Consoles, which are always in sync with each other, and have remained so for the last 4 years.
..........................
FULL DETAILS
RECAP: Our station is located in the Sierra Nevada Mountain Range about 8 miles southwest of Yosemite National Park at an altitude of 3,590 feet.
The accuracy issues of the MBP pressure sensors as compared to the VP2 pressure sensors seems to be an artifact related directly to our location's altitude. Thus... MBP users living at lower elevations and/or who experience less active atmospheric pressure changes may not be experiencing the same issues as we have reported in this thread and will conclude reporting here based on this location's pressure data.
We had to realign the offset again at 2:00 p.m. PDT this afternoon when the MBP's latest pressure readings were off from the two VP2s by almost 00.10 inHg. As we approach hour number (5) since that latest matching adjustment, the MBP pressure sensor is already -00.02 inHg lower than the latest VP2 pressure readings. (Remaining on the same value and indicating a "Steady Pressure State" despite a notable upward spike reported by both VP2 consoles over the same time period)
If the pressure over our location continues to increase, the MBP will eventually rise above its current reading. However, if the rise stops and then trends back down, then the MBP will likely remain at its current result, basically failing to report the 2 point rise and fall this afternoon. Instead showing no change over this entire time period, and once again failing to identify a pressure trend in progress.
My conclusions based on all these days of comparison testing is that if a Meteobridge PRO user is running a VP2 station with their meteobridge PRO, the pressure sensor providing the most resolution and accuracy of barometric pressure changes over time is hands down their VP2's pressure sensor. If like us, they have split their data uploading duties between a dedicated weather computer/software sending ALL data from their VP2 and the MBP sending ISS data from the VP2 in addition to MBP pressure data instead of VP2 Console pressure data, having this inconsistent pressure data history across all weather sites is going to start looking odd and confusing for anyone tracking such a station via multiple weather sites on the internet.
If a station operator is only using the MBP to upload all of their data to their various weather sites on the WWW, then they won't be confusing their users with two different pressure readings, BUT based on our conclusions, they will also not be posting a "complete" history of their location's pressure data over time, and will be displaying pressure graphs lacking the detail and accuracy of any neighboring stations using a VP2's Console pressure data. (Unless you graph them up, side by side like we did, you are not going to see how big the difference truly is!)
Unless something can be done to improve the sensitivity of the MBP sensors, they are never going to capture ALL of the available pressure data occurring at your station location, will falsely report a steady pressure state while also failing to report the true barometric pressure at any given point in time.
If the MBP can be updated to allow users of the VP2 console to bypass the MBP pressure sensor and instead use the pressure data from the VP2 console for uploading to Weather Nets and storing within its historical data, at least they would have a more accurate option. Granted, these users will no longer be using the pressure sensors inside the MBP which kind of seems like a waste of resources. but given the lack of resolution and accuracy we observed with our MBP's pressure sensors, this resource isn't a viable one in the first place. (We still suspect that our MBP has defective pressure sensors, or at the very least sensors that are terribly out of spec)
Questions and requests for the developer... Any thoughts on what can be done here?
•) Can the sensor sensitivity be increased to the point that the MBP tracks these smaller pressure trends to the same degree that the VP2 Console's sensor does?
•) Is it possible to provide a new user option that allows VP2 console owners with a data logger the ability to use the VP2 Barometer instead of the MBP's internal pressure sensor?
Thanks in advance for your thoughts on all my recent comparison reports and final conclusions.
[This will be our last post on this topic unless additional information is requested]
Cheers.
[]Rick.
[EDIT >> 03/24/16]
Just one day after we had concluded our VP2 vs MBP pressure sensor comparison, we saw the biggest disparity between these two sensors of any day so far. At the peak of the disparity, the MBP was under-reporting the high pressure spike by a whopping -00.23 inHg before it slowly rose over several hours to just +00.01 inHg shy of the correct high of the day as reported by our two VP2 consoles! The VP2 Barometer in both of our VP2 consoles tracked the quite rapid rise, along with the recorded high of 30.33 inHg. The MBP on the other hand took nearly 2 hours to reach the same pressure range, but was not synced with the VP2's high for the day.
So... during the period of time that the MBP was playing "catch-up" we had the MBP driven feeds on the web posting our current pressure as 30.13 inHg, while at the exact same moment in time the VP2 fed stations were posting the CORRECT current pressure of 30.33 inHg, which ended up being our high for the day.
Once again, the event described above demonstrates the primary issue with the MBP pressure sensors, which is LAG! The painfully slow reaction time our MBP pressure sensors display day after day when compared to the VP2s. While once the pressure stabilizes and becomes steady the MBP eventually catches up, and with some offset tweaking can be made to sync with the VP2's steady reading, the MBP has NEVER accurately tracked a rapid pressure trend, either a downward trend or an upward trend.
