Meteobridge PRO Barometer? - Response and Accuracy?

[D2149]

I've had my MBP for a few days now and I assume that it has had a chance to settle in. So I am starting to become a bit concerned with the barometer in this little box as compared to my two VP2 consoles. It isn't tracking well with the other two sensors, and over time can get off with them by as much as 0.20 inHg

Both the VP2 console barometers are always in sync with each other and I have rarely found the need to adjust the standalone unit in our kitchen to produce the same reading as the data logger equipped console in our living room/weather computer station. So, I am not use to having a problem like this. The MBP feels pretty warm to the touch even though it is in an open and well ventilated area. Could that be the problem? My Box Climate reading is currently 43.1 C. Is that within the normal operating temp range for this unit? I know that Davis makes a big deal about direct sun on the VP2 console which could mess with the BARO sensor inside. Made me wonder whether chassis heat could be effecting the accuracy of the MBP pressure sensor.

Any advice from fellow MBP/VP2 owners? Have you noticed a similar problem with pressure readings between your VP2 console BARO and the MBP BARO? I obviously can't keep changing the offset 10 times a day to keep all our pressure uploads the same and I am not crazy about the idea of plugging the MBP into a VP2 console thus bypassing the major feature I bought the box for. Its own independent RF system for talking to the 6163 ISS.

Would definitely love to hear from other users and the developer regarding this issue.

Thanks!

[admin]

What is your stations altitude? MPRO does not use the exactly the same formula
for sea level computation as Davis does, so depending from outdoor temp
there might be differences which have nothing to do with sensor accuracy.
When you compare station pressure (without the sealvel compensation) directly,
is there still a drift?

[Mattk]

.... It isn't tracking well with the other two sensors, and over time can get off with them by as much as 0.20 inHg ....

0.2 inHg as in nearly 7mb? No not seeing anything even close to that difference. At the most maybe around 0.5mb (<0.02inHg) and one must consider the VP2 Barometric pressure nominal accuracy is +/- 1.0mb (0.03"Hg)

[miraculon]

My "box climate" is 49.4C. It is quite consistent and I was able to calibrate out this box heating to display indoor ambient.
My barometer tracks very well with my local METAR station's barometer reading.
The CWOP QC shows a close track with my surrounding stations.
I see no problem at all with the baro sensor on mine.



Greg

[D2149]

What is your stations altitude? MPRO does not use the exactly the same formula
for sea level computation as Davis does, so depending from outdoor temp
there might be differences which have nothing to do with sensor accuracy.
When you compare station pressure (without the sealvel compensation) directly,
is there still a drift?

Thanks for the replies!

We are at 3590 feet of elevation and that is what I entered into the STATION section of the MBP interface. We just had a storm come through and it is on its way out now. The forecast is calling for another dreaded high pressure ridge coming in later today and lasting through the week, so this will be a good time to track the pressure rise. It has been very stable so far this morning, which gave me a great opportunity to sync up all three barometers.

At the moment I have had to put the following negative offset into the pressure correction field [-0.130] to get the MBP BARO reading the same as the two VP2 consoles. This is the 15th minor adjustment to that offset I have made so far.

I got them all reading the same about an hour ago using the above offset for the MBP. Right now the VP2 consoles are reporting 30.19 and the MBP was also reading 30.19 inHg. What is likely going to happen over the next hour or two is that the VP2s will continue to move off the 30.19 inHG mark while the MBP will still be reading 30.19 inHg. This is what I have been observing and what prompted me to post this thread. There is always the possibility that this last tweak was the final one needed, and now all three will track nicely, so we shall see.

In order to get a real good sense of how off the MBP has the potential to get compared to my two VP2 consoles, I am not going to do any more tweaking to the MBP's offset. I will give this a few days to see how off it gets with the two VP2s.

So far, the impression I get from my MBP's barometer is that it is VERY slow to react to pressure changes compared to the VP2s and tends to just sit at the same pressure for hours without any changes at all. Meanwhile the VP2s have been up and down several points during that same time frame.

Has anyone done a side by side comparison of their MBP BARO readings with the VP2 console's BARO over the course of a few days to see if they stay the same over time?

That is what I am referring to, and since our reports are spread out between our dedicated weather computer using the BARO from the VP2 console and the MBP reports using the BARO from the MBP, I am trying to avoid seeing radically different BARO readings over the same time periods on different weather sites displaying our weather data.

For comparison over the next few days...

The banner below in my signature line is data from our weather computer [iMac 27" and LWC software] talking to our VP2 console.

Also on our own home page:

http://basslake-weather.com

My MBP data can be found here on our new Weather Cloud page:

https://app.weathercloud.net/d0773552089#profile

P.S. Since I have been composing this message, the VP2s have fallen to 30.18 inHG while the MBP remains at 30.19 inHG.

[admin]

As you are located rather high the formula that converts station pressure to
sea level pressure is making lots of a difference. As air temperature and humidity are
also used for sea level computation, VP2 algorithm and Meteobridge algortihms can
make a difference for sure. This starts with which outdoor temperature to take (how
long averaged?) etc... With stations on just 100 feet altitude this does not make
a big difference. However, I asked you to go on sensor accuracy with pure pressure numbers
exactly for that reason.

[D2149]

As you are located rather high the formula that converts station pressure to
sea level pressure is making lots of a difference. As air temperature and humidity are
also used for sea level computation, VP2 algorithm and Meteobridge algortihms can
make a difference for sure. This starts with which outdoor temperature to take (how
long averaged?) etc... With stations on just 100 feet altitude this does not make
a big difference. However, I asked you to go on sensor accuracy with pure pressure numbers
exactly for that reason.

Please explain "Pure Pressure Numbers". Are you referring to the pressure reading within the brackets on the LIVE DATA page? That is currently reading 26.35inHg.

What can be done to resolve this issue of disparity between the VP2 pressure readings and the MBP? Could our sensor be defective in the MBP? Can the sensor be adjusted other than using that offset variable in the STATION settings page? Would "fudging" our altitude value lower make your calculations more accurate and better track the VP2 readings?

I kind of need to figure this out if I am ever going to have the confidence to allow the MBP to be our main weather data uploader. Too bad there isn't a way to get the VP2 Baro data off the console without having to totally abandon your RF receiver to the ISS.

Have you considered adding a feature that would allow the MBP to receive the "Retransmit" signal from a VP2 console? Not sure if that feed includes BARO or not. We are not using Retransmit between our two VP2 consoles because I use the kitchen console to also monitor ISS output strength since it is farther away from the ISS by a good 50 feet.

[D2149]

I broke down and adjusted the offset to -0.150 which took that extra 2 points off the MBP and it is now reading 30.20 inHg like the VP2s are. Whether that last adjustment will finally keep the MBP in line with the reference pressure reading coming from the VP2s, we shall have to wait and see.

[D2149]

The MBP is once again +8 points off from the VP2 consoles. It is reporting an 8 point higher pressure reading than the VP2s and has not moved much over the day, while the VP2s have been very active.

The VP2s are currently reading 29.92 inHg and falling, while the MBP has been pretty much stuck at 30.00 inHg and steady the entire day! I think it fell to 29.99 for a few minutes about an hour ago, and then went back up to 30.00.

What I would really like to do at this point is use my second VP2 console to provide its pressure data to the MBP along with all the ISS data coming from the ISS. I just need to buy a second data logger and I will be able to do this.

If I connect the MBP to my VP2 console using the USB option, will it pick up the PRESSURE sensor data from the VP2 console and send that out instead of its own?

This is the only solution I can think of aside from possibly requesting a warranty replacement under the assumption that the Barometer in my MBP is defective.

Getting the pressure from our second VP2 console seems like the only solution at this point. PLEASE tell me that I will have the ability to use my VP2's console pressure data coming from the data logger like we do with our dedicated weather computer!

If this isn't an option, it NEEDS to BE! At least for folks who live high up in the mountains like I do, at an altitude where apparently the MBP Barometer along with its pressure calculating system are failing miserably to report the "actual" pressure for this location!

Being able to use the VP2 console pressure data instead would at least give users like me a viable alternative.

And in the meantime, here goes another -9 points added to the offset to get the MBP to report the "Correct" pressure. [-0.220] is how much offset I need to use today to get the MBP to read out the same pressure as the VP2s. On its own, it moved a total of -00.01 lower today while the VP2s fell nearly -00.10 inHg over the same period.

[admin]

Looks like you are not able to understand, that reason is most probably different
calculation formula. I asked you two times to please compare station pressure data,
so that we can compare raw pressure data first instead of mixing up everything.

Did you realize that the MB PRO has two pressure sensors (thb0, thb9)
and could you imagine to post that readings (raw pressure in brackets)? Putting
then the station pressure from the Vantage also onto the table will be a starting
point...

[D2149]

Looks like you are not able to understand, that reason is most probably different
calculation formula. I asked you two times to please compare station pressure data,
so that we can compare raw pressure data first instead of mixing up everything.

Did you realize that the MB PRO has two pressure sensors (thb0, thb9)
and could you imagine to post that readings (raw pressure in brackets)? Putting
then the station pressure from the Vantage also onto the table will be a starting
point...

On Mon Mar 14, 2016 3:40 pm I had asked for clarification as to what PURE PRESSURE was referring to. Assumed it was the value within the [brackets], but wanted to make sure that this was the data you wanted to see for comparison and not something else before I provided it. You can review this thread and see that you never replied to that request for clarification until today. Please keep in mind that I am still learning this system and all of its features, names etc. I have had this system for a little over 1 week. How long have you been involved?

The comparison data you requested:

Indoor 33s 89.5% 19.7°C 42% 1016.8hPa (888.0hPa) 67.5°F 42% 30.02inHg (26.22inHg)

System #9 23 sec 39.6°C 13% 1023.8hPa (895.1hPa) 103.3°F 13% 30.23inHg (26.43inHg)

Davis VP2 Console #1: 30.05 inHg
Davis VP2 Console #2: 30.05 inHg

(NOTE: I adjusted the offset on the MBP 4 hours ago to make the MBP pressure reading the same as the VP2s - Since that last offset adjustment which had all three sensors displaying the exact same pressure reading, the VP2s have seen a .03 rise over those 4 hours, while the MBP's "Indoor Pressure Sensor" has remained steady at 30.02 inHg. -- Can't speak to any changes with the "System" pressure sensor since I was not monitoring that reading, but will begin to track and include it from now on.)

In response to your other comments...

NO. I was not aware of the (2) pressure sensors in the MBP unit. Is this documented somewhere in the manual or the Wiki? If it is, I obviously missed it.

But this would explain why those two pressure readings in the LIVE DATA are never the same value [pressure in brackets] with the #9 line's pressure readings also not abiding by the offset setting I added. Neither unadjusted sensor is reporting a barometric pressure remotely close to our actual pressure today, and the offset controlled sensor (Indoor) seems incapable of tracking the subtle rises and falls over the last four hours that both VP2 consoles accurately reported, in line with several surrounding stations in our area including two NWS bots.

[admin]

Thanks for providing the numbers for both sensors. The pressure correction on station
tab does not work on thb9. I will change that with next update.
However, when you have a correction of -.2 inHg defined, both pressure
sensors show raw data around 26.43 inHg +/- .01 inHg which looks rather fine
to me. I doubt that both are wrong and show same numbers.

What station pressure does your VP2 console show? I don't remember if the VP2 console can
show anything different from sea level pressure. If not, please adjust altitude to zero for
a moment and have a look if this is close to thb9, which I would assume.

Having done this, we know if it is a sensor problem or a computational difference
to work on.

If it is a plus to have the pressure sensor to notice short term changes (like a door
being shut etc) might be controversional. So far the Bosch sensors in the MB PRO
are not known to be extremely slow.

[D2149]

Thanks for providing the numbers for both sensors. The pressure correction on station
tab does not work on thb9. I will change that with next update.
However, when you have a correction of -.2 inHg defined, both pressure
sensors show raw data around 26.43 inHg +/- .01 inHg which looks rather fine
to me. I doubt that both are wrong and show same numbers.

What station pressure does your VP2 console show? I don't remember if the VP2 console can
show anything different from sea level pressure. If not, please adjust altitude to zero for
a moment and have a look if this is close to thb9, which I would assume.

Having done this, we know if it is a sensor problem or a computational difference
to work on.

If it is a plus to have the pressure sensor to notice short term changes (like a door
being shut etc) might be controversional. So far the Bosch sensors in the MB PRO
are not known to be extremely slow.

Thanks. It was obvious to me that the #9 sensor was not connected to the offset section. That was why I didn't perceive it initially as being an independent sensor, but simply an unadjusted reading from the same pressure sensor, or possibly a piece of debug code that was left in for some reason. (since the readings were not exactly the same as the Indoor values)

Not sure what you are inferring with the "Door being shut" comment? That choice of words seems to imply that what I am asking for out of your hardware is somehow unusual or impractical? Okay...

I am certainly not asking for an unrealistic level of sensitivity out of your sensors. Just a professional level of sensitivity, which based on the retail price of the device shouldn't be considered an unreasonable expectation. I'd like to see a pressure sensor in the MBP that doesn't miss a 00.10 inHg pressure drop over the course of 4 hours. A rather active period of pressure change in .01 to .03 inHg increments that was clearly reported by both of our VP2 sensors, and all but ignored by the MBP yesterday.

The two issues I have with the MBP sensors are 1) Lack of sensitivity to minor pressure changes in the range of 00.01 to 00.02 inHg and 2) The inability to maintain sync with a professionally calibrated and maintained reference sensor for more than a few hours. Sync adjustments that are always made during a period of stable/steady pressure conditions.

The primary reason for the loss of sync is that the VP2 sensors track the small pressure changes with the appropriate UP or DOWN arrows on the display, while the MBP reports STEADY over the same time period. If those pressure increases or decreases add up to a significant pressure change over the course of the day, that often leaves the MBP reporting the current pressure inaccurately by 00.10inHg or more compared to our reference readings from the VP2 #1 Console.

[Edited for length]

[admin]

I think I asked for a very simple thing and after reading your very long post
I am still not sure what the difference between the station pressure coming
from the Vantage and the two MB PRO sensors is.

There is one sentence hidden in your explanations that states:

They are beyond reproach in my opinion and even when zeroed to sea level are still 00.24 inHg lower than your MBP sensors are reporting.

This sounds like you did set station altitude in the console to Zero. Really, did you do?
Or is it more a hypothetical discussion point which you then did not do as you think you already know what the result will be?

Please allow me not to believe that raw pressure data from two independent sensors in the MB PRO have nearly the identical failure of .24 inHg.
With all respect, this does not sound reasonable. Sensors used are from Bosch with a good accuracy in the specs, not something from China.
I will check if conversion to inHg might do too much rounding which flats out the minor changes you are looking at.

To work on the sea level calculation formula, I am reaching out to understand the computation used by Davis which seems to be based partly on lookup tables etc. So it will not be easy to get an extremely close match. This is work to do on my side.

MB PRO also calculates altimeter pressure as requested by CWOP. Did you try a test feeding to CWOP? As we worked a long time
on that computation I am rather sure this computation is fine. Do you have altimeter reference for your location (Airport)? Otherwise it will
not help much.

[D2149]

The VP2 pressure reading with altitude set to 0000 is 26.42 inHg. (30.06 inHg at our elevation) So, it does appear that your altitude calculation is the source of the error.

We are not using the MBP to upload to any of our major accounts. The VP2 pressure is what is being sent to all of these.

You can compare local stations to ours on this CWOP page. Our data is at the top.

http://www.findu.com/cgi-bin/wxnear.cgi?call=DW2149

The difference in readings right this minute is the unadjusted/offset 30.24 inHg from your #9 minus the 30.07 inHg from my VP2 #1. (00.17 inHg difference) At the moment a little less than the .24 inHg seen earlier today, but still significant. If this is not the sensor's fault, then it must be the altitude calculation you are doing which is failing to pull the pressure down to the 30.07 inHg it should be at right this minute in time. (Or push it up if you consider the all zeroed starting point pressure above)

EDIT: The VP2 pressure just fell to 30.06 inHg along with our #2 console in the other room. The MBP sensors remain unchanged.

In addition to this over reporting unless the offset is engaged, the sensor fails to track .01 to .02 inHg changes at the same real-time resolution that the VP2 sensors do. EDIT: The drop of 00.01 inHg reported by the two VP2 sensors, but missed by both of the MBP sensors is a perfect example of this difference in sensitivity. Which later could be interpreted as less resolution of pressure changes when viewed on a historical graph.

I provided you with the additional details in my previous post so you could see that the VP2 sensor being used for reference can be trusted and the differences between its performance and the sensors in the MBP are real, not imagined. I type at 120 WPM, so I apologize if my posts are too long for this forum. I don't anticipate that I will be doing much more posting here after this submission, so no need to freak out!

As far as the Davis Hardware...

I have no knowledge of what Davis uses for their own pressure calculations. All I know is that I have never had any complaints from anyone about our pressure readings coming from the Davis hardware, and I personally find the performance and accuracy to be excellent and exactly on the level of detail I would expect from a professional instrument. All the remote NWS stations are also Davis 6163s so that is a good enough endorsement for me.

If you need anything else from me, let me know via email.

Its looking like the MBP is not going to work for our intended purposes until you have the time to figure out how to get your pressure data to sync and track with the same speed and accuracy as the VP2 pressure sensor. A good goal to shoot for in my opinion, since anyone coming off a dedicated computer/software rig is going to likely be using the VP2 pressure sensor there, and will not want to basically downgrade their Barometer performance when they move onto the MBP.

I would definitely encourage you to make this a top priority.

Take care.