I am trying to measure the accurate altitude of a hoist elevator using 2 MPL3115A2 (Altimeter - pressure sensors) realtime and publish it to the cloud.

[pari1]

I have two pressure sensors (MPL3115A2) with each connected to a Particle Photon, the sensors are placed next to each other at the same level.

In barometric mode, one sensor reads 95043 Pa, while the other read 95405 Pa. This translates to a difference is 362 Pa.

In altimeter mode, one sensor reads 1742 ft, while the other sensor at the same level reads 1658 ft. This translates to a difference of 84 ft.

I also observed that these differences (pressure & altitude) keep varying over time. I'd appreciate if anyone with experience on these or similar sensors could shed some light on this variance in measurement.

I am trying to use these sensors to get the altitude of the construction elevator in realtime, any suggestions on accurately measuring the elevator height would also be appreciated. This solution seems to be plug and play (Easy on a construction site)

[david_diaz]

Hello Pari,

Despite the MPL3115A2 is factory calibrated, sources of offset errors could be due to device-offset variation (trim errors), mechanical stresses (mounting stresses), shifts due to temperature and aging.

In this case, I recommend setting the Altitude Data User-Offset Register (OFF_H) to implement the relative measurement by altitude-reference adjustment. This register mentioned provides user adjustment to the vertical height of the Altitude output.

As depicted in Figure 6 on page 9 of the application note AN4528, combined with offset removing, the measurement error becomes smaller with the measurement range reducing. This ensures effective and sufficient accuracy for each application case.

Please review the application note mentioned above in order to get more information.

Feel free to let me know if this information is useful for you.

If I misunderstood your question, feel free to let me know.  I will be glad to help.

Have a great day.

David Diaz.

[marcuslp]

Hi David,

With regards to my colleague Parikshith's questions, I would like to clarify, essentially we have two questions:

1) Regardless of whether we apply an offset we are getting very strange fluctuating results. Please can you suggest why this may be happening and what the resolution might be.

2) According to web literature these sensors will allow us to gain precision altitude readings exceeding our intended height use (up to 60 floors of a building, each floor being 3 meters). Please can you or one of your colleagues confirm that what we are trying to achieve with the MPL3115A2 is (or should) be technically achievable. We understand the reading when using any type of altimeter sensor will not be 100% precise (and is affected by outside influences) but do you see any technical reason why we may not achieve an accuracy of within 3meters (on a 60 floor building) if we use a calibration device as part of the solution.

Appreciate your support. 

Regards,

Marcus

[david_diaz]

Hello,

Accept my apologies for the delayed response.

This deviation could be generated due to some factors:

1. As we know, the MPL3115A2 is factory calibrated for sensitivity and offset for both temperature and pressure measurements and the default values are used on power up.

This calibration procedure is done individually for each part in full production. This results in each sensor calibrated with different coefficients from each other. Each part goes through a set of temperatures and pressures so that the coefficients are trimmed/created for each part.  There will be slight device to device variation.  The accuracy mentioned on the datasheet is for the singular part.  When two parts are used side by side, it actually can be such that one is at the far extreme from the other.

As I mentioned before, using internal-offset correction registers, you can adjust the offset values to easily remove these errors after power up for the highest possible accuracy.

2. Please be aware about your manufacturing process, the reason of this is because there are several factors/processes that can damage the diaphragm of the sensors like the soldering process (could be stressed by high temperatures for long periods of time), cleaning process (chemicals or liquids should not be in contact with the hole on top of the device), quick temperature changes can affect it, apply pressure below/above the operational range (below 20kPa and above 110kPa could damage the device), use different media than clean dry air, etc.

Once again, please refer to the “Offset Correction” in section 7.18 of the MPL3115A2 datasheet.

3. I would also suggest placing an RC filter of 10 ohms and 1uF between VDD and GND.  In another application when trying for the best accuracy, this reduced some of the spectral noise density and improved the part output overall.

I hope this information will be useful for you.

If I misunderstood your question, feel free to let me know.  I will be glad to help.

Have a great day.

David Diaz.

[pari1]

Hello David,

I would like to present the below summary based on logged values from two MPL3115A2 pressure sensors, placed next to each other (i.e at the same level).

Summary:

1. In altitude mode, the difference in altitude measured by two sensors varies from 2m to 35m. This means at certain times the difference in altitude measurement is 2m and at other times it is 15m and this difference goes upto 35m.  Is there a way to solve this issue.

2. Similarly, the sensors temperature measurement is not consistent, for example one sensor measures 86.56°F and the other measures 84.31°F, a difference of about 2.25°F, this difference varies. Is there a way to solve this issue.

For your information, I have bought MPL3115A2 sensors from Sparkfun electronics.

Best Regards

Pari

[pari1]

Hi David, Thanks for assisting us on this.

My Application:

I am trying to publish the accurate realtime altitude of the Hoist Elevator Car installed on an 80 Storeyed building.  (need accuracy of 1meter). 

This Hoist elevator would be operational for 3 years until the project is complete.

This building is in NY. I know the the GPS Location of it. It could be windy as we go to the higher floors.

My Approach:

Use two sensors.

        The first one at ground level which is exposed to ambient air.

        The second one is installed on the Hoist Elevator Car which is also exposed to ambient air.

        Both these sensors are collecting data realtime and publishing to the cloud via (Arduino +3g modem).

        I would calculate the difference in pressure from both the sensors and use it to calculate height.

My Questions:

1. Is this approach correct ? Is there a better approach to solving this.

2. Do I need 2 sensors for reliability ?

3. Would I be able to accurately determine the altitude across different seasons ?

4. I have 2 pressure sensors (MPL3115A2) beside one another, they are also giving different temperature readings why is this ?

Thanks for helping us out.

Pari