MPXV7002GP accuracy at the same pressure

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

MPXV7002GP accuracy at the same pressure

1,038 Views
tonibenitez
Contributor I

Hi. I am designing a pressure control system with 10 MPXV7002GP sensors. In the design of the PCB where the sensors are placed I have followed the instructions on the datasheet (3 On-Chip Temperature Compensation, Calibration and Signal Conditioning).  I am converting the analog signal with a controller with 10bits ADC so I have  ( 0- 5V  --> 0-1023 bits )

Following the datasheet, I have applied this criteria to scale:

0.5 V  --> -2000 Pa  --> 102.3 bits

2.5 V -->         0 Pa  --> 511.5 bits

4.5 V -->   2000 Pa  -->  920.7 bits

And finally I have implemented the auto zero, taking atmospheric pressures every second during 60 seconds to find the average. Then I substract this average in every reading.

I know that with 10bits ADC I can only note changes in pressure of 5 Pa.

I am testing the sensors at atmospheric pressure. The steps are:

1- Start and apply auto zero

2- Every 30 seconds take the atmosferic pressure.

3- During 1 hour I save the lowest and highest pressures.

4- Save the difference between the lowest and highest pressures.

My question are:

Is it normal that at atmospheric pressure I have differences between 10 Pa to 30 Pa between the lowest and highest pressures?

Normally the pressures are never 0 Pa but  -1, 4, -6.....jumping every 30 seconds normally by 5 Pa.

Some real example:

PRESSURE

DIFFERENCE

10:45

12:15

Diff P

P min

P max

14:05

17:36

Diff P

P min

P max

17:45

19:03

Diff P

P min

P max

SENSOR P00

10

-6

4

10

-6

4

25

-6

19

SENSOR P01

15

-6

9

15

-7

8

15

-1

14

SENSOR P02

10

-5

5

20

-10

10

24

-5

19

SENSOR P03

10

-4

6

19

-9

10

24

-4

20

SENSOR P04

15

-5

10

15

-7

8

34

-8

26

SENSOR P05

10

-4

6

15

-11

4

24

-6

18

SENSOR P06

10

-5

5

14

-7

7

14

-1

13

SENSOR P07

10

-4

6

15

-11

4

25

-6

19

SENSOR P08

20

-16

4

39

-12

27

14

-4

10

SENSOR P09

14

-5

9

10

-6

4

29

-6

23

Is this normal?

Have I followed the correct steps?

Thanks

0 Kudos
Reply
2 Replies

662 Views
TomasVaverka
NXP TechSupport
NXP TechSupport

Hi Toni,

Such a drift is most likely related to the noise that is inherent to piezo-resistive pressure sensors.

Could you please post here your complete schematic? Are you using the recommended capacitors as shown in Figure 3 of the MPXV7002 datasheet? I would also recommend taking a closer look at our AN1646 which provides some additional guidelines to filter out the noise.

I hope it helps.

Best regards,

Tomas

PS: If this answer helps to solve your question, please mark it as "Correct". Thank you.

0 Kudos
Reply

662 Views
tonibenitez
Contributor I

Hi Tomas

Yes. I followed the instructions in the MPXV 7002 datasheet. As I commented. I have 10 MPXV7002GP sensors. The Schematic is this:

pastedImage_0.png

The 10 sensors are connected to a master PCB via connectors. The lines are 0.5mm and I haven't a GRD plane in the sensors' PCB. There are a common ground plane for everything in the master PCB.

I have taked a look at AN1646 and now I have doubts if it would have been better to applied for filtering the schema in the figure 2 instead of the one I applied.

What do you think?

If it helps I am using as controller an Arduino YUN.

  pastedImage_1.png

In this article also speaks about filtering via software and I want to try it, since I have a built prototype and it is easiest than modified the design. But I want to clarify how I must to apply it in my case. The steps I do are:

AUTO ZERO

First I apply auto zero. Save the result so I can subtract it in every reading.

READING 

I am reading data every 30 second. At 0 seconds and 30 seconds. Until now I simply read the data recollected at those times. But now following the AN1646 instructions I suppose I have to do:

  1. If I want an average of 60 samples, In every interval of 30 seconds I have to read every 500 ms and the do the average.
  2. This average would be the reading with the filter applied for every interval of 30 seconds.

Am I right?

Another last question. In the AN1646 in the section  EFFECTS OF NOISE IN SENSOR SYSTEM It says this:

The transducer bridge produces a very small differential voltage in the millivolt range. The on-chip differential amplifier amplifies, level shifts and translates this voltage to a single ended output of typically 0.2 volts to 4.7 volts.

I have applied an interval of 500 to 4500 mV as working interval for the MPXV7002GP.

Is that correct or I must change this interval?

Any help would be appreciated.

Thanks.

0 Kudos
Reply