Vacuum sensors failing

[pathill12345]

I have purchased five of your sensors, 2 MPXV5050V, 1 MPXV4115V, and 2 MPXV6115V (not delivered yet).  The first three of the five have failed after working for awhile.  They have been mounted on a small pcb sold by Modern Devices.  I have followed the recommended circuit in Fig. 4 on your data sheet (for the first 3--there are some changes to Fig. 4 for the MPXV6115V).  I have checked the power supply and it is operating at a steady +5V. The sensors worked great for awhile. They are feeding analog input to a NUMATO USB breakout board. Based on the pressure, the GPIOs on the board are set to turn a vacuum pump motor on or off and to control some solenoid valves.

My question is:  Does the sensor mounting orientation affect the life of the sensors? I have them mounted on a vertical surface of an enclosure (the little pipe that connects to tubing is horizontal).  I am grasping at straws.  I thought maybe the gel inside the sensor was flowing away from whatever it is protecting.  The total time that the sensors were used before they failed was less an 8 hours but they have been mounted as described for weeks.  I need to solve this or find a different sensor.  Any ideas?

[pathill12345]

OK, I think I have resolved my issue. I apologize if it has caused concern for anyone besides myself because the sensors were not failing. It was my circuit that was failing (the one that the sensors were embedded in). I had some protective diodes that were installed, by me, backwards. This caused the small 100 nF capacitor (see figure 4 in the sensor data sheet) on the supply voltage side of the sensor to fail. When the capacitor failed, the source voltage was shorted to ground.  I assign some blame to the vendor that sold me the diodes because they had a small label on the package that said the white band indicated the anode end of the diode, which is wrong.  I should have known it was wrong.

Thanks to those who submitted comments trying to help me.  I apologize again for wasting your time.

[reyes]

Hi,

You mentioned that devices are failing, but, can you please describe the failure? Is the output fixed at 0V or 5V at any pressure? Did you test the sensor by itself once it present the failure (without connecting the NUMATO USB breakout board or the vacuum pump motor)?

Answering to your question “Does the sensor mounting orientation affect the life of the sensors?”, it should not. Pressure sensor are mechanical components (due to their internal MEMS) and have internal gel, but none of these component should be compromised for the mounting orientation. I have never heard from a customer with a similar problem before.

Have you performed the same test with the sensor in horizontal position?

One thing that I’m worried about is, make sure that the pressure on the sensor’s port is always negative (vacuum) it should never be positive, this could degrade the internal structure of the sensor causing permanent damage.

Regards,

Jose

[pathill12345]

Jose,

Thank you for your response. I rechecked the MPXV4115Vby disconnecting its

output from the NUMATO analog input. Today, it is showing on the sensor

output line a voltage of +5 volts when power is applied to the device it is

in and the vacuum level is 0.00 inches of Hg. When I run the software and

apply about -10 inches of Hg to it, it diminishes slightly. My code that is

controlling all of the hardware says the output of the sensor is 4.6 volts

and computes the vacuum level as zero. The 4.6 volts is computed by my

code (JAVA) by:

SensorVoltsOut = (((double)result2/1023.)4.6); *

The variable "result2" is the NUMATO A/D output, which apparently is 1023

because (1023/1023)*4.6 is 4.6 volts.

This was today. Before, the error was in the other direction because the

vacuum levels were like -30 inches of Hg, which is impossible for the

little vacuum pump.The output of the sensor had to be a very low voltage

close to zero to get this result. The vacuum level for the MPXV4115V is

computed by:

• ReservoirVacuumLevel =((

(SensorVoltsOut/5.0) -.92)/0.007652)0.2953;

The factor 0.2953 converts the results to inches of Hg.

I am experimenting with replacing the vacuum sensor controlling the vacuum

pump with an adjustable vacuum .switch that shuts the pump off when the

vacuum reservoir reaches the desired level. It looks like it will work

pretty good. I will still need the other sensor to control hardware that

has to be controlled by software.

No positive pressures are present. All air tubes are either zero or

negative pressure. All the vacuum levels are within the maximum levels

that your sensors can tolerate. I did get rid of the MPXV5050 because it

was getting close to the upper (lower in vacuum terms) limit.

I will try some more testing tomorrow.

Patrick

[reyes]

Hi Patrick,

For the case of the MPXV4115V, around +5V on the output when 0.00 inches of Hg is correct. Check the image below:

 

If you are measuring pressure level on the range of -30 inches of Hg, then the MPXV4115V or the MPXV6115 are better options than the MPXV5050V, in the case of the MPXV4115V, when reaching -30 inches of Hg (-101kPa), you should get around 0.7V

Regards,

Jose

[pathill12345]

Jose,

Re-reading the electrical parameters for the MPXV6115V, I notice a source

current limit of 6 mA (typical value).

I have never inserted a series resistor in my circuits to limit the

current. Where does it go? I calculated I need 833 ohms to limit the

current from my power source, if I am interpreting correctly the source

current limit. Does it go between pin 3 and ground?

Patrick

[reyes]

Hi Patrick,

Yes, you can use a current limiting resistor, it should be connected between Power source and Vs pin 2 of the sensor. 820 Ohms resistor should work fine.

Regards,

Jose

NXP Semiconductors

[pathill12345]

Thanks for your timely help. I was afraid that putting the resistor there

would reduce the supply voltage below +5v but I guess the input resistance

is very high. I have been trying to measure it on one of the ruined sensors

but that might not be reliable and I am using a cheap multimeter.

Anyway, thanks again for your help. I wrote my question assuming you worked

for NXP. Sorry if it sounded a bit demanding. I am ready to hook up one of

the new sensors and try it out. NPX should modify their figure 4 to show

the value and placement of this resistor.

Patrick