Cbart42
Posts: 4
Joined: Tue Apr 30, 2019 1:41 pm

Pressure logging project

Tue Apr 30, 2019 2:02 pm

I am fairly new to Pi usage and want to do a Pressure logging project to sense and record pressures over time from a set of Pressure Transducers (0-3000PSI). The logger will need 2 sensors to be recording pressures from seperate banks either by same Pi or one for each side. Whichever way works best. The desired transducers come in the following options which would work best for Pi connection and recording? I was thinking the 624 as it could be powered directly from Pi 5v pin (I think) but unsure if that's best route or using one of the others. Any suggestions would be greatly appreciated. Not sure exactly what else I will need to get desired logging from sensors.

Input/Excitation Voltage:
625 (2-wire): 9 to 28 VDC
623 (3-wire): 9 to 30 VDC
624 (3-wire): 5 VDC (fixed)
626 (3-wire): 14 to 30 VDC
627 (3-wire): 14 to 30 VDC

Output:
625:Output: 4 to 20 mAFull scale output: 16 mA ± 1%Zero output: 4 mA ± 2%

623:Output: 0.5 to 5.5 VDCFull scale output: 5.0 VDC ± 1%Zero output: 0.5 VDC ± 2%

624:Output: 0.5 to 4.5 VDC ratiometricFull scale output: 4.0 VDC ± 2%Zero output: 0.5 VDC ± 2%

626:Output: 1 to 11 VDCFull scale output: 10 VDC ± 1%Zero output: 1.0 VDC ± 2%

627:Output: 0 to 10 VDCFull scale output: 10 VDC ± 1%Zero output: 0 VDC ± 2%Supply Current:15 mA maximum (no load)

PhatFil
Posts: 1421
Joined: Thu Apr 13, 2017 3:55 pm
Location: Oxford UK

Re: Pressure logging project

Tue Apr 30, 2019 6:11 pm

It might be easier to use an arduino to read the upto 4.5v signal generated determine the pressure and just send that to your pi over serial i2c, spi, or even wifi or b/tooth.. otherwise you will need to protect the 3v3 gpio with a ?voltage divider? and suffer the proportional loss of resolution as a result?

User avatar
omegaman477
Posts: 148
Joined: Tue Feb 28, 2017 1:13 pm
Location: Sydney, Australia

Re: Pressure logging project

Thu May 02, 2019 11:17 am

Cbart42 wrote:
Tue Apr 30, 2019 2:02 pm
I am fairly new to Pi usage and want to do a Pressure logging project to sense and record pressures over time from a set of Pressure Transducers (0-3000PSI). The logger will need 2 sensors to be recording pressures from seperate banks either by same Pi or one for each side. Whichever way works best. The desired transducers come in the following options which would work best for Pi connection and recording? I was thinking the 624 as it could be powered directly from Pi 5v pin (I think) but unsure if that's best route or using one of the others. Any suggestions would be greatly appreciated. Not sure exactly what else I will need to get desired logging from sensors.

Input/Excitation Voltage:
625 (2-wire): 9 to 28 VDC
623 (3-wire): 9 to 30 VDC
624 (3-wire): 5 VDC (fixed)
626 (3-wire): 14 to 30 VDC
627 (3-wire): 14 to 30 VDC

Output:
625:Output: 4 to 20 mAFull scale output: 16 mA ± 1%Zero output: 4 mA ± 2%

623:Output: 0.5 to 5.5 VDCFull scale output: 5.0 VDC ± 1%Zero output: 0.5 VDC ± 2%

624:Output: 0.5 to 4.5 VDC ratiometricFull scale output: 4.0 VDC ± 2%Zero output: 0.5 VDC ± 2%

626:Output: 1 to 11 VDCFull scale output: 10 VDC ± 1%Zero output: 1.0 VDC ± 2%

627:Output: 0 to 10 VDCFull scale output: 10 VDC ± 1%Zero output: 0 VDC ± 2%Supply Current:15 mA maximum (no load)
I would choose the 624, and look for a suitable ADC hat or I2C board that can range 0-5VDC. The measured range 0-3000PSI is a wide range and depending on the accuracy needed go for an ADC with at east 16-24 bits of resolution.

With these sorts of transducers the quality (cleanness) of the supply voltage is critical, be sure to implement the suggested power supply conditioning and filtering circuit recommended in the Data Sheets.

Always consider over pressure events. Although most of these style of transducers have a 10x burst pressure, anything over 150% of FSD usually damages the sensor, or at least will require you to re-calibrate.

Also, depending on what you are doing with the pressure reading, temperature compensation may be required. I am not sure of these pressure sensors are temperature compensating. Even if they are, they are usually only able to compensate internally for a moderate temperature range, beyond that you need to externally compensate in H/W or in the S/W.

Example: if your are trying to determine the volume of compressed air in these banks, based on the water capacity of the storage tanks and the fill pressure, due to Gay-Lussac's Law: The Pressure Temperature Law, you will need to correct for pressure increases due to higher temperatures. Gay-Lussac's Law: states that the pressure of a given amount of gas held at constant volume is directly proportional to the Kelvin temperature.

Temperature compensation for liquid media is less of an issue. However most if not all mechanical style sensors (as is the 624) are calibrated for a given temperature (usually 20º or 25ºC) so if accuracy is your thing, and the environment (or media) is outside the 624's calibration temperature, then you may want to include a temperature sensor as well.

Please let us know more about the application, as when you mention 3000PSI and 'banks' I am lead to think its a diving/medical air/gas application, which my many years in the Navy have given me a great interest.
..the only thing worse than a stupid question is a question not asked.

Cbart42
Posts: 4
Joined: Tue Apr 30, 2019 1:41 pm

Re: Pressure logging project

Fri May 03, 2019 8:11 pm

Thanks for the info so far!!!! I have been researching ADC's to read the analog signal from the transducer, I have a couple in mind. I plan on using the 624 transducer so far.

For ADC I might use an ADS115, its inexpensive and appears to be workable solution, however I am never sure as I am still very new to this.

Also I need to figure out the power conditioning and filtering to make sure the transducer gets a stable fixed 5v. What is best way to handle that?

The application is for a Medical Gas Manifold. I need to monitor and log the bank pressures over time to investigate possible unauthorized use or loss of gas during facility down time and get accurate information on usage both daily, weekly, even hourly might be nice. Facility is small and their pressure recordings for me are terrible. Maybe once a day if at all, with many discrepancies in the data. I cannot sit and watch it all day either. Bonus objectives would be external access of data either through local network or website to have real time graphing of current usage and bank status. If it works out I can use this in many other facilities.

Overpressurization shouldn't be a problem as the attached H- cylinders are never filled past 2500 PSI usually they're in the 2300 range.

Thermal changes shouldn't be a problem for this facility either as manifold is located indoors but I will have to take that into account if I use this for facilties with manifolds outside.

Cbart42
Posts: 4
Joined: Tue Apr 30, 2019 1:41 pm

Re: Pressure logging project

Fri May 03, 2019 9:12 pm

Update: the 624 transducer is almost 2x times the cost of some of the other ones (624=$300 vs 625=$125) so to keep cost down I need to change to a different transducer.

Using the 625 with output of 4-20mA cuts transducer cost in half so now its finding ADC interface for Pi that works with that. I guess that will require separate power supply for powering transducers but that should still be cheaper in long run. I need two transducers in the end.

Any thoughts?

Cbart42
Posts: 4
Joined: Tue Apr 30, 2019 1:41 pm

Re: Pressure logging project

Fri May 03, 2019 9:18 pm

This looks promising for interfacing with transducer,
can even get it with inputs half mA/half VDC and allows for external powering:

https://widgetlords.com/products/pi-spi ... 6898176393

This might be a winner, any suggestions are obvious problems using it with the 625 transducer I might be over looking?

JohnsUPS
Posts: 123
Joined: Fri Jul 06, 2018 2:13 am
Location: USA

Re: Pressure logging project

Sat May 04, 2019 5:07 am

If you go with the 625 transducer with the 4 - 20ma output, it is to your advantage that it has a current source output. Current outputs are inherently less noisy and can run fair distances in an electrically noisy environment. This is good, especially if you're connecting it to an ADC.

You can easily scale the output of the transducer by simply putting a scaling resistor across the output to create a voltage for the ADC.
if you drop a 165 ohm resistor across the output of the transducer, you will have 3.3 volts out at full scale (@ 20ma out), and 0.66 volts out at zero (@ 4 ma out). Other scaling (for a display for example) can be done in software.
Electrically, the typical way this is done is by locating the scaling resistor close to the ADC and not at the transducer. This is so that any long runs of wire between the transducer and ADC input are done using a current mode connection which is much less susceptible to noise. This is why transducers such as this (and many others) are offered with current outputs.

You mentioned the ADS1115 - I've used it and like it. What is nice is that it is very low power, and can run off of 3v3. If the ADC is powered from the Pi (3.3 volts), the I2C bus can be connected directly to the Pi without voltage level conversion. Using the 165 ohm scaling resistor allows you to directly connect your transducer as well. With 16 bits and either two differential or four single ended inputs plus a comparator alarm function, it seems like just the ticket for your application.

Return to “Automation, sensing and robotics”