Help with Scaling 4-20 mA to 0-5 VDC transducers

[esday]

I’ve got a 4-20mA pressure transducer that I’m connecting to an analog acquisition system that takes 0-5 VDC. I’ve got a 250 ohm resistor bridge properly wired in but I’m having trouble coming up with the right linear scaling to output a proper signal. I’ve got the calibration report from transducer that I can calculate my slope and intercept for the milliamps and (absolute) 0-15 PSI, I.e. without the resistor, but I can’t calculate it properly when converting to VDC. I’d also like to try to output kPa but I can do that separately.

[altenbach]

Please include a link when cross-posting from reddit

[aputman]

Well, you have a 0-5V input and a 0-15 PSI output.  That equates to a slope of 3 (rise over run).  Obviously, those are ideal values and not what your sensor will output.  

You can use the Map Ranges scale to enter the calibrated 0 value and the calibrated max value of your sensor.  Use Ohm's law to convert the milliamp value to volts. 

So let's say your calibration report says that 19.93 mA = 14.98 PSI.

Plug 19.93 mA into an Ohm's law calculator with a 250 ohm resistor value and you'll get 4.9825 V.

Enter that into your scale for the maximum value.  Do the same steps for the minimum value. 

[DavidBoyd]

It's important to remember when converting a current loop transducer to a voltage input (typically via a 249 ohm 1% resistor), that your 0-15 psi functional range DOES NOT map to 0-5Vdc, it maps to 1-5Vdc.  Current loop solves two very old, real world problems: 1) it reduces noise pickup on long wiring runs, and 2) it provides an automatic off-scale indication for an open wiring fault (0 mA is NOT a valid value, ever).

Unlike what was posted elsewhere, your idealized mapping is not 3.0 psi per volt, it's 3.75 psi per volt, and you have a (-)3.75 psi offset, as shown below.  If your current loop is disconnected, you'll properly get a negative pressure reading, which you should flag as a fault.

Hope this helps!  Best regards,

Dave

(Note that in the MAX custom scale dialog, you can edit the sample graph X-axis values as an easy way to visualize that your slope/intercept values are correct.  Here I'm showing 0-5V, but you can change the scaling to 1-5V.)

[JB]

Since my applications require HW to interface the signals anyway, I use current loop receivers (RCV420) to convert the 4-20mA signals to 0-5V.

[santo_13]

---

@JB wrote:

Since my applications require HW to interface the signals anyway, I use current loop receivers (RCV420) to convert the 4-20mA signals to 0-5V.

---

Nice one, interesting to see a monolithic solution, it would be nice to see something that can output to +/-10V to maximize the resolution and accuracy of the typical DAQ inputs.

[AnalogKid2DigitalMan]

I've been a fan of Dataforth for various conversions and isolation needs for DAQ interfacing, they have 4-20mA to 0-10V and -20 to +20mA to -10V to 10V units. They do employ filtering so one needs to pay attention to the needed Bandwidth.

https://www.dataforth.com/current-input-signal-conditioner-search

-AK2DM

[aeastet]

Analog Devices is also another company that provides this type of signal conditioning.

[aputman]

I've found these devices to be extremely customizable and compact.  They take a wide variety of input signals and will convert to a similarly wide variety of outputs, configurable via DIP switch. 

https://www.phoenixcontact.com/en-us/products/signal-conditioner-mini-mcr-2-ui-ui-2902037

[JB]

To round off this interesting information on converters, I'd like to mention that I've used a signal calculator for an application.

It is capable not only of measuring and generating signals of various types, but also of performing arithmetic operations (+, -, * and /) on two inputs and generating the result on an output.

Configuration is performed via a utility supplied with the converter.