# 4-20mA Output

#### gavo

Joined Dec 13, 2007
5
Hi All, I need to generate a 4-20mA output from a resistive probe, I have connected the probe to 5V through a 100K resistor and reading it into the A/D of a uP, works great!, then I need to add the 4-20mA output, I have connected the probe to a 741 opamp and I am getting weird readings, it is also changing the value I have on my uP when I connect the 741, does anyone have some ideas on a simple 4-20mA output?

#### beenthere

Joined Apr 20, 2004
15,819
You need a special circuit to make the 20 ma loop - http://controlsignalconverter.com/.

If the probe signal is going to a microprocessor, what is the need for the current loop?

Can you post the circuit with the op amp so we can try to spot the problem?

#### gavo

Joined Dec 13, 2007
5
Hi,

I have attached the circuit using the 741 opamp, the unit works great showing the correct value on the lcd without the opamp circuit, once connected it changes our measured value on the LCD.

We need to output this data to 4-20mA so we can intergrate our design into a software package which only excepts 4-20mA or 0-5V input. We are getting a very small voltage output from around 100mV to around 1.2V on full scale (our 100k trim pot we have tied to +5V put on our uP A/D pin so we can adjust calibration), we even tried to put a trim pot in line with the 4-20mA output connector on the opamp circuit board to adjust the span, this helped adjust the 4-20mA output range but the measured value stayed the same even if we pulled the probe out which then goes to +5V, any ideas would be appreciated

#### Attachments

• 3.7 KB Views: 117

#### beenthere

Joined Apr 20, 2004
15,819
Your 741 is not going to output 20 ma. In the arrangement you have, it is trying to be a voltage follower. R7 will give some additional voltage gain. With a 4 -20 ma loop, any current below 4 ma is a fault condition.

You would have a better circuit by using the voltage input to the A to D. Replace R7 with a trim pot, of about 50K value. Tie one end to the output, the other to ground, and the wiper to the inverting input (pin 2). By adjustment, you can alter the gain until the full output of the probe equals a 5 volt output from the op amp. If you want to eliminate the positive offset, use a dual supply. With only a single supply, the op amp can't pull its output to zero (a 741 probably can't get to 1.5 volts)

Unless your probe has a low impedance output, you might also get better results with a FET input op amp, like an OP-27, or an OPA134. 741's are a thirty year old design, and many better devices exist.

#### mrmeval

Joined Jun 30, 2006
833

#### gavo

Joined Dec 13, 2007
5
Hi,

I have substituted the 741 for an OP27 and replaced the 250R resistor with the 50K pot, my measured value on the uP now stays the same, bu the mA output is now between 0.4 when the prope is out and 0.78mA when the calibration probe is put in, there is movement but values make no sense, any ideas why?

#### gavo

Joined Dec 13, 2007
5
Would it helo if we increased the trim pot to 500k?

#### gavo

Joined Dec 13, 2007
5
Hi beenthere,

I have tried another method I connected a power supply to my probe into and varied the voltage from 0-5v to simulate the probe the device worked once, I tried it again and nothing, is there perhaps something I am missing on the op27?

#### beenthere

Joined Apr 20, 2004
15,819
If you have the op amp connected to a more realistic voltage supply, like + and - 9 volts, then the op amp's output could swing from zero up to 5 volts. As I said before, the op amp circuit you posted is a voltage follower with gain. It will not operate a 20 ma current loop. That is why I set you the reference to the 20 ma loop converter.

If you decide to use the voltage input to your A to D board, then the 50K pot connected as I suggested will allow you to work with the probe. If you want to use the current loop, you must obtain a specialized circuit to convert the probe voltage to current. A single op amp cannot do that.