Siganl Conversion 2

[galvani]

I use a PLC with 0-10v analog inputs. My devices (pressure transducers) have a 4-20 ma output. To convert, I put a 500 ohm resistor inline with my signal to generate a voltage drop. This results in a signal of 2-10v proportional to the ma signal. This works fine for my application, but I am cutting off some of my resolution.

Now the question...

Shouldn't I be able to use an OP AMP instead of the resistor workaround to generate a true 0-10v signal from my ma device? It seems as if I should be able to set an OP AMP up as a difference amp with my device on one input and a constant 4 ma siganl on the other. Since the output of the OP AMP should then be proportional to the DIFFERENCE between the two signals, I should get 0 - 10 v output (with an appropriately sized feedback resistor).

I should, but I can't get this to work in my mock-ups. Any ideas? Is there a better way to do this?

Thanks,

g

 

[nbucska]

Make a diff.amplifier with a gain of 5/4.
Connect its + input to the 500 Ohm, the -
input to +2 V reference.

<nbucska@pcperipherals.com>

 

[electrageek]

What is the resolution of your PLCs analog input 0-10V A/D converter? Usually they are 12 bit. If you check the actual resolution all the way through the system maybe the addition of OpAmp is not worth the effort. If you have money you can buy an ActionPak or Mighty Module that will do the same job. (4-20ma to 0-10Vdc convertor) I have used them many times. Another approach is to get either a 4-20ma PLC input card or replace the transducers with the 0-10Vdc output type. Otherwise follow nbucska's advise.

 

[jbartos]

Suggestion: The above postings do not address any stability aspects, namely if there happens to be a feedback loop one cannot just add another amplifier since one could potentially end up with a positive feedback.

 

[electrageek]

Jbartos
Good point, but generally pressure (PID) loops, if he has one, are fairly slow compared with the speed of an analog convertor. To get positive feedback would require the reversal of polarity not just adding an amplifier. But your right, you do have to watch the polarity.

I personally would rather get rid of extra devices, just for reliability sake. That is why a direct connection from the transducer to the I/O is the best.

 

[jbartos]

Suggestion: Check on the stability of even and odd number of amplifier stages with overall feed in Referece:
E. James Angelo, Jr., Electronics: BJTs, FETs, and Microcircuits, McGraw-Hill Book Co., 1969
Fig. 13-3 Graphical analysis of the Amplifier in Fig. 13-2. Essentially, an odd number of amplifier stages creates the stable amplifier with overall feedback and an even number of stages leads to the bistable device. However, by the proper design of two stage amplifier feedback, it can be the stable two-stage amplifier.

 

[electrageek]

That one of the things I enjoy about Eng-Tips there is always something to learn. I will read up and hopefully become a better engineer. I have never run into this problem but maybe it has been hidden away causing instability and I didn't see it. Thanks jbartos