zero / span circuit

[AdamMayotte]

Hello,

I am trying to design a circuit which uses op-amps and resistors that would be able to make zero and span adjustments to an input signal.
Can anyone here help me with the circuit design or maybe direct me to a site that has that type of info.

Thanks,
Adam

 

[MikeHalloran]

I'm sure someone here can help ... provided that you help, too.

For instance, it would be helpful to know the range occupied by the incoming signal, where you think its zero is, and how 'stiff' it is.

And it would be nice to know what span range you want on the output signal, and what you're going to drive with it.

And, it would be helpful to know how fast the incoming signal changes.



Mike Halloran
Pembroke Pines, FL, USA

 

[VEBill]

'zero' is offset.
'span' is gain.

To get a sensible answer, you'll need to provide more details regarding your exact application.

 

[Comcokid]

And also, what you have available for powering the opamp. Example - if the input signal goes below ground you will need a negative supply.

 

[AdamMayotte]

I want to know the position of moving arm.
I was thinking of providing feedback of the arms position with a pot attatched to the moving section of that arm.
As the pot value changed I wanted a output, proportional to the change of resistance, in terms of a current of 4-20mA.
SO..
a change in input resistance(doesn't matter how much) to provide a 4-20 mA output.
I am using a 12 VDC battery as my source.
Speed of incoming signal would be 1 degree of rotation per second (max).
I can provide negative source if needed.

Thanks a lot for your help,
Adam

 

[VEBill]

If there is a computer involved (at the other end of the 4-20mA loop ?), then do the calibration in software.

1 degree of rotation per second (max). People will get very bored watching the arm move that slowly. ;-)

 

[Skogsgurra]

What you need is a very basic opamp circuit with offset and gain adjustment.

The good old (around 30 years now) collection of opamp circuits at National Semiconductor

http://www-micrel.deis.unibo.it/~benini/ELEI/Reading/AN-31.pdf

is a good start.

Use the second circuit (Non-inverting amplifier) and make R2 adjustable. Use it to set the gain (1/span) you need. Then use a couple of resistor to add your input signal and an adjustable offset signal before inputting to pin 3 (Vin).

Since you seem to have a 4-20 mA signal, you will have to make a voltage out of it forst. Use a suitable resistor for that. 250 ohms will give you 1 - 5 V, which usually is a good range.

The resulting output is a voltage, so you may want to make a 4-20 mA signal out of it. Use the precision current source on page 9 for that.

You may also find several other approaches if you study all the circuits given in the reference. It is a very complete collection of analogue circuits created by people that lived opamps years after years.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[AdamMayotte]

Thanks so much for that page of circuits, its awesome.

I was wondering.
After I am able to set my offset to 1 volt, if I adjust R2 and change the gain won't that change my offset as well?
I want to have an output of 1-5 volts regardless of my input voltage.
From what I can tell if I have a gain of 1 and an input of 1 volt, I will have an output of 1 volt...which is good for my zero (offset).
Then if my max input is say 7.5 volts and I adjust my gain to .666 in order to provide an output of 5 volts, that will change my offset to .666 volts when I have a zero input of 1 volt, won't it?

Thanks again for the help,
Adam

 

[AdamMayotte]

woops, I didn't do my math right...I guess I could only have a min gain of 1.

 

[LionelHutz]

Is this a 1-off circuit. If so, Phoenix Contact has a module that takes a pot input and provides either a 0-10V or 4-20mA output. I'm not going to do the search for it but I know it's there.

http://www.phoenixcontact.com/

 

[ScottyUK]

PR Electronics also has one, the PR5111A. It is dead easy to use and they will configure it for you at the factory if you tell them what you want. We've been using them for a while now and they have been so reliable that we have almost forgotten they are there. Helpful and knowledgeable staff too.

http://www.prelectronics.co.uk/filer/5111uk.pdf

----------------------------------

Sometimes I only open my mouth to swap feet...

 

[Skogsgurra]

"I guess I could only have a min gain of 1"

Yes, for the non-inverting, but a simple voltage divider on the input helps. An inverting amplifier can also have gains < 1.

Agree that there are quite a few midules that does almost anything you need. BASI is another make. They can cive you a version that takes 100 data pairs to produce any characteristic you need.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Skogsgurra]

...and, yes. Of course. You get interaction between offset and gain. If you can't live with that, you have to combine an amplifier to get G*Vin and an adder to get G*Vin + Voffset. A simple matter and there are dual as well as quad opamps at no cost at all.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[AdamMayotte]

Well...

Thanks a lot for everyones help.
I just really like playing with this stuff but unfortunately I'm not too good at it yet. It doesn't seem to want to come naturally.

Thanks all,
Adam

 

[AdamMayotte]

Hello it's me again.

I have sat down and designed what I want and from what I can tell it works well. I built the circuit in a simulator and it runs as I would like.
I am wondering though. How do you know what parts to buy to make it? There are 1000's of different types of op-amps and it seems a little hellish to try and determine which one is the best.
I would like to post the circuit but can't seem to figure that one out either. All I've been able to find is a way to refer people to a website photo...which I don't have.

Adam

 

[Skogsgurra]

Good to hear!

You can post pictures in several ways. Keith Kress has written an FAQ about that: faq238-1161

Which opamp to use? Go for internally compensated "ordinary types". No expensive ones, no fast ones, and - if you only have one single supply - ones that work with a single power supply. That leaves quite a lot to chose from, the LM324, the TI 081/082/084 bunch, LM358. If you want to go nostalgic, you can always use the '741 and '301, but they will need a dual supply.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[AdamMayotte]

Thanks for parts suggestions.

Here is the picture:

 

[AdamMayotte]

Sorry,
I will give an explanation.

Purpose:
to provide a 1-5 vdc output indicating an arm position

As R_load changes in resistance (indicating a change in direction of an arm) there is an increase in voltage to INPUT.
INPUT provides a voltage to the GAIN circuit which multiplies the INPUT voltage to provide a scaled output. If the GAIN voltage needs adjusting a voltage can be added by the OFFSET circuit.
Both the GAIN and OFFSET voltage are added to provide a summed OUTPUT to the voltmeter.

 

[Skogsgurra]

Can't see what R6 and R7 are feeding. Guess the summation point of yet another inverting opamp. I guess that the use of course/fine rheostats (yes, they are not potentiometers) is dictated by the simulator. You should not do that IRL. I have a few things to say about the practical side of the schematic.

1. Use the potentiometers as potentiometers. That will give you better range and better control.

2. Do not use 3 Mohms if you are going to use standard opamps. They have bias currents close to microamps - and that means temperature dependent errors in the volts range. Use resistors in the 1 - 100 kohms range. Much less errors and less noise pick-up.

3. Do the simulation using a 741 and then with a TI084. Compare results, you will understand what I mean with errors due to bias and high resistance values.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Nice posting job Adam!

Now do as Gunnar suggests and pick a few common amps to stick into your simulator. Your simulator should have them already included. You can run a Montecarlo on them and see the wild "outback" ends of your choices. The cheapest is probably an LM358. I go thru about 1,000/year. I'm not thrilled with their warm up drift which is substantial, but if your circuit's application allows for a brief warm up you'd probably be good to go.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com