Signal Generator 1vp-p with a DC offset

[Chris_P]

I posted this same question at the end of another topic, but I thought I'd start a new topic with it. Sorry about the double post.

I have built, with a lot of help, a digital thermostat using an LM335 temp sensor and an LMC662 opamp as a differential amplifier, to give me the full 10 bit accuracy on an AVR ADC over a restricted temp range. But calibrating it is a puzzle to me. My understanding of electronics is pretty basic. I have been reading and studying a lot but I am still stuck at a level where most of it goes over my head.

What has been suggested, and would be ideal is to build a signal generator with a 1V p-p amplitude that I can offset. Basically I need to go from 2.532V to 3.532V pretty accurately. Once I have this I can calibrate my thermostat circuit to output 0 - 5V.

Any suggestions on how I build something like this would be greatly appreciated.

 

[pc88]

How about start with something like this circuit:

https://www.electro-tech-online.com/custompdfs/2007/06/K23.pdf

Then to add an offset voltage you can use another op-amp like this page describes for the "summing amplifier":

http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/opampvar5.html

To generate the offset voltage, just use a pot (like 10K) connected between Vcc and GND. The wiper terminal will be your offset voltage.

 

[Chris_P]

How about start with something like this circuit:

https://www.electro-tech-online.com/custompdfs/2007/06/K23-1.pdf

Then to add an offset voltage you can use another op-amp like this page describes for the "summing amplifier":

http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/opampvar5.html

To generate the offset voltage, just use a pot (like 10K) connected between Vcc and GND. The wiper terminal will be your offset voltage.

Thanks for the tips. I sort of get the idea with the first diagram to build a function generator. All I really need would be the square wave generator. What I am not sure on is how I restrict the wave to 1V peak to peak? Is that something so obvious I should see it?

How does the summing amplifier provide a DC offset? I'll keep reading ....

 

[pc88]

You can provide the offset and the 1v p-p amplitude using just one op amp. The summing amplifier use of the op amp performs a weighted sum and amplification. So, if you are generating, say, a 9v p-p signal, just amplify it by 1/9. In the example above, set R1 = R2 = R and R3 = 1/9 R. Then

Vout/R3 = V1/R1 + V2/R2 or Vout = 1/9*(V1 + V2).

V1 is your signal and V2 is a voltage you've established using a pot.

For a good in-depth treatment of op amps try this out:

https://www.electro-tech-online.com/custompdfs/2007/06/slod006b.pdf

It also goes over some very useful circuit theory which will make your electronics life a lot easier!

 

[pc88]

I should mention that in practice there will be some complications due to
real-world operating limitations of op amps. For instance, an op amp cannot
produce a voltage outside of its supply rails. So, if you start with a 9v p-p
signal and want a 1v p-p output with a 0-9v variable offset, you might have
to use two op amps: the first to attenuate the input signal to a 1v p-p level
and the second to add the offset. Also, things are a little more complicated if
the op amps you want to use are single supply instead of dual supply, but the
basic idea will be the same.

 

[Chris_P]

Thanks for all the info pc88. I am reading the info on op amps that you linked to. I think I need the dummies guide. I'll start by building the square wave generator and then go from there.

 

[Roff]

Do you have an accurate voltmeter? If not, you won't be able to calibrate your circuit.
I personally don't think the square wave generator/scope method will be accurate enough. You can't measure 5 volts accurately on a scope. I would just generate the two DC reference voltages and connect them one-at-a-time to your thermostat circuit. Measure the output with your voltmeter.

 

[Chris_P]

Do you have an accurate voltmeter? If not, you won't be able to calibrate your circuit.
I personally don't think the square wave generator/scope method will be accurate enough. You can't measure 5 volts accurately on a scope. I would just generate the two DC reference voltages and connect them one-at-a-time to your thermostat circuit. Measure the output with your voltmeter.

I was wondering about this Ron. I have only just bought a scope and was working out how to use it yesterday. Soon came to realise that they are not that accurate. So I ordered a new multimeter, one that measures 0 to 5 in mV. And I'll do exactly as you have recommended, make 2 reference voltage points and calibrate to them. I'll probably play with the other idea too, a good learning experience. Thanks.

 

[Roff]

I was wondering about this Ron. I have only just bought a scope and was working out how to use it yesterday. Soon came to realise that they are not that accurate. So I ordered a new multimeter, one that measures 0 to 5 in mV. And I'll do exactly as you have recommended, make 2 reference voltage points and calibrate to them. I'll probably play with the other idea too, a good learning experience. Thanks.

You should buffer your references. see below.