0-10v calibrator, basics, idea, questions, target

[sraken13]

Hello, I have nothing to with electronics but I have good electrical background. In my past Ive been programming atmega8 ~~

I want to ask simple questions and I will appreciate any answer that can lead me in close future to creating my own regulated dc power source, 0-10V, mV steps, some % accuracy for calibrating purpose.

on what components to base?

I know about adjustable voltage regulators like LM317 for example, or precision voltage reference chips with ability to trim the output but not in such a range of 0 to 10
I wish to know a bit more of basics and design of accurate voltage calibrator, lets assume; no display, no measure, output accuracy 1%, 0,5%, 0.2% or even less; 0,05 or 0,02 or 0,01. I have budget but absolutely no clue. Throw some ideas, datasheets, circuits or complex solutions for my better understanding, eventually I will grow up. Thank You.

[MosherIV]

Hi

Welcome to the forum.

So you are after something like this:
https://www.rohde-schwarz.com/us/product/r6240a-productstartpage_63493-9189.html

They are not power supplies, they are a 'voltage source'

You would be better posting in the metrology section.

Have a look at what this guy has produced
https://www.eevblog.com/forum/metrology/new-pdvs2mini-dc-voltage-reference-from-ian-johnston-reviewed/

Basically, you need a high stability precision voltage reference.
This then is the voltage reference for a high resolution DAC.
The DAC then needs to drive a voltage buffer/amplifier (try a transistor in emitter follower mode).
Once all this is working, you characterise the output for linearity and program the set point voltages into a look up table whuch is used to program setpoint voltage for the DAC.

[sraken13]

ah so i need DAC  gonna study about them.

I found forum index: Best out-of-the-box 10V reference, seems to be useful for me.
https://www.eevblog.com/forum/metrology/best-out-of-the-box-10v-reference/

that r&s 10 µV / 1 nA resolution ;DDD a bit too much for me and my needs.

main purpose of mine is sourcing mV to simulate thermocouples and sourcing V to analog inputs of plc. I already own professional equipment to do that,
but want to learn more in depth even if that thing will never get out of the breadboard. 

[Andreas]

a good entry would be Conrads article:
http://conradhoffman.com/mini_metro_lab.html

Build or buy that is the question.
Ian sells 0-10V calibrators at a very attractive price.
Considering that you will probably need years until your first working sample is built.
https://www.eevblog.com/forum/metrology/new-pdvs2mini-dc-voltage-reference-from-ian-johnston-reviewed/msg2647110/#msg2647110

with best regards

Andreas

[iMo]

If you want to learn something you may try the PWM source below (an example).

The PWM 5V signal is generated by arduino (your atmega8 for example).
At the output you will get around 1.3-10.00V with PWM duty 0-100%.
In the simulation below the 1kHz 5V signal with 50% duty creates aprox 5.000V at the output.
The real voltage will be +/- few mVolts off because of many factors.

Mind the precision and stability and is not great. You may play with it and improve it with more skills.

PS: add 100uF/25V capacitors at the input and output of the LM317.

[ndzied1]

Attached is a circuit I found and breadboarded and it seems to work. Besides 0-10V output you can also get 4-20mA output.  I'm actually trying to work on a small PCB version that can be powered from 24VDC or 3x9V Batteries.

[fcb]

The 317 output (OUT) is 1.25V higher than the adjust (ADJ) pin - if you set the ADJ pin to 0V, you will get +1.25V out of your circuit.

So with a simple single rail 317 regulator circuit you can make a 1.25V to 10V voltage source.

With a nice op-amp, voltage reference a few passives you could make an OK voltage source with PWM control.

Don't forget that the accuracy of your PWM output will be related to your microcontroller voltage rail unless you use external switches for the PWM.  There are some schematics of a Fluke calibrator reference generator somewhere out here, that makes fascinating reading if you want to build super-precise voltage source controlled via PWM.

[Kleinstein]

A LM317 is not really suitable for a reference source, even at the lowest level. The TC is relatively poor, there are no separate sense pins (even if just inside the case) and the voltage reference and power stage are in the same case. So the temperature will change with loading.

About the lowest level to consider would be an LM723 regulator with an external power transistor (even if small like BC548).
A zener reference like LM329 (still relatively cheap) and OP (like OP07) should be more precise. For good accuracy without a good meter at hand, I would consider a ready build cheap ref circuit (e.g. from China) build around some AD58x chip supplied with measured actual value. With some luck the readings are really from that unit.