A/D converter with double integration

[Lean40]

Hi,

I'm a beginner in labview and i want to make a double intergration a/d converter. I started doing somthing like this, does it makes sense? What to do next?

[GerdW]

Hi Lean,

---

@Lean40 wrote:

I'm a beginner in labview and i want to make a double intergration a/d converter.

---

We notice the "beginner" status from the way you spell LabVIEW...

What is a "double integration A/D converter"? Can you provide information about your goal/requirement?

---

@Lean40 wrote:

I started doing somthing like this, does it makes sense?

---

Not for me...

• What's the point of the rather unused shift register and why do you need that additional feedback node after the loop???
• Why do you need to multiply with 1e-3 with each iteration inside the loop when you could do that once before the loop???
• What is the expected result of your routine?
• ...

Please sketch the algorithm you want to implement in your VI!

---

@Lean40 wrote:

What to do next?

---

I recommend to start with some beginner tutorials to learn LabVIEW basics before starting to implement your algorithm.

You should know how to use arrays, several other datatypes, shift registers/feedback nodes...

[Henrik_Volkers]

Do you want to simulate a dual slope AD converter ?

https://en.wikipedia.org/wiki/Integrating_ADC

LabVIEW is a software to read/control AD converters you usually already have as hardware.

If you don't have a dual slope converter, but want to have,  for example, the benefit of line noise suppression (if the first integration is set to line frequency periode... or NPLC (number of power line cycles *) ), you can do that in software by sampling n times faster (n=10 or 1000 or ..)  than line frequency with the AD converter (DAQ) you have and do an average (~ integration) on the data in software.

Still the most precice DMM/voltmeter references use dual slope A/D converters...  (like HP/AG/KS 3458 or Fluke 8508/8588 )

*) NPLCs greater than 1 are usually measured as an array of 1 NPLCs and averaged in the instrument 😉

[altenbach]

In terms of the LabVIEW "language", your VI is pure gibberish. You create a signal once and graph it once, then you apply the exact same calculation 1000x seconds (invariant input, same result!). After you stop the VI, the chart 2 will get updated. Most likely yout feedback node got auto-inserted because you created a circular dependency.

LabVIEW is a programming language, not a circuit simulator, but you can of course implement all the math required to simulate the behavior of such a converter.

Start as follows:

• Do a few basic LabVIEW tutorials. Check the learning resources listed at the top of the forum
• Look up the math to model such a system.
• Implement it in LabVIEW.
  • What i the signal over time?
  • What are the controls?
  • What are the indicators?
  • How does the user interact with it?
  •  How do you validate correct operation
  • etc.