ADC Gain PCI-6221

[Hosehead]

I am wondering about the gain used in the ADC for my PCI-6221.  I know it is auto-selected depending on range and the range depends on my max/min inputs; in fact this article is really great in understanding how it works.  This article has an example which starts with the sentence: "For example, the PCI-6030E has an ADC with input range of 10 V."  and then proceeds to give the ADC gain for each range, very straightforward.  However, I am missing what seems to be one key piece of information: what is the "ADC input range" of my board?  I cannot find it anywhere in the specs.  (When I looked at the specs for the example board, the PCI-6030E, I could also not find anwhere that gives this 10 V.)

Thanks for any help.

[Dennis_Knutson]

Page 1 of the specs.

[memoryleak]

Very interesting article 🙂

Depending on the card there may be a number of gain/attenuation stages, as well as an unknown amount of super-sekret analog magic designed by people far more clever than I am.  By the time the signal gets to the "ADC" (in quotes, because the line can be pretty blurry on some of our products) it doesn't necessarily look anything like the input signal (although in the case of the 6030E, the article seems to suggest it's merely attenuated).

The end result is something fairly complicated, but also something you don't have to worry about at all! (Buying the card gets you the driver, which happily abstracts all that away).  If you give it your max/min input values, your card will select the best settings and return you a proper value.

Is there something special you're trying to accomplish by manually playing with gains?  It may be more helpful to answer that question 🙂

Message Edited by memoryleak on 09-21-2009 09:21 AM

[Hosehead]

---

Dennis Knutson wrote:
Page 1 of the specs.

---

Can you be more specific?

(If you are referring to the "Input Range" I don't think this is correct, at least not according to the article I mentioned.  In the terminology of that article that's "Signal Range" and the formula is:

Gain =  "ADC input range" / span of the "Signal Range"

So for a +/- 10V input "signal range" and an "ADC input range" of 10 V, the Gain is 0.5.)

[Hosehead]

---

memoryleak wrote:

Very interesting article 🙂

Depending on the card there may be a number of gain/attenuation stages, as well as an unknown amount of super-sekret analog magic designed by people far more clever than I am.  By the time the signal gets to the "ADC" (in quotes, because the line can be pretty blurry on some of our products) it doesn't necessarily look anything like the input signal (although in the case of the 6030E, the article seems to suggest it's merely attenuated).

The end result is something fairly complicated, but also something you don't have to worry about at all! (Buying the card gets you the driver, which happily abstracts all that away).  If you give it your max/min input values, your card will select the best settings and return you a proper value.

Is there something special you're trying to accomplish by manually playing with gains?  It may be more helpful to answer that question 🙂

Message Edited by memoryleak on 09-21-2009 09:21 AM

---

I'm not trying to play with the gains but since the gain is a kind of signal modification I would just like to KNOW what is happening to my signal in between the transducer and computer.  The gain is the most basic modification. 

I think it's strange that this information is not readily available, the rest of the specs are so very detailed. 

What I hear you saying is that there's a bunch of stuff that goes on in these boards and nobody except some inside NI experts can know or understand it so just stick with the specs and ignore the "how" and you'll be happy.  Then it seems the article I quoted is either misleading (at best) or outright wrong (at worst). 

[memoryleak]

Hosehead,

I'm still traversing the schematic for the 6221, but until I'm finished:  The converter for that particular card requires it's input to be at most 5V, given that you can approximate the overall gain used for the various ranges (until I learn more).

The specs are still the final word on accuracies, and the driver will mask all the scaling of the input values.  

As for the information not being readily available, there usually isn't much demand for it.  If there's something specific you're concerned about regarding the inner workings of the card please let us know 🙂  (most people are just concerned with the overall performance and capabilities)

[Hosehead]

---

memoryleak wrote:

I'm still traversing the schematic for the 6221, but until I'm finished:  The converter for that particular card requires it's input to be at most 5V, given that you can approximate the overall gain used for the various ranges (until I learn more).

---

OK, thanks for this, looking forward to your reply.

---

memoryleak wrote:

As for the information not being readily available, there usually isn't much demand for it.  If there's something specific you're concerned about regarding the inner workings of the card please let us know 🙂  (most people are just concerned with the overall performance and capabilities)

---

Totally understandable.  My surprise was more directed to missing what I thought was a fundamental and simple characteristic - the gain.  But I see how NI has tried to make it as painless as possible for as many people as possible, and the way NI presents the accuracies I agree the detailed information is not necessary.

I appreciate your helpful responses.