PXIe-4135 (20W) Noise specification / Board revision

[andrew_psemc]

In the PXIe-4135 (20W) specifications, there's a note about 150fA (femtoamps) noise (0.1Hz to 10Hz, pk-pk typical). The end note is note 12, which says: "Specifications in this row are typical for the following PXIe-4135 (20W) revisions: 157420C-03L, 157420D-03L, and 157420E-03L". In other words, rev. C, rev D, and rev E.

There are have been several revisions after E. I believe F and G now, maybe more.

Could someone comment on whether this specification note 12 is:

1. simply out of date (that is, it also applies to later revisions F and G)

or

2. up to date (that is, something about later revisions F and G do not have this lower noise characteristic)

?

[ChrisR. · Accepted solution]

Note 12 refers to the entire row.  Specifically, it's pointing out that specs in that row that are warranted for later revs. are only typical for the older revs, C, D, and E.  Thus the noise spec that you seem concerned about is not affected by the note, as it's called out as typical anyway.  Only the absolute accuracy gets demoted from warranted to typical.

That particular row was absent from the original spec for the PXIe-4135.  We added it to improve our comparison against competitive devices, many of which include a median filter in their specs' fine print.  We couldn't tighten warranted performance for products that had already shipped without testing to the new tighter spec.  Thus the new line in the spec is warranted for new products, that get tested to it, but typical for older ones, that shipped without the tighter testing.

[andrew_psemc]

Thanks Chris. I understood your response, but I'd like to rewrite in my own terms - partly for you to check my understanding, and partly for anyone else who stumbles across this thread.

In short, the revisions C, D, and E are older and the end note 12 doesn't pertain to newer revisions F and beyond.

Under Definitions of the specifications:

"Specifications are Warranted unless otherwise noted."

The particular column I pointed out (Noise 0.1Hz to 10Hz peak to peak, Typical) is already marked typical. Therefore, noise is not warranted, regardless of version, regardless of end note 12.

The end note 12 makes all other columns (non-noise) typical (instead of warranted) for specific older revisions C, D, E.

The reason for this end note 12 was a retroactive update to Table 4 to match a filter setting in competitor's products. The filter is explained as end note 11.

When Table 4 was updated to include this new filter setting, previous earlier revisions (C, D, and E) had already shipped (could not be tested). Only the newer revisions (F and later) could have the warranted columns validated with the new filter.

Let me know if I got any of that wrong. Thanks again!

[ChrisR.]

You got it!