How Do I accurately measure digital Polses using a compact field point

[Timmar]

Hi All,

My Goal Is to measure the speed of an item by measuring the difference between operation of 2 switches.

Pulse can be between 100 msec and hours.

I require > 100 uSec Accuracy @ 100 msec, 1msec @ 1 Sec etc...

This architecture is fixed, What I am looking for is a method.

I am used to using M-Series NI-DAq's with their synchronus digital sampling or counter acquisition,

I have looked at a cFP-DI-304 but it looks like it is an asynchronus sample, dependent on CPU Loading.

The cFP-50x is promising but I can't find many code examples and there are a few "sorry for the missed pulse events" notes.

Does anyone have any suggestions or helpful hints for me?

Thanks in advance.

[Chris V]

Hi Timmar,

Why are you leaning towards Compact FieldPoint?

While Compact FieldPoint has many great use cases, it seems to me that you are simply looking for high speed digital acquisistion.  If that's all you want, I would recommend one of our newer X-series cards (not appreciably more expensive than the M-series, just higher performance) or our C-Series devices (CompactDAQ or CompactRIO depending on your application).

Have a great day,

Chris V

[Timmar]

As I said in my post, the architecture is fixed, I understand that there are many alternatives, most of them are an "Up-Sell",

Very few have extended temperature range, LV R/T,  small form factor, fair price and industrial 24V operation.

We have an existing solution that uses custom hardware and I am looking to improve the accuracy, reduce size, improve reliability and reduce overall system cost,

Compact Rio is too expensive to achieve these goals.

I experimented with a cFP-DI-304 (on change method) yesterday an got reasonable results around the 1mSec mark.

I don't have a sig-gen capable of the 12V required to stimulate the inputs at a a precise rate, so I wasn't able to confirm it's accuracy.

What I want to find out is how accurate the on-change timestamps are? Do they go beyond the quoted 1kHz sampling mode? are they prone to event loss like other NI devices?, Does CPU load effect function.

How about cFP-500/502?

I would like to be able to help myself but I haven't been able to find much useful information on the inner architcture of the cFP backplane, or associated modules.

The most I could find was that the data is transferred a bit like a shared variable.and is subject to losing samples if cpu loading is too high.

In the case of event based acquisition, I am hoping that it capures the edges from a high accuracy timebase and places it in a shared variable /buffer for collection by the software.

Thanks

[Chris V]

Hi Timmar,

I confirmed that this kind of edge detection (within the specs you are requesting) will not be possible with cFP.

R&D suggested the 6624 board for this kind of application, although I realize that it obviously not ideal for your application.

Chris V