Please recommend a counter/timer card for a encoder to find phase angle

[waruna1993]

Hi, 

We are Using PXIe 8135 controller and PXIe 1082 chassis. I want a card compatible with this that can acquire data from an encoder.

Our current requirement is to get data from an encoder with 1000 pulses per revolution rotating at a rate of 6000 rpm.

So the card should be able to read 100,000 pulses per second.

We are using this setup for a vibration analysis. To find the phase angle. (We already have a vibration card).

To get the exact phase angle we need to observe the rising edge of each encoder pulse with a minimum time error.

Therefore sampling at the Nyquist sample rate won’t be enough.

If we can go for a higher sampling rate, it will be the ideal scenario.

I’m thinking of purchasing PXI 6608 (Counter timer card for this purpose).

I would appreciate you guys expert opinion on whether this card will be suitable for our application or not?

And can I use a digital I/O card for this purpose instead of a Timer/counter card?

If so, could you please recommend a card for this application.

I look forward to hearing from you guys

[Kevin_Price]

Only time for brief thoughts:

1. a PXI-6608 is likely suitable but the newer PXIe-6612 is better in several ways.  The onboard FIFO is an especially important improvement for 100 kHz data rates.

2. You might also want to consider something from the PXIe-63xx X-series multifunction family.  That'll give you 4 of the same kind of counters the 6612 has, plus quite a lot of additional I/O and functionality that *isn't* available in a pure counter card.

3. Counters are a little different kind of beast than other I/O.   Nyquist considerations often don't apply in the manner that folks here seem to expect.  For example, a period measurement task would have an effective sampling rate of 100 kHz and a measurement resolution of 10 nanosec with the boards I suggested.   You don't have to "sample" at 2x or 10x the pulse rate, instead the pulses *become* the sample clock while an internal 100 MHz timebase establishes your time measurement resolution

4. No, you don't want to use a dedicated DIO card for this.  However, for future flexibility in other measurement scenarios, I'd give strong consideration to a multifunction board from the X-series family (63xx).

-Kevin P

[waruna1993]

---

@Kevin_Price wrote:

Only time for brief thoughts:

1. a PXI-6608 is likely suitable but the newer PXIe-6612 is better in several ways.  The onboard FIFO is an especially important improvement for 100 kHz data rates.

2. You might also want to consider something from the PXIe-63xx X-series multifunction family.  That'll give you 4 of the same kind of counters the 6612 has, plus quite a lot of additional I/O and functionality that *isn't* available in a pure counter card.

3. Counters are a little different kind of beast than other I/O.   Nyquist considerations often don't apply in the manner that folks here seem to expect.  For example, a period measurement task would have an effective sampling rate of 100 kHz and a measurement resolution of 10 nanosec with the boards I suggested.   You don't have to "sample" at 2x or 10x the pulse rate, instead the pulses *become* the sample clock while an internal 100 MHz timebase establishes your time measurement resolution

4. No, you don't want to use a dedicated DIO card for this.  However, for future flexibility in other measurement scenarios, I'd give strong consideration to a multifunction board from the X-series family (63xx).

-Kevin P

---

Thank you very much. I'm thinking of buying PXIe 6363. 

You are a life saver. 

[SteveD123]

Why not contact your local NI sales engineer? It is their job to answer this kind of question.