external reference for USB 5132

[Penny]

Hello,

I have a couple (hopefully) general questions about using an external reference clock on our USB 5132 digitizer.  We want to lock it to the other oscillators in our system, all of which are presently locked to one 10 MHz reference.  If I connect our existing 10 MHz reference to the PFI1 input, my questions are:

1.  Is the 5132 now locked to our system?  Or, do we need to specifiy in the setup software that we want to override the 5132's internal reference and rely entirely our external reference?

2.  In order to have the best possible lock, do we need to tell the 5132 to sample at a frequency that is evenly divisible by 10 MHz (e.g. 20 MHz, 10 MHz, 5 MHz, 2 MHz, etc.)?  Similarly, are we worse off if we want to sample at say, 13 MHz?

Thanks for any thoughts.

Best regards,

Penny

[Izzzzz]

Hello Penny,

1. Yes the 5132 is locked to your reference system 10MHz clock (with some offset due to delay in the time it takes to transmit the clock signal from your system to this USB device) if you selected the PFI line as the source of an external timebase. Please take a look at the following example Help>>Find Examples>>Hardware Input and Output>> Modular Instruments>>NI-Scope>>Features>> niScope EX External Clocking.vi.

2. No, although I would expect that if you enter in 13MHz and your device was not capable of doing this it would coerce it to the closest value. I would suspect that only integer divide downs from the clock in question are valid such that 12. 5MHz is the value it would coerce to if you were to request a 13MHz value (with a 50MHzexternal clock). The 5132 can not do RIS so sampling at 20MHz with a 10MHz SampleClock Timebase Source is not feasable. 

Regards,

Izzy O.

Applications Engineer

National Instruments

www.ni.com/support

[Penny]

Thank you Izzy.  I believe that I understand both of these answers but will take another look at our setup to make sure.

Best regards,

Penny

[Penny]

Hi Izzy,

I took a closer look at the VI that you included - thanks.  If I translate it to C#.NET and want to run our USB 5132 at 10 MHz with a 10 MHz external sample clock, then if I look at http://zone.ni.com/reference/en-XX/help/370592N-01/scopepropref/pniscope_sampleclocktimebasesource/ I would add the following 2 lines to my C# code:

sampleScopeSession.Timing.SampleClockTimebaseSource = "PFI 1";

sampleScopeSession.Timing.SampleClockTimebaseDiv = 1;

Does it sound like we are all set here?  Do we need this line too, or would this be too much information:

sampleScopeSession.Timing.SampleClockTimebaseRate = 1.0e7;  // 10 MHz

Maybe it is enough to just specify the Source as "PFI 1", connect our 10 MHz reference, and then everything else will fall into place by default?

Thanks again,

Penny

[Izzzzz · Accepted solution]

Hello Penny, 

You will need all three. The "sampleScopeSession.Timing.SampleClockTimebaseRate = 1.0e7;" is specifying the rate of the SampleClockTimebase Source. The only piece of the code which you will not need is the "sampleScopeSession.Timing.SampleClockTimebaseMultiplier " because your device does not support RIS.

Regards,

Izzy O.

Applications Engineer

National Instruments

www.ni.com/support

[Penny]

Hi Izzy,

OK thanks!

-Penny

[Izzzzz]

Hello Penny, 

Just to clarify and correct my above statements.

1. The method I sugested does not lock your USB-5132 clock to the 10MHz clock external clock it replaces the 50MHz clock on board with the 10MHz external clock. Thus your maximum sample rate will be equal to the external clock you connected to the PFI line if you decide to utilize the method I have suggested.

2.  Instead of coercing the sample rate down like I suggested the card will acutally coerce the sample rate up. In the previouis example if you asked for a sample rate of 13MHz then the card would result in giving you a 16.67MHz clock (50MHz/3).

Regards,

Izzy O.

Applications Engineer

National Instruments

www.ni.com/support

[Penny]

Hi Izzy,

OK thanks.  For now we are just sampling at 10 MHz so I believe we are all set.  We just set the source to PFI1 and it is running.  It stops running if we disconnect the cable.  The data confirm that the sample rate is 10 MHz.  Good to know what our other options are if we need to sample lower.

Best regards,

Penny

[Penny]

So if I can just 'sum up', if I just specify

sampleScopeSession.Timing.SampleClockTimebaseSource = "PFI1";

with no other commands, and then send in our digital 10 MHz reference to the PFI1 port, I think that we are now successfully locked to the 10 MHz and sampling at 10 MHz.  I think that this is because:

1.  We now have our external clock.

2.  The default value of sampleScopeSession.Timing.SampleClockTimebaseDiv is 1.

3.  The default trigger configuration is "immediate". 

Does this sound like a reasonable thing to do?  Or instead, do you think we should specify the TimebaseDiv, the SampleClockTimebaseRate, and then somehow specify that the digitial PF1 is our trigger source?

Thanks again,

Penny

[Izzzzz]

Hello Penny, 

I would still set the timebase rate in order to tell the card that the signal coming in is 10MHz. You do not need to set the TimebaseDiv since the default is 1 and you wish to sample at 10MHz. Since the PFI line is aleardy in use to bring in an external clock you typically cannot specify it as a trigger source, and should just leave the default trigger source as immediate. 

Warm Regards,

Izzy O.

Applications Engineer

National Instruments

www.ni.com/support