Krypton Synchronization

[ThaDrewid]

Hi I am trying to synchronize our NI daq system with a Krypton K600 camera-based measurement system.  Our system: PCI-6052E measurement card, 4 SCXI-1001 chassis with 1102s and 1520s in them.  I also have an SCXI-1302 50-pin connector block.  The Krypton system requires two TTL signals: one to indicate a clock signal and one to indicate when to start and stop archiving records.  I am generating these two counter signals with Dev1/ctr0 and Dev1/ctr1.  What I want to do (or think I want to do) is acquire the current value of Dev1/ctr1 in addition to all the other channels so that when the Krypton system sees this signal go high, my program will too.  Thus, my program will know which records to archive to a file.  My question is what is the best way to do this?  I tried jumpering the ctr1 terminals into Dev1/ai0 in both RSE and NRSE configurations but this doesn't seem to work well as an unwanted voltage gets fed back on the ctr1 line.  Do I need to dedicate one of the channels on our 1102 cards for this purpose instead?  Thanks in advance for your help.

-drew

[Sean_C.]

Hello Drew,

Are you Daisy Chaining you 4 SCXI chasses?  If so, AI0 - AI3 will be reserved for your 4 chasses.  You should be able to wire the counter output to AI4 of your DAQ device.

Let me know if you ahve any further questions.

Regards,
Sean C.

[ThaDrewid]

Yes we are daisy chaining the chassis.  I switched the analog input channel to Dev1/ai8 and seems to work (sort of) for a voltage supply but not for acquiring the counter output voltage.  Hooking a voltage supply up to connector pins 4 & 19, I can acquire the value in my program but there is a -0.1 V offset in the signal.  I don't observe this in either MAX or my multimeter.  If I just run jumpers from my counter output (43) to ai8 (4), and the digital ground (33) to aisense (19) I am not seeing the proper voltage on ai8.  Instead all I read is the -0.1 V offset.  Doing this also halves the voltage output from my counter output.  Any suggestions?

-drew

[Sean_C.]

Hi Drew,

The SCXI Chasses each multiplex all signals to a single differential line.  This means the AI0 - AI3 AND AI8 - AI11 are being used by the SCXI Chasses.

Also, if you are using AI Sense as the reference voltage for your AI channels, make sure that you are measuring in NRSE mode.

Try using AI4 and see if the problem still occurs.

Regards,
Sean C.

[ThaDrewid]

Can't seem to get any reading on ai4 in either NRSE or RSE configurations.  I have a power supply wired to ach4 (11) and aisense (19) or aignd (1) and I am measuring with MAX but all I get in either case is +/- 10V. 

-drew

[Sean_C.]

Hi Drew,

Can you verify that you can read correctly from this channel without the SCXI Chasses connected?  Try connecting your power supply directly to AI 4 on the PCI-6052E (pins 28 and 27) and read it in RSE Mode.  If this works, I would start looking for bent pins on all cables, SCXI-1346s, and your DAQ device.

Also, have a look at the following Knowledge Base:  http://ae.natinst.com/operations/ae/public.nsf/web/searchinternal/09c41ee1e479893286256b02007cc76f?O... .
How are you connecting your SCXI-1302?  I am assuming that you are using the SCXI-1180.  If so, what are you connecting the 1180 to?  Is it a SCXI-1346, or a SCXI-1349?  You cannot use a SCXI-1349 to connect your 1180 if it is also connected to a parallel capable module (such as the 1520).

Let me know your results.

Regards,
Sean C.

[ThaDrewid]

I am using the SCXI-1180.  It was connected to the SCXI-1349 in chassis 4 which is probably why it wasn't working.  I tried it with each of the 1346s in chassis 1,2, and 3 and I only get a reading when we are plugged into chassis 1 in RSE mode.  Is it normal that this would only work with chassis 1 or should i be looking for bent pins in the hardware connecting 1 and 2? 

BTW, I can't open the link you sent.  I cannot seem to locate a host by the name of ae.natinst.com on the internet.  Is this an internal site?

-drew

[Sean_C.]

Hi Drew,

I apologize, I pasted the wrong link.  Here is the correct one:

http://digital.ni.com/public.nsf/websearch/09C41EE1E479893286256B02007CC76F?OpenDocument

You should use the SCXI-1180 in the first chassis.  One of the features of the SCXI-1346 is that it will shift all AI channels up a position (AI0 is routed to AI1, AI1 to AI2, etc).  It does this so that each chassis is routed to a different channel on your DAQ device (rather than all chassis routing to AI0).  Therefore, if you are trying to measure a channel on chassis 4, it's AI channels will have been shifted up 3 times.  I hope this helps.  Please let me know if you have any further questions.

Regards,

Sean C.

[ThaDrewid]

Ok, now I am having problems generating a square wave pulse train to control the krypton camera with.  I am setting up my task as follows:

daqmxCheckErr(DAQmxCreateCOPulseChanFreq(m_kryptonSyncTask,
                "Dev1/ctr0", "",
                DAQmx_Val_Hz,
                DAQmx_Val_Low, 0.0,
                1.0, 0.5));
                // this is commented out for testing...
                //gpDataProducer->GetArchRateHz()/2, 0.5));

    daqmxCheckErr(DAQmxCfgImplicitTiming(m_kryptonSyncTask,
                DAQmx_Val_ContSamps, 1));

    daqmxCheckErr(DAQmxCfgDigEdgeStartTrig(m_kryptonSyncTask,
                "/Dev1/ai/SampleClock", DAQmx_Val_Rising));

    daqmxCheckErr(DAQmxStartTask(m_kryptonSyncTask));

I have the counter 0 output wired to an input on my daq and am acquiring the 1 Hz signal at 20Hz.  See the bitmap image attached to this message for the waveform I am seeing.  It is not square and varies between -2.5 and 2.5V when it should vary between 0 and 5.0V?

-drew

[Sean_C.]

Hi Drew,

Can you reproduce this behavior from a test panel in MAX?  Try setting up the pulse train generation from MAX and wire the counter output to an analog input.  Then view that input channel in MAX.  Does it still vary from -2.5 to 2.5 Volts?  Is it a square wave?

If you observe the same behavior in MAX, try performing the same test without the SCXI hardware connected.  Does this solve the problem?

Let me know your results,

Sean C.