Troubleshooting: analog input tracking

[BME_genuis]

I have a problem of 2 analog input channels tracking one another (Channels 6 & 7).  Channel 6 is wired to my potentiometer however Channel 7 is not wired up in the NI 1300 terminal block.  The voltage range is 0-5V.  There is actually a slight difference in their readings but they follow each other (i.e. 1.45 & 1.48).  These channels are configured in Labview 5.1.1.  Only Channel 6 is used in my code but it worries me that something is not working right especially because now Channels 12 & 13 seem to be tracking one another.

1)  Are the channels physically connected to one another?

2)  How can I check if this is a hardware or software issue?

Thanks guys!

[Dennis_Knutson]

ghosting

[BME_genuis]

Dennis,

That makes sense since I am using multiplexing mode in PCI-6071E.  If ghosting is an issue with subsequent channels why not just skip a channel for each analog input.  The reading for the channel that is wired up and being acquired is completely accurate so apparently the ghosting does not have a detrimental effect on the first channel.

This makes me wonder if this is happening also in my analog output module 1163.  I have two channels wired up subsequent to one another to control a gas flow controller.

-Scott

[Paul_C.]

Hi Scott,

Ghosting is typically only an issue with higher sampling rates due to the speed of the convert clock and interchannel delay.  All these things can be accounted for to reduce ghosting.   You could skip a channel to reduce the effects but you would lose channel count.   In addition, the 1163 is a digital card and shouldn't suffer the same ghosting effect as analog input.  The same goes for analog output on our DAQ cards.  The m-series and e-series cards use an individual digital-to-analog converter (DAC) for analog output and don't multiplex.   This means they don't have any ghosting effect.

I hope this helps,
Paul C.


Message Edited by Paul C. on 07-01-2008 09:26 AM

[BME_genuis]

I am running the DAQ at 50 Hz.  Does ghosting show up as two channels not only track each other but show almost the same value.  Can you tell me the correct way to wire up leads to NI 1300 to account for these issues?  One pair of channels (channel 13 & 14) that are tracking one another are both wired up to NI 1300.  They are leads from two limit switches however only when one switch is triggered do you see any response in the channels or in other words when limit switch 1 is triggered you see 5V in channel 13 & 14 but when switch 2 is triggered there is no response.

[Paul_C.]

Hi,

The correct way to wire the NI 1300 can be seen in the manual found here.  Take a look at pages 6 and 7 for instructions / pictures.  I assume you have Switch A to channel 13 and switch B to channel 14.  If you turn on switch B, can you see the signal on channel 14 at all?  My first inclination would be to verify that every part in the system is working properly.  I would first try and verify that both switches work with channels 13 and 14 seperately regardless of ghosting.  After that is done, I would work on the ghosting problem. I'm assuming that when switch A is turned on your can see the signal response on both channels 13 and 14.  Are the wires shielded and seperated?  Have you tried any of the suggestions in the previously posted article? 

Regards,
Paul C.

[BME_genuis]

Paul,

When you turn on Switch B you do not see a response in Channel 14 at all.  The two switches are wired in parallel with Switch A connected to Channel 13 and Switch B connected to Channel 14.  Why would I see no response with Switch B in Labview 5.1.1?  Is that channel not configured correctly and if not how would I go about that?

-Scott

[Paul_C.]

Hi Scott,

Are you sure that the switch is working properly?  Can you change the switches out and verify?  This would eliminate that as a possibility.  It sounds to me that you might not be properly sampling from the channel 14.  Have you tried using test panels in Measurement and Automation (MAX) on a single channel (13 and 14 individually) to verify?  This can be a simple way to verify the signals coming out of the switches without the introduction of possible noise due to multiplexing.

I hope this helps,
Paul C.