Re : MHDDK analog I/O examples don't work on a PCI-6251

[phl]

Hi,

        I am using the MHDDK package to control a PCI-6251 board on a VxWorks OS. The digital I/O sub-system seems to work correctly, but the analog I/O sub-system doesn't. The analog input values are always the same (about -8.0e-4) for all the 16 inputs, and the 2 analog outputs give always 13,93V. Using the aiex1.cpp (On-demand acquisition) and aoex2.cpp (multi-channel On-demand generation) examples gives exactly the same results. Does anyone have some solution to this problem?

Thanks,

phl

[Nacer_M]

Good afternoon,

Do you properly connect signals? and do you use the good mode ? (DIFF, RSE, NRSE)?

http://digital.ni.com/public.nsf/allkb/1FFCBE3451ABEE09862562970059881D?OpenDocument
http://digital.ni.com/public.nsf/websearch/D509679FFAE2764386256297005D0C9D

[phl]

Thank you for responding.

     In the operational configuration, my PCI-6251 is connected to a SCC68 connector block with 4 SCC Lowpass Filter modules providing 8 filtered inputs signals affected by a static gain of 0,5. So in this configuration the NRSE mode is necessary with the AI SENSE connected to the AI GND (the SCC-LP modules are Single Ended modules with an input range of +-10V [2 differential inputs per module] and a static gain of 0.5). So the analog inputs corresponding to the filtered signals are set to +-5V range and the other analog inputs are set to +-10V range and are all configured in NRSE mode.

     But the problem is that the example programs of the MHDDK (aiex1 and aoex2) don't work for the analog inputs and also the analog outputs. Of course in that case the analog inputs are set to DIFF mode with a range of +-10V and the inputs located behind the SCC-LP modules should give erroneous values, but the results are exactly the same in all cases and the same for all the inputs values (the 16 inputs in NRSE mode and the 8 inputs in DIFF mode). For the 2 analog outputs, they are blocked to 13,93V in all cases. Are you sure that the initialisation sequence of the example programs are correct for a PCI-6251?

Thanks,

phl

[Nacer_M]

Good morning,

There are two cases:

1)Your board is damaged, but you must check in a computer with DAQmx which is free (in Windows XP) :

http://joule.ni.com/nidu/cds/view/p/id/965/lang/fr

If in MAX, you have the same problem, i invite you to contact NI.

2) It's using MHDKK and i invite u to see this note:

http://digital.ni.com/public.nsf/allkb/2D93070A3DDEFD7186256C59007289E6?OpenDocument

In this case i can't do more, and i hope that a specialist of MHDDK will help you.

Nacer Mokhtari

NI France.

[Tom_W_[DE]]

Hi phl-

The NI 625x devices require a couple of extra settings that are shown in the examples but are commented out by default.  Is it possible you overlooked them?  Diego describes these two changes (including a bug in adcReset() and aiConvert polarity) in this post.

Message Edited by Tom W [DE] on 03-17-2008 09:16 AM

Message Edited by Tom W [DE] on 03-17-2008 09:16 AM

[phl]

Hi Tom,

Thanks for your reply.

The two functions described in this post (adcReset and aiPersonalize) have been correctly encoded and called (the Mseries examples given in the MHDDK are already configured for the 625x devices). And the same examples ported on a Win32 system, executed in a command window and using the "Windows WDM driver" work correctly. So the examples given in the MHDDK work correctly.

In fact, the problem is that the industrial PCI bus I use don't have any +12V and –12V support. As the PCI-6251 device uses the +12V for the analog converters these conversions don't operate. The solution is to apply an external supply for the +12V. Is there a solution to supply the +12V directly on the device, or is it necessary to supply the +12V by the mean of the PCI bus?

Thank you in advance for your answer.

phl

[Tom_W_[DE]]

Hi phl-

Thanks for the update.  Unfortunately, there is no recommended or safe way to supply the +-12V rails to the device other than for the PCI slot itself to supply them.  So, I would recommend updating to a new power supply and/or motherboard as needed.

Message Edited by Tom W [DE] on 03-28-2008 09:15 AM

[phl]

Hi Tom,

          I just want to tell you that the problem is solved. I have installed a DC-DC converter in the control system box to provide the +12V on the PCI bus. Now all the functionalities of the board work correctly.

          I hope this experience may be useful for anybody who is confronted to that problem.

phl

[Tom_W_[DE]]

Hi phl-

Thanks for the update, and glad to hear you're up and running now.  Let us know if you run into any other problems.

Thanks again-