No FPGA Response on LSB Weights and Offsets?

[Bassett_Hound]

Ah...I think I have a few other suggestions and need to back track a little ....

So, from looking at your pics, it jolted my memory that the IO nodes (start, stop, and read ) will have priority over the property nodes ( meaning they could block them and give the error your seeing ).  And from the picture, since there is nothing that determines the flow of events its possible that the Start DO lines and property nodes are executing at the same time.  A simple solution would be to just use a sequence structure where in one frame you read the calibration info, and then in the next write to the start line ( mos to fhe examples should do that )

Other than that you might check to see what cable you are using to connect to the expansion chassis. The recommended cable for the expanssion chassis is the SHC68-68-RMIO. You can also use the SH68-C68-S. Any other cable might cause crosstalk that could case module communication errors.

Hope that helps,

Bassett

Message Edited by Bassett Hound on 09-13-2007 03:14 PM

[Mr._Jim]

Hi Bassett,

Yep, I'm using the SHC68-68-RMIO cables.  They're a little long for what we're doing, but I don't think I'm running them near anything noisier than a usual office environment.  Holding the cable up and out of the way of anything doesn't seem to change the result.

Now - the picture you saw is of the very first implementation I had.  I've since removed the invoke nodes, but I'll try setting up a sequence structure and see if that helps.

Jim

[Mr._Jim]

Good morning Bassett, and happy Friday,

I've added the sequence structure (attached picture), and sadly, I've had little success.  In fact, the vendor ID and serial stopped coming through, too.


I switched the 9237 modules in slots 1 and 2 again.  No response from the slot 2.
I switched the expansion chassis on the same cable, on the same connector.  Still no response.
I switched to a different cable on the same port.  Still no response.

No matter what, it's always slot 2 on connector 0.  So given those results, I'm betting it's probably the card somehow.  (Either a bad connector pin or something else?)

Should I call up and get an RMA, or is there something we didn't think of?

Thanks again,

Jim

[Bassett_Hound]

Hey Jim,

Sorry, I was out of the office on Friday.  Does slot two work on Connector 1,2 and 3?  If it doesn't I would suggest calling in just so you can get on the spot help.  I'm sure we can figure it out but it would easier to have a dedicated engineer to communicate with.

Bassett

[Bassett_Hound]

One small minor correction...When I said RMIO I really should have said RDIO.  I believe the Applications Engineer working with you should be able to take it from here...

[Mr._Jim]

No problem, Bassett.  It was all definitely helpful nonetheless.

If anyone else has this issue in the future, and I sure hope not, the bottom line is that the SHC68-68-RMIO cables are the wrong ones for this application.  Apparently some of the "wrong" pins are tied to ground (at least for the PXI-7813R and probably others).  When you use certain lines for different modules, you can cause other modules to become unresponsive, hence the intermittent appearance of the issue.

As Bassett remarked in the last post, the SHC68-68-RDIO are the correct cables.

Thanks again to everyone who helped me with this.

Jim

[SLee2004]

I am running into the same problem.  I have a cRIO chassis with 4 9263 modules.  3 modules are ouputting LSB and Offset values.  The second slot outputs 0s for LSB and Offset calibration values.  I have tried to recompile the FPGA VI several times and still have the same response.  I know that my code works because it is used on several other test cells.  Please let me know if you find a solution.

[Mr._Jim]

Hi SumTumWong,

For me, it was an especially elusive problem due to having the wrong cables.  Are you sure your cables are correct? (see my last post in this thread) The RMIO cables will fit perfectly and appear to be OK in every way, but you'll have very strange, intermittent problems.  For the same reason, I'd also check your pins and the general health of the cables - if you mash a pin, you might have problems, though the problem should be pretty consistent in that case.

For what it's worth...

Jim

[Mr._Jim]

Oops... I forgot to ask an important question: Are you using cRIO, or an FPGA PXI card with expansion chassis?  If you're using cRIO exclusively (not PXI), then obviously my comments about the cables won't apply to you.

[SLee2004]

I am using an RIO NI-7813R card.  I am 100% sure the cables are correct because I have the same setup in other test cells.  I checked the pins, and they are in good condition.  I have added a delay time between downloading the code and reading the calibration values.  I don't know what else to check.