Digital Sidecar Connection Issue

[Russ Beavis]

You might want to check out the FRC Diagnostic Window app from WPI's Think Tank. It's a very useful tool for "directly" twiddling outputs from the DSC and reading analog inputs. If you're suspicious of your hardware setup and don't want to separate SW issues from HW issues, this app at least gives you a (hopefully) solid SW app to start with.

http://thinktank.wpi.edu/article/138

If I'm understanding your description (I don't have the bandwidth to download the pix right now), why are you moving the digital I/O module from the cRIO and using a 15 pin serial cable instead of keeping the module in the cRIO chassis and using a 37 pin cable? Interesting design choice... Maybe trying to save a little weight?

Have you confirmed that your 15 pin cable passes all 15 signals straight through instead of swapping pairs of signals? I'm not familiar with any 15 pin serial cables but there are certainly 9 and 25 pin serial cables in both straight and crossover configurations.

Also, there are (I believe) 20 or 40MHz SPI signals going from the cRIO chassis to the digital IO module. Passing those signals through a long 15 pin serial cable could be problematic. You may be getting inadequate comm through that cable.

Good Luck!
Russ

[Happyisgood]

You just made me think of something that I never considered before. The 15 pin serial I'm using is a VGA cable... meaning it may have some inactive pins. I'm not trying to save weight, I'm trying to save time. The serial cable that we need would take a few days to get however the VGA is readily available.

EDIT: On the cable all the pins are present, I don't know how to check if they're active or not.

[Alan Anderson]

Originally Posted by Happyisgood

The 15 pin serial I'm using is a VGA cable...

Then it's not a 15 pin serial (which I've never heard of either). It's a VGA cable. (The 37-pin cable you're avoiding also isn't quite a serial cable, though I'm pretty sure it has some serial signals on it.)

EDIT: On the cable all the pins are present, I don't know how to check if they're active or not.

The way to check is to use an ohmmeter to verify 1) continuity between corresponding pins on each connector, and 2) noncontinuity between all pairs of pins on a single connector.

But you really shouldn't be abusing the system that way. Why aren't you using one of the official 37-pin extension cables?

[Happyisgood]

I disagree with your statement that this is "abusing the system."

I am using a cRIO that DOES NOT have said 37 pin cable. If there was a way to get it where we wouldn't be spending a fortune and waiting an extended period of time for it to be delivered I'd get it. The VGA is a 15 pin M-F cable, and according to every bit of documentation that I could find it does function as a serial cable and should work in our situation.

Anyway, for that program suggested by Russ. All the comm on the table is functioning properly, however simulating a signal through the PWM still results in the Jag not going solid. Is there a chance I wired it wrong? I'm going to try the Spike that's attached...

[Russ Beavis]

I like the idea of testing relays. You don't even really need to connect any Spikes, though - there are debug lights for every FWD/REV relay output that should be able to tell you whether the sidecar is working as expected. Just twiddle all of the relay outputs to see if things are hooked up. If your setup can drive all of the relay outputs, you should be able to twiddle the PWMs as well (twiddling the relay outputs is MUCH more complicated than twiddling PWMs and GPIO since it requires a number of synchronous signals).

What's the state of the Robot Signal Light?

http://decibel.ni.com/content/docs/DOC-3120

Russ

[Happyisgood]

I probably should have noticed this before, there is one light on the dashboard that should be lit up that isnt. Notably the FPGA and COMM Ok light is not on although my system is fully enabled. I dont know what that means. I'm thinking it's probably a Labview update that I'm missing. I'm going to update and re-image to see if that helps. Yes my driver station is enabled, yes I have the robot enabled button pressed.

EDIT - By the way, the spike didnt work
EDIT AGAIN- Light (if it was connected) would be on
EDIT AGAIN AGAIN - Light would be on (confirmed by attaching a light.) Reimaging the cRIO did not work. I know there is comm between the DS and the Sidecar because all the Digital IO relays turn green and read a value of 1 when I attach it. I still can't manage to get that FPGA check to clear. Any suggestions?

[FourPenguins]

You may wish to try this http://www.chiefdelphi.com/forums/sh...sidecar+weight on a temporary basis to see if your cable is the problem.

[Happyisgood]

This may sound odd. I tested this thing alot. I found that attaching a jumper allows the center pin to put out 6V (as is proper.) Using a

Please note, if you've looked at those pictures, they are not standard PWM cables they are simply 3 pin signal cables we soldered one end to make pins and the other end already had a 3 pin clip piece. These SHOULDNT be a problem, we've tested them.

Anyway, this is really starting to irritate me, I know the sidecar is functioning, but I can't get a signal to either of my Victors or my Spike.

I've narrowed it down to three things...

a) Faulty Signal Cables
b) Two victors AND a spike all broken (unlikely, they worked before)
c) Signal issue (I doubt it, the "signal light" is putting out a constant voltage)

[Greg McKaskle]

I'm not an expert on this, but I'd suspect that the 40MHz signals from the backplane are too distorted by the cable to make the DIO module function in all of its modes. I've seen people use extender boards to get a proto board into the chasis, but never a 5 ft cable. As mentioned, if you can try it without the cable, with the digital breakout mounted upside down, or with a cable on the slow side of the DIO, it will confirm whether this is the issue.

Greg McKaskle

[Happyisgood]

Ok the problem is solved. It was the cable, I'm going to go out and buy a proper cable to go with the system. For all of you who helped so much with this problem thankyou.

I'm going to go buy a 37 pin M-F cable for use with the cRIO digital sidecar.