SIGNAL LIGHT WONT WORK [Archive]

vhcook

05-02-2010, 17:20

Have you confirmed your wiring matches both the Robot Power Distribution (http://www.usfirst.org/uploadedFiles/Community/FRC/Game_and_Season__Info/2010_Assets/2010_Robot_Power_Distribution_RevA.pdf) (with correct polarity) and Robot Data Connectivity (http://www.usfirst.org/uploadedFiles/Community/FRC/Game_and_Season__Info/2010_Assets/2010_Robot_data_enc.pdf) diagrams, paying particular attention to the power and signal wires to the digital sidecar?

Is the digital module on the cRIO in slot 4 (with 1 being the slot nearest the lights)?

Are the lights green on on your cRIO?

Does anything else on the digital sidecar work (PWMs, sensors, whatever)?

kick3nitx360

05-02-2010, 17:34

yes all that is done correctly. But the light that is inside the digital sidecar where you need to plug in the signal light is also not blinking

jsasaki

05-02-2010, 17:45

Oh ! I had the same problem last night. Do you have two of the 40 pin sidecar modules? If so try switching the two modules and if it still doesnt work try another 40 pin cable and if it still doesn't work try re-imaging the c-Rio. If it still doesnt work i'm not sure.

Al Skierkiewicz

05-02-2010, 21:23

Let's try a different approach. Are the three power supply LEDs lit on the sidecar? If not is there a breaker in the PD that is feeding battery to the sidecar? Have you rechecked the wiring at both the PD and the sidecar? It is a frequent mistake that wires at the PD end are not stripped the required 1/2-5/8 inches that is needed for correct insertion. Something less allows a user to insert the wire so far that the Wago will clamp down on the insulation and not the bare wire. Double check that the wire is correctly inserted at the sidecar end as well.
If the battery LED is on and the other lights are not, try removing the Robot signal light from the sidecar. If they come on with the RSL removed, you may have shorted the wiring for the RSL. If they do not come on, it is possible that the card is defective.

peter1626

14-02-2010, 16:20

Hi, i am another programmer from 1626.
Have you confirmed your wiring matches both the Robot Power Distribution (with correct polarity) and Robot Data Connectivity diagrams, paying particular attention to the power and signal wires to the digital sidecar?
Yes
Is the digital module on the cRIO in slot 4 (with 1 being the slot nearest the lights)?
Yes
Are the lights green on your cRIO?
Yes
Does anything else on the digital sidecar work (PWMs, sensors, whatever)?
Yes the entire system works except for the signal light. On the digital sidecar we have 4 victors, 1 spike, 1 sensor, and 2 servos. All work correctly.
If so try switching the two modules and if it still doesn’t work try another 40 pin cable and if it still doesn't work try re-imaging the c-Rio.
We have tried 4 modules, 2 40 pin cables, 2 digital sidecars, and 2 cRIOs. We also have reimaged the cRIO numerous times.
Are the three power supply LEDs lit on the sidecar?
Yes
If not is there a breaker in the PD that is feeding battery to the sidecar?
Yes the sidecar is in the 30 amp spot with a breaker
Have you rechecked the wiring at both the PD and the sidecar?
Yes we have rewired the entire system 4 times
If the battery LED is on and the other lights are not, try removing the Robot signal light from the sidecar. If they come on with the RSL removed, you may have shorted the wiring for the RSL. If they do not come on, it is possible that the card is defective.
The 3 power/ battery lights are on the digital sidecar. The Signal light still doesn’t blink and when we remove the signal light the small green led does not blink.

Please HELP! We don’t know what else to do!
Thanks for your help!

samir13k

14-02-2010, 16:48

attempt to download the original code to the cRio once again. it should not make a difference as it should not be changable in the code, but in FIRST, sometimes you just have to try anything.

peter1626

14-02-2010, 16:53

attempt to download the original code to the cRio once again.
Just tried this, it did not work!

What is the current image name that should be on the cRIO?

apalrd

14-02-2010, 17:29

Try picking up your robot, flipping it over, and banging on it, then pick it up and shake it. This will remove all metal shavings, which can short parts of the DSC. We have had to do this twice already on our robot, we know it happens when the +5 light on the DSC goes off and nothing works, but the DSC still has power lights on.

Alan Anderson

14-02-2010, 18:52

Is the digital module on the cRIO in slot 4 (with 1 being the slot nearest the lights)?

Yes

Can you take a picture and post it? The only thing I can think of that matches your symptoms is having the module in Slot 6 instead of Slot 4.

Pausert

14-02-2010, 20:30

Check to make sure have the Digital Sidecar power wired + to + and - to -. The digital sidecar will function (partially) even if its supply power is backwards, but the signal light will not function in this circumstance.

Al Skierkiewicz

14-02-2010, 20:47

Peter,
You have tried everything I would have gone through. That really leaves only the side car as being defective. If the PWM signals that pass through are functioning normally and are disabled when the robot is disabled then it can only be the side car.
A software expert can answer this better than I but the default code is written such that the RSL functions when the robot is disabled and flashes one of a series of codes that indicate certain conditions on the robot. When enabled, the RSL light should come on solid.

"There are four states of this status LED, listed here in order of priority. If multiple occur, the highest one listed is what is displayed.

1. Fast Blink (200ms on / 200ms off) - System error: No driver's station communication, bad cRIO Image, bad team ID, extensive communication errors.
2. Steady on - Everything ok
3. Fast-Slow Blink (200ms on / 900ms off) - System disabled either by system watchdog, user watchdog, or Driver's Station set to disabled and low battery (<12 Volts)
4. Slow Blink (900ms on / 900ms off) - System disabled"

samir13k

14-02-2010, 23:41

Just tried this, it did not work!

What is the current image name that should be on the cRIO?

image v19 would most likely be on your cRio. If you installed the latest update (out since fridayish?) you should currently be at v20 which you will need to do. download and install both the Labview and DS updates. I dont believe that this will fix this problem if you are already running v19, but its a mandatory update non-the-less

kick3nitx360

15-02-2010, 11:42

Here are the pictures you asked for.

http://i349.photobucket.com/albums/q394/kick3nitx360/IMG_0052.jpg

http://i349.photobucket.com/albums/q394/kick3nitx360/IMG_0053.jpg

kick3nitx360

15-02-2010, 11:51

yes samir, we are using v19 when we image the cRIO.

samir13k

15-02-2010, 12:08

when you wire the signal light, do you have the jumper wire in there? you need to run positive to one of the outsides, negative on the middle one, and positive on the other outside... you can do this by soldering the positive to split into two, or by just putting a jumper wire from one side to the other...

here comes cheesy paint drawing:
http://i213.photobucket.com/albums/cc160/samir13k/RSL.jpg

kick3nitx360

15-02-2010, 12:34

Mr. Lim

15-02-2010, 12:53

Sounds like a bad DSC...

Can you grab last year's DSC and swap it out to test?

If this one is bad, open it up, clean it out, and maybe lightly touch the ICs surrounding the RSL connector to see if anything is getting excessively hot while powered on.

To be honest, I don't even remember whether the little LED near the RSL connector is supposed to blink in unison with the actual RSL.

Can anyone here confirm?

Al Skierkiewicz

15-02-2010, 12:57

There is an LED on the sidecar that should blink in unison with the RSL. It will function without the RSL even connected. However, there is a "fuse" (possible solid state fuse) in series with the drive circuit. It is possible that the fuse has opened/failed. Please try a different DSC before we go much further.

samir13k

15-02-2010, 13:08

Sounds like a bad DSC...

To be honest, I don't even remember whether the little LED near the RSL connector is supposed to blink in unison with the actual RSL.

Can anyone here confirm?

Yes, there is a little green LED on the DSC right next to the output for the RSL power leads.

I'm starting to run out of ideas here. Theres a rare possibility that you have two bad DSC's maybe? it sounds like you have two control systems, maybe build the benchtop system with the second system, and see if that works?

It also looks like you are running some pretty thin wire to power your DSC. Maybe try something thicker? i believe that i use 16AWG on ours. I reccomend doing the same thing on your Analog Breakout. The wire you are using on that is meant for pneumatics.

Jon Stratis

15-02-2010, 13:10

When you try a different DSC as Al suggests, i would make one more suggestion - make the wiring as simple as possible to test it. By that, i mean keep the DSC off the robot, run two power wires to the PDB and the connector to the cRio, and that's it. then power on the robot and let it boot up. That way, you can be absolutely positive none of your other connections are causing any problems.

If the LED on the DSC works in that situation, start making other connections, checking the LED at every step. If the LED dies at some point, at least you know what connection it was!

My gut instinct says you have/had a short in the wiring for the signal light that blew the fuse Al mentioned... even if that short was later fixed, without the fuse the light still won't work!

peter1626

15-02-2010, 14:19

I was just wondering if it could possible be my cRIO image:confused:. Is there any way I can program the light to blink with labview just to eliminate the chance that it is a hardware problem:confused:

peter1626

15-02-2010, 16:22

We have tried different modules in the cRIO, 4 DSC, 2 cables from the cRIO and 2 lights. We have re-wired AGAIN this time with the wire you recommended however this still did not work! >:( I am currently in the process of updating and will let you know if that works! :)

Jon Stratis

15-02-2010, 16:45

Peter, lets go back to the beginning here, because I'm suspecting more and more the fuse Al mentioned, and a short somewhere in your wiring. If, for example, you wired up the signal light incorrectly, such that it permitted such a short, and successively tried all 4 side cars with this same wiring, it's possible you blew the fuse in all 4 side cars. Even if you had fixed it (the second signal light you tried), because they were already blown it wouldn't help.

Where is your team located? It could be handy to have another team nearby assist with some debugging. For example, if their signal light works, you could try replacing their side car with yours - if the light now doesn't work, then the problem is in the side car. If the light does work, then we know the side car is not the issue and can start tracing it back from there.

kick3nitx360

15-02-2010, 17:46

We only tried the first DSC with the signal light we didnt try the other ones. However, we are located in Metchen New Jersey and would be very happy to have another team come out and help us with this urgent issue. We would also be willing to go to anyone near by. We are all at our whits end and really need the help.

Al Skierkiewicz

15-02-2010, 19:05

Peter,
Try another DSC without inserting the RSL connector. You should be able to see the LED on the board flashing when disabled and steady when enabled. It is entirely possible to have a shorted RSL that will hold the LED on the board in an off state. Turn off the robot before you remove the connector, count to ten and then turn the power on. It this is solid state fuse, it will not reset until the power has been turned off. If that works, use a VOM to test the connector and RSL wiring. It should not be less than 500 ohms in both directions.

chris1592

15-02-2010, 22:51

Are we talking about the orange beacon?

Mike Betts

15-02-2010, 22:59

Are we talking about the orange beacon?

Yes.

Alan Anderson

16-02-2010, 00:43

Are we talking about the orange beacon?

Yes...though we're also talking about the little green LED on the Digital Sidecar that blinks along with it.

Mark McLeod

16-02-2010, 08:35

I do have to say...

Most of the programmers think of the RSL as the LED on the Digital Sidecar, because it's all we have until the electricians wire the Big Yellow Light (BYL).
Most of the electricians think of it as the big yellow light connected to it as a repeater, because they get to wire it.

We all know it is both...

It's labeled "RSL" on the Digital Sidecar.

peter1626

16-02-2010, 08:58

Try another DSC without inserting the RSL connector
We tried this and none of the other 3 rsl lights on each of the dsc work.

As one of you suggested I sent one of our mentors home with a DSC that does not blink and he will bring it to another team he knows and have them try it for us. When he returns with it I will let you know weather the other team got it to work or not!

peter1626

16-02-2010, 09:00

Is there any way to "force" the light to work with code or something else??

Mark McLeod

16-02-2010, 09:02

We tried this and none of the other 3 rsl lights on each of the dsc work.
Umm, that means it's not wired to the Power Distribution Panel.

White Wago on the DSC to a pair of red/black wagos on the Power Distribution Panel - mind that the wires aren't caught on the insulation instead of the bare wire.
Here's a photo (http://www.chiefdelphi.com/forums/attachment.php?attachmentid=8455&d=1264980108)

As a side note: We call those three lights to the left of the Wago connector the power status LEDs, that's why they are labeled "bat, 6v, 5v"
The RSL is only the LED to the right of the Wago connector.

Mike AA

16-02-2010, 09:24

I wish I had been here earlier. In the picture you can clearly see there isn't anything connected to the wago connector for the DSC.

Alan Anderson

16-02-2010, 09:50

I wish I had been here earlier. In the picture you can clearly see there isn't anything connected to the wago connector for the DSC.

What picture are you looking at?

There are minor wiring issues in what Dave showed us, but an obviously empty Digital Sidecar power connector is not one of them.

peter1626

16-02-2010, 10:00

Mark McLeod;921167]Umm, that means it's not wired to the Power Distribution Panel.
I was not talking about the Power indicator lights i was taliking about the 1 RSL light on each of the 3 DSC totaling 3 RSL lights that dont work.
AND
It is wired EXACTLY as you show in the picture.

I wish I had been here earlier. In the picture you can clearly see there isn't anything connected to the wago connector for the DSC.
If this was true the 3 power indicator lights would not be on also the entire DSC works EXCEPT for the RSL led and the big orange light.

There are minor wiring issues in what Dave showed us
If you are talking about the small wires, I have previously mentioned that we have corrected this issue by changing the wires to 16 AWG as recomended.

Mark McLeod

16-02-2010, 10:13

Sorry I misunderstood your "3 RSL's".
The RSL is automatic as far as we're concerned. Our code cannot control it.

Does your code or the default code work after you download it?
If it doesn't do anything then I'd have to agree with you that there is an issue with the cRIO running that you have to track down. The RSL may be out because there really isn't any robot code.

Al Skierkiewicz

16-02-2010, 11:13

Mark and Alan,
I will defer to you two on this. Is there a step in the procedure that then saves the image to the Crio instead of merely loading it? i.e., the image then reverts to a previous saved image when the power is cycled. (like some step is being skipped in this case) I seem to remember something odd about this last year, but as you know, I am not the software expert.

Mark McLeod

16-02-2010, 11:48

There are generally two different ways to download our own code to the robot.

One is temporary where it resides only in memory and only as long as your environment, for instance LabVIEW, is running and connected to the cRIO.
When this is stopped the RSL will go out altogether.

The other way is to permanently download the code and that only starts to run after the cRIO is rebooted. The RSL should always be on after any reboot completes.

The symptoms here could happen, for instance, if the user temporarily ran the code from say LabVIEW, then changed User accounts to start the Driver Station. The act of changing user accounts stops all processes.

I'll have to go play with the cRIO to replicate when the RSL is out altogether. My memory may be faulty on when exactly the RSL is out, because it's become an automatic recognition if I see it in person.
The RSL is basically out whenever no code is actually running on the cRIO.

DSST\neal.ian

16-02-2010, 12:11

One reason for us that we weren't able to make this light work was because we didn't have the jumper on the light. You might want to check that.

peter1626

17-02-2010, 09:32

Yes, we do have the jumper cable on the signal light. However, its not so much the actual light that we are worried about, the little LED signal light that is in the digital sidecar doesnt even blink.
Once again the problem is not our BIG ORANGE LIGHT it is the little green LED that is on the DSC the one that blinks in sync with the big orange one! If that little green one does not blink there is no chance that the big orange one is going to blink!!!!

And for the last time we have our jumper on the light, but it does not matter because we are not plugging in the big orange light!

PS Just to let everyone know the DSC worked for another team, however when we hooked it up to our robot it still did not worked:confused:

We appreciate all the help guys!:)

Al Skierkiewicz

17-02-2010, 09:39

Boy Peter, that is starting to look like something in the Crio or the cable, if it worked on another robot. Mark might have more insight after he reports on his tests.
Just to be sure, that digital module is in slot 4. It looks like it is from the pictures.
B. The Digitial Sidecar must be connected to a NI 9403 module in Slot 4 of the cRIO-FRC. If it is connected through any other slot, the light will not function properly.

Mark McLeod

17-02-2010, 11:11

From a software perspective, the only reliable ways for the RSL to remain off seem to be:
Having the Safe Mode DIP switch set on the cRIO
The cRIO failing to boot properly - need a null modem cable and the cRIO Console Out DIP switch set to read on a PC
Improper cRIO image formattingI'm just concentrating on things Teams would normally do, or situations they might expect to get into, not catastrophic HW failures or unusual SW corruptions.
At least that's all I got to repeat the problem in the short time I mucked with it. There are probably other ways as well. I'll look some more when I get a chance.

Alan Anderson

17-02-2010, 11:30

Having the Safe Mode DIP switch set on the cRIO

We'll have to add that to the standard troubleshooting script.

Jon Stratis

17-02-2010, 12:58

PS Just to let everyone know the DSC worked for another team, however when we hooked it up to our robot it still did not worked:confused:

We appreciate all the help guys!:)

Well, i think we've managed to eliminate a lot of possibilities thus far. I would recommend trying the following:

1. Check the dip switch Mark mentioned... It's news to me (i've never had a need to pay attention to it), and seems like it could easily account for the problem.
2. Eliminate your teams software. See if that other team would be willing to let you load your software on their cRio, and see if the signal light works. for safety's sake, pull out all the breakers to any motors before doing this (so the robot doesn't do something strange and break itself). This would eliminate your software.
3. Load the other teams software onto your cRio and see if the light still refuses to work (with the same precautions as in 2). This would indicate a hardware/firmware issue.
4. try replacing the digital module and the cable between it and the cRio.

I'm going to put my money that step 1 will fix the problem... the rest is just to give us more information to figure out where the problem exists.

kick3nitx360

17-02-2010, 18:32

http://i349.photobucket.com/albums/q394/kick3nitx360/Untitled.png

This is how are pins are set, is this correct???

Alan Anderson

17-02-2010, 22:42

This is how are pins are set, is this correct???

That photo shows all the cRIO's DIP switches set to the off position, which is the way they should be for most teams' normal operation.

It also shows no power or data connection, and it has obviously been edited, which is a little confusing.

Al Skierkiewicz

17-02-2010, 22:55

Dave,
Since we are grasping at straws anyway, the next step would be to exercise the switches with the power off. Every once in a while, a switch gets a little dirt inside and it needs to be cleared. Hey, it's worth a shot and only takes a few seconds.

samir13k

17-02-2010, 23:24

this thing is driving all of you guys nuts im sure,
im not even on your team and it bugs me!

whenever you download code or re-image, has it said its been successful?
have you attempted to download code/reimage from a different computer?
have you been downloading wirelessly or through cat-5?
make sure you are using a normal cat 5 or cat 6 cable, not a crossover.

at this point, im just throwing out things, hoping one simple little mistake is all that it is. I believe you used multiple cRios to test this, its hard to believe that more than one would have the same issue... Mark, Al, or Alan, would it be probable to use the little reset button on top of the cRio before imaging and downloading code? If so, is there a proper order of operations to do such? ie hold for a certain amount of time, or until a certain LED flashes in a certain pattern?

I am fairly confident in this teams abilities as to getting everything wired properly and connected correctly by this point, we have gone through most of the hardware errors that i can think of. I believe its something in the code/cRio that is giving the problem.

Maybe go through your Digital cRio modules and slot 4 in your cRIo, and check for anything that may have shorted them out. One shorted module may have shorted out the slot in your other cRio maybe? also causing the shorting of all the other modules that were plugged into that slot? i dont know how practical that is, but its a possibility. Once again, can Alan or Mark back up this theory? It would be unfortunate, but the warranty may cover it if that happens to be the problem. Also check the DB37 cable for any shavings inside that may have put two of the pins in contact. That may have also fried the modules/slot...

At this point in the season, would it be practical to send the cRio back to NI? they overnight replacements for you i believe if you are having issues and send the other one back.

Al Skierkiewicz

18-02-2010, 00:03

Samir,
I defer to Mark or Alan on software/Crio questions. I am just approaching this from a troubleshooting angle since that is what I play in real life.

Alan Anderson

18-02-2010, 09:13

I'm stumped.

All I can think of to try right now is to start over and ask all the usual questions again, on the slim chance that whoever answered them the first time didn't fully understand what they meant or didn't look closely enough at the system to answer them correctly.

Question #1: when the system is powered on, what is the state of the three power status LEDs lit on the Digital Sidecar? BAT, 5V, and 6V should all be on brightly.

mfalk

18-02-2010, 10:59

If the LED on the DSC works in that situation, start making other connections, checking the LED at every step. If the LED dies at some point, at least you know what connection it was!

We ran into this same problem and methodically checking every connection resolved it for us. It turns out that someone plugged the pneumatic pressure sensor into the wrong location.

Mark McLeod

18-02-2010, 11:08

samir13k

18-02-2010, 11:33

Have you installed the cRIO gaskets (http://digital.ni.com/public.nsf/allkb/FA1B856FC4EB6F9D86257673007935A1)?
The reason I ask is that process includes cleaning out any accumulated metal dust that may have collected inside the cRIO.

I had a similair thought, but i found it rather odd that both cRios were performing in the same manner. I am correct as to you having two cRios right?

Al Skierkiewicz

18-02-2010, 11:49

OK,
I am reaching back into 2009 on this. It seems to me that the DSC power LEDs can come on dimly even if the power supply wires are not connected. Does anybody remember that? An old mentor of mine used to say..."When you have checked everything else, check the power supply one more time." I would pull the connector out of the DSC and measure the power at the DSC side of the connector. Tug on the wires and see if they pull out.

Mark McLeod

18-02-2010, 13:04

The DSC gets a very little power from the 37-pin cable that can make some of the lights glow, and even that dimly, in the absence of power through the Wago.

Think about what you haven't tried changing, maybe because it was so obviously right by just looking at it.
Maybe a new Wago connector with new wiring back to the Power Distribution Panel, and get someone different to wire it.

peter1626

18-02-2010, 17:13

IT WORKS!!!!!!!!!!!:yikes:
We removed the entire electronics board from the robot and removed EVERY WIRE AND COMPONET! (For the 3rd time) I then wired only the power board, cRIO, cRIO module 9403, and a DSC. The signal light still didn’t work so I imaged the cRIO and the light work! I changed the wire from the power board to the DSC to 14 AWG (it was tough to fit in the wago connectors but I did it)!

Thank you all for your help, especially 1257 for inviting us to their shop and testing our components!