Half Sidecar PWM Ports Working

[afclsn]

This is team 4474, posting during our second FRC season. It seems that the first 4 PWM ports in our sidecar don't respond correctly. Ports 1 and 2 output "forward" to out motor controllers while ports 3 and 4 output "reverse" at all times. Ports 5 through 10 work just fine. We've tried two different sidecars, several different serial cables, and two different cRIOs with the exact same results. The battery voltage was near full when we tested it as well. Help would be greatly appreciated!

[Nirvash]

Is the sidecar wired to and recieving 12volts from the PDB and are all three voltage lights on it fully illuminated with the DB37 cable disconnected? ( Cable that attaches to the crio)

[kevin.li.rit]

This is team 4474, posting during our second FRC season. It seems that the first 4 PWM ports in our sidecar don't respond correctly. Ports 1 and 2 output "forward" to out motor controllers while ports 3 and 4 output "reverse" at all times. Ports 5 through 10 work just fine. We've tried two different sidecars, several different serial cables, and two different cRIOs with the exact same results. The battery voltage was near full when we tested it as well. Help would be greatly appreciated!

Just to help us help you here are some basic questions.

How are you determining that they are outputting reverse/forward all the time?
How are your motors wired?
How did you determine ports 5-10 are working properly?
What speed controllers are you using and are they wired properly?

[Alan Anderson]

...the first 4 PWM ports in our sidecar don't respond correctly. Ports 1 and 2 output "forward" to out motor controllers while ports 3 and 4 output "reverse" at all times.

I am unable to imagine a failure that can cause this symptom. I suspect something else is going on. My first guess is that your motor controllers are miswired with the motor terminals connected to power and the power terminals connected to the motor.

Ports 5 through 10 work just fine. We've tried two different sidecars, several different serial cables, and two different cRIOs with the exact same results. The battery voltage was near full when we tested it as well. Help would be greatly appreciated!

What do you mean by "serial cables"?

[afclsn]

Is the sidecar wired to and recieving 12volts from the PDB and are all three voltage lights on it fully illuminated with the DB37 cable disconnected? ( Cable that attaches to the crio)

To be honest, we haven't checked this( we hind of forgot about this) but we kind of assumed that this what the issue, since we ran four motor controllers for like 1/2 hour on our drinetrain (mecanums) with this issue, but using the ports that were working.

Thanks for reminding us of this, we will be sure to check it tomorrow since our meeting is now over. Thank you! We will let you how how it goes.

[afclsn]

Just to help us help you here are some basic questions.

How are you determining that they are outputting reverse/forward all the time?
How are your motors wired?
How did you determine ports 5-10 are working properly?
What speed controllers are you using and are they wired properly?

We determined that they are outputting forward/ reverse by attaching a cim through a victor and then trying the pwm in each port (disabling the bot and changing the code to account for each new port) when in ports 1&2 spin one direction with no driver controls, and there is a solid green light on the victor. Ports 3&4 spin the other way and display a solid red light, and the rest work as expected with driver controls(we were able to successfully run our mecanum drivetrain in the working ports) The morons are wired to the victors following the wiring diagram, and they work in ports 5-10 (which was determined by by using them to control our robot which behaved as expected) and since everything worked as expected and we followed the wiring diagrams I am pretty sure everything is wired correctly. Any thoughts? Thanks for the help.

[kevin.li.rit]

As previously stated I would check to see if the 12 volts is going into the sidecar.

[afclsn]

I am unable to imagine a failure that can cause this symptom. I suspect something else is going on. My first guess is that your motor controllers are miswired with the motor terminals connected to power and the power terminals connected to the motor.



What do you mean by "serial cables"?

First, as I described in my other replies, the we followed the wiring diagrams, and successfully used them to drive our drivetrain around for about 30 minutes.

Second, by "serial cable" I meant the db37 cable used to connect the digital sidecar to the digital io module in the crio. Sorry I used wrong name.

Any other ideas?

Thanks.

[afclsn]

As previously stated I would check to see if the 12 volts is going into the sidecar.

We will definitely check this tomorrow, as this is a new assembly of the electronics and it very well could be the issue, I just didn't think to check it.

[Al Skierkiewicz]

We had a similar failure but do not expect this to be your exact issue. We started to lose PWM outputs one after another until half the board had no output. We replaced the DSC and checked the bad one only to find a crack. The way the board is laid out, the crack cut the lines to one of the output devices but not the others.

[Toa Circuit]

Check to see if they really are getting a forward or reverse signal. Green on the Jags is forward, red is reverse.

Why do I ask? I'm guessing you're using ports 1, 2, 3 and 4 for your drive motors. 3 and 4 are on one side, correct? If so, then everything is working as it "should", just not what you are expecting. (Since on a normal FRC drivetrain, the left side is a mirror image of the right side, and the motors are rotated 180 degrees...) You'll need to change your code to drive motors 3 and 4 inverted, or on the OUTPUT side of the corresponding Jags (The side with M+ and M-), make the motor's red go to M- and the motor's black go to M+.
If that isn't what you're doing... then something very strange is going on indeed.

EDIT: I guess I didn't understand your problem correctly. You're saying that the problem isn't inversion but lack of control?

[afclsn]

We had a similar failure but do not expect this to be your exact issue. We started to lose PWM outputs one after another until half the board had no output. We replaced the DSC and checked the bad one only to find a crack. The way the board is laid out, the crack cut the lines to one of the output devices but not the others.

2 digital sidecars did the same thing, including a brand new one from the KOP. We also opened up the digital sidecar and do not see any cracks/debris/etc.

any other ideas?

thanks.

[afclsn]

Check to see if they really are getting a forward or reverse signal. Green on the Jags is forward, red is reverse.

Why do I ask? I'm guessing you're using ports 1, 2, 3 and 4 for your drive motors. 3 and 4 are on one side, correct? If so, then everything is working as it "should", just not what you are expecting. (Since on a normal FRC drivetrain, the left side is a mirror image of the right side, and the motors are rotated 180 degrees...) You'll need to change your code to drive motors 3 and 4 inverted, or on the OUTPUT side of the corresponding Jags (The side with M+ and M-), make the motor's red go to M- and the motor's black go to M+.
If that isn't what you're doing... then something very strange is going on indeed.

EDIT: I guess I didn't understand your problem correctly. You're saying that the problem isn't inversion but lack of control?

Yes. The problem is that the motors just spin and we have no control of them when we should be able to control them using the joystick. This is only the case for the first four ports too, since we can control them exactly as we like in the rest of the ports. We had our drivetrain running on the other ports because the first four ports will not work. The motor controllers are wired as they should be. It is really strange. We e had problems with the sidecar last year, but then it was just that metal shavings got into the sidecar and shorted it out. This is not the case this year (we opened it and checked). Any ideas, because right now we are stumped. Thanks.

[Alan Anderson]

any other ideas?

Only of the "are you sure it's plugged in?" sort. For example, what revision of the cRIO image is installed?

[kevin.li.rit]

afclsn, Can you take a picture of your side car and electronics layout for us?

Thanks.

[Toa Circuit]

Try making some code that allows you to directly control each output (As a Jaguar... like jag[1].Set(joystick.GetAxis(1)) ). See if the problem is really hardware or just your code not doing what you think it does.

[afclsn]

Try making some code that allows you to directly control each output (As a Jaguar... like jag[1].Set(joystick.GetAxis(1)) ). See if the problem is really hardware or just your code not doing what you think it does.

Yeah, we did this before we tried it with code for the mecanums, and it had the problems with the simple code.

[afclsn]

afclsn, Can you take a picture of your side car and electronics layout for us?

Thanks.

I will get one up shortly.

[afclsn]

We will definitely check this tomorrow, as this is a new assembly of the electronics and it very well could be the issue, I just didn't think to check it.

Just checked, all three power lights are green.

[afclsn]

I will get one up shortly.

Here is our layout, without the motors connected.
We connect the black motor wire to the M- port, and the red motor wire to the M+ port.https://docs.google.com/file/d/0ByDtApfHHB3xSUhWbjBrVXl5WDdVUkItZkRHWEtXTEZ0V0hZ/edithttps://docs.google.com/file/d/0ByDtApfHHB3xRTh1UUtLeWNsWFE1LU8zMjNtck9fd1FOMXRr/edit

[kevin.li.rit]

Here is our layout, without the motors connected.
We connect the black motor wire to the M- port, and the red motor wire to the M+ port.https://docs.google.com/file/d/0ByDtApfHHB3xSUhWbjBrVXl5WDdVUkItZkRHWEtXTEZ0V0hZ/edithttps://docs.google.com/file/d/0ByDtApfHHB3xRTh1UUtLeWNsWFE1LU8zMjNtck9fd1FOMXRr/edit

I think you forgot the attachment. Correction it doesn't look like those pictures are public.

[afclsn]

I think you forgot the attachment. Correction it doesn't look like those pictures are public.

I noticed this when I posted it, but due to the securities on our school network, I couldn't easily fix them. Here they are now.
http://imgur.com/5nqNRqG
http://imgur.com/1Ta5JIg
I case the photos don't work
http://imgur.com/5nqNRqG
http://imgur.com/1Ta5JIg

[arizonafoxx]

I noticed you are using the ribbon cable. Could you have a backwards ribbon cable? I know those were floating around and its also easy to make backwards if you not careful. I am not sure of the symptoms of a backwards we have never had one backwards, but I know other teams have. Does anyone know if this could be the symptom they are seeing?

[afclsn]

I noticed you are using the ribbon cable. Could you have a backwards ribbon cable? I know those were floating around and its also easy to make backwards if you not careful. I am not sure of the symptoms of a backwards we have never had one backwards, but I know other teams have. Does anyone know if this could be the symptom they are seeing?

This was the pre fabricated ribom cable from the kit, plus we tried it with another pre fabricated cable (an actual cable, not a ribon cable), so I don't think this is the problem, but all these cables were new this year so it is a possibility. I will check it tomorrow with the ribon cable we used last year and know works.
Thanks.

[afclsn]

So thanks for all the suggestion. Last Friday the sidecar started working again, and it was set up exactly as it was Wednesday and Thursday. This is good because the sidecar now works, but is also bad, because we don't know what went wrong. If you any other ideas as to what went wrong pleas let us know so that in the case it happens again we and other teams have a reference to go off of.

Thank you,
Team 4474

[Al Skierkiewicz]

Bet the multipin connector was not fully seated at one or both ends.

[Bill_B]

I was going to mention that driving your mecanums for a half hour might be just about the limit for a battery to handle. I know it is tempting to keep the testing going, but you really have to watch the power level during testing and driver training runs.

[Jared]

Bet the multipin connector was not fully seated at one or both ends.

Use some 4-40's to hold the ribbon cable into the sidecar and digital module. We've had this happen in the past to us.