OMG - Problems problems

[Crossle86]

I have a test robot and control system in my office and have done quite a bit with it with no major problems. Today, I went to one of the teams I support because they were having problems. I spent time with them going over their setup and code and man what an eye opener! Here are the 3 main issues I observed:

1) They are having significant connectivity issues with the RC phone not detecting the motor controllers. No good pattern I could see, but changing the battery seemed to calm it down for a short period of time and then it again failed detection of the controllers. While trying to identify a pattern, the following problems were observed:

2) An opmode is running and operating as expected. Without any apparent cause, the STOP button on the DS is replaced by the INIT button, but the opmode is still running and continues to run. INIT button is not responsive. Observed this several times. No exceptions thrown.

3) A simple opmode feeding the L/R joysticks of a single controller to some motors has very strange behavior. The left JS and its motors work just fine. The right JS and its motors do nothing. Putting in telemetry to see the value returned by gamepad1.right_stick_y shows this always returns zero...even though the display of the right JS y value on the RC screen shows values in response to stick movement (though at stick center, the DS value = .5). The gamepad1.left_stick_y works just fine. What gives?

4) While tinkering with the controllers to try to get a handle on #3, the DS screen stops displaying the little controller icons above user 1 and user 2 labels top right of DS screen (controller icons not visible). No amount of fooling with the hub, the cables, the controllers, reboot phone etc etc fixes this problem and we were stopped on any further trouble shooting (and class time ended). Ideas?

[Mitch]

1) They are having significant connectivity issues with the RC phone not detecting the motor controllers. No good pattern I could see, but changing the battery seemed to calm it down for a short period of time and then it again failed detection of the controllers. While trying to identify a pattern, the following problems were observed:

This particular one is being discussed in the sticky at the top of the forum. I'm trying to get some teams to try what I believe to be a fix for this. See:

http://ftcforum.usfirst.org/showthread.php?4993-Update-on-USB-Disconnect-Issues-(unable-to-detect-USB-modules-during-scan)&p=19017&viewfull=1#post19017

It does involve taking apart your motor controllers and soldering a capacitor (and is not "official"), but for me, it has completely eliminated this problem.

/Mitch.

[FTC5024]

GamePad

This particular one is being discussed in the sticky at the top of the forum. I'm trying to get some teams to try what I believe to be a fix for this. See:

http://ftcforum.usfirst.org/showthread.php?4993-Update-on-USB-Disconnect-Issues-(unable-to-detect-USB-modules-during-scan)&p=19017&viewfull=1#post19017

It does involve taking apart your motor controllers and soldering a capacitor (and is not "official"), but for me, it has completely eliminated this problem.

/Mitch.

We are having issues with the gamepads not being able to be seen when using the USB Hub. We have to disconnect the hub and only use one gamepad. The tried it with full battery charge, different gamepads ect. but nothing seams to work. Try it later and it works. Not sure what the problem is but if you can disconnect the hub and it works it must be something in the hub design that we don't understand. Power requirement verses power supplied by phone maybe?

[Mitch]

I haven't looked at the gamepads but I *have* seen issues getting my gamepad to connect sometimes. My mentor-bot has only one joystick, so I'm not even using a hub.

One of our teams managed to destroy a hub last weekend.

In the DS application, I've noticed that the green glow indicator is really hard to see once you have done the START+A sequence to pair the joystick. I'd really like to see a more visible indication of connectedness (entire joystick icon could be red if not connected, for example)

/Mitch.

[Jonathan Berling]

2) Next time this happens, can you collect the log files?

3) Did you go into settings on the driver station and select your gamepad type? If you did, and it still wasn't working then I'm not sure what the issue is.

4) It won't appear until you hit Start-A. The issue in 3 could cause the driver station to not register the Start-A key sequence.

[Crossle86]

2) Next time this happens, can you collect the log files?

Will do.

3) Did you go into settings on the driver station and select your gamepad type? If you did, and it still wasn't working then I'm not sure what the issue is.

Yes, absolutely. Checked everything...

4) It won't appear until you hit Start-A. The issue in 3 could cause the driver station to not register the Start-A key sequence.

DS gamepad icons are always present on my test setup though you won't see the green glow until you do the Start-A/B sequence. In this problem, the white gamepad icons do not appear at all...just the words user 1 and user 2. The DS app seemed completely unaware of the controllers though the phone seems to know they are around. If you open the settings menu on DS app, the controller joystick will move the selection on the settings menu...but the app does not respond at all to the controllers as input devices to the app itself. A reboot of phone did not cure this. More testing to come.

[Crossle86]

Update on these issues

[QUOTE=Crossle86;19127]I have a test robot and control system in my office and have done quite a bit with it with no major problems. Today, I went to one of the teams I support because they were having problems. I spent time with them going over their setup and code and man what an eye opener! Here are the 3 main issues I observed:

1) They are having significant connectivity issues with the RC phone not detecting the motor controllers. No good pattern I could see, but changing the battery seemed to calm it down for a short period of time and then it again failed detection of the controllers. While trying to identify a pattern, the following problems were observed:

The usb connection fail or drop is still a major problem for this team. I'm going to move that issue over to the existing thread on usb connection problems.

2) An opmode is running and operating as expected. Without any apparent cause, the STOP button on the DS is replaced by the INIT button, but the opmode is still running and continues to run. INIT button is not responsive. Observed this several times. No exceptions thrown.

Could not reproduce this problem today. Suspect joystick issue below may have caused this.

3) A simple opmode feeding the L/R joysticks of a single controller to some motors has very strange behavior. The left JS and its motors work just fine. The right JS and its motors do nothing. Putting in telemetry to see the value returned by gamepad1.right_stick_y shows this always returns zero...even though the display of the right JS y value on the RC screen shows values in response to stick movement (though at stick center, the DS value = .5). The gamepad1.left_stick_y works just fine. What gives?

This turned out to be a bogus controller. I asked the student 'did you swap controllers', he answered 'yes, sure'. Well not so sure.

4) While tinkering with the controllers to try to get a handle on #3, the DS screen stops displaying the little controller icons above user 1 and user 2 labels top right of DS screen (controller icons not visible). No amount of fooling with the hub, the cables, the controllers, reboot phone etc etc fixes this problem and we were stopped on any further trouble shooting (and class time ended). Ideas?

The controller icon disappearing and reappearing is a behavior of DS 1.2. DS 1.1 did not exhibit this behavior. Not sure how I missed the DS update but my lab phone had 1.1 on it. Now with 1.2 I see the icon behavior on my lab phone so this is a non-issue.

[EV713]

There is a new thread related to this concerning the icons above User 1 and User 2 being shown incorrectly.

http://ftcforum.usfirst.org/showthre...0038#post20038