Top Ten Things NOT to do with the 2009 FRC Control System

I invite your thoughts on the following list.

Top Ten Things NOT to do with the 2009 FRC Control System:

  1. Don’t use the system without reading and following the documentation posted at http://www.usfirst.org/community/frc/content.aspx?id=10934

  2. Don’t install the updates in an incorrect order: update LabVIEW, re-image the CRIO and update the Driver Station firmware.

  3. Don’t mount the cRIO or Axis camera on a conductive material: the enclosures for these devices are grounded.

  4. Don’t disconnect the power to the Driver Station while downloading the firmware upgrades via a USB thumb drive.

  5. Don’t use the Driver Station Digital Inputs without external pullup/pulldown resistors.

  6. Don’t try to operate the robot without the dongle connected to the Driver Station.

  7. Don’t use the incorrect Ethernet cable types on the cRIO MDC: Port 1 connects to the robot wireless with a straight-though cable and Port 2 connects to the camera with a cross-over cable.

  8. Don’t install the modules in any alternate order: NI9201 in slots 1 & 2, NI9403 in slots 4 & 6, NI9472 in slot 8.

  9. Don’t use 1/4-20 nuts on the Power Distribution lugs: M6 nuts must be used.

  10. Don’t try to use servos on the Digital SideCar without connecting the jumpers at each PWM connection.

The heading seems a bit alarmist. I usually think of Don’t lists as things that will break something. So 5-8 and 10 seem a little overblown. Also:

  1. Misleading. You can update in whatever order you like. Should be, “Don’t run different software versions (or update levels) on your PC, cRIO, or Driver’s Station”

  2. Somewhat misleading. You’d probably be fine mounting the cRIO or camera on a small piece of metal, as long as the metal itself is isolated from the rest of your robot’s frame.

  3. Incorrect. You can use the inputs just fine without pullup or pulldown resistors by actively switching the signal wire from +5V to GND. Or by actively driving the inputs with an external circuit. The only problem is simply connecting signal to ground or 5V and expecting the input to change state when you break that connection.

  4. Silly, as the robot won’t operate without a dongle. Perhaps, “Don’t lose your Enable Dongle” would be more meaningful.

  5. Incorrect. The robot wireless, drivers station, and most modern laptop and computer interfaces have auto-sensing technology, and would operate fine with a cross-over cable. So the only cable that absolutely has to be correct is the crossover between port 2 and the camera.

The heading seems a bit alarmist. I usually think of Don’t lists as things that will break something. So 5-8 and 10 seem a little overblown. Also:

  1. Misleading. You can update in whatever order you like. Should be, “Don’t run different software versions (or update levels) on your PC, cRIO, or Driver’s Station”

  2. Somewhat misleading. You’d probably be fine mounting the cRIO or camera on a small piece of metal, as long as the metal itself is isolated from the rest of your robot’s frame.

  3. Incorrect. You can use the inputs just fine without pullup or pulldown resistors by actively switching the signal wire from +5V to GND. Or by actively driving the inputs with an external circuit. The only problem is simply connecting signal to ground or 5V and expecting the input to change state when you break that connection.

  4. Silly, as the robot won’t operate without a dongle. Perhaps, “Don’t lose your Enable Dongle” would be more meaningful.

  5. Incorrect. The robot wireless, drivers station, and most modern laptop and computer interfaces have auto-sensing technology, and would operate fine with a cross-over cable. So the only cable that absolutely has to be correct is the crossover between port 2 and the camera.

Before being so quick to criticize, one might consider the identity of the original poster, and why he may be providing the information that has been posted. Vince knows what he is talking about, and there is logic to each of his statements.

Wisdom is knowing the difference between “the way to do things” and “the right way to do things.” Just because an action may be physically possible does not mean that it is smart to do. Vince is wise. Listen to him.

-dave

.

I’d promote items 9 and 4 because they are the most painful. 9 will cause permanent damage, 4 will cause serious headache. 3 (the camera) is in a good place. If the camera isn’t isolated, it is possible to do some damage.

Then I’d put 5, since it can lead to some wonky behaviour. The rest are good reminders, and some could probably be rephrased as “Don’t forget to ______ if you want to _______”.

Cool list though.

Excellent list…some things I didn’t know about such as the grounding issues. Good engineering practices with the 2009 control system:

Don’t plan to consider switching programming languages mid-season. Don’t even say ‘maybe if a xyz mentor comes in we’ll switch’.

Don’t assume that just because an interface exists on the cRIO (such as SPI) that it is available on the Driver’s Station (personal experience here).

My first impression when reading Kevin’s post was:
“Dude! That’s Vince you’re talking about! I think you should probably just shut up and listen!”

Dave said it much better than I would have.

-John

Excellent list…I would add to be sure and only cut the cord of the EXTRA power supply for the Axis camera supplied in the KOP, NOT the one in the box with the Axis camera…you will need that one for using the camera as a stand-alone for working with Vision Assistant 8.6.

Maybe I’m the only one who hasn’t heard of him, but who is Vince and why shouldn’t we question him?

Vince Wilczynski

I don’t know. Call me crazy, but i really think a DON’T list should have a list of things that are highly foolish and dangerous to do that might actually, you know, permanently damage things. Half of Vince’s current list is much better posed as a troubleshooting list, since they’re just things that keep the robot from working correctly. I figure a DON’T list should look something like, ohhh:

Call me crazy, but in my opinion, a list like this should be reserved for things that you REALLY shouldn’t do because they’re dangerous to your equipment (in this case) or your life. It should be a checklist of things you can run through to make sure you’re not about to turn your thousand dollar investment into so much magic smoke. It should not be a list of oddly worded good ideas, or odd things that might prevent your robot from working. Those sorts of things seem (to me) to properly belong on “Top 10 tips on using the new control system” or a nice troubleshooting list, respectively. Admonishing teams to “Don’t forget to read the manual” seems silly when you should be telling them not to turn their cRIO into a very ostentatious paperweight.

To that end, here’s my proposal of a more useful list of "Don’t"s that I’d actually be comfortable giving to teams.

  1. Don’t connect reverse polarity power to the Power Distribution battery lugs.

  2. Don’t connect reverse polarity power to ANY power terminals on any device.

  3. Don’t connect power to the motor output side of a speed controller.

  4. Don’t mount the cRIO or Axis camera on a conductive material: the enclosures for these devices are grounded.

  5. Don’t disconnect the power to the Driver Station while downloading the firmware upgrades via a USB thumb drive.

  6. Don’t try to operate the robot without an appropriate Enable/Disable dongle connected to the Driver Station.

  7. Don’t touch the pins on the cRIO backplane if you’ve removed a module; they are static sensitive, so you can damage them. Corollary: Awlays cover the connectors for removed modules.

  8. Don’t use 1/4-20 nuts on the Power Distribution lugs: M6 nuts must be used. 1/4-20 nuts will permanently damage the lugs.

  9. Don’t use unmodified keyed PWM cables from the kit. They can only plug in backwards and may damage your sensors.

  10. Don’t turn on your Driver’s Station while the joysticks aren’t centered. They automatically center when powered up and your robot may move unexpectedly.

Now, 6 and 10 there won’t actually inherently smoke something on the control system, but I figure it’s also important to tell teams that paper-clip disable switches are a bad idea, and that their robot might develop a mind of its own if they’re not careful.