View Single Post
  #2   Spotlight this post!  
Unread 30-04-2010, 19:28
Jon236's Avatar
Jon236 Jon236 is offline
Registered User
AKA: Jon Mittelman
FRC #2648 (Infinite Loop)
Team Role: Mentor
 
Join Date: Jan 2004
Rookie Year: 2000
Location: Windsor, Maine
Posts: 741
Jon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond reputeJon236 has a reputation beyond repute
Re: 2010 KoP Improvements

Quote:
Originally Posted by EricVanWyk View Post
Moderator -

Can you please pull the 2010PD thread out of "Feedback Thread: Robot Control System" and place it here? I am much too interested in the results of that conversation to conflate it with this one.

Thank You!
- Eric



Jon -

Long story short, the 2010AB provides more accurate sensor data than the 2009AB, except in two corner cases that I believe are not found in FRC.

The 2010AB significantly improves on the accuracy and the noise of the 2009AB, at the cost of a greatly reduced maximum output current. Additionally, it features reverse voltage protection. Now the all of the blue-plastic boards (DSC, Breakouts, PD*) can handle cross-wired input power.

The 2009AB can happily push just under 3 amps all day long. The 2010AB can theoretically push half an amp, if the correct conditions are met and the electrons find your lightning-goat sacrifice pleasing.

But, the noise is now well below a single-count on the ADC. There will be a notable decrease in fuzziness of some sensors, such as potentiometers. Also, I added some low pass filters, so noise coupled in from e.g. motors should be somewhat reduced. Unfortunately, this does limit the usefulness of any analog sensors that operate in the kilohertz plus range. I don't think any are really in use in FRC, so I'm happy with it.

If anyone does use analog sensors that are faster than a few kilohertz or use more than half an amp, please let me know.

* The PD protects itself, the camera, the cRIO, and the router when cross-wired. Attached Victors, Jaguars, and Spikes will likely destroy themselves.
Thanks Eric....as proof to the robustness of the breakouts, we had the pneumatic break-out 24 v supply cross-wired inadvertently. A splice went to the cRio. When powered on, the only casualty was the wire...the cRio and pneumatic break-out did well.

So good work Eric!

The suggestion on the previous post for a break-out panel that plugs into the cRio modules is intriguing. It eliminates several failure points. I realize that it is too late to get something like that into play for 2011, but for 2012 and beyond, it would be worthy of consideration.
__________________
Jon Mittelman

Senior Judge Advisor New England & Israel 2014-2015
Infinite Loop Mentor 2011-2015
TechnoTicks Mentor 2000-2011
Championship Chairman's Award 2009 Team236 TechnoTicks
Judge 2010-2015 Championships
Senior Judge Advisor New England District Championship 2014-2015
Judge Advisor Tel Aviv Regional 2007-2015
Judge Advisor Pine Tree Regional 2013
Maine Regional Planning Committee
New England District Planning Committee
Lead Inspector Microsoft Tel Aviv Regional 2006-2008
Judge & Lead Inspector GM/Technion Tel Aviv Regional 2006
Judge UTC Hartford Regional 2006
Reply With Quote