2010 Lesson Learned: The Negative

[Chuck Glick]

My biggest complaint: bumpers.

When rules on bumpers get close to the same length as the robot rules... you have a problem.

1. If they have to be mandatory, then at least let teams choose their own colors and only ban red and blue. Allow teams to show their own team colors. Red and blue mandatory bumpers this year took away from many teams image.

2. Make bumpers optional. If I want to risk my 6 weeks of hard work then so be it. I don't like being told that I have to protect something I worked on. Current bumper rules take away from many frame options. Take the rule back to 2006, where teams could opt to use them and there was the added weight bonus if you chose to.

3. If you keep them mandatory, supply the noodles in the kop. It isn't fair to teams in cold weather states to have to pay a premium compared to teams in warm areas.

[AdamHeard]

Originally Posted by thefro526

This is true. We were shoved into a goal during our second qualification match and couldn't get out. After the match I looked at the goal lip and how it caught on the wheels and saw that the only way a team could've prevented this from happening was to either build a robot too talk to go into the goal, or by giving your chassis about .5" of ground clearance.

Not really, low ground clearance would just mean your frame gets stuck. It was really a tough spot to be in, as you had little room to move forward and back to wiggle out.

[synth3tk]

Originally Posted by Chuck Glick

3. If you keep them mandatory, supply the noodles in the kop. It isn't fair to teams in cold weather states to have to pay a premium compared to teams in warm areas.

The price we have to pay during the winter for pool noodles is outrageous. It was almost cheaper for us to buy online than to purchase from a local shop/store.

[yarden.saa]

Originally Posted by yarden.saa

Even not one judge came to out pit. how do the other teams won in prizes at the chanpionship when none of the judges came to our pit? I DON'T KNOW.

In the WPA key staion there was only one computer that set the routers to Curie field and this computer didn't work so they had to set our router manualy.

The field issues were exactly like in Israel Regional ):

We got penalties on nothing when other teams didn't got penalties when itg was obvious they should get penalty.

I left the CMP with bad taste in my mouth.

We borrowed router because the routers from Israel are not useable because of the settings change and the router was not working well. Just after we replaced the router with another one in thespare parts and set the settings manually there were not field issues to my team.

[Cynette]

Thursday Matches in Atlanta: As much as we all liked having extra qualification matches, I did not like what it did to the "intensity" of Thursday. Too short of time to practice and make final repairs on the robots, too short of time for inspections, team members trying to get pits organized while competing, pit scouting, match scouting, giving conference presentations and trying to attend the conferences. Our team also organizes the All Rookie Meet and Greet and we had invitations to deliver. Even with a detailed team schedule, our team members all ended up trying to do several things at once which was very stressful!

I am not a fan of the red and blue bumpers, but I have to admit that they win out over the flags! But my team really missed it's red camo bumpers. Or even the choice of putting red camo numbers on the bumpers. We would have made sure they were visible from 300 feet, so I still don't understand why we were limited to a solitary color.

The changing dimensions in the game manual. We build a whole field and also make a game test for our students. The manual said the tower was this tall, the drawing said it was this tall. The manual said the target was this diameter, the drawings said it was this diameter. We were convinced that the bump height changed multiple times until it finally settled down to one dimension in the manual and +/- 1/4 inch in the team update. We rebuilt field elements and revised the team test more than once. In the end we all had to go with eh, its close enough.

Not allowing teams to measure field elements at competition! Our robot hanger lifted the robot to the same dimension every time - guaranteed to be at least an inch above the platform - according to the drawings. At our second regional, it wasn't tall enough. We asked to measure the field so we could make a spacer for the robot hanger and were denied.

Little things - not showing the Chairman's videos at each regional. We are zero for two for the past two years at different regionals. And the list of WFFA nominees should be shown as part of the WFFA presentation.

[Vermeulen]

Originally Posted by synth3tk

The price we have to pay during the winter for pool noodles is outrageous. It was almost cheaper for us to buy online than to purchase from a local shop/store.

During Build, we can't find them at any price. We had to steal them from last year's robot this year.

[Sunshine]

Originally Posted by Vermeulen

During Build, we can't find them at any price. We had to steal them from last year's robot this year.

Do what we do and buy them in the summer and store them. problem solved.

I'm hoping we can use the same ones we built this year again

[Vikesrock]

Originally Posted by Vermeulen

During Build, we can't find them at any price. We had to steal them from last year's robot this year.

Seems like we're a rare breed. A mentor for our rookie mentee found some at a dollar store. We ended up with enough to build all 4 sets of bumpers (2 for each team).

[synth3tk]

Originally Posted by Sunshine

Do what we do and buy them in the summer and store them. problem solved.

I'm hoping we can use the same ones we built this year again

We usually don't meet during the off-season, and almost didn't have enough money for a team this season. But that's a good suggestion, and hopefully we can pick up enough this summer to last us for years!

[billbo911]

Originally Posted by synth3tk

The price we have to pay during the winter for pool noodles is outrageous. It was almost cheaper for us to buy online than to purchase from a local shop/store.

I go to the "Dollar Tree" store during the summer and by a case of 40 for a dollar a piece. I did that two years ago and we still have some left over. The funny thing is, we store them in a high school engineering class room and for some strange reason, they seem to grow legs and walk away.

[Bruceb]

Last year I found noodles at a pool supply store here in Wisconsin in the middle of winter. I found them this year at Walmart.
I REALY LIKE THE BUMPER COLORS MATCHING THE ALLINCE COLOR. Yes, I was yelling. I think it is the best thing I have seen to help teams and spectators identify the alliance partners. Hope we keep doing it. I don't mind the idea of being able to customize the middle third.

I think the control system stinks. Way to complicated and buggy.

Bruce

[Karibou]

Originally Posted by yarden.saa

Even not one judge came to out pit. how do the other teams won in prizes at the chanpionship when none of the judges came to our pit? I DON'T KNOW.

This definitely isn't the first time I've heard complaints about this. I understand that there are only so many judges, and there are so many teams to visit, but I would definitely be upset if we didn't get to say hello to the nice people in the blue and green shirts. Even if a lot can be said about a team by their actions (which is a good thing), not being able to explain your team organization and history, or even being able to hand over a business plan is a huge loss and a huge disappointment.

[apalrd]

Originally Posted by Karibou

...but I would definitely be upset if we didn't get to say hello to the nice people in the blue and green shirts...

Blue shirts, yes I would be upset
Green shirts, not so much. I would actually rather not talk to them.


About the complaints on the control system:
1. The system is two years old. There are bugs. We don't like them, but they are still there and haven't been worked out yet.

Scenarios that are true: About half an hour before "ship" (bag), we begin to do some drive practice in the basement on the practice field and the arm (that arm with 650ft-lbs of torque at stall) suddenly goes out of control, slams into the chassis control bar, and then pushes that into the tower (and is disabled). Having no idea what is going on and wanting to not fix it then (drive practice is more important), we disconnect the motor wires and continue. We didn't see it, but the kicker is also acting funny (but its hidden by the hood). A few days later, we opened the bag for a two hour unbag time before our week 1 district. We power on the robot, but do not enable it, and probe the voltage of the arm potentiometer (and chassis string pot) using a multimeter. Both are about where they should be. In software, we then see that the voltage is actually returning 0. We can't figure out what is wrong. A PWM cable did not come disconnected, because the sensors have power. Could it be a loose connection between the Analog Bumper and the Analog Module, could it be a bad Analog Bumper, could it be a bad Analog Module, etc. We send a runner to grab a spare analog module and bumper from the practice bot upstairs. We look at this one, and see that the analog module has actually come out of the cRio. We have checks in software to see if the sensor is out of range (like shorted to ground or +5), but when the module ejected the Get Avg Voltage returned the last good voltage, which was legitimate. We re-inserted the analog module, glued it in with silicone, rebooted, and were fine. Kinda annoying thing to happen.

Numerous practice occurrences: We find that, suddenly, all of our Victors and Spikes loose comm with the cRio. We are enabled, but something is wrong. What could it bee? A loose power connection to the Digital Sidecar? A loose DB37 cable (two possible loose connections)? A loose digital module? It turned out to be metal shavings in the DIO side had shorted out the +5 line (which left the disable circuit unpowered).

Michigan State Championship: During inspection, we find that there is a short between ground and the frame. Pulling the cRio connection we find that it is the cRio (we guessed cRio first). After removing the cRio, verifying that the plastic sleeve on the bolt is intact, we begun pulling modules until we discovered it was on the digital sidecar's line. We looked for possible shorts on the digital sidecar, removing each PWM, Relay, and DIO until none were it. We then saw that one of the bolts securing the digital sidecar was touching a screw for the DB37 cable. If we wiggled the connection, we would loose the continuity between the frame and ground. After finding this and covering the bolt with electrical tape, were now left with a robot that had all of its digital IO disconnected and its cRio in pieces.

More MSC: While re-assembling it, we broke something on the Analog Bumper or Analog Module (we don't know which). We looked at the kicker and the arm, and both were oscilating. Graphing all of the used analog inputs we found that all of the analog inputs were oscilating, by around .4 volts. We replaced the analog module and bumper with our only spares, which we got from our practice bot that was in Jim's truck.

Kettering: We found that Kitty's kat box had no input at all. It simply wasn't being found by the Classmate. Apparently there is a bug in the Cypress driver that will cause it to crash under the right conditions, and it's more likely to happen when connected to the FMS. The solution is to reboot the classmate twice, but full Windows reboots take more than the 5 minutes allowed for team-caused field delays (However, the FTA at Kettering was nice and let us reboot the classmate if it happened). We implemented a check for loss of Kitty's box (pull DIO 8 high and check if it goes low), but we then found that if it has the Cypress board and looses it it will continue to send the last good data, and that check will fail (however, it the Classmate boot without getting any data at all, it will send blank data and this check will pass). We then transfer the major functions (kick, deploy claw, flip robot) to buttons on Matt's gamepad.


33 control system failures this year:
Analog module fell out of cRio (friction locks are not good enough)
DSC poorly designed so mounting bolt can touch DB37 screw and create ground short which is almost impossible to track down
Analog module or bumper broke for no apparent reason
Radio spontaneously reboots going over the bump (poor reliability of radio, mounting orientation helps this)
Radio power cable loosens going over the bump (poor power connection design, a friction based solution will NEVER work in FIRST, except Lunacy where stresses were really really low)
Cypress board/driver has issues, causing loss of Kitty's Kat Box when booting (Greg McKaskle says its around 1 in 10 times)
No App switch often has to be pressed to download code (lack of any hint of optimization in WPI code)
Builds take more time then a match (LV-RT design flaw causing it to re-compile entire library every build)
FMS Lock (design oversight; in past systems a FMS lock would have been reset when the FMS cable was unplugged because the DS would reboot on power loss, now its very obvious that it exists and is hard to get rid of)


Lack of blame: If you call/email NI for support, they can redirect you to WPI if it could be a software problem (who could likewise redirect you to NI if its a software problem), or any number of other sources. It's hard to blame the fault on a single company, so it's hard for any one company to give you support on products that aren't entirely theirs.

[Jaxom]

Originally Posted by JaneYoung

Dark numbers on the bumpers - red and blue. Don't use dark colors. Couldn't see the dark numbers on the webcast.

I couldn't agree more...black numbers (however wide) on blue bumpers Just Doesn't Work. But what I don't understand about that is that R15 is quite clear -- "...in a contrasting color....". Now, I guess there can be some interpretation of that, but the point of the rule is clear -- people need to see the numbers -- so I really can't understand why black on blue was acceptable. While inspecting in the St. Louis regional, we made teams renumber their bumpers in contrasting colors. Meaning white. One coach asked if he was expected to go out & buy white paint -- we told him yes, if that was what it'd take. Unsurprisingly, when the announcement went out "anyone have white paint?" the answer was "yes".

The simple answer to this problem is as someone else suggested in the thread -- make R15 say "white numbers on all bumpers". Done.

[Justin Montois]

Originally Posted by pfreivald

Our Overdrive robot extended outside our bumper zone a maximum of 4", and then only while acquiring the ball, so that we would not incur incidental contact penalties. Watching the games and counting actual incidental contact from various trackball manipulators, I think *every single match* would have been a 0-0 tie if they were calling it properly...

...which means that teams are so used to penalties not being called for certain rules that they don't even worry about them... So why have them?

Ball manipulation devices often touched and not receiving a penalty for this is the CORRECT call.

Incidental contact penalties do not exist.

G37...
http://www.usfirst.org/uploadedFiles...0Game-RevG.pdf