Power-related problems can be partially mitigated by thinking about how you program your robot. For example, in a motor "move" command you could specify to run the motors for, say, 2 seconds or alternately for 5 rotations. On a freshly charged set of batteries these might perform equivalently. But the time-based move will not perform consistently at different power levels since the robot will travel further in 2 seconds on a fully charged battery than on a poorly charged one. The rotation-based move, on the other hand, should be repeatable. That command will force the robot to go the full specified measured distance; it will just take longer when the batteries are lower on power. This was one of the major advances of the NXT over the RCX - every motor has a built in rotation sensor. When used properly it really does help. - Glenn From: First Lego League in Virginia and DC [mailto:[log in to unmask]] On Behalf Of Mark FLL Allen Sent: Wednesday, November 04, 2009 4:29 PM To: [log in to unmask] Subject: Re: [VADCFLL-L] Some answers to RE: [VADCFLL-L] more problems All, I received an off list question about how much the batteries can drop off. Our experience was that a freshly recharged brick was about 8.1 - 8.2 volts. After 30 minutes this would be down to <8.0 and repeatability became an issue. We literally plugged it in at practices every time it stopped running including for short programming changes. A warm fully charged brick was very repeatable within the limitations of the robot design. On a side note it is worth mentioning that this really is not due to a deficiency in Mindstorms but rather a physics issue. If the robot turns on drive wheel rotations and the non-drive side is sliding (friction), as power changes the robot will not turn exactly the same even though the drive wheel rotations do. Forward/Back motions should not be effected if based on rotations (time is a different issue). For next year (I hope everyone returns) consider pre-season sessions testing the effects of different designs on variability vs. power. Below are some ideas I sent to a coach after last year. Best regards, Mark ------------------------------------------ A few quick ideas and exercises that helped my kids get better with the robot piece through the years. Navigation; Pick a spot on the table a long way away from base that can be reached without any turns. Program robot to go straight to get to that point. Place a piece of paper on the table (tape it down). Run the robot say 5 times for each kid aiming at that spot. When the robot stops mark a dot on the paper using some reference off the robot - perhaps use a different color for each kid. How much variability is there? Why? Does it differ by the operator? Also, if they are using time versus distance to navigate run a similar exercise comparing the two (hopefully they are not using time for general navigation). Ask them when using time makes sense. Examples include when you need to get to a fixed object like a wall. Now program the robot to get to about the same spot except line it up against a wall in base (note a way to line it up carefully) have the robot go forward, make a turn, and go forward again to about the same spot (doesn't have to be exact), and repeat the paper marking exercise. Which method has more robot variability? Which has more operator to operator variablity? What does this tell you about the margin of error for completing a task on the table? How else can you improve navigation? Examples to consider include travel along a wall, backing into a wall to line up, using bumpers and side rails. Can you get to the Ice Core Drilling rig more effectively one way versus another. Also, what happens as batteries run down? How can this be minimized? Robot Design; Considering the learnings from the navigation exercise, what should be considered in the robot design. Is more weight better or worse? Which wheels work best? Try thin wheels versus wider wheels. Discuss the needs for a good robot. From my experience the kids can do very well if a robot can reliably travel forward/back, turn left/right, and move an arm up/down repeatably. One other very important consideration is coming up with a way to quickly change attachments. Programming Only one piece of advice - have everyone go through the ORTOP tutorial over the next few months. Make a quick exercise after each section. Similar to quickly changing attachments, moving from one task to the next quickly is very important. The kids should explore ways to make changing programs easier or combining them using what they lear n going through the tutorial. The level of sophistication can very depending on the age/capabilities of the kids. Confidence Finally, after going through all this above (which will probably take weeks or months), have the kids try to accomplish a task from this year's challenge that they thought was too hard or struggled with. I'll bet if they go through everything above they'll make tremendous progress. If you get through all this or have questions, just give me a call or send me a note and we'll keep them challenged. -----Original Message----- From: Mark FLL Allen <[log in to unmask]> To: [log in to unmask] Sent: Wed, Nov 4, 2009 2:58 pm Subject: Re: [VADCFLL-L] Some answers to RE: [VADCFLL-L] more problems Folks, Our team found the most repeatable method is to use rechargable batteries (or the pack) and recharge before every run. The repeatability, in years of testing, was much higher than we could ever get otherwise. New non-rechargable batteries deliver high power with poor repeatability. At practices, we had two sets of batteries and recharged one set while using the other. At tournaments, task one after a robot run was to recharge. Best of luck to all, Mark Allen Coach Team Fire Phoenix 2004 - 2007 -----Original Message----- From: Laura Dysart <[log in to unmask]> To: [log in to unmask] Sent: Wed, Nov 4, 2009 1:27 pm Subject: Re: [VADCFLL-L] Some answers to RE: [VADCFLL-L] more problems Thanks Curt for pointing out I needed to reply all: all- this battery issue is a little upsetting- we did have the problem below with battery- after putting in new batteries- our student did adjust power level- seems like our team is capable of making adjustments to the robot when it doesn't work based on the battery- but our team will NOT have a computer at the tournament on Saturday to make adjustments there- I'm a little worried about the battery- makes me wonder if we should stop doing practice runs..... I've thought of bringing my husbands lap top but its a Mac and we've been using the school's computer which is not a Mac. I have told them that no matter what happens on Saturday, they have seen their robot do the mission successfully. At one point we had it 10 out of 10 trial runs- changed the battery- have since made adjustments but it's never been the same since. know it's a lesson in batteries- but are there any tips I am not thinking of/ Thank you- Laura Dysart On Nov 4, 2009, at 11:42 AM, Curt Tran wrote: > I just want to share some of our experiences during the past few > years with > all the rookie teams. > > > [Robot Acting Differently] > Please be aware that the NXT motors are running on battery pack! > Just like > any computer, the NXT software and CPU on the brick should run > consistently > every time. Since this is Robotics, beside the computer chip and > software, > there are also moving parts (motors), sensors (that need > adjustment), and > hundreds of LEGO pieces (that might come loose). We have a student > screaming this Monday "Did someone make any change to the Robot; it > was > running perfectly when I left on Saturday!" What has happen was > that he > wrote the program with a half charge battery on Saturday afternoon, > and we > have fully recharged the battery before they met again on Monday! > The fully > charged battery will allow the motor to run a bit stronger and will > throw > off a (precision) program that worked perfectly the day before. > Just like > any other battery operate power equipment, a fully charge power > drill will > run differently versus one running on a low charge battery pack! > Don't be > surprise when the kids have finish all their missions perfectly the > Friday > night before just to find out nothing run right the next Saturday at > the > competition! We blame it on the table, lighting and mission models > at the > competition. But, it can also be that fully charge battery on > competition > day! > > It is really tough for a rookie team to deal with all the > challenges, we > been there and know exactly what you are going thru. As coaches, > parents, > teachers, we need to stay calm and not stress out as competition day > is > quickly approaching since the kids are just watching us to see if we > will > give up and throw in the towel or not! > > > [Block Skipping] > Please be aware that the 'Motor' block has a 'Wait for Completion' > checkbox, > while the 'Move' block has a 'Next Action' selection of 'Brake' or > 'Coast' > (on v2.0 only). These selections have significant affect on a > program. As > an example, if you have the 'Motor' block to move an attachment (or > a wheel) > for 1 rotation, but something is blocking it and the motor cannot > turn a > full 1 rotation, with the 'Wait for Completion' check, the program > will hung > since it is still waiting for the motor to complete the 1 rotation > before > move on to the next block. However, if you don't check the 'Wait for > Completion' and the 'Power' setting for the 'Motor' block is not > strong > enough to lift three loops, the program will just continue with the > next > program block and continue running like the 'Motor' block never even > get > executed. > > > [Debugging Program] > The best way to debug a program is to use a 'Wait' block (the hour > glass > icon). Just change the default 'Touch Sensor' to the 'NXT Buttons' by > select it from the drop down list box. Now you can make as many > copy of > this wait block as needed and place them anywhere after any block of > code. > The robot will stop and wait until you press the 'Enter' button > before it > continues on to the next block. You can delete these 'Wait' blocks > after > you are done debugging and the program is ready to go as expected. > > > [Software Versions] > Please be aware that there are different version of LEGO Mindstorms > NXT > Software (v1.0, v1.1, and v2.0), and there are many different > version of > Firmware (v1.0, v1.05, v1.07, v1.21, v1.26, v1.28) out there. You > will need > to make sure you have the right version of NXT Software matching up > with the > right Firmware version. We have discovered it the hard way when our > students program on different PCs which have different Mindstorms NXT > Software and the brick (having different Firmware). In one case, we > have > download Firmware v1.28 from a PC with NXT Software v2.0 to a brick, > and > then download program written on an older NXT Software v1.1 to the > same > brick. The robot starts acting very strange, and it did even lockup > sometime. The following are the best match up that we have found (any > additional comments on this are welcome): > > Mindstorms NXT v2.0 - Firmware v1.28 (best). > Mindstorms NXT v1.1 - Firmware v1.05 (okay). > Mindstorms NXT v1.0 - Firmware v1.0 (very old). > > To check the version of the NXT Software, run the NXT programming > software, > click on HELP and select "About LEGO MINDSTORMS Edu NXT..". It > should say > "LEGO MINDSTORMS NXT v2.0" or "v2.1" under the robot picture on the > right > side. > > There are two way to check the Firmware version. On the brick, when > first > turn on, press on the right arrow about 4 or 5 times and select > 'Settings'. > Press the right arrow 2 times and select 'NXT Version'. The first > entry in > the display 'FW' will tell you the firmware version on the brick. > The other > way is to connect the brick to the PC running Mindstorms NXT > software, then > open any program and click on the 'NXT window' icon (it's the one on > top of > the 'Download' icon). If the brick is connected, the Firmware > version is > display at the bottom right on top of the 'Close' button. (This is > also > where you can manage the memory on the brick and delete any unused > programs > by clicking on the 'Memory' tab.) > > (For Mac user check out this blog for installing NXT v1.1. and NXT > v2.0: > http://foresthillslego.blogspot.com/2009/10/mindstorm-nxt-20-installation-on > -mac.html) > > > [Dead Battery] > If you insert a fully charge battery, but the brick will not turn on. > Remove the battery and check the spring where it's making contact to > the > positive and negative posts of the battery. Sometime they are bent, > and > will not make full contact. Just adjust it a little and this might > fix the > problem. We also have a dead battery where it displays fully charge > but > just not working at all. You can only test this by putting the > re-chargeable battery in another NXT brick. This is where 6 'AA' > batteries > come into play, but be aware that the weight of the batteries can > throw off > a programs on the robot that been programmed with the re-chargeable > battery. > > > [Dead Brick] > We have encountered a few problems with dead brick. When download the > firmware, make sure the PC and the NXT brick's batteries are fully > charge. > If the firmware download failed during update for some reason, the > brick is > dead and will not turn on. To fix this the brick will need to be > reset (by > pushing the hardware reset button with a paper clip) and download the > firmware again. If your NXT is clicking when you insert batteries, > push the > hardware reset button for five seconds before you insert the USB > cable. This > will ensure that the brick initializes correctly for the firmware > download. > > [The hardware reset button is located within the LEGO TECHNIC hole > below the > USB connector on the NXT brick.] > > > > Best wishes, > T. (Curt) Tran > 2007, Coach Team #3563, Rabid Llama Lords. > 2008, Coach Team #324, Scitobor. > > -----Original Message----- > From: First Lego League in Virginia and DC > [mailto:[log in to unmask] <mailto:[log in to unmask]> ] On Behalf Of Phil Smith III > Sent: Tuesday, November 03, 2009 3:16 PM > To: [log in to unmask] > Subject: Re: [VADCFLL-L] more problems > > I'm sorry you're so frustrated. I can certainly believe it, > especially this > close to Regionals! > > At the risk of being called a Pollyanna (or worse), let me suggest > that you > make this into a learning exercise for the team: > > - What could (not *should*) they have done to avoid/plan for this? > ("Have a > spare brick" being the obvious, if not necessarily practical, answer) > > - What can they do *now* to get around it? (Do any of them own a > brick? Can > you buy one and then return it once this one is fixed? Have you > called LEGO > for support?) > > - Could it be a battery issue? Have you recharged it since the > problem? > > - Can you try it on a PC instead of a Mac? > > FLL is all about solving problems. This is beyond the level of > problem they > SHOULD have to deal with, of course, but that doesn't mean they > can't be > encouraged to step up and try to figure it out. > > And I hope you'll come to competition anyway -- the table runs are > 1/4 of > the points, and the whole point is the experience, not "winning". In > fact, > they might find another team with an idea, or a spare brick to lend... > --> Phil Smith III > Virginia State Judge Advisor, 2007, 2008, 2009 > Judge Advisor, Northern Virginia Regional tournaments, 2007, 2008, > 2009 > Division 1 Judge Advisor, Virginia State tournament, 2006 > > Coach, The Capital Girls, Oak Hill (retired) > Team 1900 (2002) > Team 2497 (2003) > Team 2355 (2004) > Team 1945 (2005) > -----Original Message----- > From: First Lego League in Virginia and DC > [mailto:[log in to unmask] <mailto:[log in to unmask]> ] On Behalf Of Sherene > Silverberg > Sent: Tuesday, November 03, 2009 2:38 PM > To: [log in to unmask] > Subject: [VADCFLL-L] more problems > > Lee Hilt kindly sent me a picture that showed us how to find the reset > button. My son depressed the button (I was unable to see it with my > old eyes) and now the brick refuses to switch on or to stop clicking > so we can't reload the drivers. > > I have to say that I went into this FLL League with great excitement. > I am now totally turned off by the entire experience. Buggy > equipment, buggy programs, drivers download pages hidden so well that > unless you are given the exact page, even Google's spiders struggle to > find it. > > I have received 6 emails in the half an hour since I sent my original > email. All these people are experiencing the self same problems as we > are experiencing with the programs that run, not resembling the > programs that were downloaded. > > This is the most appalling situation I have ever experienced. Lego > needs to get their act together and sell equipment that works. Driver > and firmware updates should be easily found. This is an exceedingly > expensive exercise and for this money, I expect far better service and > equipment. > > My team has spent more work trying to fix Lego issues than they have > on developing their programs. It is most frustrating for my team to > write a program, have it run perfectly 3 times and then on the 4th > time have to execute a program that does not resemble the one that ran > perfectly 3 times prior. > > Now we have a dead brick just because we attempted to reset it. > > Just in case you haven't realized it yet, I am fuming. > > Sherene > > -- To UNSUBSCRIBE or CHANGE your settings, please visit > https://listserv.jmu.edu/archives/vadcfll-l.html and select "Join or > leave > the list". > > -- VADCFLL administrative announcements are sent via VADCFLL-ADMIN-> L. Visit > https://listserv.jmu.edu/archives/vadcfll-admin-l.html to subscribe. > > -- To UNSUBSCRIBE or CHANGE your settings, please visit > https://listserv.jmu.edu/archives/vadcfll-l.html and select "Join or > leave > the list". > > -- VADCFLL administrative announcements are sent via VADCFLL-ADMIN-> L. Visit > https://listserv.jmu.edu/archives/vadcfll-admin-l.html to subscribe. > > -- To UNSUBSCRIBE or CHANGE your settings, please visit https://listserv.jmu.edu/archives/vadcfll-l.html> and select "Join or leave the list". > > -- VADCFLL administrative announcements are sent via VADCFLL-ADMIN-> L. Visit https://listserv.jmu.edu/archives/vadcfll-admin-l.html to > subscribe. -- To UNSUBSCRIBE or CHANGE your settings, please visit https://listserv.jmu.edu/archives/vadcfll-l.html and select "Join or leave the list". -- VADCFLL administrative announcements are sent via VADCFLL-ADMIN-L. Visit https://listserv.jmu.edu/archives/vadcfll-admin-l.html to subscribe. _____ To UNSUBSCRIBE or CHANGE your settings, please visit https://listserv.jmu.edu/archives/vadcfll-l.html and select "Join or leave the list". VADCFLL administrative announcements are sent via VADCFLL-ADMIN-L. Visit https://listserv.jmu.edu/archives/vadcfll-admin-l.html to subscribe. _____ To UNSUBSCRIBE or CHANGE your settings, please visit https://listserv.jmu.edu/archives/vadcfll-l.html and select "Join or leave the list". VADCFLL administrative announcements are sent via VADCFLL-ADMIN-L. Visit https://listserv.jmu.edu/archives/vadcfll-admin-l.html to subscribe. -- To UNSUBSCRIBE or CHANGE your settings, please visit https://listserv.jmu.edu/archives/vadcfll-l.html and select "Join or leave the list". -- VADCFLL administrative announcements are sent via VADCFLL-ADMIN-L. Visit https://listserv.jmu.edu/archives/vadcfll-admin-l.html to subscribe.