1726 progress - week 4 [Archive]

TheOtherGuy

30-01-2009, 00:06

Finally got everything moving at once tonight! Here it is:

http://www.youtube.com/watch?v=U6PchiI4bDQ

drive motors are going slow due to some semi bad programming (half as fast as they should be, we JUST started wiring a pistol grip RC controller). The turret locks on but is a little slow, so we'll also need to work on some PID constants. Overall though, we're quite pleased with the intake and hopper system.

Comments / suggestions / questions appreciated!

Don't forget to check out our photo gallery and youtube page:
1726 photo gallery (http://photos.project1726.org/thumbnails.php?album=34)
1726 youtube page (http://www.youtube.com/user/frc1726)

and for those at school with youtube blocked:

http://photos.project1726.org/albums/userpics/10010/normal_IMG_9079.JPG (http://photos.project1726.org/albums/userpics/10010/IMG_9079.JPG)

Vikesrock

30-01-2009, 00:09

Wow, nice bot!

Is that fire rate still with a single CIM driving the shooter? Also have you gotten a chance to test how accurately you can vary distance based on shooter wheel speed?

Arefin Bari

30-01-2009, 00:11

That is one beautiful robot.

TheOtherGuy

30-01-2009, 00:11

Wow, nice bot!

Is that fire rate still with a single CIM driving the shooter? Also have you gotten a chance to test how accurately you can vary distance based on shooter wheel speed?

Thanks! yea, this is a single CIM powered turret at ~3600RPM. We're still controlling the speed with a joystick but we're planning on spending all of tomorrow getting it to shoot distance-wise.
EDIT: sorry, I forgot to mention that we have indeed tested distance and even with a joystick it is pretty darn controllable :)

Woody1458

30-01-2009, 00:26

can the helix be run in reverse? So that you can pull in a empty ball and spit it out without having to empty your helix.

CCCP

30-01-2009, 00:28

Yes it can :D :D :D

Michael Corsetto

30-01-2009, 00:47

This is an amazing robot, especially for being (mostly) done by the end of week 4. Does your shooter adjust takeoff angle or wheel velocity to change the range? What percentage of shots do you make from that range (8-10 feet from the looks of it)?

Excellent robot 1726, thanks for keeping us all updated with your progress!

=Martin=Taylor=

30-01-2009, 00:50

That looks amazing. Maybe I'm going to have to stop doubting what that camera can do. :ahh:

I wish we were as fast and organized :(

TheOtherGuy

30-01-2009, 00:52

Thanks! Unfortunately we just got the camera tracking on the turret today, so we really haven't had much time to adjust anything. We're planning on creating a "leading" program based on the speed of the turret encoder, but that might take a few days to implement considering all the other things that need to get done (traction control in auto?). Stationary we can make every shot from any distance, although our shooter speed is still controlled by humans and as such keeps fluctuating slightly :P
We'll (try to) keep everyone posted on our progress with the turret!

Woody1458

30-01-2009, 00:57

That is really awesome! great job NERDS. I can't wait to see it at work in Silicon Valley Regional.

s_forbes

30-01-2009, 00:57

That is really awesome! great job NERDS. I can't wait to see it at work in Silicon Valley Regional.

You'll have to wait a while, this one's only going to LA and AZ!

Woody1458

30-01-2009, 01:01

Darn! well Ill see you on TBA i guess

Also when I said Silicon I meant Sacramento, and I based that on seeing you there in 2007

Akash Rastogi

30-01-2009, 01:24

What a crazy buncha NERDS and your robots....

Great looking bot guys. I'm loving the helical loader. Is the entire thing being painted blue and white?

s_forbes

30-01-2009, 01:39

The current plan is to make an improved drivebase, but painting it beforehand. The rest of the wooden parts will likely be painted blue and white as well.

Neat thing about the shooter though: since it's been operated so much, there's a nice little oil streak along the wood in line with the chain. There's also lots of oil marks on it from where we grabbed it with dirty hands... not sure how well paint will stick to it. We could always just build another one, we might actually have enough time this year to make the robot look good. :)

CraigHickman

30-01-2009, 01:40

Woah. That's one impressive bot!

adengler

30-01-2009, 02:01

Impressive robot all around! Best of luck at the regionals!

dtengineering

30-01-2009, 02:03

That looks great, and if the tracking gets tuned in over the next couple weeks it should be deadly in competition.

Rather than painting it blue, however, might I suggest a nice bright aniline dye and some lacquer? Our drive modules a now a nice bright red baltic birch... and as soon as our second toughbox arrives (yes... we were short shipped one in the KoP and FedEx tracking says it will arrive tommorrow) we'll get it mounted and take some photos.

Thanks for the update, great to see that things are going well and that beautiful helix is working well.

Jason

GaryVoshol

30-01-2009, 06:37

Oh man! You make the scorer's job hard. :rolleyes:

Since I won't see it in AZ or LA, I'll expect to be watching it in Atlanta. Preferably on Sat afternoon.

Doug Leppard

30-01-2009, 06:50

Good job, thanks for showing us.

sdcantrell56

30-01-2009, 06:54

That is an awesome robot and especially this quickly. It is eerie how similar our designs are even down to the wood, but somehow you guys are ahead of us by at least a week. With all of this time to debug stuff you guys are going to be unstoppable. Good luck and I can't wait to see the finished product.

Enigma's puzzle

30-01-2009, 08:32

i however can wait to see the finished product, ill find a way to wait until the Championships, that is unless you are on our team.

But wow, im intimidated.

pheadxdll

30-01-2009, 08:46

Awesome looking robot. Just a quick question!

How are the balls in the spiral harvester fed to the top?

MrForbes

30-01-2009, 09:07

thanks for the kind words....

The balls are fed to the shooter by a small roller wheel just under it, at the top of the helix. This is another part we need to work on, the helix drives the balls up against this and the balls can sneak out around the part that is supposed to stop them. Today's challenge for the mechanism team is to design a cure for this problem.

The picture is a few days old, showing the roller on the prototype top part, looking up from below.

http://photos.project1726.org/albums/userpics/10010/normal_IMG_0199~0.JPG

Brandon Holley

30-01-2009, 09:24

1726...GREAT JOB!!

Very impressive robot.

Kudos to you guys for getting this done this quickly.

GO NERDS

spc295

30-01-2009, 09:58

:rolleyes: now the programmers get the robot right

MrForbes

30-01-2009, 10:01

What's interesting about this year, is that the programmers have already had the robot for most of the build season, and they got a head start by spending a few work sessions getting the new control system up and running before kickoff. We have a much bigger programming team than we've ever had (4-5 students vs. 1) and they are taking advantage of many of the features that LabView offers.

Donut

30-01-2009, 12:41

That's a deadly looking robot there! Have you gotten to testing while the trailer moves around yet (or is your strategy to wait for opportune moments)?

1726 has been one of the top teams at AZ since their inception (even with last year's electrical troubles they still had one of the best hurdlers there). With the experience of LA before the regional this year (and another 2 weeks to work on the robot) you're going to be difficult to contend with :ahh:

Oh, and will that whole side panel for electronics be removable (or have a quick release cover)? That looks like a nice easy access design.

Sean Raia

30-01-2009, 13:05

WOW! i never thought of feeding the ball to the turret like that! I like how you mounted the electronics, make sure thats stable because if the gets bumped it looks like it might fall off.

Nice design though!

pitzoid

30-01-2009, 13:15

Nice job NERDS! Can't wait to see you guys on the field in AZ.....

JesseK

30-01-2009, 13:50

Innovation with wood, and a near 360 degree turret it looks like; great job guys!

Now, it's time for some color!

XXShadowXX

30-01-2009, 13:55

what do you use to turn the turret; servo, motor, or is connected right to the robot's frame

Vikesrock

30-01-2009, 14:00

what do you use to turn the turret; servo, motor, or is connected right to the robot's frame

Their turret is powered by a Banebots RS545 through a 36mm 64:1 transmission. It has a V-belt pulley on the output that drives their "turret base" which is actually a very large V-belt pulley.

Pictures can be found here:
http://photos.project1726.org/thumbnails.php?album=34&page=1

Specifically the turret motor and pulleys:
http://photos.project1726.org/displayimage.php?album=34&pos=15

JoeArnold

30-01-2009, 18:20

Nice.

falconmaster

30-01-2009, 23:12

As usual, Awesome! Just Awesome! I wish I were you! Way to go!

TheOtherGuy

31-01-2009, 00:56

Thanks for all the feedback!

I sure hope the programmers get the robot for the last week this year, that would be a first! Although we're mostly still waiting for the rest of the robot to be "adjusted" before continuing :cool:

So we basically did a small distance test today with PWM values and distance; we managed to get very consistent shots considering the variety of conditions our balls are in. Plugged into a jaguar, here are the basic results:

output(0to1) Distance
.3-.4 0-1ft
.4-.5 2ft
.5-.6 3-4ft
.6-.7 6-7ft
.7-.8 8-9ft
.8-.9 9-10ft
.9-1 9-10ftNot very exact (off the top of my head, I'm pretty sure it's right) but it is much more accurate than it sounds. We'll map this out and create some relationship between the motor speed and distance. We're planning on using the height or size of the target as the distance calculator. If we have time we'll add some form of revolution counter to the shooter for more a more determinate angular speed than PWM values (3500rpm is kinda fast for an encoder... I saw 111 used a reflection sensor on their 2006 robot in the "Behind the Design" book? perhaps that'll work...).

EDIT: I have been reminded to mention that the distance calculated was between the front of the robot and the back of the trailer (space between the robot and target trailer). Which may or may not be important :rolleyes:

CraigHickman

31-01-2009, 01:02

Not very exact (off the top of my head, I'm pretty sure it's right) but it is much more accurate than it sounds. We'll map this out and create some relationship between the motor speed and distance. We're planning on using the height or size of the target as the distance calculator. If we have time we'll add some form of revolution counter to the shooter for more a more determinate angular speed than PWM values (3500rpm is kinda fast for an encoder... I saw 111 used a reflection sensor on their 2006 robot in the "Behind the Design" book? perhaps that'll work...).

A good way to sense speed on a high velocity wheel like that would be to put a small magnet into the wheel (drill a small hole and put an iron nail in, slice the head off), and set up a hall effects sensor. Each rotation should be VERY noticeable, so you should be able to get some speeds out of that....

Wayne Doenges

31-01-2009, 07:06

Good looking robot and a real nice shooter.
I have one question. Will it track, and score, a moving target?

MrForbes

31-01-2009, 09:55

We're working on that. Still need to get several other things squared away first, but scoring on a moving target is one of our prime goals, so we will surely try our best to make it work. Our first test of this ability was kind of disappointing.

Doug Leppard

31-01-2009, 13:08

What size roller did you use? What is the material around the roller?

AHS1599

31-01-2009, 14:18

That bot looks really "screwy" :) Nice job!!

MrForbes

31-01-2009, 21:43

What size roller did you use? What is the material around the roller?

The shooter and helix rollers are 4" ABS. The shooter roller has gum roughtop conveyor belting (wheel tread), the helix roller has shelf liner from Walmart, attached with aluminum strips to hold it on.

http://photos.project1726.org/albums/userpics/10010/normal_IMG_0442.JPG

here's the only picture I could find just now...and we redid this, the flaps are each only about an inch wide now, instead of 2".

It might have some wear issues, so we may be looking for something else to use?

Doug Leppard

31-01-2009, 21:59

The shooter and helix rollers are 4" ABS. The shooter roller has gum roughtop conveyor belting (wheel tread), the helix roller has shelf liner from Walmart, attached with aluminum strips to hold it on.

here's the only picture I could find just now...and we redid this, the flaps are each only about an inch wide now, instead of 2".

It might have some wear issues, so we may be looking for something else to use?

Thanks that was helpful. Ours is a 5 inch running at 2500 rpm. It is not working yet other than the proto type. So these numbers give us a hint how we might be doing.

MrForbes

31-01-2009, 22:01

The shooter is set up with chain drive, 15 tooth motor to 22 tooth roller, from the CIM. The helix is driven by an AM Stackerbox with a CIM on it (s_forbes from our team seems to have come up with the idea of putting the CIM on that transmission, and AM put it into production), I think it's 3.57:1 ratio, driving a 9 tooth sprocket, and a 22 tooth on the roller.

Manoel

31-01-2009, 22:10

Very impressive!

Looks a lot like our mechanisms (helix and shooter are very very similar), except you guys are testing and debugging and we're staring at a pretty SolidWorks model yet to be built... ;)

RandomStyuff

01-02-2009, 13:32

Wow, extremely impresive!

How did you make the CIM motor move by angle?

sdcantrell56

01-02-2009, 13:38

Wow, extremely impresive!

How did you make the CIM motor move by angle?

I am confused. The CIM motor is not moving at an angle anywhere on there robot. The CIM is chained to the shooter wheel with 2 sprockets. I'm only answering because apparently we have created near identical robots without any form of collaboration. :ahh:

s_forbes

01-02-2009, 13:54

Distance is changed based on motor speed, which has proven to be consistent enough to work during gameplay. Fortunately, with the shooter wheel spinning slower it still spits balls out at the same rate (about 4 balls per second).

I'm only answering because apparently we have created near identical robots without any form of collaboration. :ahh:

haha, I hope that means we're on the right track. :p

EDIT: The turret is turned by another motor (a Banebots), if that's what you're referring to. It's turned by a v-belt pulley.

Oh, and I almost forgot; here's a video of some of the testing we did yesterday: http://www.youtube.com/watch?v=I9wl1x9Wf_Q

RandomStyuff

01-02-2009, 13:54

I am confused. The CIM motor is not moving at an angle anywhere on there robot. The CIM is chained to the shooter wheel with 2 sprockets. I'm only answering because apparently we have created near identical robots without any form of collaboration. :ahh:

EDIT: nvm, just a misuderstanding... i get it now...

sdcantrell56

01-02-2009, 13:58

Distance is changed based on motor speed, which has proven to be consistent enough to work during gameplay. Fortunately, with the shooter wheel spinning slower it still spits balls out at the same rate (about 4 balls per second).

haha, I hope that means we're on the right track. :p

I hope we're on the right track as well. It would seem like it from our preliminary testing and your much more finalized testing. The only difference between our bots is somehow we made ours way more complicated which is not really a great thing. HAHA. I must say it looks really cool though. We are able to change our shot angle by adjusting the hood of our shooter though so we do not have to decrease speed.

=Martin=Taylor=

02-02-2009, 00:10

How high is the bottom roller off the floor and what is the traction material on it? Spray on rubber?

MrForbes

02-02-2009, 00:12

Akash Rastogi

02-02-2009, 00:31

It's about 7.5" off the floor iirc, and it is spray on rubber. We also used spray on rubber for the roller that feeds the balls into the shooting turret, but it didn't do the job there...so we added some roughtop.

The bottom intake roller works great.

I couldn't find any today at Lowes but really wanted to get some for a team we're mentoring. Any other places its available?

MrForbes

02-02-2009, 00:33

We got it at a building supply store that only has locations out West...but try this

http://www.plastidip.com/diy_where_to_buy.php

mcf747

02-02-2009, 01:27

Nice tracking on the turret, but is it also adjusting speed too? It seemed like the balls were hitting the target from different distances but there was no solid clip of video with distance change just splices?

See you at LA.

Matthew Forman
Mortorq Team 1515

GUI

02-02-2009, 01:32

The shooter was not programmed to adjust for speed in the video, but the code is in the works.

TheOtherGuy

03-02-2009, 01:58

New video! We made a little bit of progress on the camera tracking (helps that the manipulators all work perfectly now :rolleyes: )

http://www.youtube.com/watch?v=Bvf33dblw2I

We were trying to add in lead tracking to account for lag in the camera, to little or no success thus far. And we still have to implement a distance calculating algorithm (programming takes forever... but we have 2 weeks this year! hurrah!).

Doug Leppard

03-02-2009, 06:50

You guys are great, thanks for the videos, it inspires our team.

BHS_STopping

04-02-2009, 00:21

Doug Leppard

04-02-2009, 06:32

Another new video, this time with distance sensing! Our programmers made a pretty nifty code which we calibrated by taking a few data points (PWM value vs. distance fired). We're thinking about some type of encoder for the shooter roller, but so far we're definitely satisfied with what we've been able to accomplish.

Anyway, here's the video! http://www.youtube.com/watch?v=eTCr2nvFn3g

And thanks for the comment Doug, any chance to help inspire another is a chance worth taking!

Team 1726 you guys are just fantastic. By ship date your shooting will be unstoppable.

Now this next part I only say because you have been so great in sharing.

I have heard you are not doing traction control. If this is true you will be passed up by teams that shoot great like you but can maneuver just that much better. Don't let that concrete floor fool you, it will be tough driving on the real stuff.

You will do well in a regional without traction control because of your great shooting. But going to championships is a different story. I know because in 2007 we made it to Einstein and it is different.

Again you have been an inspiration to push my team harder. You are so ahead of us.

45Auto

04-02-2009, 07:11

Doug Leppard

04-02-2009, 08:25

I guess once the games start we'll find out who was right, the teams who concentrated on shooting or the ones who concentrated on traction control - we're going to be with 1726 on the "shooting" side!

When my son and I went to an MIT recruiting time when he was applying to MIT someone asked this question "Should I get As for grades or should I take AP classes? Do one or the other?" The MIT recruiter said "Do both."

If you want to be at the top of Lunacy do both well. In my estimation the top teams will have excellent driving ability and excellent shooting. Do both well to make it to the top.

We are doing both. We have two software teams, one doing traction and one doing shooting. If I could only do one well, it would be shooting.

MrForbes

04-02-2009, 08:56

We are planning on working on traction control too, we just haven't gotten there...hopefully Doug's words will be taken to heart by our programmers, and they'll get the encoders wired up and get working on implementing traction control by the end of the week. We have done only a little driving on the FRP with the trailer attached...need to find more carpet and buy more FRP as we only have an 8 by 12 area which is just too small to do anything. We also need to build a dummy trailer to tow behind the robot while towing the "real" one behind another robot....

So much to do, so little time, and that's when we're ahead of schedule!