Author Topic: Team Brain Damage  (Read 55025 times)

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #405 on: September 28, 2019, 10:48:33 pm »
I spent the whole day in the shop today. I finished up the flipper assembly this morning and did some testing with the new flipper on the old base to make sure it would self right and to see how the flip power is. The self righting is solid, but the flip power is tough to gauge without cleats.  After I made sure everything was holding up ok and I didn't introduce any new rope snags or the like I started in on one of the new side frame rails.

Pictures:
https://imgur.com/a/SoVCA1z?

Video:
480 FPS flipping a target with no cleats - https://youtu.be/4aY1Bi5ZdrA
480 FPS Full power self right test (still uploading) - https://www.youtube.com/watch?v=bHwjCM8ATbE
real time flip and self right test (still uploading) - https://youtu.be/2DQZ1DQi4SY

Tomorrow will be wrapping up the frame sides and new top plates. I might also replace the rear frame piece since it is the only remaining one, and I realized today it has taken a fair bit of damage so some of the screws no longer work on the corners. Should be no big deal.

-Zac
« Last Edit: September 28, 2019, 10:56:15 pm by zacodonnell »

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #406 on: September 29, 2019, 10:25:55 pm »
I spent the whole day in the shop again today. I didn't *quite* make it to doing another flip with the cleats installed, but I did get the frame replaced and both drive sides set up. I haven't tested them under power yet but they're buttery smooth on the bench. I need to add the corner gussets and cleats to the frame, and install the top plates then it's back to the test box to see what breaks. Gonna be close on weight, but I think it's manageable. I might have an hour or so to work tomorrow and Wednesday, and maybe three hours on Thursday. I'll have to pack too

Pictures:
https://imgur.com/a/SoVCA1z?

-Zac

rcjunky

  • Roboteer
  • ***
  • Posts: 134
    • View Profile
    • Canadian Carnage Robotics
Re: Team Brain Damage
« Reply #407 on: September 30, 2019, 12:27:47 am »
Have you considered doing long wide outriggers like I use on Llama Sheen to make it more stable on flips?
Andrew Burghgraef
Canadian Carnage Robotics

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #408 on: October 02, 2019, 10:50:26 pm »
More progress tonight. I did a little more testing and I'm pretty sure there's a loose wire somewhere. When the flywheel gets going and the vibration is high, I can hear one of the controllers disconnecting and reconnecting. Until tonight I wasn't sure which one, but now I'm sure it is one of the drive controllers, likely the flywheel side which is flaking out on me. When recalibrating the controllers to make sure that wasn't the issue I noticed that the power connector on that one looked like it had been unplugged a bit. I'm not sure if it was already that way or if I did it while messing with it, but the former would explain the symptoms.

I'm going to do some testing with the flywheel spinning the other direction too. I think the robot rearing up is mostly caused by the rotational momentum in the flywheel transferring to the robot after the flipper arm hits the stops. Rotating "downward" toward the front should stop the rearing up, but might cause it to tumble forward more. I'm hoping the cleats will prevent issues there.  I did bend one of them sideways slamming things around in the test box so I might trim them back a bit just to leave a little more support. That would save me some much needed weight too.

I'm leaving Friday after work to make the 3+ hour trip to King of Prussia, so I have tomorrow evening and I may end up taking Friday off of work to make sure I'm ready. Cutting this one close!

Too lazy to get the pictures this late so you'll have to wait. I did take some though

-Zac

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #409 on: October 09, 2019, 10:32:50 pm »
Back from Franklin. I had a lot of fun and enjoyed hanging out with everybody again. I even got to help out a couple of teams and did a hilarious demo against Wumbo.  The robot didn't do very well, drawing two lifter bots that pretty much controlled me the whole time. 2 minutes against a lifter bot is just not very long. I'll need a better solution in the future.  The extra wheels worked well, and the bot remained fairly mobile. The weapon clutch worked ok, though it failed to fire a few times. I think the problem has to do with wear on the ti shaft that the flywheel rides on, but I'm not sure. It is a couple day and $100 project to replace that part so I didn't do it for this build.  I do need to improve my testing procedure though - in real combat with turning and accelerating the systems aren't as reliable as they are in my test box (big surprise). Of course I finished it around 5:00 on Friday so not a lot of testing happened :)



Updated build report with the last wrap up and weight reduction steps:
https://imgur.com/a/SoVCA1z?

Video Jasmine captured from the livestream of my fights:
Heracross, loss by decision: https://drive.google.com/open?id=1aGFZE4xS-e7Rg7GgPZ3hZm7s-NKsMjXO
BEEEES??!, loss by decision: https://drive.google.com/open?id=1fmeg1yaRU-fIDF2EFwDQkF6Iz5iRR7Io
Sportsman rumble, win by enjoyment: https://drive.google.com/open?id=1_wnMBE8IkfoskB6eCNH4avfxzpEqeEV7

If anybody has the video of the Wumbo demo I'd love to see it. It was not planned so I didn't get a recording. I haven't had time to clip it from the live stream yet.

-Zac
« Last Edit: October 09, 2019, 10:36:20 pm by zacodonnell »

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #410 on: October 09, 2019, 10:35:18 pm »
I should also mention that the live stream and fight list seemed to go well this year. What did you guys think about it? Jon worked really hard on it, and I pitched in here and there. I think it makes it way easier to stay connected with friends and family who can't come. I know I had a handful of people from home watching now and then.

There are a lot of folks that work hard behind the scenes to pull these events off. If you get the chance, thank one of them sometime. This event seemed to run very smoothly, with pretty much non-stop action. I am impressed that everybody was able to keep their bots running with no postponements and no test box. Kudos to the builders for making it happen too

-Zac

rcjunky

  • Roboteer
  • ***
  • Posts: 134
    • View Profile
    • Canadian Carnage Robotics
Re: Team Brain Damage
« Reply #411 on: November 28, 2019, 01:09:12 am »
I seen you're not signed up for Moto yet, I assume ya just missed the first window? Next window is this Friday. I kinda want to see how Llama can take the hits from Magneto, especially back driving my weapon with that much force, not exactly something I can test.
Andrew Burghgraef
Canadian Carnage Robotics

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #412 on: November 28, 2019, 03:26:43 pm »
Yep, just missed the first window. Iím hoping to get in on the second. Not sure if magneato is gonna make this one. I donít want to take somebodyís spot unless I can make a real try at it.

I have a list of small upgrades planned for hyperpolarized. Mostly they center around a better idler and improving maintenance on the bot. That should help it hold a heading much better and reduce wear on the roller as well as help with the between fight scramble. It was the scariest field of feathers Iíve ever seen together last year, this year is unlikely to disappoint

Zac

rcjunky

  • Roboteer
  • ***
  • Posts: 134
    • View Profile
    • Canadian Carnage Robotics
Re: Team Brain Damage
« Reply #413 on: November 29, 2019, 07:26:12 pm »
The crazy starts in 1/2 hour, hopefully ya get in, if not gotta wait list your way in
Andrew Burghgraef
Canadian Carnage Robotics

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #414 on: November 29, 2019, 07:28:47 pm »
Iím ready. Thanks for looking out for me

Zac

rcjunky

  • Roboteer
  • ***
  • Posts: 134
    • View Profile
    • Canadian Carnage Robotics
Re: Team Brain Damage
« Reply #415 on: November 29, 2019, 08:18:59 pm »
Every year round 1 is nuts and round 2 gets shorter and shorter. This year I think round 2 was harder then round 1 to get in. The site was down for me until there was like 5 spots left between the bigger classes, good thing I was already in. As with every year, its gonna be good times.
Andrew Burghgraef
Canadian Carnage Robotics

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #416 on: November 29, 2019, 08:37:43 pm »
WOOOOOOO I made it in! Thanks to the multiple people who sent me reminders. For the record, I spaced out last year on the first round completely, but this year I was paying attention and still wasn't fast enough on the first round. It got a bit shaky there when the BDB went down but we made it.

I may drop the sportsman later if it looks like I won't get to it, but for now it is registered. If you know somebody who is trying to get into that class, then consider this inside information for all three of the people who watch the sparc forums :)

For Hyperpolarized I'm planning upgrades to these things:
1. Better idler setup. The old one was a nightmare to maintain, it wore down in almost every fight and depended on a complicated disassembly process to repair. I tried to make it as small as I could and expected to never have to mess with it but this shell is just a tiny bit uneven on the top so it gets obliterated at the speeds I'm spinning. Clearances on the top for this bot are so much tighter than the last few that the idler is way more involved. I'm hoping a better idler will allow it to hold a heading better. It's hard for you guys to see, but this robot slews to the side really bad while driving. I made it super flat so it would be super stable, but the goal was to make it way more mobile than the last version. In the end it is much faster and more stable, but with the oversteering problem it wasn't actually any better at going where I want when I want.  That's going to be vital with all of these terrifying bots.
2. Better lexan mounts for the top. The old pie wedge shaped plastic pieces don't have enough overlap with the steel shell. That means the screws are mounted too close to the edge of the plastic, and the slightest impact causes the plastic to crack around the screw holes and break apart. This makes the shrapnel guard into shrapnel, and allows the screw to come loose. That's bad because it then drops between the shell and the mid plate and gets caught on the weight/cooling holes in the mid plate. That binds the shell up. Not a huge thing at full speed other than having metal bits in the innards, but if it is stopped it can jam the shell up and screw up the roller. I'm pretty sure this is what cost me the two fights at the end of moto last year. I actually have a pic of the scrape mark where the screw gouged the mid plate and then hit one of the holes. I don't know when it happened exactly, but I can see that being a problem. I'm going to rework the midplate holes to avoid alignment with the screws and also improve the mounting for the screws. My current plan is to weld little threaded spacers to each of the shell spoke braces and do a single continuous lexan circle above the spokes, with rubber gromments for the mounting holes. That should eliminate the cracking issue other than on a pretty significant hit, and mean the screws can't fall into the bot as easily.
3. A better center post setup. The current post was a last-minute retrofit of the Threecoil flywheel as a bearing surface. It is soft and only mostly round. It worked remarkably well for being so last minute, but I want to redo it out of a real machined part. Basically the bearing inner race I got was a very loose fit on the bearing and it led to way too much wobble. This is probably .002-.005 bigger and it was way better, but I don't trust it. Basically I'm thinking I'll make a single steel bowl shape and get rid of the center aluminum plug. That should save a lot of weight and allow a custom fit surface for the bearing.
4. Better skids around the edge of the shell besides the idler. These are wearing down reasonably fast as expected, but I didn't do a good job planning how much maintenance would be required and how hard it would be. Right now I basically have to pull the electronics box apart to get to a nut on the other side of the screws AND remove one of the standoffs that blocks the wrench. It's a huge hassle, and a threaded mount would have fixed it. Basically I'm thinking I'll rework the corner blocks on the electronics boxes into a combined standoff and frame mount that I can also use to mount a skid. It'll be annoying to make four of them, but compared to the effort last time it shouldn't be too bad.
5. A better power switch / LED setup. The current setup works fine for drive, but requires me to hold the charging switch while I turn on the weapon to prevent the giant arc from the weapon controller caps filling. This spark is what cost Thomas Kenney on Shaka - eventually it burned through the plastic of the whyachi switch and grounded on the mounting screw to the frame (from some other wiring damage that caused the frame to ground). It led to a bunch of bad electrical gremlins and eventually a meltdown. The charging switch is good, but a bit of a hassle, and the voltage is too much for the LEDs and resistors I had on hand, so I'm going to do a more legit job with that this time. At moto I depended on the weapon controller sound to know the weapon was on (plus the drive controller LED) and that is nerve racking at the show. I just ran out of time to finish it.

To make all this happen I still have about a half pound in the budget, and I can save another half with the center post redo I bet (haven't run the numbers yet). Unfortunately a legit idler is going to take up enough space that I don't think my power switch setup will work. I have a new design I want to try that is loosely based on a fingertech switch but includes a charging circuit and some compliance to allow for an extra tight hold even in a high vibration environment.  I've been happy with the whyachi switches, but they are so expensive now and so big that I'm looking elsewhere. The little black four-position style with the copper screw just sounds too unreliable to me so I'm toying with my own solution.

Looking forward to the build and to moto. It's going to be vicious and a lot of fun. I'll try to post here as I have progress but it is likely to be sporadic through the end of the year. We shall see!

-Zac

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #417 on: December 05, 2019, 07:37:00 am »
Well... I decided to drop Magneato. I'm going to focus on Hyperpolarized. Hopefully that works out to be a good choice.

-Zac

rcjunky

  • Roboteer
  • ***
  • Posts: 134
    • View Profile
    • Canadian Carnage Robotics
Re: Team Brain Damage
« Reply #418 on: December 06, 2019, 01:20:14 am »
That let Ravi get a spot, thanks for the heads up. Given I didn't see this for 5 hours because night shift, clearly pays off to monitor this forum :P I should have some content to share soon. I'm looking forward to seeing how your changes work out, I just like seeing shell spinners work well ;)
Andrew Burghgraef
Canadian Carnage Robotics

zacodonnell

  • Roboteer
  • *****
  • Posts: 575
    • View Profile
Re: Team Brain Damage
« Reply #419 on: December 07, 2019, 06:57:27 am »
Excellent, that's just the kind of thing I was hoping would happen.

-Zac