August 16, 2018, 08:16:21 am
Welcome, Guest. Please login or register.

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - zacodonnell

Pages: 1 ... 31 32 [33] 34
481
Progress Reports / Re: Massacre & Friends
« on: February 06, 2015, 05:20:00 pm »
The bearing around the weapon pulley setup is pretty cool. That's a clever way to get more support out of the same height so you can keep the bar low.  Are you planning to change the thickness / shape of the bar to avoid the propeller effect? The tie in with the rest of the frame makes me think that center post warpage is not going to be in the top 10 failure modes for this bot.

-Zac


482
Progress Reports / Re: Alex's bots
« on: February 04, 2015, 07:20:14 pm »
Stainless top and bottom will be a waste of weight. Garolite, UHMW, lexan, or aluminum would all provide a better bang for the buck there I think.  The base could be a frame member like Kyle said so something a little sturdier there like aluminum is probably a good bet.

I second the motion of leaving a little space to run wires and stuff... They always take up more space than you expect and you didn't leave much in that bot.  The 775s are way overkill IMHO, especially at 14v.  A pair of 550s will be more than sufficient (this is what I run on Magneato at 5S A123) and save a lot of weight. You won't have enough bot to get the 775's torque to the ground, or it will be so squirrely to drive that you will gyro over upside down trying to attack.

On devour I put some wings sticking up from the wedge to help deflect horizontal spinner blows. I don't think I ever actually had to fight a bot that would be affected by it but I think the idea would translate well to your bot.  The easiest blow to take is one that deflects off of your steel wedge :)

@Ellis: one reason to avoid lipo is that it would make the bot ineligible for the Franklin Institute event in Philly, the one closest to Alex.

-Zac

483
Progress Reports / Re: Bone Dead Robotics
« on: February 04, 2015, 07:12:12 pm »
The center mount is looking better Brandon, but I still think you could make it stronger without adding much weight. Right now it looks like you screwed blocks of plastic to the base, then screwed down through the top block of plastic into those blocks.  It would be a lot stronger, and not rely on plastic threads holding as much, if you just passed bolts straight up from the bottom, through the blocks on the base, through the top part, and then put nuts on them.  This is basically how I did Tripolar and it held up very well.  Leaving the plastic a little space around where the bolts go through will help prevent pull out too.

A hardened shell would hold up better... but a warped shell might never spin.  Whyachi can do the hardening but it is an expensive process to fixture the part so it won't warp and use a hardening process that reduces warpage.  I've never been brave enough to actually do it and have stuck to an unhardened 4130 shell for all my 30 pound bots.  They do eventually warp, but with enough clearance they can maintain their shape for a while if you design them right.  The warpage over time is the reason I couldn't spin with Tetanus in the Motorama final last year.

-Zac

484
Progress Reports / Re: Novarobots: Iron golem 30lb Sportsman
« on: February 04, 2015, 07:05:26 pm »
"For now I'm going to rely on hot glue and hope"... Heh, made my whole day.

You may have mentioned this before but why use an OSMC and arduino control board over a relay and TDE switch? Are you expecting to use variable speed control frequently on the hammer strikes? Prototyping a fancy-shmancy control system of some kind?  Leaving the extra pin available for a bump sensor or the like is always good.  I'm now using 9 IO pins in Magneato and I have ideas for more. Every time I have to cut the shrink off and redo the board I worry about messing something up...

-Zac

485
Progress Reports / Re: Twisted Sick Robotics
« on: February 04, 2015, 07:00:08 pm »
The UHMW will open up around the shaft pretty bad, even in 3/4" form I'd bet. To combat that you might be able to make some small flanged aluminum blocks that you can use to mount the shaft, then sink them into the aluminum and bolt them in place to spread the load.  If you leave enough meat around that I don't think you'll break the UHMW - it is really tough.

The shock problems are breaking your drive motors right? If you put some foam between the drive motors and the frame rails that keep them from tearing off the gearboxes it could help a lot. You are awfully tight on vertical clearance though so there may not be space.  Definitely use a metal pulley for the motor this time :)

-Zac

486
Progress Reports / Re: Twisted Sick Robotics
« on: February 03, 2015, 05:52:38 pm »
What size / type of belts are those? I didn't think you could do an idler for the back of v belts but that setup looks solid to me. A 2600 watt motor is a lot to put through the belts. I'm interested to see how it holds up.

-Zac

487
When he says "pete sells" he means http://www.kitbots.com.  It is for new builders after all!

-Zac

488
Questions / Re: TinyESC- Possibly Blown?
« on: January 29, 2015, 12:55:27 pm »
The only other thing I can think is a broken trace on the board between the solder joint and the controller chip. Sounds like you were pretty thorough...

-Zac

489
Questions / Re: TinyESC- Possibly Blown?
« on: January 28, 2015, 09:09:33 pm »
If the ESC is showing no signal by blinking the LED then the processor is probably ok (since it is executing the "slow blink" code) and the part of the controller responsible for driving the motor isn't being told to do so so it may be fine too.

Do you have other things (like the other tinyESC) plugged into the RX so you're sure that it is linked to your radio, powered up, and pushing signals through the pins? If so, try swapping the plug positions between the working part and the non-working part in case it is the port on the rx.

Also, you probably already checked but is it possible you flipped the plug around on the RX?  I did this when troubleshooting the other day and it totally messed up my decision tree :)

-Zac

490
Progress Reports / Re: Team Brain Damage
« on: January 28, 2015, 09:03:37 pm »
Initial prototyping suggests that a dual UHMW loop of shame, with one on either side of the blade, should be reasonably effective at helping to bot roll over without making it overweight. I'll have to decide what attachment points I'll use. I'm not sure if I want them to be mounted only to the top plate or only to the bot. I do know that I'd rather not have it be half and half because it makes working on the bot more annoying.

I seem to have run out of my favorite material - ~.1" UHMW sheet.  I have a bunch of the .15" thick black sheet that I made the wings on Magneato out of but I don't have any more of the thinner stuff that should be better for this use case. Nothing Mcmaster can't take care of, and I'll certainly find a way to use it.

-Zac

491
Progress Reports / Re: Team Brain Damage
« on: January 28, 2015, 08:03:16 pm »
Actually last night I was just thinking that. This bot is so front heavy and lopsided that a loop of shame will probably work great. I haven't decided if one across the width will be better than one on either side of the blade front to back yet.  I'm hoping to try something out tonight.  My 2 ounces should give me a couple of 1" wide .1" thick UHMW strips that will be stiff enough to do the job.

-Zac

492
Progress Reports / Re: Novarobots: Iron golem 30lb Sportsman
« on: January 28, 2015, 07:54:25 pm »
I'm enjoying the photo stream. Making packs is a lot of work isn't it?  I switched to an 80w weller soldering iron with 3/8" chisel tip when I started making my batteries and have had no issues since. A 100w is a monster. Every time I make packs I wonder how steve hill could do it for so little on top of the cost of the cells, shrink, bus bars, etc.  I use folded over 14ga wire for my bus bars because I have it and have put 200A through it briefly.  No biggie. I think your descriptions of the cell polarities might be backwards. I can't tell from the pics if you got the wire colors the other way but definitely worth double checking before you plug one of those bad boys in :)

I'm amazed that your little HF lathe can handle cuts like that in a steel gear.  My little tag lathe sure couldn't. I have a 7x10 little machine shop one now with carbide insert tooling and it is like that soldering iron - I don't know how I did without it.  I might have to reevaluate the top armor on my flipper between you and Mike... At least with just one I could hope to avoid him in the brackets for a little while.

-Zac

493
Progress Reports / Re: Team Brain Damage
« on: January 26, 2015, 09:08:05 pm »
Tonight I spent some more time on Chop. I installed the spare weapon controller I had then did some more testing. It spins up fine, and after I remembered to reprogram the controller for 100% reverse it spun down fine too.  Unfortunately when I did some invert testing it promptly smoked so I've got no weapon again.

This robot has now eaten three weapon controllers and two drive controllers.  Sure, the drive controllers were just calibration problems, but I've ignited two weapon controllers. I'm also being mean to the weapon a little by asking it to self right while inverted, but I didn't expect it to just smoke immediately.

I'm going to have to think a while about what controller to use as a replacement. I have a unidirectional one from Scurrie that is monstrous but could fit in the battery bay. The reversible controller was to help the bot self right but it seems the higher speed gearing on the weapon and rear bubble in the top plate are conspiring against me to prevent that.  It won't self right if it only spins one way, but if it won't anyway then a unidirectional controller is fine. The alternative is to reduce the speed of the weapon and hope the additional gearing is enough to get it to self right again... which would mean ordering another one of the controllers I just lost 2 of in as many days.

Good thing I still have 3 weeks to decide!

-Zac

494
Progress Reports / Re: Team Brain Damage
« on: January 25, 2015, 10:30:48 pm »
I double checked the charger and the test was 441mah out of the weapon battery.

I then tried to do some self right testing. It had a bit more tendency to drive the bot forward instead of flipping it over than before. Possibly in part due to the bubble in the top plate changing the impact angle when inverted. I'll have to look into that further.  At the end of my testing the weapon went out for some reason. I did a couple of quick checks on the bench and the weapon controller is showing no signs of life now... It definitely has power (verified by meter on the exposed solder joints) and the "power switch" jumper shows the contacts shorted, so something else is going on.  My spare controller shows a blinking LED when it gets power and nothing else, so something is wrong with the input power. The solder joints look pretty shady on the main leads so I'm going to try reflowing them tomorrow to see if that fixes it. If not I'll swap in the spare. This controller was only ever used on Devour and Weakened Warrior (in two exhibition matches) so I'm not sure how well tested it is.

-Zac

495
Progress Reports / Re: Team Brain Damage
« on: January 25, 2015, 05:04:05 pm »
Today I spent most of the time in the shop working on redoing the electrical system in Chop.  I pulled out the Tiny ESCs so I can send them back to Kurtis. He agreed to send me a replacement upgraded pair for free which is pretty cool.  After thinking about the bot a little bit more though I decided that I wanted some more ground clearance, which meant bigger wheels. Bigger wheels meant more current draw and I was already a little iffy on using the TinyESCs to drive around a 6 pound bot in a 16' box at 15v... So I decided to swap in the Old Reliable BaneBots 12-45 controllers that drove Scurrie in all those fights I mentioned above.  These are way overspec for this bot but I have about 4 ounces available and they only add about 2, and being overkill in the speed controller department has served me well in the past. Also it means I get to play with the bot this weekend instead of waiting for the TinyESCs to come in.  I'll keep them as backups in case these controllers get bashed.

After I got it all set up I took the bot out for a spin and... it drags horribly left.  That makes sense I guess, as the weapon motor and weapon controller are way heavier than the battery and drive controllers.  The difference is so much that the bot visibly leans to that side on the foam wheels.  This makes the back corner drag and so the bot slews to that side when I'm driving forward.  I added a skid in the middle to hopefully make the drag be more centered, and that helped some, but it is still tough to drive forward.  It handles like a dream backwards though, so I may have to do my long distance driving that way once Chris has their weapon stopped :)

I did a 3 minute test with as many spinups as I could get (without hitting a target... I need to put another floor plate down if I'm going to be hitting my test box with the weapon) and I tried to drive continuously the whole time. This bot isn't going to be doing any pushing so that's probably about as heavy as the load is going to get on the drive. I was nervous about the load on the drive motors since the wheels are bigger than they have been in Threecoil and the voltage is a little higher too.  I was also interested to see how much battery capacity I used with the bigger battery and faster weapon.

In the end the hottest component was the weapon motor at 104F and the drive motors and battery were fine.  I used about 350 mah during the test too, though it will be a lot higher when I'm actually hitting other bots and self righting.  I still need to verify the self righting works but I think it will.

Video:
http://www.youtube.com/watch?v=HL8mcuoPZGI&feature=youtu.be

Pics:
https://picasaweb.google.com/107279152885550274671/ChopBlockMoto15#6108403026580036578

Now I get to decide what to do with the other 2 ounces and maybe run some more tests.

-Zac

Pages: 1 ... 31 32 [33] 34