Announcement

Collapse
No announcement yet.

Team Nuts' Build Diary

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Team Nuts' Build Diary

    I've got too many robots to warrant separate build diaries, so I've decided to compile them all into one..
    ..And by too many robots, I mean I've built around 35 across every weight class since starting in January last year (average of about one every 2 weeks!), 17 of which are currently still active.. Below are all the working ones plus a few old chassis

    WP_20150615_002.jpg
    My general philosophy is to build robots as cheap as possible from anything I can find... as such most of my machines take on a fairly rustic look, and building 35 in a year hasn't bankrupted me :L

    Heavyweight - (Big) Nuts

    Original build diary: "£50" Heavyweight

    WP_20150615_009.jpg

    Nuts is a rather odd creation that cost me about £30 to build in the end (largely due to a lot of very generous parts donations). It's a flail 'spinner' based off my fleaweight (on top in the photo) of the same name (It was going to be called Big Nuts, but that starts to sound a little too crude ). It is not meant to be competitive, but is seriously fun to drive. It had its first fight with the flails fitted yesterday:


    (Video kindly supplied by Harry DT)

    Featherweight - Richie

    Original build diary: Richie: The £50, 3 week Featherweight

    WP_20150408_003.jpg

    This was built in early April last year ahead of the RoboChallenge event in Yeovil (a few miles from where I live). It has since had many tweaks and rebuilds, but still only ever has about £50 worth of components in it at any one time. I entered it in the champs this year, coming roughly 19th (I missed qualifying for top 18 by two seconds). I'm currently considering going over the £50 mark and fitting the lifting mechanism with a 2kw brushless (currently runs off a 12v 550 motor).
    Also started work on a £30 axe robot called Johnny, more on that in the future hopefully

    Beetles - Bork and Naff

    WP_20150306_003.jpgWP_20150418_004.jpg












    Bork was an assbot thrown together for Whitwick out of a Cybots toy and bits of an old desk... Still did remarkably well and actually pushed Soundwave into the pit in the competition..
    Naff was another beetle thrown together ahead of an event, and despite looking kinda sleek, it was even less competitive than Bork. I'm hoping to build a proper beetle at some point; I have all the parts for Big Bully, which is planned to be 4WD with a lifter.

    Antweights and smaller

    Original build thread (RW101): Another New Antweight..

    I have too many of these to talk about them in detail, so I'll just list them:

    Ants:
    Active
    BullyDozer
    Why Wait?
    Skutter
    Bad Idea
    Tinny
    Lionel
    Retired
    Flail Whale
    Chocolate Bot
    Click Clack
    Fleas:
    Haggis
    Nuts
    Chug
    Smash and Grab
    Nanos:
    Minidozer
    Sir Lift-a-lot
    Nibble
    Spinni
    Nullity
    Scrat
    Scampi

    I've built various other robots, either as just test rigs or non competition concept machines, which bring it up to about 35.

    Electronics

    The electronics for every one of my robots has been homemade, and I am planning on selling various bits of electronics now in development.. More on that as it happens..

    Thats it for now, I'll probably have more updates fairly soon as I'm on holiday after exams, so I've got a bit of catching up to do!
    Last edited by Rapidrory; 15 June 2015, 22:21.

  • #2
    Rory

    I can let you have an unsensored sensored 42mm alien power motor to play with

    Chris

    Comment


    • #3
      Hi Rory,

      On the Johnson 600 motors you used in Richie, how much extra length did they add to the overall argos gearmotor unit? I'm thinking of upgrading but I might not have the width in my chassis.

      Cheers

      Dan

      Comment


      • #4
        Ok, so this update isn't directly about robots, but it is related.

        Recently I've been interested in using these 800W scooter motors for a Heavyweight. To test them out, I bought one to fit in a miniature railway locomotive that I've been restoring for a friend. After buying a few other bits and spending a few evenings bolting it all together, I ended up with the following setup:



        It has a 10:1 gear ratio to the wheels which gives it a top speed of 6 mph and a load of low speed grunt. Alex and I spent all of yesterday throwing it round the track trying to stress test it, but even loading it with about 400kg and driving at full whack for a few minutes it barely even got warm. The first half of this video shows it running around the track quite nicely:



        I know that running a train and a combat robot are hardly equivalent, but I'm none the less very impressed with it's performance and am confident that a pair of them will throw 100kgs around quite happily after perhaps a bit of battle hardening (although they don't come with a fan and the magnets appear to come glued in place so not sure what else would need doing..)

        Another reason for doing this restoration was so I could trade it in return for their milling machine which they have never uses since they got it 20 years ago... Hoping to retro fit it with some CNC gear so I can do fancy spinning disks

        Attached Files
        Last edited by Rapidrory; 2 August 2015, 22:51.

        Comment


        • #5
          That could be an insanely nice machine with some heavy restoration work! Very jealous.

          Comment


          • #6
            Agreed. The jelly is real.

            Comment


            • #7
              Yeah it all seems to work fine, just needs a good clean up and a few more milling bits (it came with 4).. weighs about 1/3 of a ton though so getting it out of the shed and back home was interesting; currently having to build a new work bench that's capable of supporting it's weight :P

              Comment


              • #8
                It seems that the 800W motors are currently out of stock everywhere. This appears to be because the miniature railways community, which also uses a lot of Bosch 750s, have just cottoned on to them as a Bosch 750 replacement and have started buying them in large numbers (by large I mean the shop I bought mine from had a stock of just 10). I spoke to one of the train designers who uses them and his advice was to ring Red Circles motorcycles and pre-order them. They're currently restocking from china so hopefully will get them back in UK stock soon.

                Comment


                • #9
                  Been a bit of a while since I've posted anything, but since the season's coming to a close I thought I'd bring this up to date.

                  The last few months have been very busy with live shows, but Nuts is still alive albeit with a few Thor induced bruises. Nuts somehow came 6th at the world championship, though I put that almost entirely down to luck and the heavily melee based format of the event. But still, not complaining; always amusing to be able to say that your robot is sixth in the "world" :L

                  The flails have been working well; although not hugely effective, they look and sound great when hitting things at full pelt, and occasionally manage to to put a dent in things (even managed to knock out Gravity's link in one round). Did have an unfortunate habit of falling off, but that was due to a nut and bolt coming undone which has now been replaced with a solid bar, so that shouldn't be a problem any more. Here's a video of it attacking a pallet ahead of the world champs:



                  The addition of a car horn has been one of my better decisions on that robot; that thing's hilarious :L

                  I do want to build a more serious heavyweight at some point, as fun though Nuts is it rarely actually influences a fight, more just runs away. My plan was to use the scooter motors as listed above, but unfortunately they seem to have been stopped being manufactured in July, and absolutely nowhere seems to stock them or the 1000W variants any more.. My plan would be to build something like shock-wave, ie 4wd with a centre mounted lifter (I'm annoyed I didn't spot it when it came up else it'd saved me a job). Probably won't start that project till the summer though as it's my final year of uni, and things are getting quite busy..

                  Richie is in a very sorry state ATM; ever since the FW champs I've more or less been running it without maintenance, and at Stevenage one of the drill gearboxes finally gave up the ghost after over a year of heavy service. The lifter had stopped working a long while before that, so there's very little that does still work on that bot. I've patched the drive up again now, but it really needs a full rebuild.

                  I'm planning a new feather, Richie 2. The plan for this would 4wd running off Dewalts, lifter running off another dewalt, and a Raex skirt to try and keep the spinners out (would be similar to the american bot uberclocker). Unfortunately, I got 18v XRP Dewalts, which while very powerful have the problem of having a separate gearbox that's a pain to mount. I've got the mill up and running so could make something up, but it's not an easy task.. My other alternative is to run a pair of NTM5060s for drive with a single stage spur gear setup... much lighter, more compact, and more crazy :L

                  Another feather project I have was instigated after a few beers, where I was persuaded to buy a featherweight spinner setup... As such I'm now in possession of almost all the parts to build a feather spinner. This may well be my champs bot next year, as I don't really have time to build Richie 2 due to uni work, and the spinner is a much simpler build. It'd be a vertical running off an NTM5050, and probably have the drill based drive units borrowed from Richie. Bit of Raex on there to keep the worst of the hits out and it should be good.

                  On to beetles.. Things have just got silly here; in another perhaps rash decision, I bought Lynx off Sam.. this was already the most I'd ever spent on a robot, and it wasn't even a working one. Since then I've put the new mill to good use and have turned Lynx into 'First Lefts' (bad pun on Last Rites) as seen below in this rather poor photo :



                  Lynx had a ~1Kw motor driving the spinner, which is crazy levels of overkill, but not wanting to spend any more on this robot (something I'm failing at as it's already up to about £150!) I had to just bolt it to the top as it's the only place it fits. I've connected it to the blade by a very heavy duty timing belt, so it should at least be able to make use of some of that power; It will spin up about as fast as I can move the throttle stick, and is unlikely to stop unless I tell it to or something breaks :L Below is a gentle test demonstrating it's scariness:



                  The drive runs a pair of brushless 40W Sk3s driven by Afro 12A... Afros are so much better than the 25A trackstars that were in there; it handles almost as if it were brushed. Big fan of the Afros and the SimonK firmware they run; makes brushless drive a whole lot easier
                  Attached Files
                  Last edited by Rapidrory; 13 November 2015, 09:02.

                  Comment


                  • #10
                    On the subject of cheap HW motors yet again (seemingly a favourite topic of mine ), I was having a look at the economy Ampflows today and comparing their performance charts to a Bosch 750Ws; the E30-150 seemed a little weak with a 750W peak, however the E30-400 peaks at 1.5KW (compared to a Bosch's 1.2KW peak), and is only about £70. However reading around I found this quote from Mario about the E30-400 here:

                    Originally posted by maddox10 View Post
                    I would compare it to the Bosch 400, but with the 400 @18V

                    It can do a heavy, no doubt. Price and weight are favorable.
                    But gearratio should be "high". Like 15/1 for 200mm wheels.

                    And no, it isn't too good to be true. It's about normal.
                    This doesn't seem to match up with the performance charts, which imply that it should be slightly more powerful than a Bosch 750 on 24V. Are the E30-400 charts wrong? Is Mario underestimating them? Am I just missing something? It just doesn't seem to add up as it is..

                    Posted the charts for the E30-400 and the Bosch below so you can see what I mean...

                    Attached Files

                    Comment


                    • #11
                      So, as some of you may have seen from the team Nuts facebook group (https://www.facebook.com/nutsrobotics/), I've been working on creating a suitable sensorless brushless drive system for the heavier weight classes (researching motor tech and control seems to be the main aim of Team Nuts these days :L). The advantages of brushless are obvious: Lighter, smaller, and cheaper than equivalent brushed motors, as well as there being a wider range available. The issue has been that brushless ESCs are unable to run these motors at low speed, and struggle to start them turning whilst under load. This means that the motor has to be much more powerful than needed, and that low speed control is often lacking.

                      Alex and I had been experimenting at flashing a TZ85A with SimonK firmware. This had better low speed control than the Trackstars that had previously been used, but still lacked start up torque. It's viable as a cheap option for featherweight drive, and there's still some tweaking to be done with the settings, but probably not going to cut it for heavyweights.

                      We were then pointed to VESCs, an open source ESC developed in Sweden by Benjamin Vedder ( http://vedder.se/2015/01/vesc-open-source-esc/). His ESC makes use of current sensing and a much more powerful processor to make a sensorless brushless motor act almost exactly like a brushed motor; low speed control and plenty of startup torque, as well as many many other cool features he has added to these controllers.

                      The standard VESCs are spec'd for 50A @ 50V, fine for a feather but slightly low for a heavyweight. As it is an open source design, I tweaked the PCB layout slightly to increase that current rating up to around 100A, allowing motors like NTM5060s to be used. I got 30 of these PCBs made, and put two of them together over Christmas.

                      During all this, I had a very generous donation of Foxic 1's chassis from Craig, which I plan to use as a test bed for brushless drive in heavyweights. I also bought a pair of ~2.5KW NTM5060s to use as the drive motors.

                      I created some test videos of the VESCs, as can be seen below.



                      I also borrowed Olaf off of Chris (Shakesc) to test them in. Unfortunately i managed to kill one of the chips on the VESC the day before, and didn't have any spare so had to run one motor off the VESC and one off the TZ85 with SimonK.



                      Once the TZ85 with simonK gets moving, it can be seen to handle similarly to the VESC, however the first and last clip shows that it struggles to start the wheel moving under load, whereas the VESC powers through, even with ~60kg piled on top.

                      My final uni term starts in a week, so I'm not sure how much time I'll get to work on these. My plan is to get a few made and out for testing (I have a few beta testers lined up), so hopefully they'll be able to find any bugs or work out the best settings without me having to put too much time into it.

                      It's all looking very promising so far though!
                      Last edited by Rapidrory; 3 January 2016, 16:55.

                      Comment


                      • #12
                        Great work brushless heavyweight would be very interesting to see in action

                        Comment

                        Working...
                        X