Klaus Is Not Your Friend

In an effort to update my online presence more often I am going to be compiling my forum posts from the wonderful BBB forum which is my new home back here. Just so there is some constant log of what I'm doing that isn't Facebook and I don't have to write everything twice.

If forums are your jam, you can attain the same information and pictures there but if the single blog post is what you desire then have at it. They won't be identical but the broad strokes of my ballistic assault on the English language will shine through. Crimes against punctuation notwithstanding.

 

We start with the result of a rainy afternoon spent inflicting Battlebots on my darling wife and pricking about in Solidworks. I came up with something really simple but pretty funky. I came up with Klaus and this is his tale.

 



A bit of a parts bin special and a low hanging fruit extravaganza on show but I am pretty pleased with how this little cutie is looking. Complete departure from the MotherLoader line but this may actually function as a result. Making a little use of the new dual brushed ESC I just snagged a beta version of from BBB it is fully brushed bar the token tickle saw which is run off one of my many 1806’s.





Featuring a lot of plastic both printed and sheet. The plan was to get away with only needing the arms getting lasercut and the rest done in house. It would be nice to have something fairly simple that is just reliable and ready to jump into all threats.





Bit the bullet and got a little bit of test printing done on the new lad with a bunch of ebay parts rocking up all at once. I got some £1 saws, some funktastic polyurethane belting and bearings waiting in the wings. I was pleased to find a BBB package in there too which featured the super cool fresh and new dual ESC.

I was determined to run brushed motors on this project to shift some of my Empire Of Dirt into service. I jumped at the chance to grab one of the beta run ESC units and pop it through its paces. The plan is to hardwire a 4ch RX straight to it just for minimum serviceability and then hard mount the brushless ESCs in a little nook near their corresponding motors and just run power lines and a signal back to base. I really wanted to tag in better discipline with wiring this as it could prove tricky due to size and the realist in me knows I will have to hot swap weapon motors as they are basically a consumable.





Around this time I just went full send on the model and just chucked out some quick n’ dirty test prints to get a minimum proof of concept. Pretty pleased with some of the guesswork clearances and such. Overall vibes are cute and angry so I’m well on track. Made myself a quick test belt and was pretty happy with how it turned out. It wasn't too bad at all but the thought of making seven more was not the most thrilling of proposals.

Slowly adding more detail and mounting points. Using all the favourite tricks of fasteners and hardware (although no shoulder bolts are present yet in a shocking turn)






This is a silicone mould filled to the brim with 30a Sunny Delight. The hub is just plain Jane ABS for now but I had the option to branch out to TPU or nylon when I get into the swing of printing all the chassis parts.. They just take a 12mm hex bore to be super standard and easy to swap out. 41mm OD and 18mm wide. The tread profile is base purely on aesthetics but I feel it will work pretty well and be quite resilient. Please ignore the pound of pork sausages masquerading as a hand...





I had revised some of the prototype print geometry and rammed them through pretty roughly to start working through the finer details. My goal was to have a wiring loom mostly completed with this as a driving test bed that I can then transplant into a proper chassis once parts were finally printed up.





I set about making a baseplate and some armour which involved trying out my latest deathtrap, a hot wire bender to fold up the rear making it all one piece of 3mm HDPE. Whilst it worked on 12v it lacked a bit of pep so I jammed one of my ole faithful Lidl Parkside 20v lithium packs and sat back to watch the fireworks. Big thanks to late, great Dale Heatherington for the idea




With that in hand I went about cutting and getting a rough idea of how I wanted the front panel to look. The wedge will mount loosely on two mounts between these printed pivot points to give me some chance at riding the floor. Forks aren’t off the table but they are a lot of work to make them work well.





Cut needs clean-up but that will come with top armour fitment I think. As it stands here it is weighing 811g on the incredibly well calibrated kitchen scales so there should be enough room to play around with setups should I wish to.




Now I'd come far enough down the road and I wasn't entirely feeling the saws so it didn’t take me long to flip up into some interchangeability! I was looking at this robot and trying to play out how it will fight in my mind alongside rumbling around MotherLoader ideas. Without going too far to the damaging darkside I do see hammersaws featuring in ML in the future. Being able to flick people up with the lifter then give them a gentle knock with a spinning lump would be a neat tactic. With this in mind I put together a minimum effort (i.e. more use of spares, spit and slight modification of the existing hammer arm) module to try out on this robot






Featuring a single tooth hardox disc I had for my antweights mounted on an Emax 2205. Much like the dual saws off a 1806 I am not expecting miracles it is just the minimum viable product for my nonsense. R&D on the cheap.

 



The orange hub houses a suitably phat ball bearing so the motor might last out the first impact. I realised that the disc is the ‘wrong’ way round for a hammer saw - part accident but I may test it this way round as it potentially could act as a terrible pokey vert stick lance if left in the fired position.





I love living in the future. That is the dual brushed BBB ESC, one 33a brushless controller for running the weapon, a bidirectional 20a esc to pop the arm up and down and the receiver all bound up into one 20.8 gram tumour. This gets wedged down the centre portion of the robot, buried deep under the axe mechanism. A neat, nifty not-yet-designed belt guard stops the pulley from sucking the wires.

 


It was at this point more AliExpress goodies were dropping onto my doormat. Such a shame to see but China wins out over the locals once again. I could have bought these hubs locally but for less money and only about a 10 day wait I got 4mm bore 12mm hex hubs, some pinion gears for MotherLoader and a bunch of xt30’s shipped 9700km.




Also worth noting I had been working under the running title of KLAUS since inception and as the time has come to chuck it at a competition and inspiration has completely failed to strike I set it in stone. Makes a welcome departure from trite _Loader traditions for now.




I squeaked out a sliver of steel from work and wrote a super quick and dirty program on the press brake so I can make a few hundred of these easily should I want to. Spares game on point. Material wise it isn't anything to write home about, it's S450 base steel and its galv coated as most of what I made at the time was. I printed some nylon mounts that bolt onto the front with captive nuts slipped into the bulkheads. I’m hoping they’ll be strong enough to form a decent method of attachment. I don’t intend this to be a vert facing wedge - horizontals and non spinners for sure. Given the way it attaches it has a decent amount of movement without being able to fold under the robot or form a point of extreme stability - I.E. the dreaded ‘thing’



Looking pretty swish if I do say so myself, serving up 2008 K2 realness. The whole mounting system was test printed in PLA-ST like everything else but will swapped over to nylon. I am hoping there is enough toughness but if it is going to feel shady I can swap them out for TPU. I thought about what sort of attachments I can have also - forks feel like a bit of a cop out but I might have to bite the bullet and try at least a little to have a chance of winning a fight.





Got it all swapped over to the cheap and cheerful esun nylon. Quite like now it looks with the white as it’s a little different for me and it helps that orange pop just a tiny bit more. Managed to sort out my drying setup too and after a few false starts I managed to get some pretty respectable parts. I also bumped the speed right down and I am loving the results. Parts feel great after a few days sucking up moisture in the oppressive humid jungle of Yeovil.




Note the difference in the bulkheads. The far one was sopping wet by comparison. It will be relegated to spares duty. I popped a few more plugs in to the wiring so swapping out panels and parts is a lot easier than the MotherLoader series. Planning on driving it hard into dangerous places so I expect to have to repair a lot more!




Despite feeling pretty confident with printing nylon after a few false starts, making the chassis of Klaus did teach me a lot of little tricks and just hammer home the fact you just can't cut corners with this and get a good part. Low, slow and dry. There is no substitute. For my setup I need to start prepping about 8-12 hours beforehand if it's been a while and a good 2-3 hours if its been over a day since its been dried out. Also tree supports continue to be a godsend.



At this point I thought it was appropriate to pop it on the scales. Bit of a pleasant surprise, weight is looking pretty good! I was aiming to have a substantial lid in 4-5mm polycarb with an alu backup for fighting hammersaws and the like.

I couldn't hold off any longer as the robot was getting pretty close to being done I gave it a little buzz around as a twitch test and it was really pretty impressed with just how spicy young Mr Klaus is. He is angry and is most definitely Not Your Friend.

 



I was planning to have a simple little life just laser cutting out the arm parts and replacing the prototypes in a relaxed afternoon but due to unintended financial penalties (i.e. being temporarily embarrassed as it was near the end of the month) the quotes I had returned were reasonable but untenable. Go Go Gadget backup plan!

 


It wasn’t too serious in the end as I managed to cobble something together for some pocket change using some aluminium flat bar and gratuitous 3d printing. It still has the 11x5mm bearing support pushed into a printed axe head. This bolts to the aluminium with four m3's which mimic the motor mount pattern so it's a universal part.




The whole assembly doesn’t look too dissimilar to the ideal and who knows, it may work a little better (and be easier to fix due to alu’s easy bending nature) It’s a tad longer than the originals which was a tweak I think will help it with the reach. The semi structural smile is still in place. After all, t’ain’t nuthin can dull Klaus’s sparkle.





I plucked out the last of the wet printed bulkheads and installed the final ones. The whole robot comes apart pretty easy so swapping out parts won’t be too painful. I am pretty proud of the weird captive nuts I have jammed in the back where space was at its tightest. The front has a little more meat so it has screw in inserts.




The lid has finally been made - I went for a slight risk with polycarbonate. Even at 4mm it probably won’t be the most robust or impact resistant but everyone loves a clear lid!




Also in my poverty inspired laser cutting drought I did away with metal forks and have elected to run with plastic ones. While they won’t be as pointy , spikey or resilient as their steel counterparts these should have a bit more give while still letting me cheaply poke around under people (and their robots har har)

I made a distinct push this time to go in and polish up all the little bits I was unsure about or had a little niggling feeling they’d fail. I have an awful tendency to get complacent and gloss over fairly glaring flaws in my robot, even with available time to address them just out of pride (admitting the infallible narcissist made An Mistake) or fear that I’d make something worse so I leave it. Well this time, motivated by a Schwarzenegger documentary I knuckled down and started fixing all the little horrors and quirks I could in the days remaining.






The arm was the biggest point of contention. Firstly I was worried about the motor for a couple reasons. Anyone who has dabbled in the fickle business of DIY brushless + gearbox knows that pinion slip can be an issue (as well as an embarrassing medical complaint) As this motor was going to have an even harder and more brutal life than normal with the blunt stops and direction changes I elected to go belt and braces and have a go at silver soldering the pinion on. It worked really pretty well and I got a nice clean joint - only inhaling a small cloud of vaporised acid as a by-product.





I was also a little worried about wire eating from the rotating back end of the 1806 and wishing to brace it back against the frame to keep it in place. I drew up a quick motor guard and slotted it over the motor while I had it apart to secure the pinion. It was really nice to have this time to go over and pick up the niggles - something I will strive to replicate. I mean, the other option is doing it properly from the ground up and I'm not about to lie to your faces and say that's on the cards.





While the O rings were great for style and acting like a clutch they slipped way too much to be consistent and reliable at self righting. After a brief consultation with the void at large I found I could scab in some GT2 timing pulleys in without modifying a thing. Technobots has these phenomenally cheaply so I just hit go and ran with it





They needed a back breaker tensioner to get enough wrap and synch up the centre distance but other than that it was pretty much a drop in solution. I drilled and tapped the aluminium pulley for the aluminium arms I made previously and had a quick play. Very responsive and much more grunt.




While the stock pulley was great, my lathe-less existence was hurting me that I couldn’t knock a bit off the boss then make up a spacer and have it central.

Luckily I live in the future where you can just hit send on a part and get exactly what’s on your screen in mere hours. I drew a 59 tooth GT2 pulley (just to ease some of the ANGRY tension I had foolishly added) to replace the 60t aluminium one and printed it in ABS. It has the interlocking geometry for the arms. I just threaded direct to the ABS which works really pretty well (it shouldn’t, though. I toyed with the idea of heat press inserts but I only had them in 6mm length and I figured a physically longer bolt was just a better idea.





I also knuckled down and printed something to give an extra little bit of security in the form of a link flap. Again just ABS it pivots on one of the mounting bolts for the lid and latches on an M3. With a little bit of Klaus branding to, after all the link can’t fall out if its stuck in a K-hole




Getting to a finished state here and I was pretty happy with how it was setup. I had a good deal of options, wedge could be mounted front or back and I could have up to 4 forks on the front. I fitted two with magnets 3 and the rest were just free floating. Weight was pretty tidy as I flew well under even fully laden with the kitchen sink setup








Can’t beat a bit of branding either! I actually found I quite liked the wedge on the back. Might be something to design in to any later versions. The idea of being able to fend off attacks from the back while I poke around with my forks at the front is pretty appealing with the culture of 3 way melees.



So ends the build of Klaus which was a pretty laid back affair - even with the mad rejigging of the weapon arm at the 11th hour. I had some goals in mind for the bot, principal being “Don’t be MotherLoader” which it isn’t and therefore is infinitely successful. I wanted something that could drive hard and BOY did that succeed. The robot drove beautifully, was blisteringly quick and respectable in the power department. This was dulled slightly by the fact I drove it horribly. Just normal fat fingers and poor responses. Slow and steady would have been the best approach but that just was not on the cards. I had it doing figure 8’s and such on the kitchen floor quite handily but obviously this doesn’t compare to actual combat or proper target seeking. I took the dual rates down progressively in each of my 3 fights and by the last one my throttle was at 90% forward, 75% reverse and my steering was down from 80 to 60%. Must have worked as I went the distance - even if I lost.

The self righting was a let down in the end BUT I will take a slight victory here as even working the handful of times per fight before the grubscrews slipped and galled the shaft is a couple times more than it would have done without the extra effort which probably bought me a collective extra 20 seconds of invaluable fight time.

The direct drive motor was never going to make it past the event but I gave up after the second fight expecting it to do anything. It is not a write off as it stands but it is very, very unhappy. I will not be doing this again. I think it counts as a hate crime at this weight class.

Overall a messy performance but not without valuable data. I liked working on Klaus and thinking about how to make it better. Simplifying will help (though it is already fairly brutalist in its approach)

I hope 2024 will bring another version of Klaus in some form. It is absolutely on my list but that is a long, boring and expensive list. I also hope this forum post plagiarism copypaste gangbang worked okay. I did jiggle around the wording a bit here and there but tried to keep the tone pretty similar.

Thank you for your time as ever. You may find more of me and my work:
- On Facebook
- On Youtube
- On the BBB Forum

MotherLoader V2

 Following on from my tepid return to the beetle scene I made another robot in the same vein as MotherLoader. I set out with the best intentions. Take what I learned from the last outing and apply it to a new build which would yield a better performance and justify my nasty habit of building a whole robot and throwing it to the wolves under the guise of R&D.

Now, I tend not to measure my success by the standard Robo-yardstick (y’know, winning fights) because that would just mean I spend valuable build time crying in the foetal position. Instead I go by how far I can push a silly idea uphill and just physically get the shape in my mind into the arena. It’s easier on the soul this way.

Bearing this in mind, would I deem MotherLoader MK2 a success by my low, low bar?

No, absolutely not.

I did not come away with a warm fuzzy feeling as I did with the first ML. There isn’t a brilliant fizz of ideas and excitement either. There is instead a grim determination to Do Better and a long list of foundations to shore up. 

Coming back to the beginning I set out to make something grand and interesting. I wanted to make a really neat light central chassis where I could bolt in weapon modules. Cool eh? I thought so and started with the drive. I needed something small, light and powerful. All things synonymous with brushless motors. I had a little experiment on the last one but nixxed it in favour of ‘ole faithful stock  brushed. Way back in 2014/15 I had a crack at brushless drive before it was cool. I actually used some old motors that have been rattling around in my parts bin since then. Brave perhaps but waste not want not. I have fleeting green moments where I’ll attempt to recycle everything, even at detriment. A small slice of 90’s Robot Wars mindset I suppose, use whatever you have lying around!

I picked up some of these dual gear 25mm gearmotors that seem popular with those not in the planetary club. These were insanely cheap and built like a fucking sewing machine inside - I don’t get why some people turn their noses up at them. 

A small, nondescript amount of fudging had the old SK3’s stripped and a new shaft cut and pressed in. I just locited the pinion on and peened the end of the shaft over. A little bit shady considering some people have to resort to acid flux and silver solder but I actually didn’t have any problems with slippage.

The weight difference between a brushless unit and just the bare motor is incredible. I scoured aliexpress for the cheapest possible BLheli32 ESC’s and these “readytosky” 35a units came up trumps. They ran about a tenner each with tax and shipping.

Giving a nod to its predecessor I elected to house the ESC’s in pockets where they sat snuggly in the printed bulkhead. Space was going to be tight as I wanted to fit a whole robot drive system into a slim horseshoe shape to leave the middle empty for modules.

I really liked how effective (and cheap) captive nuts were in the previous version so I carried this forward to this version. I just made sure they were a relatively tight fit in the plastic and they worked great! These ones are M6 for the 8mm shoulder bolts I was using as driveshafts.

Here was an oddball experiment that made sense at the time but did not get carried past a couple early versions. I was a little wary of having only one trackwheel driven, as MK1 had two motors per side driving the track. My worry stemmed from the track having greater inclination to bunch up, twist or warp when only pushed along from the back instead of pushed and pulled by two driving pulleys.  These spur gears gave an artificial drive (no added colours or flavourings though) to the front pulley so it would behave similarly to a known good system. That was the theory. The practical reality meant it just added a lot of complexity to setup that really didn’t require it and the added friction of PRINTED PARTS ON PRINTED PARTS ON PRINTED PARTS (on shitty bearings on bolt shafts)  just made it an amp suckin’ mess. I don’t feel it's an inherently bad idea just poorly executed working under my constraints. 

I was making great use of my new Ender 3 v2 printer for quick and easy prototypes in PLA by this point. I'm sure you can tell by the vibrant filament choices. Below you can see the neat little retaining clip that holds the ESC down and guides the wires too. Functionally identical to a ziptie but a little fancier. Barrel nuts are also starting to come into play now! You can see a test clip here

My first set of custom tracks makes an appearance now. These were an experiment in material reaction and technique and while it really failed to make anything other than a floppy desk novelty I class it as a relative success.

This is pure silicone rubber cast into a 3d printed mould. The profile itself is really simple. It’s just a HTD5 belt with some knobbly bits on. I made my drive pulleys a sensible size, found a centre distance that worked and made myself a belt to fit.

The mould is a pure negative of what I wanted to make so it was just a case of sealing it up and pouring it in. It’s not a great design but it was fantastically easy to demould.

On the plus side it was incredibly grippy, but being so pliant was a downside that wasn’t entirely unforeseen. It would flex and flap a lot, wanting to ride off the pulleys almost immediately.  Something better would have to be done, and it was. I shall go into more detail shortly.

The frame for the modular weapons was trailed out in some funktastic rainbow PLA (the absolute cheapest) It is sporting two 800mah 2s packs shoehorned in there to give it some guts. It was also pleasingly symmetrical at this point.

The first module I experimented with was a quick 'n dirty axe. Now there is absolutely no point building an axe in the beetle class because I do not believe they can ever cause meaningful damage. I have intentionally worded that rather prickishly to try and provoke videos of nasty axes being thrown my way and more builds! Better living through incendiary comments.

I dug out one of my old Rotolink nightmares from the V1's drive and paired it with an shorter can SK3 of the same vintage that I am using for drive.

Slapping on a lot of extra reduction seemed to be the way to go in order to keep it controllable and have enough torque to reliably self right. This are custom gears printed in ABS. The first stage is Mod1 which steps up to Mod 1.25 for the second. The axe mount is also built into the final gear.

Before I move on too much more I will just touch on the drive pulleys. They are a standard printed affair with a recess for the spur gear profile and they have the same thin profile 8mm ball bearings as the last version. These in turn run on shoulder bolt shafts which double up as a sort of pin and mount for the outer armour.

I tried to keep the classic 'Loader' look going as I feel that profile works really well to buffer damage and contain it to the front of the robot. It was at this point (with only one module built I add!) I gave up on the multiple weapon idea. While it is something I would absolutely adore to revisit at some point it wasn't exactly gelling right and I was struggling. In the face of ripping it all up and starting from scratch I made the decision to modify what I had and build in the axe setup. I just pulled the geometry I needed off the module and stuck it on to the overall shape I was running with. It was surprisingly effective.

I ran out some test frame parts in PLA (incidentally the right colour now). Now that I didn't have to accommodate the module interlocking geometry I could cut a good chunk of width out of the robot as a whole and it only ended up a few millimetres wider than the V1.

Now that I had my shape, I discovered I could slip one of the 450mah 4s packs I had left over down one side, leaving the other wise free to have the gear mechanism for the axe and the RX/BEC rats nest.

I printed up the axe gears in some of my favoured ABS blend in a more fitting colour. With the benefit of hindsight I might call ABS too soft for gearsets of this nature in future and might tend towards PLA plus or ST, perhaps even a nylon of some description would be more suitable. Makers Muse did an excellent video showing off the best materials for printing gears. I may well take heed next time. They didn't not work however, though the thinner Mod 1 running on the steel pinion was showing some deformation.

After this I cracked out the Ender and a spool of incredibly cheap TPU filament. TPU is the hot new wonder material for anyone who likes a bit of wobble in their life. It has absolutely remarkable elastic properties and will bend and deform am incredible amount before returning back to shape. People have been using it for a while now and I finally got it together enough to give it a whirl and I am really impressed.

The frame is stiff enough to actually function as a frame but still really flexible. I loved that I could put as much pressure on parts as I could with my hands and have them spring back. It was easily twice, perhaps 3 times more flexible than HDPE which is already pretty ductile. It leant itself really well to my practice of barrel nutting and using through holes and captive nuts to attach panels. I printed them with a 4mm wall and a 25% infill which got me a part on par with ABS/HDPE in terms of weight but without compromising on strength. I did have some under extrusion issues however and some of my print quality wasn't stellar. I attribute some of this to my lack of experience with both the material and printer and fairly damp filament. I noticed the more time went on, the worse some of my problems got. I gave it a quick dry in the oven and it reduced the imperfections but they weren't gone.

It was at this point I turned my attention back to the drive system, focusing on the tracks. King of the antweight class, Mr Peter Waller has had really good success with composite tracks in his 150g robots. He prints a flexible core with super soft silicone cast around it. I saw no reason why this wouldn't scale with a bit of modification

First off I returned to my drive pulley models. I completely removed the stupid spur gear string that was strangling and stressing my drive out. As the motor sits offset inside the frame in one of my signatures (aping off Biteforce yet again) It does require an indirect drive to the pulley. I stuck with spur gears for this, printing myself a Mod1 pinion and drawing another spur gear onto the HTD5mm pitch pulley. These rear ones are a 2 part affair, as picking all the support out from under that final flange was hellish and unfortunately lead to the gear profile being sloppy with bits of plastic hiding like plaque between the teeth. I pinned the flange on with lengths of brass wire and glued the assembly together. A solution to a problem that doesn't need to exist - totally on brand for me!

Donated to my efforts by Tom from Bakerbots was some really flexible TPU. It was free, the right hardness for my needs and right in front of me. The only downside was the colour. Used to make puppet parts in another life it was a rather off-putting fleshy beigey colour. Oh well, needs must!

I knocked up a rough belt profile to see if it would work as I required to and I was really lucky. It was perfect! The right amount of squash and stretch. Again I did the hand strength tension testing and couldn't break it.

I quickly added some geometry to the model to give the silicone something to form around and interlock to in the form of some bands and increased gaps. I ended up using a pretty special method to fill the moulds in the end. It was a mixture of using a syringe to inject silicone around the edges and then a sharpened pokey bit of wire to feed and pull silicone through the loops. What I was afraid of was they were just going to catch bubbles and leave cavities devoid of any actual silicone so I wouldn't get the benefit of the added latching. It would be more sensible to have a vacuum chamber, but I did all right with a pokey stick. The mould is just ABS and the silicone is the same as I use for my moulds, 25a shore from Easycomposites.

Probably the most impressed I have been by a single part! There is a lot of work involved in making it (3x 3d printed parts plus the silicone casting) and its a 24hr turnaround but it sure is pretty. Very happy! The tread is a V, just done to give it a bit of an interesting look, ever so slightly inspired by an M48 Patton tank tread but only in very broad terms. There wasn't much thought around traction behind it but it's an area I would love to explore at some point.

I quickly pumped out a second to see how it would go and left to go see Tiger 131 run at the Tank Museum while it cured. Hey, I'm nothing if not committed to the track life aesthetic. I picked up some lovely beers while there (I'm a sucker for that sort of thing)

Here was a brief segue back to the interchangeable weapons idea. Instead of having plug in modules I wondered if I could skate by just by having different waggly things tagged off the output of the weapon. Could my healthy reduction on the axe be enough for a lifter? The maths said "maybe" so I started to have a go with a trial lifting arm with a grabby hook setup tipped to follow. In the end I decided not to pursue this much further as it would have taken time I didn't have to fuck around with things I couldn't be sure would be helpful.

I printed another axe in a more suitable colour of PLA while I waited for the laser cut Hardox version to arrive. It was starting to look like a robot!

As it was reaching the later stages I began hunkering down to start cracking out the wiring loom and making armour. Unfortunately disaster struck as I tried to pull up my batteries from storage charge. No connection. Huh? Well shit, the tab on the positive output had somehow fractured. Anyone who's had this misfortune before knows well there is nothing that will let you solder to a lipo cell tab that hasn't already been tinned. If this was an antweight I'd junk the cell but I had no other viable batteries so I had to bodge it. I built up a connection using wire wrapped tight around the remaining stub and held the assembly together with adhesive. Test yanking the cables didn't cause any loss of connection and it charged okay so I ran with it. Ultimately I think this played a part in the downfall and I can chalk it up to experience: Don't Do Shady Shit.

In break with tradition I had set out to make the armour for my robots by hand. Normally laziness prevails and I will get them cut (or make the robot around what sizes I can buy off the shelf!) but I thought I'd have a play with some hand tools. Feels weird not to just push a button on a machine and let G code to the rest! I picked up the worlds worst hand plane, a sheet of 6mm HDPE and poured myself an exceedingly generous rum and coke. Time to go to work.

I marked out the shape on the plastic itself with a combination set and a knife blade. There was a little bit of vernier scribing to get my hole placements too. I scored the HDPE deeply enough that I could split it along the cut line then come in and finish it with the plane. I loved this method and will absolutely try and use it again.

The idea behind going for HDPE as a fairly integral part of the structure is to keep with the theme of wobbliness. I wanted to just absorb impacts as much as possible and come back from them. This said I was planning to have an aluminium plate on the front for facing spinners, especially Fingertech beaters as 4mm aluminium at a 45 degree angle seemed to work really well for just soaking up damage without letting the weapon dig in too far. I also cast a new pair of tracks, this time printed in white TPU of the same shore and black pigment added to the silicone.

Now children, this is how we desecrate a corpse. Yes I finally needed to crack MotherLoader V1 open to steal some metal gears for the weapon on V2. Those painfully cheap Rotolink 22mm only have 1 metal stage so I thought I'd best nab some so the axe didn't become too sad too soon. As I can't see it ever running again other than "hey it turns on!" funnies and historical value it can have all plastic gearboxes without harm done.

The final wire up and (dis)assembly had begun. here you can see the removable link and power mount which is just the kit from BBB mounted up onto a printed back plate. This was made from ABS for simplicities sake and is pretty much the only non TPU print in the frame.

The weapon motor fits through the bulkheads on a printed plate so it just pushes through. The ESC for that is also mounted in a recess in the bulkhead.

A very cheap Chinese pinion gear was part of the assembly. It was amusingly long (what it lacks in hardness it makes up for in length) so it had a meeting with a Dremel before assembly.

Once chopped it was able to be mounted. One bonus to the gear aside from the price tag was the double grubscrews. Less good is the fact due to the lack of boss its been rammed through the teeth, nuking the useful surface.

At this point the lasercutting turned up! I had some antweight bars and discs cut, alongside the obvious axes.

The front is the 4mm plate of aluminium I mentioned before. It has no wedging duties (saved that for the HDPE) So it was just there for ablative mass, bulking out the front 10mm thick where the nasties would hit it.

And this is where I start beginning to call last orders on Motherloader. The front is mounted, it has all the required safety bits (heck I even got ahead of myself and built in a locking bar - thanks me, you're the best!) Once again I became really pleased with the Tonka Truck-ness of it all. It was a very pleasant thing to have in the hands. Aided no doubt by the twist and shake that the more fluid frame and materials gave it. I would have no qualms launching it off the balcony and into the back garden of the chip shop below. A good sign for combat? Or perhaps just some unresolved personal issues.

The swing of the axe is intentionally limited also. Loose theory is that if left fired it is more use at beetle-waist height as a sort of flailing hook rather than fully fired into the floor lifting the tips of the wedge up. A sort of half thought through funny really. It was ultimately irrelevant.

Just showing off how brightly the arse of my robot lights up when you turn it on. Like some sort of insane firefly.

I took some time the evening before the competition to print out some bits for Tom, taking up my two ABS printers and so I just ended up printing my link flap doors out of PLA on the ender alongside another frame for one of his kitbots. They were a quick sketch in Fusion while it preheated and I just sent it with spares.  PLA meant they would be a consumable after all. 

 

I made the lid out of some more 3mm HDPE, same as the baseplate with the Stanley knife and plane method. Unable to help myself I decorated the lid with some old stickers from where I used to work. Juvenile I know, but one simply must have those "anger Volts"

 

Plonking the robot on the scale the night before the competition was perhaps bit of a chad move. However, it was all perfectly fine as I was sailing clean under the American beetle limit (3lb) and drifting several nautical miles from the balmy shores of the UK ruleset (1.5kg) This gives me a good indication of where I can better stick weight next time - hey, there might even be enough for a Fingertech beater. Shame they don't come in yellow. And that was that for the build! I paced this one better I think, more time wouldn't have really helped the problems that befell the robot. Well maybe months would have, but I wasn't crying out for a few extra weeks days or hours to twiddle around. I had a goal and managed to meet it in style with a lot of shit thrown at the wall and prototyping pushed out of the way early on. Parts got to be ordered and tested in good time, rejected or carried through to the final bot and that's no mean feat! I feel like I am cutting my losses from the word go sometimes but I have found constantly taking the path of least resistance (while staying true to the design vision) helps me to finish instead of chasing my tail moving goalposts.

I trundled on over to Bristol once again to push this thing through combat! As BBB is hugely well organised it had assigned pit benches and I was set up next to the formidable and impressive Luchador! This was the third version I think and it was probably one of the most aggressive non spinners going.

My first draw was pretty good, I had Angel slice which was a Mr Kippling themed (yes) horizontal spinner. And let me tell you, it is exceedingly good all right. It has had a big improvement in terms of spin up times and overall robustness it seems. It was going to be tough.

Speaking of tough, as it was a 3 way melee for the first round I had to face off against the ever present arena marshal extraordinaire, Mr David Weston and his robot Lilith. This was the less obviously dangerous robot but it was low, powerful and quick. It was everything that MotherLoader wasn't.

 

Thankfully I came away almost completely unscathed due to Dave quickly dispatching Angel Slice with a neat stack in the corner with ML providing a wobbly assist and anaemic love bite with the axe.

With Slice counted out up against the wall Lilith outflanked and overpowered me instantly. The power is impressive and the speed more so. 

I was unceremoniously dumped in the pit, quite a fitting ending. I never quite got going in this fight and I was certainly struggling to pick a direction. This had reared its head a little in testing but I think I'd managed to fool myself. It's easier to trick yourself that you're going the direction you told the bot to go when in fact it's picked one and you're along for the ride. It felt weak and listless much more so than when it was floating around the kitchen floor. I chalked it up to unfamiliarity with driving it and gave myself a mental pep talk to Do Better next time. 

Unfortunately my next time was against this sexy Spanish slice, a wonderfully anodized antagonist - Nemesis (sorry, NEMESIS). This was a truly beautiful machine which I think was a first build? The wheels were really clever in their construction and felt superb in terms of grip. The fellow also had an array of weapons to choose from as well spun off a decently sized inrunner. 

Watching the fight back I had marginally more control than it felt like at the time but I was outclassed by the pure speed of Nemesis. Not to be unkind but I feel like the driver is was outclassed by the speed of Nemesis too. It can't have been easy trying to direct that manic pinball whacking about the arena. Luckily for me it had trouble spinning up, but it was enough of a danger with the powerful drive. 

In a limp wristed display of control I managed to nudge the pit button on (only) my second attempt which spelled doom for Nemesis who after whirling another lap around the dazed MotherLoader got caught and went down. 

Victory by default is victory all the same but it felt a little underserved due to my performance. 

So with one fight won I still had at least one more before I was sent packing. Unfortunately that fight was against one of the major bruisers of the scene, Baby Shrekt. An odd and confusing name for quite a mean green machine but it harkens to it's parental featherweight which bore the name Get Shrekt. The novelty you can get out of green plastic truly is unending. I have known Sam rather a long time now, but strangely never fought before at any weight class. Couldn't have happened a few years earlier when he built axes, could it? Bastard.

Now I wasn't as bemoaning or otherwise piss riddled as others can get when drawn against a Shrekt because I was quite excited to see what it did to TPU and my particular brand of nonsense. Vertical spinners are an inevitable fate in life, much like death, taxes and getting food poisoning from a Harvester salad bar. If I wanted to test how my robot reacted one of the most feared and successful spinners this was a good way to go about it!

And what a way to go! 

Never really getting a grip early on just let me be controlled entirely by Shrekt who proceeded to punt me with astonishing force into the ceiling 4 (four!!) times and bounce me off the walls twice. 

The axe was sadly no longer attached, snapping at the join 'twixt shaft and gear but the metalwork was undamaged past some nicks. I would hazard a guess if this gear was TPU it would have survived but may have had some intrinsic problems being a gear due to being so wobbly. 

The base, only being 3mm HDPE held up well also! It has some road rash where the disc skittered across it but it did its job of taking the Shrekt-shaped bullet for the Lipo battery. 

The wedge survived the uppercuts well, with the aluminium bending outwards but not beyond usefulness. In a touch of hilarity it made a barrel nut *disappear* and pulled the bolt through TPU, HDPE and aluminium. It yanked the bulkhead out of whack too, evident by the wonderfully jaunty angle the front standoff is now sitting at. I'm sure that once tension is taken out (i.e. bent bolts) it will spring back to normal. 

The rear ABS panel absolutely ate shit though. It totally bought it and snapped, not outside the normal failure mode for the material at all. I had just got very lucky previously I think. I can't say I won't use printed ABS again for structural parts because I do rather like to flirt with the devil but it will be done with greater care and caution. 

The tracks also got a bit of a pasting. This was good as it let me know how that method of construction reacts to forcibly being separated by some angry Hardox. Relatively well. The silicone surface was peeled back but it wasn't removed and could still function. It also didn't have more than localized delamination which is what I was worried about. I didn't want one cut or slit to undo the integrity of the whole track, very happy to report this doesn't seem to be the case! The TPU core just seems to have flexed and absorbed any impact that came it's way which is just classic TPU at this point. 

My little bit of pride in all of this tragedy is that even after the pruning and air miles courtesy of Shrekt it was still on and kind of functional at the end. I clarity this statement by saying it didn't work any less or any more than it did at the start of the bout. Yes that is a low bar but I just about limbo under it if I suck my gut in. If the axe was still attached I could have righted my myself and been subjected to more hits.  

Perhaps an overreach but I like to imagine with a more reliable drive and less brittle weapon/righting system I could have kept tanking hits and there is a certain glimmer of a route to victory via the bridge of attrition. A man can dream in the non-spinner game. 

So there we have it. MotherLoader V2 is done and now I can reflect on it as a whole. I'm proud of it as an object and a shape that physically exists and of individual parts of it but I can't say I'm impressed or happy with it as a whole. Compared to the original it had the same 1:2 ratio but without feeling like it deserved to be there. A sad, bitter truth is that it failed the minimum requirements for being an active participant in it's fights and that is something I thought I was long past. I've broken it down in a loose list format. 

The bad:

Drive.

Yes, as everyone with eyes could tell you it struggled in a way robots shouldn't struggle. I definitely felt like it was lacking a consistent low end with the "pick up" point never feeling the same. I noticed this in testing and thought I had it mostly ironed out to an acceptable state with setting changes to the ESCS and transmitter fiddling. I hadn't gone far enough down that path or perhaps hadn't been looking in the right places. I also feel the battery was partially to blame with it's hack job of a broken tab meaning it possibly wasn't fully able to open the Current Tap for my thirsty motors. 

 

The motors I feel were a wrong choice. They probably would have worked very well for whatever I bought them for, 6 or 7 years ago but just because the shoe fit doesn't mean I should have worn 6 inch heels. I just don't have the legs for them. I think they were too fast and too angry for my setup and bogging them down with all the shit I pull just turned them into heating elements which occasionally spun. Going to something slower and more modern would potentially help, not that motors have a sell by date but there are more suitable types so its worth exploring (I may have some on order already!) 

The power side of things was also very disappointing. I would assume on a good day they'd supply the current more than adequately but they were having the worst days imaginable. I think they were a bit shagged after months of neglect and the broken tab was just the shit icing on the arse cake. They will be relegated to desk duty then inevitably the bin. I have the weight for something a bit beefier so I might up the capacity a little to give me more grunt. Size is a bigger issue but I'll burn that bridge when I get to it. 

The weapon fell foul of the same issues with motion as the drive, just slightly more obviously so as it was hard not to seem like a total wet blanket. This mechanism will be completely rethought and I have some painfully silly solutions to trivial problems. I can safely say I will not be revisiting the pod/module idea for a new version. 

The Good:

The printing I hold in high regard, TPU is a complete success and I will use it again without pause. 

Tracks were wonderful, it was a great benchmark to go forward and play around with traction patterns and all sorts of interesting miscellany there. It's relatively cheap to experiment so the world is my oyster. 

I was also very pleased with the shape of it all. Not only was it fairly pleasing and a relative signature look for me now it worked well defensively. The construction method was solid and reliable, even with the transition to flexible filament. I would love to find a more rigid material to print with as it would be nice to have some solid parts. Maybe it is time I popped my nylon or CF infused cherry. 

Luckily there are a few months of long winter nights to percolate and otherwise dick around with ideas before the next BBB competition in February happens. Watch this space for another yellow wobbly disaster trundling ever onward. 

Until next time, Space Cadets.

Peace xo.