Anatomy of a Frog

The Preamble Ramble

I’m not saying I am making this thread under duress, but Joe Brown knows where I live and I’d rather write a post and avoid having a 3D-printed horse(shoe crab)’s head turn up in my bed. Imagine the cost!

So here we are, I am Nick, I have built a few frog-based amphibian robots with various lifty-bits attached. A build log of some sort has honestly been long overdue so I’ll prolly be updating these posts a bit with new info if/when I remember to mention it. I’ll try and keep it to a post per-robot so i’ll stick this fancy index here in the first post for later.

Indexy bit:
v1
v2
v3
v4

Proto-frog


[The lineage of Lance-a-frogs in Robot Arena 2. Some day i’ll make a horizontal hubmotor version. Promise!]

Frog originated in the venerable game that is Robot Arena 2, way back in 2010, and there the idea lived for many years. It wasn’t until 2017 that I had started to get a proper interest in building a real-life bot, and at the time the main source of information, encouragement, guidance I was receiving on this was from a fellow builder of RA2-based tomfoolery, the wonderful Mr Craig Croucher. Heavyweights, he said, were stupid, and there was an event coming up in early 2018 that was tailored towards newcomers in a smaller weightclass. So that was the plan. I was to make a beetleweight robot, and compete at UWE Bot Brawl 2018. I just had to make a bot from scratch, based only on a robot design made in a creaky old video game. Simply build a beetle.

The early Frogs were designed in SketchUp because it was something that was free, which is always a nice feature! It also was very ‘2D’ in how stuff was made which made sense to a complete novice like myself, it felt approachable which broke down the ‘scary’ aspect of the initial design process.


[The First Draft of a Sir Lance-a-frog. It was definitely a learning experience.]

It was a process that took a few months, with the aforementioned guidance of Craig. Never outright saying ‘no’ to any idea or questionable design choice, but instead giving insight to why x or y could be a better choice overall and letting me make a final decision. He was very patient with someone learning the ropes, and it was and still is deeply appreciated.

Eventually though, around September 2017, there was a design. One that I felt would work, but importantly one that I felt I, a buffoon, could make and achieve myself at home. It felt tangible, it felt possible, it felt like the right time to start building a Sir Lance-a-frog.


[Fun Fact: the early ‘lances’ were just long DIY caulk nozzles from B&Q, because it had to have a lance on the robot. it’s in the name!]

4 Likes

Sir Lance-a-frog v1

aka “Baby’s first robot”

The original frog was a very safe and simple design, based on the ‘standard’ components of the time (1.0Ah 3s lipo, 25mm motors for drive, 3/8” Banebots and a thicc servo), not doing anything flash or tricksy, because my tools then (and still now) were very bog-standard stuff. The lifter was a simple 3mm mild steel plate hinged at the top and lifted by the chonkmeister of a servo I found off HobbyKing. Somehow totally non standard which made finding replacement servo horns impossible, but just worked. ESCs were BB10s, again just for their simplicity and tendency to just do the thing asked of them with no questions asked.

All versions of Frog have followed the same basic principles: 15mm HDPE for the chassis walls, 5mm for the top and baseplates, and then a mixture of thicknesses for the other parts. Wheelguards have varied between 5, 10 and 15mm versions, Front wedges and wedgelets have always been 5mm, and all have used M5 nuts/bolts/fixing because it felt like the most robust solution for the size and thickness of HDPE. When it came to actually constructing the bot, all parts followed the same guide: print the design 1:1 scale onto paper and then stick it to the slabs of HDPE, before cutting them out in a very efficient manner.


[Yes, that is many many holes drilled around the perimeter of the chassis walls. Yes, I had to sand everything down afterwards. Yes, this is how the first frog was made. Yes, I know this is a challenging image to process]

I’ve used a multi-tool since then to cut out the bulk of parts. I learnt!

Another aspect to being a simple design, was it allowed focus on making it a solid design overall. I always tend to favour robustness over all: Build your robot to take a hit, be able to react to that hit and take more if needed. By tanking a hit and being able to keep going, you give yourself time and opportunities to get a lucky counter, or just to wear your opponent down. This design choice lent itself well to making it out of thicc bits of HDPE, which would be fair easier to manage, to repair or to maintain at an event, than mild steel or other materials.


[CHASSIS ONLY]

As far as competing? It did reasonably well at UWE 2018, managing to win it’s first two fights, my first fights in robot combat, before losing the next two. The latter fights are where I learnt my passive skill, my innate ability to find open pits and drive down them. Something something good habits forming.

The next event at Robots Live! Burgess Hill was a proper smack in the face, as Frog proceeded to lose it’s first three fights in varying manners, including another pit excursion, before picking up a solitary win.

By this time I’d already noticed a few things that I wanted to improve on v1:

  • making the front of the robot a shallower angle, to help get underneath opponents
  • changing the lifter from 3mm mild steel, to 2mm gr5 titanium for weight-savings and hardiness. Mild steel bent far too easily on spinner hits, which would leave large, awkward curls and cuts.
  • Increasing the amount of internal space, as with large, bulky components there was barely any room for the delicious electric spaget that makes every robot do their thing.


[He silly. Fun Fact #2, I stopped putting feathers on the robot when they got sucked into the drum of Swag Demon during a rumble. Imagine if the FRA had heard about this!]

It quickly became apparent that making these changes to v1 would prove very difficult – so the decision was made to create v2, and v1 would hop onto the shelf to enjoy retirement, with a record of 3 wins, 5 losses over 11 fights at 2 events. Not the worst start.

In summary: Got my foot in the door and left me wanting to do more.

4 Likes

Sir Lance-a-frog v2

aka “the Bugglebots 2018 one”

Construction was almost identical to before. Design parts in SketchUp, print them off, stick to HDPE and attack with tools on the foldable workbench outside.

As with the previous changes mentioned above, another change was the colour of Frog. This is where I started experimenting with painting my bots, as whilst plain HDPE can look nice in certain situations and usages… it was a bit too plain for my liking.

Turns out gunmetal grey is a lovely colour to work with, and can deceive more than one person into thinking your robot does actually have metal parts on it.


[I really liked this version for how clean it was. This is the BBB3 version, and Fun Fact #3, the wedgelets that it ran at Bugglebots and beyond were 99% to cover up damage the bot recieved at BBB3. Vanity!]

I really liked the visual simplicity of this version, and in comparison to v1. The robot as a whole felt a bit sleeker than before, and felt like a good test bed for ideas. Wedgelets on the front, some ‘proper’ lances made from HDPE to actually try and poke and prod and be awkward with. Thinner wheelguards and just generally little refinements over it’s lifetime.


[Mmm, copper.]

Something that did come to the fore in this version of Frog in particular, was that it could take a fair whack and keep going. It still fundamentally drove the same way as v1, which helped in continuity with the driving skills I’d developed so far. But it turns out whilst the robot did have a few of the previous issues solved, the true war-crime of the link placement had never been addressed, which certainly cost Frog once or twice, but thankfully never in a prominent, recorded and streamable situation…


[Feck]

Bumlinks. Not even once.

This frog stuck around for a while, doing the bulk of it’s fights in 2018 and the beginning of 2019 over the course of four events (BBB 3, Bugglebots S1, FRA Euros 2018, and UWE 2019). I achieved my best ever finish in a beetle comp at it’s first event, managing to reach 4th place overall at BBB 3. It then lied dormant until 2023 (we do not talk about the spinner version), where it came out of retirement to compete at the 2023 BBB Champs. Y’know, just a casual 1714 days/4 years 8 months between events. As you do.


[“I still got that frog in me”]

Once again though, there were a few things I wanted to try and change on the robot. Whilst the weapon was a big part of the reason for this version, it again never really got to flip or lift anyone. Bots would ride up the front but a lift, or even tipping them over was still difficult. So with the intent of trying something new for the second season of Bugglebots, I wanted to try something a bit new and out of my comfort zone.

Frog v2 currently sits at 9 wins, 9 losses over 27 fights at 5 events. Judging by the previous gap between events, look out for it in 2028!

In summary: My beautiful tanky boy.

11 Likes

The Phrog has returned! All hail the robo-phrog! Fun fact: the exploded image is my google wallpaper.

1 Like

spoilers i guess but loved that shrimech mechanism at the last champs, cant wait to see how you did that! also really digging the look of that horizonal version of frog

@Kraken4me Thank you very much, that poster is up there with one of the best things i’ve made, loved putting it together.

@morgan_ls Thank you. I will go into detail regarding that srimech when I get to writing about v4, people seemed to love it. And the horizontal frog, maybe if someone twists my arm enough :wink:

But without further ado, It’s time for:

Sir Lance-a-frog v3

aka “One of the UK’s many beetlweight 4-bars”.

4-bars are cool.

In the early 2000s, on the rare occasion I was able to catch a showing of Battlebots on Comedy Central here in the UK, there was one bot that always stood out to me – Biohazard. It always stood out as such a cool robot design, super low with the side-skirts, and the lifting arm that would just effortlessly tip and hoist other robots over. And after my first year in robots and using the flappy style lifter design, I felt like a 4-bar would be more effective overall, and to take things up a notch. It was something I initially wanted to try and adapt Frog v2 to have, yet after hmm-ing and haa-ing for a while, eventually settled on making a new version, that could incorporate a few other tweaks:


[Experimenting with bits on v2, post UWE 2019. Maybe it could have worked ok, but to give the idea a proper go, it was time for a new chassis.]

  • A redesign of the back of the bot to reduce the mount of panels and parts to manufacture, give more area for the back panel to bolt into. Why? V1 and V2 had two angled back panels, which had to screw or bolt into the chassis rails or the base/top plates themselves, and they never felt 100% secure, and there was not enough space to use the same fixing method on both. Simplifying this would allow me to use one method that was more permanent.

  • Incorporating a link door into the top plate of the robot. Why? The link door was located on one of the back panels, so was now moved to the top plate to retain easy access to the link. And there was no way I was going back to a bum link.

  • Having a spare top plate that could be set up to have a weapon setup similar to v1 and v2. Why? A contingency plan, should the 4-bar be irreparably FUBAR’d at an event, I’d be able to pivot and still run an active lifter, rather than be stuck with no active weapon.


[Chassis bits. Cut out with a multitool. No death by a thousand drill holes this time!]


[Once cut, sand everything down to the dimensions needed with one hand, because the other is needed to pull the trigger on the drill. Make all the dust. I was wearing masks before it was cool!]


[:clap: CHASSIS :clap: ONLY :clap:]

The 4-bar itself was designed with the aid of some knock-off Meccano, as designing something like this from scratch was something I needed to do visually. I’m not smort enuf to just go into CAD and smash this out in a few minutes, it had to be something I understood and could look at in real life. With the Meccano, it helped me get a good idea of how long he arms would need to be in order to get the motion I wanted, AND to fit into the constraints of Frog’s internal height. When I had gotten my head around that, then I was comfortable attempting to design the actual linkage in SketchUp – not an ideal program for testing linkages due to the lack of proper joint support, another reason for wanting something to work on IRL.


[I’m glad this made sense to me back then, because looking at this in December 2024 (and taking the photo of it now) it’s a bit “huh?”]

It was also decided at this point to bolt the arms to the chassis walls, rather than having it’s own rail in the middle of the bot to bolt to. This was done to again keep the part count and the weight down. Figured the space was there and the mechanism would fit there, so that’s how I proceeded. Trying to keep the weight down nearly buggered things up however, when I decided to lighten the chassis rails with two large holes, and nearly had to scrap and start over due to picking the worst place to put these new “speed holes”. Surely these will not cause problems later.

[Video of the first test of the 4-bar. Cuppa in progress (not pictured but heard)]


[v3, functional, with the new bent wedge design. 5mm HDPE bent with a jet lighter. Job done.
I do not remember WTF I did to the lance between cutting it and here. Why is there a hole there, past Nick??]

V3 proved to be quite adaptable during it competitive lifespan, or rather, I was less wary or hesitant of adapting it. At SXS it had a version of the 4-bar that re-used the titanium plate from v2, and it briefly fought as a front-hinged flipper at BBB Champs ‘22, because why not? It could also self-right in this form, imagine being able to self-right during a fight, crazy…

v3_fronthinge
[He do a front flip]

Overall, these were only relatively small tweaks compared to v2, and in theory allowed me to reuse certain parts and pieces from v2, on v3. However, this was only done on certain situations and occasions.

Internally, v3 was very much the same as v1 and v2 before it at first. There were gradual changes, namely the move to brushed 22mm motors as 25s became a little harder to source and keep reliable. Towards the end of it’s run, I also moved to the BBB Dual ESC, which helped in simplifying the internal spaget.

Frog v3 is also where I started to get :ribbon: 𝒻𝒶𝓃𝒸𝓎 :ribbon: with the liveries and paint jobs. Whilst v2 had started the paint trend, and I’d enjoyed the gunmetal/copper look, It became a great source of satisfaction and distraction to decide what new look I would give Frog for an event. And a lot of these ideas were spur of the moment ones that ended up being fun. “What if I made decals for this version?”/”Oh hey, this event is super close to Halloween.”/”Froggy Chair my beloved.”
Doing robots in general has given me such an outlet to be creative and to get ideas out and into the world, and it is a great feeling to have an idea come to fruition, expressed in the way you wanted it to be. I’m very proud of what I’ve done and always wanted to be one of those builders that is making something cool-looking, even if it’s never ultimately super-competitive.


[So many looks, which one is your favourite?]

For a robot built in 2019, it was always fun to try and keep it running, push it to do as many fights and events as possible. Making it all the way into 2023 feels impressive to me (even though I didn’t compete for two years thanks to da Virus), and for a while it lived up to the reputation of being a tanky mofo… there was plenty to be proud of with this robot. The Last Chance Rumble in Bugglebots S2, the run to Top 8 in it’s return at Robot Rebellion in 2022 showed what it was capable of on a good day.

I will say though, in my opinion the tankiness definitely dropped off in the last few events it ran. The deep runs felt less possible towards the end, for many reasons. There are and were a ton of new and exciting robots that’ve appeared that have done some super stuff (Rudimental, Luchador, EMP and Chucky do the lifty/flippy thing far better than Frog ever did), existing robots improved a ton, drivers matured and got/continue to get the best out of their robots (Abracagrabra, Babrog, Attitude Adjuster spring to mind), and spinners in general became so potent even compared to where they were a few years prior (Gizmo, Propane, Icebreaker to just name a few) - and Frog v3 never really made a step forward to keep up with that. There was never a real ‘magic bullet’ that was going to get Frog to go from ‘Top 16’ to ‘Easy Top 3’.


[So bemdy]

And then it got smoked. Remember the speed holes from earlier? In it’s final competition fight at Robot Rebellion 2023, the right-side of the chassis was was basically torn in two by Baby Dead Bod, a horizontal spinner that likes to throw hands and wheels/motors at its opponents and itself from time to time. Over the course of one relatively short fight, it tore that large, major section of the chassis off at its weakest points, leaving a mounting point for the 4-bar hanging on by the smallest amount of warped HDPE. Turns out if you give a horizontal with dutty strength a nice opening to hook into, it can absolutely ruin your shit, fam. Aside from also bending an arm of the 4-bar, snapping the mount for the wedge, tearing through the holes for the chassis bolts, and generally just carving that side of v3 up. Elsewhere, there were also large gouges taken out of the rear of the left chassis wall, which compromises the ability to secure the rear panel of the robot. The paint was also scratched. What an asshole!

The thing is, aside from a fat dent in a new lipo I’d bought specifically for the event (disposed of at the event), internally it was untouched. It worked. With the wheelguards prised off the wheels, it drove. The servo still worked fine, the receiver was fine, the ESC was fine, it’s just the chassis it was attached to was all kinds of buggered. Yet it’s a point of pride that I got it into a whiteboard an hour later.
Whilst it has been “repaired” to a functional, driveable state since, these are not repairs that will hold up to any sort of scrutiny, in a competitive situation fighting spinners, so v3 currently sits in retirement alongside v1.


[Punished Frog]

At 16 wins, 12 defeats, and 36 fights over 8 events, it’s definitely the one version with the most longevity out of all the frogs.

In summary: Get 4-bar’d, Son!

5 Likes

Frog is love, Frog is life.

Love a 4 bar, and just the frog in general, the sword too to coral (however you spell it) people. and the paintjob makes me jelly every time. hope to see more frogage at events im at next year