Barróg - Featherweight Lifter Build Diary

[Maxamuslead]

Clearly eion needs more penny's XD but srsly chopping boards behave very differently compared to HDPE. It much more brittle where hdpe bends more then shatters e.g. Gabriel valley carbide. Reason why we love it
Absorbs a tonne of energy

[Shooty]

So this quick update is mostly to ask about gears, sprockets, and how to mount them, but first, an update on the new machine. Barróg Doom should be constructed a lot quicker than the original, with the base, back and sides being cut in the 30 minutes I had free last weekend to work on it.


A shot of the new walls, placed next to one of the old ones to give a sense of scale.

My main question to all of the builders on the forum is on the matter of mounting sprockets, gears and axles. I've purchased all of the relevant gears and sprockets off of Beltingonline.com, but I was hoping to ask for advice on how to mount them. I'm planning to use 10mm rods for all but the largest gear, which I was going to use a 25mm steel pipe similar to Giles Ruscoe's Mantis. What are common ways to mount these gears to the axles, and how to mount the axles to the bulkheads? What steps can I take to avoid the gears from misaligning? Should I use bearings and, if so, what kind? What would be the recommended way to machine these parts to minimize loss in strength while improving weight efficiency? Should I use the 10mm rods, or go up, or even down?

Any help people can chip in with this would be much appreciated, thanks in advance.


The parts in the flesh. Definitely bulkier than they need to be, but it's in how much should be shaved off that I don't know.


All of the parts seen from an angle. I'm thinking securing the sprocket to the small gear may be good, though space would need to be left of course for the chain.


A side on view of the transmission system.

[Ocracoke]

What are common ways to mount these gears to the axles, and how to mount the axles to the bulkheads? What steps can I take to avoid the gears from misaligning? Should I use bearings and, if so, what kind?

There are two kinds of shafts. There is a dead shaft, where the shaft is static and the gear slips around the shaft or has bearings inside the shaft to allow it to roll and then there is a live shaft, where the shaft is mounted by bearings on either end and the gear is fixed to the shaft. In Kaizen and Jibril, the axe is a dead shaft but in Jibril's drive system, it is a live shaft.

The material that the shaft is mounted in is important too. HDPE deforms way too easily (though I will confess, Kaizen's axe was a dead shaft in HDPE with a cap on one end to prevent it falling out).

The gears I have come with grub screws in them which I drill a indentation into the shaft and then threadlock them in. If yours haven't, I suspect welding it would be the best option.

Given your design, I would go for a live shaft for your first stage, mounted on flange bearings (https://www.gearsandsprockets.co.uk/...lock-bearings/ for example) and then perhaps a dead shaft for your final stage to the arm but, weight allowing, pillow bearings might work here.

What would be the recommended way to machine these parts to minimize loss in strength while improving weight efficiency? Should I use the 10mm rods, or go up, or even down?

10mm rods are OK in my opinion, depending on what it is made out of. I would have said a lathe to remove some of the bulk from the gear hubs would have been the way to go on losing weight.

Always make a cardboard or wooden template first and its a case of just being precise with your markings. Noting I have never done a chain drive, the sprocket and chain needs to be perfectly aligned so I would mount your motor with the sprocket on first, measure from the edges of your robot as to where the sprocket is horizontally and then mount your second sprocket on the subframe to line up with the depth of your first sprocket.

So if, after being mounted, the motor sprocket, from the far left edge of your robot, is say 280mm in, the second sprocket also needs to be 280mm in. I am sure others who have done a chain drive before can advise further.

[McMullet]

Lathe to remove weight from the gears sounds like a good plan. It's not the most efficient in terms of strength-to-weight ratio but it does mean you keep everything symmetric. I guess drilling holes would maintain more strength for the same mass reduction but unless you get them really accurate you might end up with some wobble. On the other hand I'm not really sure how much a problem a bit of eccentric mass would be when you're reducing the rotation speed so much...


I was wondering if could be worth you putting in some kind of direct solid connection between the motors and driven sprocket to keep them aligned? As it is, you're going from the driven sprocket, down the mounting bracket, then along the base to the motor (potentially also having to measure out to the edge then back to the motor). It seems like there's a lot of room for measuring/marking errors, and for things to flex in operation, which could leave you with misalignment and derailed chain, one way or another. A bracket that goes straight from the motors to the first driven shaft would give you a nice direct load path and minimal scope for error in manufacture. Might even be worth moving all the motor mounts onto the weapon bracket for the same reason.

I'm loving the new design - it's really clever bit of mechanics to double up the lifter/grabber function. Looking forward to some more updates!

[Eidg]

I spare weight by drilling holes in the gears, if they don`t turn as fast as a couple of rpm it should not be a problem.
You have to leave a couple of mm material around the holes.

[Shooty]

Well, I've been pretty negligent with updating this build diary, and as a result this has ended up being less of a description of my build methods along with asking for advice along the way, and more of something like this:



Nonetheless, I present Barróg Doom, now just a wedge and recharged battery away from completion!



As a result of this, for this post I'll be breaking down some of the key new areas of difference between both Barróg's, namely the new weapon system, as well as smaller differences in the general shape and construction of the machine.

THE WEAPONRY:

For this I'm just gonna rattle out a bunch of pictures with the important notes underneath them.


The new twin drill lifter setup. I had intended to run two 14.4V motors, but for Bolton I'll be running on two 10.8V motors instead. Probably the trickiest part of the bot to get right, as a single misalignment could strip everything and destroy the entire setup. It was managed in the end however, and both move as one when wired to the same ESC.

A shot of the new gears and sprockets. Some beautiful machining on display here, and most likely why this bot was significantly lighter than Barróg surprisingly. The drill press, as well as some new drill bits and spade bits meant that everything was in line, and it all fit together mercifully simply.


One of the earliest shots of the assembled weapon system. Many bolts had to be filed down so that everything fit together, and spacers were still required so that everything stayed in line. 3D printing to the rescue once more!


A good look at the chain in place within the robot. Placing the drill motors after everything else was in allowed the chain to be nicely in line and placed after any differences between CAD and real life (Thanks again for the mounts Roboteernat, incredibly accurate!). Also shows off just how surprisingly compact the entire system is; the main challenge left was how to wire the bot so the weaponry didn't chew the wires up in the process.


One noticeable change made between the CAD and real life was the widening of the radius on the claw mounting points. This allows the plates to sandwich between the main drive gear as well, reducing the risk of misalignment.


Locking bar setup for Barróg Doom. The bar prevents the gear from moving too far forward or reverse. Simple, but effective.

GENERAL ASSEMBLY:

Some of the main changes are already visible up top, such as the new rollcages on the side to prevent getting beached, as well as a bracer piece in the center of the machine to strengthen the green front piece at its weakest point. Smaller, more subtle changes have been made up front as well. The base has been recessed 20mm so that a hit bending it upwards won't jam on the front piece holding it up, and the holes have been kept to act as failure points so tears don't extend too far up the machine. The entire front plate is hinged, and bungee cords are being used for Bolton as an experiment to hold it down. My theory is that they will rob vertical spinners of much of their bite they usually transmit up to wedges, allowing it to stay viable for longer, while maintaining some downward force on my end. The theory could be complete pants, but hey, that's the scientific method for you.


The new base. The bungee cords will be mounted better than what they currently are, promise.

Barróg utilized the internal space within it to have loose wiring, which worked surprisingly well given that the bot could twist quite dramatically and not have cables come loose. Barróg Doom has nasty gears to chew things up however, and so the wiring has to be a lot tighter here. After some tinkering around everything was made to fit to the bulkheads, and the new link holsters used for Léim Thart were easily copied over to the new bot. Time will tell whether everything works in the end, or if something explodes. Plenty of fuses used at least!


The wiring on the left bulkhead. Somehow this is as simple as it will ever look, all the other connections plus the receiver cables turn it into a real junction.

Barróg Doom will make its debut at Extreme Robots Bolton along with a refurbished Barróg. I'm very excited to see how the improvements will play out, and will be doing plenty of research at the event to see if I can go any bigger than I have. To finish off the post, here's a general video of tests from Barróg Doom, including a drive test, a weapons test, as well as a self-righting test:

https://youtu.be/UuynlfCzbV0

[Ocracoke]

Much approval, well done. See you tomorrow.

[dotDominic]

This looks excellent Eoin! Very promising testing videos - that weapon really looks the business now. Good luck at Bolton.

[Eidg]

Nice work especially the gears which you made lighter.

[COAESystems]

I like this new design, looks like it will perform well! I'll be at Stevenage Robots Live but will mix it with Extreme Robots next year.

[MrkLeigh]

Love the new design, will be interested to see how those wheels hold up. I may get a few myself.

[McMullet]

I was wondering how you were progressing with this! Looks great. It's amazing how updating the build diary is the one thing that's easier to procrastinate about than building the robot. Thanks for the write up though, good to see the WIP shots and how everything works!

I note in this picture that the screws/nuts holding the hubs to the platewheel are on alternating sides.... any reason for that other than an obsessive need for symmetry (which I can totally understand)?

[Shooty]

Late as always, but here is the update from the bots from XR Bolton.

DAY 1:

We arrived at the event in good time, and both machines were looking ready to go. Unfortunately upon arriving to the venue we discovered that despite having more tools with us than some of the heavyweight teams, one of our transmitters was left back in Ireland, and I still knew very little about programming the new Spektrum Dx6e, as did most other teams around. Ultimately before the first fight, we ended up using a 6i for controlling Barróg 1 driven by my dad, donated by the Botwrights (Binky, Crackers N' Smash), and a more familiar Devo 7E from the Raes (Team Tartan Heart, Nessie), both of which we were grateful to have. I also was finally able to get my hands on the two extra 30A BotBitz ESC's I bought earlier, so it was only at the event that I was actually able to run both bots together.


Both bots ready to go in.

Because we were ready so early in advance, we both got to drive the machines around the arena before the first show to warm up as it were. Unfortunately after only a few seconds of trying to use the lifter on the new bot it stopped working, and once we took it out the problem was apparent: the 30A fuse protecting the lifter had blown. I've used 30A fuses to protect my 30A ESC's in the past with Barróg 1 with minimal problems, so I assumed that they would be fine with a similar twin drill setup. However I failed to account for the fact that the lifter would have far more stalling torque applied to it, and as a result every lift was in danger of exceeding the amperage loads. Without a more powerful ESC there was little I could do, and if I removed the fuse I'd only end up buggering one of my ESC's, so I decided to just replace the fuse, knowing it was a short term solution.

In the flurry of rebuilding Barróg Doom it unfortunately wasn't in one piece for the first fight, so I ended up just driving the first Barróg in the fight instead. (Credit again to Team Ironclads for the footage)

https://www.youtube.com/watch?v=sn1IWsOTtIY

Some observations from this fight:

• Fairly standard fight, went the full three minutes, didn't get hit by a vert, which is always good in my book.
• My largest contribution in this fight was righting the wheelie bin.

POST-FIGHT:


So after the fight Barróg Doom was fully reassembled and ready for testing, OG Barróg only needed a quick charge to be safe as well. Both machines were entered into the second warm-up to see how they handled. Unfortunately 20 seconds after starting Doom was shoved into one of the corners, splitting one of the side rollers, and somehow getting the machine stuck either on the wall or the floor. While trying to free myself, I ended up blowing the fuse on one of the drive sides, and lost the lifter fuse as well in the process. Meanwhile my dad was still getting the hang of driving a featherweight, and got flipped by Explosion multiple times in the process. We did get to find out however how OG handles getting flipped; the plastic top cracked slightly, but the wheels survived the sustained flipping, something nice to know as I never got to test out that part of the machine throughout all our fights.


For some reason Boom getting stuck caused one of the rollcages to split. Very strange indeed.

For the second fight, both machines were armed and ready to go in when Murphy’s Law reared itself again. Shortly after driving Barróg Doom into the arena the machine stopped responding to my radio setup, causing me to frantically take the machine out before the fight started. Dismayed, I ended up again driving Barróg for another fight, still unsure as to what happened to Barróg Doom.

https://www.youtube.com/watch?v=-gmVdQzEMKM

Some observations from this fight:

• The strategy remained the same as the last fight, survive until the pit drops, then get as many as possible in there. I personally counted three pittings in the fight above, so I’d consider the strategy a success.
• At 2:35 Binky gets a nice charge in on Barróg, but like similar fights previously Barróg still has maintained its almost unexplainable characteristic of being able to shrug large hits off. It’s nice to know that the machine’s chassis still holds up to these hits 20+ fights later.
• One thing different on the Dx6i compared to the Devo 7E I’m used to was the mixing already programmed in, limiting turning down to 40% and acceleration and reverse at 60% of what’s possible. This makes it easier to control, useful given how my dad was using it to learn how to drive, but also meant the machine never got in the powerful pushes I knew it could manage.

POST-FIGHT:

While it was a good fight for Barróg, I was still very disappointed at how Barróg Doom stopped working the way it did, especially as I was now halfway through the event with not a single fight under Doom’s belt. Upon outside inspection, it seemed the Orange Rx used in the bot was having issues, occasionally starting and stopping when large numbers of machines were active. Since radio issues are the worst possible failure for any bot, I figured I needed to upgrade as soon as possible to prevent it happening again. I spent the rest of the day brute-forcing my way through the Dx6e, teaching myself how to program it, getting my desired mixings through, and replacing the entire radio system on the bot.

[Shooty]

DAY 2

Most of the morning was spent chatting to teams about the night before, and both Barróg’s were in a good position to continue fighting. For the first time, both machines were able to enter the arena together for the first time at 100%, in a melee with just a single spinner compared to the other fights with multiple.

https://www.youtube.com/watch?v=YU2QUPpBPbY

Some observations from this fight:

• Within the first 5 seconds, Barróg ends up on its side. I would have been terribly disappointed for my dad given he’d be stuck on his side right at the start of his first fight, but fortunately the machine was tipped back over 2 seconds later, so everything worked out in the end.
• My strategy for this fight was to survive until the pit dropped, then see if I could get Get Shrekt in there, giving me 90 seconds to get used to how the machine handles. I was very happy with how it drove, though decided to use the lifter only sparingly.
• Dad was getting used to driving Barróg around for the first time in a proper fight. He found driving and turning together to be very hard to do, as well as which way to turn the machine when it wasn’t in line with him. Fortunately he ended up improving very quickly, which is always great.
• Once the pit goes down, I go for Shrekt, a move I almost immediately regret as one hit shatters my front end, rips the bungee cords right out of their hooks, and leaves me upside down. The machine self-rights despite the damage though, something I’m very proud of finally achieving!
• Less than 10 seconds later, Barróg then collides with Shrekt, splitting the top into two very busted pieces and disabling the main weapon, something that takes me a few seconds to grasp if the audio from the video is to be believed (I was next to the camera for this fight).
• At this point, I just go hell for leather at Shrekt. The machine takes a few more whacks at the front, but asides from the front piece and a chunk from one of the side pieces nothing else seems to be getting damaged, and I’ve self-righted multiple times without losing the fuse on the lifter.
• After a few attempts of pitting Shrekt, the bot flips me one more time, where I get stuck on top of another bot after self-righting. Unable to move, attempting to wriggle free with the lifter finally blows the fuse in it, rendering me immobile at last with 10 seconds to go. Also somehow Barróg made it to the end of the rumble as well!

POST-FIGHT:


Ow

Well, suffice to say, that fight was the most damaging fight I’ve been in so far. For Barróg Doom, I’d lost the top and front piece, the latter of which was divided into four pieces, the steel wedge was written off completely, both hinges for the front piece, one corner at the front, both bungee cords, and the lifter fuse. For Barróg, the front corner was nearly coming off, and the front piece was also broken into four pieces, tearing out the linear actuator in the process and breaking it (again). So like any sane person, we decided the best course of action was to try and repair the machines for the final melee.


The top piece on Barróg, there was already a crack near the top, but this hit just absolutely totaled it.


The actuator gearbox casing was utterly destroyed as well, splitting my weapon into two for the second event in a row.

For Barróg Doom, I took a template of the cuts on the broken pieces, and made spares at the event outside with a jigsaw, while Dad went to get new hinges and bungees at the nearby hardware store. For Barróg, we took out the remains of the actuator, and opted to just have the pre-cut horizontal config front piece fit in place to save time, held down again only with bungees. Plastic wedgelets were fit to each machine quickly, and somehow after a quick battery charge both bots were ready to go back in! Helping all the construction efforts was the fact that our pit neighbours left on the Sunday, giving us a whopping two full pit spaces to work with. Perhaps a bit overkill for two featherweights, but if any two feathers needed the space then it was these two!


The impromptu cutting station set up outside the event to get two new tops installed.

With a final congratulatory boastful expletive directed at Sam Price (Shrekt’s builder), both bots were ready to go in to the final melee, hoping to get as much field data as possible.


Both bots (somehow) ready to go back in.

https://www.youtube.com/watch?v=8BYnd6rQT6g

Some observations from this fight:

• Though you’d think I’d have learned a lesson from the previous melee, my targets again are the spinners in the melee. Unfortunately, staying in too close, along with the plastic wedgelets getting worn out quickly means that I can’t quite get under any bot with enough force to push them like I love to do.
• After a minute of driving around, I seem to lose one of my wheels on the right side, affecting my driving, the first wheel failure I’ve had since Manchester last year. The extra pressure put on the remaining right wheel followed by the massive landing from the floor flipper seems to end the remaining wheel, and I’m now driving on one side of the bot.
• Ultimately not much else interesting happens in the fight, Barróg is pitted shortly after the halfway point of the match, and Barróg Doom joins it with 30 seconds to go in the fight.

POST-FIGHT:

Overall a fairly average fight, and I was just happy to make it into the arena at all with both bots. The wheels were the biggest mystery to me, I was not used to them failing at all. Outside of a crack on the top piece of Barróg Doom no other major damage was sustained, so at this point we just watched the remaining fights, said our goodbyes, and packed up for the ferry ride home.


POST-EVENT THOUGHTS:

• The biggest letdown of the event for me was the HDPE chopping boards. Previously I theorised that the plastic would only chip away upon hits, but both bots had their tops rendered useless after one hit from Shrekt. I always knew this plastic wasn’t proper HDPE, but I always thought its unique properties would have some useful applications at least. A new, different type of plastic was ordered after this event.
• I massively underestimated the amps the lifter would draw, leading to me blowing weapon fuses constantly. A new weapon ESC was needed before the weapon would be able to do anything at all.
• The Spektrum Dx6e was a challenge to set up, but once it was working the machine handled beautifully, and I really prefer driving with it over the Devo 7E thanks to its larger size and stiffer resistance from the sticks.
• The wheels on Doom performed well for the most part, but two were lost in the final fight, and I had to retighten some retaining screws among the other wheels. Steps will need to be taken to make sure that my wheel arrangement will work better in the future. No damage to Barróg’s wheels though, they seem to defy all odds by continuing to work at this point.
• The steel wedge got too beat up while trying to get underneath things, and took too much off with it upon nasty hits. The HDPE wedgelets can be effective, but get damaged too easily, and will almost never win a fight against another bot with a ground-scraping wedge. A better option is needed for events in the future.

Bolton was nonetheless a great event, and like always, thanks to the community these events are fun no matter how well the bot performs (special shout to George, Sean and Alex for the drinks). I’m guessing most people reading this already saw how I did at the FRA FW champs only a month later, but I wanted to go into detail here about exactly what went right and wrong with the new build. I don’t necessarily see myself as an amazing bot builder, but I feel at this point I have a good sense of what needs to be fixed and how after an event, which in my opinion can be just as valuable.


Barróg Doom after the event, with plenty to work on.

[Shooty]

I note in this picture that the screws/nuts holding the hubs to the platewheel are on alternating sides.... any reason for that other than an obsessive need for symmetry (which I can totally understand)?

Cheers for the compliments! Probably the latest reply possible, but yeah, the need for symmetry is the only reason it's like that, makes no difference otherwise. We actually can't fit in a regular alan key into the socket head screws to loosen them, so my dad actually ended up cutting an existing key we had short to fit, not an easy job to do by any means!