Mr. Stihl Your Girl

[Patrick (Patrick Faulkner)]

I’ve always had a love of saw based robots, with dead metal being my favorite robot as a kid. I’ve never made (actually finished) a saw robot, but I’ve had a few attempts previously.

Struck with inspiration at FF, I hastily scrawled down an idea for sportsman featherweight, entirely based on a pun.

The design intent was to have a dual chainsaw/waggle stick control bot, using as much actual chainsaw parts as possible. The chainsaws are effectively decoration, and the idea was to use the teeth to get more purchase on slick HDPE or metal edges, and provide an entertaining looking robot.

Using cheap brushed mini chainsaws seemed a good place to start, and a few days later my idea started to take shape

I intended to use as many parts as I already had lying around to both reduce cost and ensure I actually get it built, so the construction will be as simple as possible. It’s my first full HDPE bot, as well as my first router based build so this might be a fairly steep learning curve compared to my preferred flat pack sheetmetal.

I doubt my initial servo based idea will have anywhere near the torque required to self right, but it’s a start. I’d love to have each “arm” individually controllable, more for show than actual use but I suppose there is also an element of redundancy in doing this. The coaxial lift/chainsaw drive is also due to the fact I’ve got a lot of 12mm shaft parts from other projects spare, so it should help me get moving quickly. I started designing a proof of concept mechanism and as usual it got a bit out of hand…

Meet Pochita, the test bed for a single module of my chainsaw drive. I got carried away with the mechanism and worked out it would fit pretty decently into an ineffective beetleweight

Some printing later and here we are

It’s the 4" bar version, and they go up to 12" on the same mount so it shouldn’t be an issue to change lengths for the feather.
Thankfully it all fit together apart from the weapon shaft, as I misread 40mm as 30mm so the drive flats are too short for it all to fit in the right place, but thats nothing a bit of machining can’t fix. If it all goes well, it should be a simple matter of scaling it up to suit the feather version.

[McMullet (Sam Roberts)]

Chainsaw feathers are the new meta!

That looks great, I love the chunky gear built into the mount and am totally kicking myself for not having the same idea.

Are you going to bring it to Robodojo?

[Patrick (Patrick Faulkner)]

The gear seems to work under testing but I’ve not had it run under power yet! I’ll stick the .STEP file in here somewhere once I’ve tried the new version of you want to have a crack at it. It’s based on my old saw bot which used a similar live axle to keep the weapon motor in the body rather than on the arm. I reckon moving toward a HTD pulley or perhaps a smaller module gear might be better for what I want it to do as it’s currently getting driven from a servo which isn’t ideal.

I think Pochita is designated for a life of decoration and testing without some serious redesign, but you never know. I’m aiming to take the feather to robodojo once it’s complete. I’ll probably be at robodojo with some antweights in November all going well, I don’t see me finishing my main beetle by then at the current rate of progress.

[IrregularJoe (Joe Earlam)]

I had a very similar idea when I saw you could buy just the replacement bars and chains on aliexpress for under a tenner! I’m so glad to see someone has implemented it, I hope we get to see it fight.

[Jack (Jack Franklin)]

This is amazing, great work so far.

Would love to see it in the arena some time

[Kraken4me (Alex Revell)]

Impressive and very spikey!

Be Stihl my beating heart…

Reminds me of Chain Bastard.

[Patrick (Patrick Faulkner)]

After a brief hiatus, I’ve picked up working on this again.
I’m going to separate my beetle weight/pochita stuff into a different thread at some point to not clutter this up as it’s grown legs and turned into its own project to get Hydra drive working on the new ESC.
In the meantime however I’ve redesigned the chainsaw mechanism into modular parts for my testing. Moving to a servo based control rather than a “dumb” motor solution was always the plan at some point, but enough 150kg/cm RC servos would bankrupt me, and mechanically coupling them through the horn has never filled me with confidence.

I had a bit of a search over the last week for different mechanisms before finding some extremely mispriced 24v servos on Amazon, looking similar to those used in Flatpack by harry. I doubt they will wow with their performance but provided it’s enough to move a waggly stick with teeth on them I’ll be happy, and more importantly if i break them I’m not out of pocket

Moving back towards the mechanism, it’s now split into a “pod” which should drop in and out of the robot for replacement, rather than trying to make new frame parts every time I create a new revision

Extremely overbuilt and I reckon I can drop a lot of mass and size out of it when I get it together and play in person, its got chunky 15mm HTD5M belts and pulleys just because that’s what I have lying around. I’m not sure of the torque required for “sawing” but I’ll stick the original chainsaw drive back on a tester some point soon and get a target value.

As it stands, I’m left with this -

Keeping the overall shape the same as the first draft, but there’s a lot of empty space due to the servos position.
There’s a 4" bar and an 8" bar on the model, the 12" I initially planned I think might look a bit silly!
Plenty of room for significantly larger batteries, it’s just a question of how much this is going to weigh out of HDPE, and how rigid it will be when assembled.
Once I’m back to work I’ll get the design finalised and on the router, hopefully only a few weeks till its moving!

[McMullet (Sam Roberts)]

Looks sweet! I did a bit of research on chain gearing for Chainway and the main thing to look at seems to be chain speed. About 10-15m/s seems to be the normal range for electric saws, the one I used as a “donor” was 13m/s but I geared it down to just under 9m/s to be compliant with the BEVs limit of 20mph. As long as your motor power is less than the rated power in the saw, and you aren’t slowing it down ludicrously, the chain tension should be about the same.

It looks like you have a similar setup to me there, I also used 15mm HTD5 belt and a 5055 outrunner. Mine was 760kV on 4s and had to be geared down about 4:1 to get the right speed (about 3500 RPM at the drive sprocket for the chain istelf).

[Patrick (Patrick Faulkner)]

That’s great information! By sheer dumb luck the parts I have here should put me in the same ballpark, my motors are only 470kv so on 4S with a 5:3 reduction puts me at about 4000 RPM. Going up to a 2:1 gets me to 3384rpm which is probably where I’ll go to in the future, I’m not constrained by space where the pulleys are.

I did manage to get some more progress on the module mockup (ignore the concrete screw keeping the pulley in place, I lost the shoulder bolt somewhere in my bag) but Antweights has taken over most of my free time for the past few weeks. Hopefully we’ll have the first spin/actuation test by the end of next week.