View Full Version : Mech - Scythe

07-13-2008, 07:39 PM
As Tyberius has suggested I'm starting a thread for my mech Scythe.

Scythe is a custom quad walker. I built a quad a few months back from scratch and it's just been sitting on my shelf. The quad was really a proof of concept and an opportunity for me to try out my new Lynxmotion Botboard. In it's current configuration it simply walks around, no control, no avoidance. It's basically just a test platform. I am currently working on the code to make it controllable via a wireless PS2 controller. I also need to modify the walking gate. It tips over on the front right leg no matter how I try to rebalance the batteries (or it flops on another leg if I go too far. It just needs a more robust program to make it really effective. Like I said I got bored with it and shelved it. I need to grab some video and throw it up on Youtube one of these days.

Here are the basics in current configuration:

MCU - Lynxmotion Botboard with a Basic Atom Pro.

Code - written in Basic Stamp. I have the whole code available if anyone wants it. By education I'm a Mechanical Engineer, so it is probably not the most elegant code. However, I commented the heck out of it, so if you're just learning, like me, it's easy to follow.

Servos - Hitec HS-311, yes I know the cheapest servos out there. If you look closely at how I designed it you'll notice that the servo motors have to do very little work to hold the leg in position. It's all about force angles people.

Batteries - 7.4V 1200mAh NiMH. I bought a couple wireless house phone batteries on clearance.

New Design:

The current bot, which we'll now name SP1 (Scythe Prototype 1) will get completely rebuilt. I'm working on redesigning the frame and legs in SolidWorks and having acrylic laser-jet cut parts. You can get that done cheap and it means you only have to assemble your bot once the parts arrive. Control, as I mentioned above will be accomplished with a wireless PS2 controller (2.4 GHz). I'm thinking about getting one of these http://www.geeks.com/details.asp?invtid=MINI-WIRELESS-CAM&cat=VID for the vision with two servos for pan/tilt function. I'll use the same two servo pan/tilt mechanism to aim the weapon. I'm planning on using SP1 as a test bed for weapons and trying to pilot a mech and firing at the same time. There is no such thing as too much stick time. Plus, I think it would be great to see how difficult hitting a target is.

I gotta say I am wicked excited about this whole deal. Just ask my wife, she's ready to beat me. I bought some bb's at Wally World today (6mm orange). I'm planning on building a mini arena to contain the bb's and then paint the inside black to make it easier to see the balls.

That's all for now. I'll post more later this week as I continue building.


07-13-2008, 07:46 PM
Good start. Do you know what servos you are going to use for the final version?

PS. I can't believe you have those pointy tips on that nice wood table... ;)

07-13-2008, 08:11 PM
I'm planning on sticking with the HS-311's. My wife and I just had out first son nearly 9 months ago and disposable income = 0 right now. I try to make up for meager components with better designs.

Thanks for the comments.


PS that table is the best area for taking nice photos. I test it on the floor. I forgot to mention the pointy feet will get changed to a flatter foot with an impact sensor, so as it moves over uneven terrain it will keep the camera/gun level.

07-14-2008, 04:17 AM
Nice quad, now get that video uploaded! ;-)

It's nice to see someone going ahead with a quad, those mech wars are gonna be great!

07-17-2008, 02:52 PM

that's a nice desing! I was thinking of making a biped but a quad could be a nice opputunity to try it out. Plus they look really cool!

Nice work of that bot!

07-22-2008, 10:04 PM
Thanks for all the compliments. The videos are uploading on Youtube right now.

Over the past week I have been busting my butt working on the programming. Right now SP-1 is a random walker (it walks straight, then turns, then turns the other way, then backs-up). I'm working on the control program. I have two avenues I'm pursuing. Avenue A is a PS2 controller to control the robot. I've been quite frustrated with this. The Avenue B is using MSRS and then sending the signal to the bot through wireless serial/blue-tooth/something. I'm leaning this way for resiliance come game day. As others have pointed out PS2 controllers in that environment could prove messy.

I also made some progress on a CO2 powered NERF dart launcher. I know the rules are limited to BB's only right now, but I still plan on building a 4 Dart shoulder launcher. I was also planning on making it a Salvo fire (all 4 darts firing at once), that's against the Shotgun rules right now as well. I still plan on building it, A. because it's easy, and B. because it's AWESOME! Check out this test video YouTube - Scythe NERF dart CO2 test 1

Basically, the launcher is a CO2 bicycle tire pump. I took the connection off the end, drilled it out a little bit, and threaded a piece of brass connector with a hose barb on it. I then connected a short length of tube I had from my 3# combat robot, which just so happened to be the perfect fit for the NERF Dart, and KaBlam! Missile Launcher. I think it would look sweet mounted up on the shoulder of a bot. I can pull the darts for competition, and then just demo it against someone in a grudge match or something later. It is too cool not to build though. Imagine 4 darts ripping off the bots shoulder and speeding towards an opponent. I imagine the combined weight of 4 darts would be enough to knock and opponent over.


07-22-2008, 10:40 PM
That's crazy cool, Matt! That's got some acceleration behind it!

07-22-2008, 10:53 PM
Those little 12g CO2 cartridges are at ~900 PSI. The little tire pump doesn't regulate the pressure at all. it restricts the flow somewhat, but as you saw not very much. I'm thinking I could machine a block of Aluminum to hold 4 darts and just plumb 4 lines into the back. The trigger on the tire pump requires very little force, and could be actuated with a simple servo, no problem.


Edit: I'm still waiting on the walker video to upload. I'll post tomorrow evening with the link. You can scroll through my videos to find it, if you're in a rush. Note: I had the MCU and battery loose on the bot, so it flopped like a fish while walking. It's normally reasonably smooth, but it does tip over on occassion. I'm working to fix that with better mechanics, and better programming.

07-22-2008, 11:58 PM
I say let 'im compete with his nerf gun!

07-23-2008, 12:08 AM
I have no qualms about nerf weapons. I'll add it to the rule set.

07-23-2008, 12:28 AM
I work with 60,000 psi intensifiers. Now that would really make a nerf projectile interesting.


07-23-2008, 09:56 AM
With that much pressure it could actualy become a weapon to knock over the eneny 'bot!

Can you give me a link for the intensifiers?

07-23-2008, 10:32 AM
More like a shrapnel grenade tossing bits of mech all over the arena...

Which might be pretty cool unless it's your bot :veryhappy:

07-23-2008, 07:17 PM
I would be very impressed with NERF if they made a dart capable of holding up to 60 ksi. Considering most metals would have a difficult time managing that pressure, a little foam dart taking it, would be very impressive indeed.


07-24-2008, 07:55 PM
Video YouTube - Scythe SP-1 walker demo

Note: all the guts are hanging loose, so it flops around a bit.

In good news I was able to get the PS2 controller to talk to the ABB. Turns out the PS2 controller was bad. I have everything connected to a wired PS2 controller for now.

I'm still planning on implementing a PC based control board. After I get it driving on the PC I'd like to try and work on an auto-targeting stereoscopic program. The bot will look for a color, trangulate the distance by comparing the two images, and voila. The idea would be once it lokks on it could just unload its ammo until the other guy went down. Including NERF missiles!!!! It may prove to be difficult to use in competition, as it would tend to lock onto someone's shirt in the crowd.


07-24-2008, 08:36 PM
I'm hiring a squad of people in brightly colored shirts to jump and dance on the sidelines for every game I play.


Nice bot!

07-25-2008, 07:17 AM
That's very nice!

Good job! Do you know how much payload it can carry?

07-25-2008, 08:39 AM
It looks like it is being trained to walk by a dancer. The way it swings its legs and moves them high and at 'point'. Very cool. Keep it up! It looks like those sharp points really grip the terrian. If you rounded them off a bit I would guess that they could slide a bit and take some of the sway out of your stalk. But then leaving the grip enables your load capibilities to be a bit higher. And it looks more manacing. Are you going to make the shins into sythes of doom?

What a novel Idea Tyberius! How 'bout a hex-mechs -with nasty BB loaded turrets on top- when it is detemined that the opponant is using colour and/or movement to auto-target, squats, drops its back door and a pod of Kodachrome worthy swarm-bots in the general shape/size of ping-pong balls rolls out and takes off in every direction.

Of course your idea is actually feasable and prolly a lot less expensive. If they all looked like Robert Palmer backup dancers and you had them all Riverdance it would prolly get the same reaction from the audience as the swarm bot diversion...

Sorry, I'm running on Diet Rockstar this morning.


08-05-2008, 10:00 PM
Mad Progress Yo's!!!

I made some good progress on the overall design of the bot. More to come hopefully tomorrow as I get everything wrapped up.

Still to be completed - Lower assembly (BotBoard mount), correct femur mounts, leg control linkages
Still to be started - turret assembly.

Overall, I'm pretty stoked with how it's turning out. Everything will be waterjet cut from 1/8" aluminum. Hopefully, I can get some pieces uploaded to Big Blue Saw tomorrow and capitalize on their free part offer. The rest I'll have them make at a later date.

Good Night,

08-06-2008, 12:36 AM
Drawings look great. I really liked the video of it walking around. Can't wait to see the whole triangulation setup.


08-06-2008, 10:29 AM
Looking good..keep up the good work..

08-06-2008, 11:28 AM
Those CAD drwings looks really nice!



08-06-2008, 12:37 PM
Sweet quad design.

08-24-2008, 10:16 PM
Well, I haven't posted much lately, but pretty much all I've been doing lately is working on my mech. For starters I was tweaking the mechanical design of the quad, squeezing in all the components, and then started looking at everyone's biped and thought they were way cooler, so I put the quad development on hold and started working on a biped. I am still working on the legs, and the kinematics. In the words of Albert Einstein "Everything should be made as simple as possible, but not simpler." That is my thought process right now. I'm hoping I didn't over simplify the walking mechanism and end up building a fancy paperweight.

I also made great progress on the weapon prototype. It's basically a take on a baseball pitching machine. It uses a small brushed motor spinning a 1.4" diameter wheel. I loaded about 25 bb's in the hopper and had a paperclip holding the bb's out of the wheel. I taped a plastic bag over the end of the barrel for testing indoors. While getting ready to fire, I heard a very quick odd vibrating noise. I thought something got sucked into the wheel, so I killed the motor. I looked in the bag and there were all the bb's. I must of bumped the paperclip enough to let the bb's get past.

In a purely theoretical case the maximum rounds per second is around 800, limited by the speed gravity pulls the bb's. The wheel itself is capable over 2,000 based on theoretical motor rpm/volt, wheel diameter, bb diameter, etc. Based upon actual conditions (bb's rubbing on the tube, temporary jamming, etc. I'd be happy with 400-500. I plan on limiting the firing rate, but I want to be capable of a high rate of fire, just in case. :D

I intend to post pics later this week. Next in weapon development is servo pan/tilt mechanism and hopper/gun mounting solutions.

08-25-2008, 12:41 AM
>> In the words of Albert Einstein "Everything should be made as simple as possible, but not simpler."

[Psst! Ya missed on that one by five hundred years or more... Try William of Ockham - "Occam's Razor" - 14th century. entia non sunt multiplicanda praeter necessitatem Probably dates back double that even with at least Alhacen around the year 900ish - but it's commonly attributed to William of Ockham]

Sorry - geek moment. I make frequent reference, even here, to Occam's Razor. :)

08-25-2008, 07:25 PM
I have a book here of Einstein quotes (great for those literature essays in college) and it gives it to Einstein. My bad.

08-25-2008, 08:36 PM
Aha! Book's Bad then, not yours. :) Lucky you didn't spring that one in the college essay. :D

08-26-2008, 11:49 AM
Hmm. Top ten in a Google search all cite Einstein as well. Is there a definitive source on this one?

08-26-2008, 12:03 PM
OT Alert:

I'm not generally big on Wikipedia, but it's at least a start to search from in this case: http://en.wikipedia.org/wiki/Occam's_Razor

08-26-2008, 12:44 PM
The actual quote is from Einstein, and is explained here (http://en.wikiquote.org/wiki/Albert_Einstein):

It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience.

"On the Method of Theoretical Physics" The Herbert Spencer Lecture, delivered at Oxford (10 June 1933); also published in Philosophy of Science, Vol. 1, No. 2 (April 1934), pp. 163-169. [thanks to Dr. Techie @ www.wordorigins.org and JSTOR]
Variants: Everything should be made as simple as possible, but no simpler.
Everything should be made as simple as possible, but not simpler.
This is very similar to "Occam (http://en.wikiquote.org/wiki/William_of_Occam)'s Razor", with the addition that it warns about too much simplicity. Dubbed ''Einstein's razor, it is used when an appeal to Occam's razor (http://en.wikipedia.org/wiki/Occam%27s_razor) results in an over-simplified explanation insufficient to meet needs or goals. It is also similar to one expression of what has become known as the "KISS principle": Keep It Simple, Stupid — but never oversimplify.

08-26-2008, 01:00 PM
>> when an appeal to Occam's razor (http://en.wikipedia.org/wiki/Occam%27s_razor) results in an over-simplified explanation insufficient to meet needs or goals.

We should start a new thread. ;)

I think the distinction is unnecessary as Ockham covered it nicely, IMHO:

"entities must not be multiplied beyond necessity" - "entia non sunt multiplicanda praeter necessitatem " *Emphasis Mine

08-26-2008, 07:37 PM
Hilarious. True genius frequents this board. Capable of designing incredible death machines and holding a detailed literary discussion in the same thread. Love it.


08-27-2008, 07:55 PM
Just out of curiousity how tall is everyone else's Mech? I'm figuring my legs will be between 5 and 6 inches tall. I'll need atleast 3 inches to accommodate the targets, plus a head, clearance for torso twist, plus others. The battery I plan to use is 5.5" long (7.4V 6000mAh LiPo). My gun with hopper is 8" from the top of the hopper to the bottom of the barrel. I'm thinking overall it is going to be 16" tall. I've never built a biped, nor seen one (other than on the computer), so I'm not certain how far off I am. I may squat the legs a little, for a lower CG. This mech may change from the tall sleak design I intended to a short squat little juggernaut.

I'm also starting to realize this thing is going to be heavy. :robotsurprised:

Cutting-Edge Robotics

08-27-2008, 08:01 PM
What my bot is going to look like: :D:p


08-28-2008, 05:38 AM
Well, I haven't started building, but I think mine is going to be tiny because of my tiny bidget. I'm thinking around 10-13 inches tall. I'm thinking of a biped brat with guns on top. :veryhappy:

Wooot! Your bot is going to have to turn completily to fire his 4 guns! Will he fall over? :eek:

08-28-2008, 06:54 AM
Yeah, Mspurk, Love to know how he will stand with all that fire power, but looks cool though.

08-29-2008, 04:15 PM
That picture was intended as a joke. I have all these high expectations, and somehow I think it is going to look more like a hacked together monstrosity.

I do want some fire-power, but that bot is just ridiculous.

08-31-2008, 01:18 PM
Big setback. Up until now I've been testing the gun by firing it into a bag (very close range) to make certain my house wasn't covered in tiny bb's. I tried firing the gun at a towel about 8' away firing ~15 balls, and it was ridiculous. I took video, but I'm not going to waste my time posting it on-line. Edit: I wasted my time, see below. My long-range accurate rifle looked more like a sawed-off shotgun. There were bb's everywhere except in the towel. Of the 15 only about5 hit the towel, and none of them had any power. It was pathetic. Here's the video YouTube - Airsoft BB gun prototype 001 - failure


I need to replace the foam wheel with an o-ring wheel, and I need to add a second wheel to take out any spin produced by the single wheel. We'll see what happens after that. If ti still doesn't work then I'll scrap it. Thus why I prototype stuff.

Cutting-Edge Robotics

08-31-2008, 01:25 PM
If I feel bad about it, could I get away with confessing that I laughed? I'm sorry - 75% of my prototypes are... prototypes too. But you have to laugh.

I always figure if I don't laugh at myself, sure as heck someone else will, so I might as well beat 'em to the punch... ;)

08-31-2008, 02:19 PM
Well, it was a nice try though.. I think it's quite inovative..

Good luck with the next version.


08-31-2008, 03:47 PM
If I feel bad about it, could I get away with confessing that I laughed?

No problem. I was laughing as well shortly thereafter. I mean really that was pathetic. Aen't no two ways about it.

Cutting-Edge Robotics

08-31-2008, 04:27 PM
Good try. Your going to get it right someday.

08-31-2008, 05:53 PM
Ok...I probably shouldn't get into this one...but, I think if you look at softball and tennis ball pitching machines, I think a lot of them use two wheels for more than one reason. The impart spin to the ball as well as propelling it forward. It also looked a bit like the balls were getting caught up hitting the inside of your barrel. Seams like it would be a bit tough to get just the right amount of sling using wheels on such a small scale. I wish you the best though.