Instructables
Picture of Lawnbot Learning Experience
This isn't an instructable  so much as a cautionary tale. What was supposed to take 5 days and under $600 to complete took instead almost 2 months and close to $1200, due to bad planning, poor directions, substandard parts, shipping times and absolutely no experience.

This is an RC lawnmower loosely based on the plans by DIY Guru johndavid400. I basically used his plans for inspiration and made up my own version. However, had I stuck to his plans I probably would have spend A LOT less money and time while trying to figure out this build.

So, first - The Electronics!

Wasted $70 - on 2 Arduino UNOs as their intended functions were included in the Sabertooth 2x25.



An interesting thing to note, is that I didn't need the Arduinos at all. All the servo encoded signals to Pulse Width Modulation conversions were handled effortlessly by the Sabertooth 2x25 motor controller board. And I couldn't figure out how to create the "Kill switch" in case my transmitter looses contact, so the lawnbot is currently a bit of a death trap, and will travel for about 7 panic filled seconds with no input from you. Develops instant reflexes in kids. ;)

Notes on the receiver:
The receiver power is really easy, it just uses the same servo cabling as all the other pins. I think it can handle up to 12 volts, but I've never taken it past 9. Currently on the sabertooth it's drawing 5 volts with no problems at all. I'm using a cheap 6 channel RC/Helicopter Transmitter, with an included Receiver from Esky, really similar to this one. http://www.xheli.com/6chesherctr.html, though mine cost ~ $50. The servo ports on the receiver part are marked: B, 1, 2, 3, 4, 5, 6.

I'm guessing the "B" pin was for power. So I used a servo extension wire (purchased in bulk from www.dealextreme.com) and just used the black and red wires for power. The white wire is usually for signal, which you don't need for power. ... I think. The documentation that came with the transmitter was translated poorly from Cantonese, so i just poked around with a probe until I figured it out.

Since I was using an avionic transmitter, all the signals were "mixed" to compensate for inflight adjustments. Unfortunately this meant I had to test each channel to see which ones were "clean" which means which ones worked without affecting any other channels. Luckily I found 3 channels that were clean so I set to work rewiring the transmitter to use the sticks on these channels. Also, I rigged the throttle as a standard elevator stick by tinkering with the inner mechanics of the transmitter so it operated just like the other stick and centers itself when I let go. Otherwise it would stay in the position I left it.

You only need the signal wires from your clean channels to run into the motor controller. From there, the Motor Controller takes the 5v signal and does a little calculation, saying 5v is full power to the wheels... 2.5v is 0 power to the wheels and 0v is full power reverse to the wheels. This allows you to use the full 5 volt sweep of the servo signal to give you a forward and reverse signal for the tank drive system. Cool eh? You don't really need to know that, you just need to remember tank drive and set your DIP switches accordingly.

Notes on the Motor Controller: 
The DIP switch settings I used were for the standard tank drive setup. 1 channel per wheel. Regarding the death trap functionality, I think it's a standard function of my avionic receiver rather than my motor controller settings, since the only signals the motor controller is getting are from the servo lines on the receiver. The receiver keeps supplying the servo signals when the transmitter signal is lost. I think this is a safety feature so your Helicopter doesn't just fall out of the sky. You have a grace period to regain control. 



Part 2 - Finding proper wheels is a must... I did not.

Yeah, don't use dolly wheels and try to adapt them to drive wheels. Important lesson. This ate up most of my time and a lot of the budget (directly and indirectly).

Wasted $30 on hard plastic dolly wheels that ended up spinning the hubs independently of the treads.



Part 3 - Remember to measure everything. I did not.

The dolly frame and wheels I purchased were completely inappropriate for the job. The Dolly frame was about a 1/2 inch too small, and wouldn't fit over the exhaust port on the lawnmower. I didn't want to modify the lawnmower at all, so I could reuse it in case the lawnbot didn't work out.

Wasted $110 - on heavy duty dolly with inflatable wheels that weren't intended to be used as drive wheels. The frame of the dolly is destroyed now and couldn't be used for the lawnbot in the end because it was a little too small.


Part 4 - Made a new frame. Have no idea how to mount the wheels to it.

Holed Angle Iron is what I made the frame out of. If I had better tools I would have probably used something a little more solid. I had to use locktite on all the bolts that held the frame together because they all shake loose when the lawnbot is in operation. I didn't have any locktite so I spray painted all the bolts in place. That seems to be doing the trick.

Wasted $20 - mounting brackets that I eventually ditched because they made adjusting the height of the lawnmower too difficult.



Part 5 - Horrible motor mount design


Wasted 1 month - trying to mount these motors, using only a wrench some nuts and bolts and a dremel tool. Come to find out the motors are from the 70's and have no compatible parts in circulation, anywhere. So, any mounting plates or drive wheels would have to be manufactured by a third party, effectively blowing the budget waaay outta the water. So I tried to make something myself, with varying degrees of success. 



Part 6 - Failure, complete with reenactment!

Took it out for a spin after finally getting the motors attached, and broke it. I didn't get it on camera, so I did a reenactment for your viewing pleasure, and my ultimate humility.



Part 7 - Looking promising...

HOLY, RESIN FILLED WHEELHUBS BATMAN!
Wasted $50 - 2 part resin to lock up the dollywheel bearings with the electric motor spindle. Almost works! 



Part 8 - Success-ish...

Oh what horrors have befallen the likes of me. The lawnbot moves! It cuts grass! It kicks some serious Butt! However... One of the wheels still managed to get mangled, partly due to my plastic resin fix, and partly due to the crappy 30 year old motor mounts.

Wasted $120 - on the 30 year old motors I purchased on eBay. I had to go out and purchase some modern motors with proper drive wheels for an additional $270. At this point I just wanted the build done. Luckily it was one of the hottest July's on record, so the lawn didn't really grow that much as I fought with this stupid thing. 



Part 9 - Threw out the old motors and wheels!

New Wheels and motors! This thing goes really fast and can handle the bumps a lot better! I bought some good motors from eBay with proper drive wheels. What a hassle free difference that makes. $170 investment

REMEMBER: Save yourselves a lot of time and confusion by using the proper tools and parts for the job... if they're available. ;)



Conclusions...All in all, it was a really fun build. But not one for people on really tight budgets. I learned a lot and I've since made some improvements to the design.
  • Durable plastic electronics box, instead of flimsy metal. (I don't know what I was thinking)
  • Routed the cables a little more efficiently
  • New proper drive wheels with a modern motor and properly aligned mounting system
  • Fresh coat of paint to keep the bolts and nuts in place, and make it look spiffy
  • Redistributed the weight so it's less likely to do wheelies and kill me
I'm going to refine the whole machine design for next summer, and hopefully have something in the end that won't frighten old people.
bremus1 year ago
Sorry to hear you wasted so much money.
you could have taken a wire off the spark plug and have it connected to ground withe a relay that you have rc control over
huf1232 years ago
Nothing is wasted. I always say that you are going to pay for your education one way or another. Whether it's college, parking tickets, or life in general, you are going to pay. I once spent $1,000 learning how to build a hydroponics garden for less than $150. I want to build a robotic lawnmower, so thank you for sharing your mistakes.
David972 years ago
(quote)"Wasted $70 - on 2 Arduino UNOs as their intended functions were included in the Sabertooth 2x25"

Ill buy them from you how much would you want for them?
mbear3 years ago
You know, sometimes seeing what didn't work is more helpful than seeing what did work. Thanks for providing this information.
phenoptix3 years ago
My kind of instructable! A seat of your pants style build, with an idea of where it will end but no real idea of how to get there! All my projects would be like this if it weren't for my other half's questioning. Thanks for sharing your journey to what appears to be a pretty cool lawn mower / robot wars entry.
rimar20003 years ago
Very well done! The balance is positive, because you have the experience.
Agreed. This was not money or time wasted.
gentlemoose3 years ago
I've been going down so very much the same path as you, working on some autonomous bots. So much failure. So many dollars. So much almost learning.

FWIW, I've been trolling Craigslist for motorized wheelchairs - quickies and jazzys, IIRC. The motors are rock-solid and they have over-engineered controllers AND very strong, stable frames. I tend to find them for under $250 complete.
Grathio3 years ago
Great writeup! I usually screw up at least as often in my projects but I'm to vain to commit the lessons to an Instructable.
omnibot3 years ago
He he .. nice.
A wonderful tribute to all those projects gone awry that we never really finnish.