Root-tiller or Claw?

How about a mini-roto-tiller or claw to prep the bed? Or incorporate SCiO into the bot?


We’re working on ways for FarmBot to prep the bed. Probably a miniature auger will be the best approach. The main problem is that the gantry isn’t rigid enough and the motors aren’t strong enough to handle high forces in the X and Y directions. So the solution has to exert forces primarily in the Z-axis (which an auger would).

Incorporating SCiO would be awesome! I’ve had my eye on that project for a while. No reason why it can’t be done!


I have a 3d printed rotor tiller i have been working on. It runs on 9 Vdc. I hacked the motor out of an Rc toy. The gearing still needs some work. Surprisingly the side load would not be that much.

1 Like

@Josh, that thing is awesome! Would be very cool to modify it to work with FarmBot. The Universal Tool Mount already provides 12V DC power, so that wouldn’t be a problem. The main concern would just be side forces I think, and maybe weight too. Looks like there was a bit of an issue with clogging when it picked up larger roots and sticks. Have you been able to prevent that with a different design? Are there any issues with the tills breaking when they hit a rock or something?

If the farmbot arm isn’t strong enough for lateral force, maybe a prong/pin attached on either side of the tiller could be added so it sticks in the ground. Of course this would mean that the farmbot couldn’t fluidly move, it would only be able to identify a weed, drop to that one spot, till, pick itself up and relocate.

1 Like

I agree with your assessment of the side load. I was watching the z axis on the FarmBot in the video you have up and noted there was some bouncing when the bot stopped. I did not realize how long the z axis was. Your first assessment would work. A type of screw would work. My Father was playing with this idea as well and bent a wire in a screw configuration and installed it in a cordless drill. It worked rather well to till the soil and removing weeds. This video is from some test he was doing. Video Link

I’d say one of the huge advantages of the farmbot project is the lack of need to prep the bed after initial setup. Tillage is a solution to simplify weed control. Since farmbot solves weed control by disadvantaging all plants not growing in approved locations and by tweaking plant separation, there is no need for tillage. All the better to let the mycelium do its job.


Richard is absolutely right. Tilling is actually really bad for the environment and should not be encouraged. It kills the mycelium colonies which essentially eliminate the need for soil additives. The mycelium breaks down dead roots, plants, insects, etc.

Best practice is NOT to till the soil!

I sincerely hope that part of the FarmBot movement becomes an educational one and we leave behind outdated ideas, like mono-cultures, tilling, and soil additives. What automation offers here is a way to create a diverse and healthy environment.

As for preparing the soil, research composting. The grass, weeds, or whatever can be composted with the dirt, creating an amazing living soil in which plants will thrive indefinitely into the future. If you see it done in Big Ag, it’s probably a bad idea and should be avoided.


I envision something like what you made but heavy enough to stay dug in and run along on its own in a straight line. The farmbot would then just pick it up and set it onto each path. The batteries and induction charging would be done at a base station that the farmbot would put it into when finished.

Or if a straight line would be a problem you could still have the farmbot guide it but still never actually have any negative kickback by having a joystick type controller that the farmbot’s specialty tool sits on top of loosely. Like upside down dixie cup over a long neck bottle. It would steer and make corrections based off sensory input instead of coordinates alone. It could even get its power from there as a fail safe, if the arm comes off the machine stops. I am thinking a magnetically connected power cord in this case so that it can find and plug itself in after a disconnect too.

I am sorry to say you lack experience with augers/milling tools or you would have worked out issue. If gantry cannot take forces in X and Y directions don’t think putting a auger in ground will be a good idea. Hits root starts rolling root around self start pulling self some random direction in ground. The issue with augers and milling tools is unless they are held well and they hit something imperfect they then put forces sideways.

There is a tool called a air spade. Yes this needs an air-compressor line this would allow force only on the Z axis. Does not break down roots. It is uses to break up surface compaction to make sure water does go in.

FuturistPlayground is right you don’t want to till soil if you can avoid it.

1 Like

Am I missing something here? The small tiller I designed is no more disruptive then blasting the ground with compressed air. The total depth is only 25 mm. As far as the Z not being strong enough it probably isn’t. But this can be resolved.

I was only playing with the idea of having a rover knock out some of the weeds in my garden. And took the design considered from the stirrup hoe as my influence. The intent is to only disrupt the surfaces. No worse then raking the ground with a hand rake. It actually started out as a Lego Robot with a rake pulled behind it. After 10 passes in one row at the right time no weeds. LittleRover


Josh the main advantage of blasting ground with air spade is the force direction you are dealing with on the frame is nice and predictable. So straight vertical against the frame. Fairly much force like the seeder being pushed into ground.

Air spade tools are fairly much undo compaction. Air spades can be fairly much the vacuum seeder in reverse. Amount of pressure and placement in relation to soil set depth.

Main reason I mentioned it Josh is that air spade tech people normally don’t consider. Yes a air spade doing only to 25 mm could go on the existing frame without needing to alter strength.

A little rover on ground like you used does not get zero compaction either. Air spades can get soil to zero compaction.

Also air spade and small tiller do have slightly different usages. Air spade depending on pressure if pressure is low enough can break up soil surface yet not break plant roots. Breaking soil surface can be important while keeping plants alive can be important of soil turns water repellent in the growth cycle. Tiller usage is normally when no plants you want are in the way and air spade usage is when there are plants you want in way but you have a issue like soil being water repellent you need to address.

Sorry to say your small tiller and air spade both set-up to do only to 25mm the air spade will break less roots. So there is a clear difference between the air spades and tillers in usage. Yes a rake will break more roots than a air spade set light.

Basically sorry Josh your statement “The small tiller I designed is no more disruptive then blasting the ground with compressed air.” is wrong. Blasting ground with air is disruptive to soil the most and less so to roots and a tiller even a small one disrupts soil and roots fairly evenly what makes it higher disruption than using an air spade. If your objective it to kill weeds air spade is not highly effective so you choose something like bashing or tiller. If your objective is to make sure soil takes in water while not killing chosen plants while doing those plants root zone air spades used at right pressure are great.


If it was powered up could it be a weed killer too?

1 Like

I am assuming it would be as the pressure will break the roots. I see now that it comes down to controllability. As the tiller is not able to be controlled verses the air spade. So the next question is how much pressure does the air spade run and what volume is required? Even if the system was to use a needle to push into the soil and push air into it. Would this be the same? Sand and soil will act like a fluid when air is pushed in it. So maybe it would not need a large compressor. I just had a thought. Would a blower work? If the air was focused like a jet engine? It could use a Brushless motor like the ones found on Quadcopters. There 12 VDC and spin crazy fast. The only concern is the amp draw can be upwards of 20 amps. I have smaller ones that are rated for 12 amps. I am not sure of the current capabilities for the FarmBot UTM. There is a lot of thrust coming off the little props. There are some jet models floating around that may be modified to work for this application as well. Just thinking of ways this could be done without having an air compressor tethered to the FarmBot.

Scary part is that is not the air pressure itself that breaks the roots with the anything like air spades. Its amount of distance it has attempted to move the roots and if plants roots cannot take it or not.

Yes needle into ground with air pushed into it does create the de-compaction effect.

Machine used on golf course and the like. This is the thing about a pressurised air de-compaction you can perform it while the bed is fill of crops.

Thing to remember a genuine airspade needs quite a huge compressor at times

So most of the compressor based soil de-compaction tools are 90 to 100 PSI and volume of air related to the size of the tool. Airspades are used for stripping soil off plants with air being supersonic. some diy versions run up to 150PSI. One of my trades is putting in cables like telephone and data. This is where I came across air spade. You can dig down to water pipes/cables/plant roots without breaking them.

I really need to define low. Under 200PSI of pressure is low for these air compressor based soil de-compaction or soil from plant removal systems. To use a airspade as a weed killer you would be talking 200PSI plus to attempt to damage the weed but also by then the soil is also long gone.


If it was powered up could it be a weed killer too?

So the answers both yes and no. If you power air spade like tech up big enough to be a weed killer its also a garden bed exterminator. Killing the weed while stripping the garden bed of soil is not what you are after for weed control and this is exactly what power up air spade tech results with. When you are needing to clean a plants roots off of soil due to some fungus infection or the like is another use of the air based tools.

Plant roots are stronger than soil in most cases. Air takes the path of least resistance what is break up the soil not the roots.

Tiller is not a selective process where air spade and other air de-compaction tech is selective.

1 Like

That robot is Awesome :grinning:

I’ve been thinking about making a Mini or full sized Power Harrow like this.

Its what we use to cultivate on our farm, But the only issue is Force! I guess you could make it do like a 20mm level then go deeper each time but I’m not sure :confounded:

1 Like

These are exactly what I am planning on building next. I am working on designing planetary gears to slow it down and add more torque to the blades. But this was going to be the next version. I like the aeration ideas as well. That will take more time to develop. I am coming in to my busy season, and will have more time this winter to tinker. O and Thank you.

From what I am seeing so far most air spades are used on tree roots. The plants that the farmbot deals with are short crop. This started out as a way to control weeds then moved into soil prep.
The information so far has shown mostly high pressure systems. The nozzles seam to be larger 10mm and up. I think injecting air under ground with low pressure would work. I will need to do some testing with air pressure and see what would work. Thank you for all the feed back.

The air spade seems to be destroying soil structure, and doing it to a greater depth than the tiller. I would see such pressure based tools focused toward needle injection of biochar slurry deeper into the ground, with the intent of not destroying any of the soil structure in place. Another way of reducing compaction is filling the holes created by precision planted, winter killed. diakon radish, which when rotted leave a hole 1 or more feet deep.

A simple application of 2 to 4 inches of compost on the surface can do much to improve soil structure. But a multiple mix of cover crops can do wonders. Under Cover Farmers.

I see the farmbot precision planting and maintenance expanded to large acreage on a similar sized but mobil platform taking care of 100’s of acres of interplanted mixed crops.

I am thinking a weed eater tool that would cut weeds, grass, cover crops right at the soil surface would be the best approach to weed control and would operate on about the same level as the little tiller, which I find a quite impressive innovation.

1 Like

There is a weight to draft ratio for all tillage implements. You need it to be heavy. From my farming experience…A ‘small’ tractor is about 3000lbs, aggressive tire treads, 25HP at least 4WD.

When you attach something to a drawbar, it’s now probably about 20 HP. We add front and rear weights to increase traction. Can probably pull 4 ft tandem.

That’s for draft. Mowing implements are lower HP and lower weight. Shallow cultivation.

When you are considering low weight single row bots for farm work, you have to think about implements being ‘driven’ and not ‘pulled’. This tiller idea slashed to the back won’t work because we are going to have to face the challenge of high draft requirements for traction.

It’s more complicated than what’s in the YouTube ‘tiller’ video.