Just watching the videos of the farmbot in action something from CNC machines come to mind. Why no mid machine storage for tools? Makes no sense really if objective is be power efficient.
Like when putting seed in having to run back to the start point every single time when changing seeds or getting more seeds is a waste of power. Putting tool storage in gantry would allow machine to be proceeding down the garden watering/spraying spot weed swap to weed tool deal with weed and then swap back to water tool all without requiring a second pass. Seed planting could be planting two types of Seed in a single pass. Lot of CNC milling machines have a tool store on the central gantry for a reason saves a lot of time and can save a lot of power.
Other thing with the tools there is no anti fouling cover over them when they are in storage. Yes it one thing to make a rack to hold the tools its a little extra to make sure while the tools are sitting in rack birds and other items cannot contaminate the connection.
Its funny you should say that. I had just been thinking the same thing.
Im looking at building a farmbot from scratch (I cant really afford the $2900 + shipping to the UK. Especially since I’d then need to mod it to suit my needs anyway) and I’d found myself wondering about that. I don’t recon it’d be hard to have a tool cup on each side, to allow a second tool to be carried so 2 jobs could be done in one pasd
The version 0.1 design actually did have tools on the gantry. See this image from the whitepaper:
There are two main reasons why we moved the tools from the gantry to the bed/tracks:
There is a lot more space on the bed to store tools. While FarmBot doesn’t need a lot of tools to function, we imagine people wanting to experiment with many different tools, seed bins, etc, especially with first generation devices, and so storing them on the gantry would quickly feel limited.
Putting tools on the gantry requires an additional motor that allows the tool bay to move so that FarmBot can select different tools. Another motor, more wheels, and another sliding component all add more cost and complexity to the machine. And when we already have a perfectly functioning 3-axis system that can position FarmBot to complete the same task (mounting and dismounting tools) it seems inefficient from a cost perspective to add another axis.
Now this isn’t to say that gantry tool storage is a bad idea, I think its a great idea, just not the best for smaller, first generation devices. But perhaps on a device with longer tracks and a smaller repertoire of tools (5 or less), gantry tool storage would be the better solution!
I said gantry storage not that everything had to be stored on it I should have been more clear.
Putting tools on the gantry requires an additional motor that allows the tool bay to move so that FarmBot can select different tools. Another motor, more wheels, and another sliding component all add more cost and complexity to the machine. And when we already have a perfectly functioning 3-axis system that can position FarmBot to complete the same task (mounting and dismounting tools) it seems inefficient from a cost perspective to add another axis.
This is the wrong presume that it requires a extra motor. Lets say you only need two tools while going down the bed. You have the holder like you have now on the edge of bed on each of the gantry towers. Yes no motor no wheels nothing fancy just a tool holder to allow a tool to be placed on the tower instead of on the bed edge. Yes the Farmbot is a functional 3 axis system this also means you can use vertical storage up the towers as long there is tool holder that goes onto the tower in a stepped pattern or gantry tool holder can be pushed out way when nothing is in it. Nothing I am mentioning here is a extra motor. Watering and Weeding come to mind as task that happens at the same time but there can be others.
Basically I see no against minimal gantry storage unless there is some vibration issue or something else that means putting a tool in a holder on the gantry means it will not stay there. There is one catch I can think of is requiring 4 end stop switches on the cross gantry access instead of 2. The two extras is to mark the start of tool storage so when zeroing carriage you are not running straight into tool storage this could also be avoided by mandating clear space at top for zeroing so no max filling each gantry tower vertically. So vertical storage on tower with 1 minor rule is fairly much make something to hold the tool that has a hinge so it can be pushed up out the way when it holding nothing.
True, another motor is not required if you only do partial storage on gantry. The tools would need to be modified so that they can be mounted and dismounted along the y-axis. Or, all tool bays would need to be facing the same (or complete opposite) directions.
Also something to consider is that if there was one space on each side of the gantry, then FarmBot could use at most two tools (not three) while away from the main bay, because it always needs an open space to dismount a tool before grabbing a different one.
Because tools are usually picked up and used for a while vs constantly swapped (at least that’s how I’ve been configuring my sequences) I still don’t think gantry storage for tools would be that useful. ie: it wouldn’t improve efficiency considerably because it isn’t saving very many trips back to the main bay. If the FarmBot tracks were really long (greater than 10m) I might feel differently about it.
I think gantry storage makes a lot more sense for seed storage, which is something we’ve been wanting to improve. Here’s why: Seeding requires FarmBot to go back and forth to the main bay every single time it needs more seeds (after each planting). This means that planting 50 plants requires 50 trips to and from the seed bin. Compare this to watering which just requires one trip to and from the bay to water 50 plants. So in the case of seed storage I think we have a lot to gain in terms of efficiency from relocation to the gantry. Another plus is that seed bins do not need to be mounted and dismounted by the UTM, so there existence on the gantry would not affect the normal flow of mounting/dismounting tools or tool bay orientation! (cc: @Gabriel)
The tools would need to be modified so that they can be mounted and dismounted along the y-axis.
So the tool storage latching is not 100 percent round? I though the tools were designed so storage could be on any side of bed by rough looking…
I think gantry storage makes a lot more sense for seed storage, which is something we’ve been wanting to improve.
A tool to pick up seed container and move it into a gantry storage. So the seed container pick up tool could pick up two seed containers and put it in the tool storage on each side of the gantry before swapping to what ever is being used to insert seeds. Save a lot of trips.
It was just me looking at this I could see reasons not to to complex gantry storage. But doing absolutely none is just bringing more overhead than it worth…
Watering and Weed bashing could technically be done in one pass. Lot of your passes currently will be designed to the limitation of only have 1 tool. It does not take much tool storage to reduce number of operations.
Also something to consider is that if there was one space on each side of the gantry, then FarmBot could use at most two tools (not three) while away from the main bay, because it always needs an open space to dismount a tool before grabbing a different one.
You have got the maths wrong here. It depends if you have tool that is designed to take stacked addons or not. Like with your seeder you could have a dibbler planter cover that could be picked up and held by the vaccum then used to put the holes in the ground then put it back to the side before picking up the seeds and dropping into the hole from height. This would be to prevent the vacuum section pulling in dirt when air is leaking past seed. Yes the a dibbler cover for the vacuum planter will only consume 1 storage position to give two tools with the vaccum planter attached reason you would not be removing the vaccum planter when you put on the dibbler cover. A storage on each side is 2 tools max if you are changing tools on the main head. 3 tools if the parts on each side of the gantry work as individual addons to the tool currently in use. Gantry Storage of 2 there is between 2 to 3 tools to use based on the configuration of the tools.
Stackable tools is something you have to consider when you have limited gantry storage.
Yes a tool item that can pick up something like a seed container and move it to the gantry storage would be a good thing. It would not be hard to modify a seed container to be holding a dibbler addon for the vacuum seeder.
Rory Aronson, I’m a big fan of adding the seed bin on the gantry riser. That was my first thought when I saw the video.
I agree having the tool storage on the bed makes perfect sense since the added weight on the gantry affects the motor requirements. Additionally, you will program the tasks to execute aligned with the tool. Watering=water tool; seeding=seeder tool, etc, and changing tools in the middle of these tasks is not truly necessary.
However, while doing a single task (seeding), it seems to make a lot of sense to have a light weight plastic bin travel with the gantry to make travel time more efficient. Travel time to Y home is better than all the way home to the bin
If you think how seeding is done in small gardens, a person carries the seeds with them as they perform the task. They do not leave the seed package on the bench and walk back and forth for each planting.
You’re right about the add-on/stackable tools math. At this time though we are not going to be making an tools like that. It is challenging enough to design the UTM and tools to mate with each other in a reliable manner that allows easy mounting and dismounting, trying to add a second layer of tool is a little out of our scope right now, but definitely possible!
Weight is a reason to limit to 2 storage positions on gantry. Even in large gardens seeds are brought out used then put back away. Main reason to keep seeds good most have to go into cold storage of some form.
I hope you will look at a tool to allow seed containers to be picked up from edge of bed and moved to gantry and put back. This is mostly allowing que. Something else on the current seed containers there does not look to be a place to locate a barcode or anything else that camera or tool could id what the container is. Current design seams to be put container in X location and hope it right.
Once starting to store tools in different places if device has been reset by power the system needs to be able to identify what it has where. Even so being able to properly id this stuff helps without gantry storage. Human puts seed containers in backwards and they told machine the right seed container numbers everything will be ok if machine can identify them. Just thought I better mention this if you are going to be redesigning.
Hey Rory, how did those CAD prototypes go I’m also keen for a simple seed storage system on the gantry. Are you guys going to release add ons like that?
I haven’t had time to work on this. You can do so though by copying the Onshape workspace and giving it a go. We do plan to release simple add-ons eventually, there are just so many other things right now that are higher priority for v1.3 hardware.
