Building a "smart" table saw?
Hello! I've got a few ideas that I'm having trouble figuring out how to implement. I'm an absolute novice in this area, so please forgive my ignorance.
What I want to do is build a table saw with a system that allows one to input a measurement, and the system will then use stepper motors to move the cut guide to the desired measurement and lock it in place. If possible I also want to be able to adjust the height and angle of the blade by repurposing other axes. Using the right gear ratios on these motors should give them the necessary torque to do this while retaining a high degree of accurracy. I'd like this to have an accuracy of .1mm, but can work with up to .5mm.
Basically my intention is to repurpose the guts of a home made CNC (using a Raspberry Pi/Arduino) to make a "smart" table saw. I believe it should simply be a matter of the right programming for this purpose. I have a 3D model of the table saw itself done at this point and can make basically any tweaks necessary to the design.
My questions are this:
Is this acheivable, in theory?
Does anyone know if someone else out there has done anything similar to this (I have yet to find anyone)?
Re: Building a "smart" table saw?
So you wouldn't be starting from an existing table saw?
It should be doable, most likely it's just the need isn't there for it to be a common thing.
I've definitely seen saw blades on huge 5+ axis routers cutting large panels, but that's something a bit different.
Setting the blade angle would be the hardest part I'd think. Getting the pivot in the right spot.. What do you plan to use for the blade arbor?
Re: Building a "smart" table saw?
Our sliding table saw has motors for blade lift and tilt, with a digital display for blade angle. It just uses two way switches, but same principal.
1 Attachment(s)
Re: Building a "smart" table saw?
@skrubol - Yes, that is correct. I purchased plans from Paoson (highly recommend that site!) and have made modifications to sizing and a few other details. My inspiration for this was indeed the industrial applications I saw using similar tech. I want to build a workshop that can manufacture things with a high degree of accuracy more quickly, and this just seemed like the logical conclusion. I will be using a high end 10" circular saw as my blade arbor.
@wizard - The safety concern is highly valid. Since the feeding of the work would still be manual, I will have an emergency cutoff in place first and foremost. I will also be making a control pad, rather than using a touch screen, and will make sure that only the "soft stop" function is available while the relay for the saw is active, so the controls cannot be activated by mistake while working. After further consideration, I'm not going for automated angle adjustment, as that is of minimal benefit in my case anyway, thanks for that! I'll post a picture of the table at the bottom, happy to show off my (at this point simple) modifications (all design credit goes to Paoson, of course).
@ger21 - I'd love to see how that is designed, if possible!
http://www.cnczone.com/forums/attach...d=375560&stc=1
So this is the current design. Thus far I've just resized it to 3'x6', and it does have a router built in, but that will likely be left manual.
Re: Building a "smart" table saw?
Quote:
I believe it should simply be a matter of the right programming for this purpose
Yep, But that pretty much describes any project from CNC to self-driving cars.. As they say...the devil is in the details.
Re: Building a "smart" table saw?
I do a lot of woodworking and if you are serious about this I would recommend buying a saw and starting from there, half of the things you are going to be building are done, and engineered for you. Adding tilt would be simple (compared to building your own), you have a solid base to mount to. And in my area (NY) you can pick up a decent contractor saw for $150 or less off of Craigslist. I checked out the plans, and I think after all the time and effort you spend building the saw you will be very disappointed. The design can't hold the weight of a properly sized motor for ripping wood. you need a full 1.5 hp for that.
But I guess the question I would ask is why you want to do this? I see very little advantage (actually more disadvantages) to doing this vs building a traditional cnc router table. The biggest advantage for something like this is the speed it can do repeated cuts at, such as ripping several thousand board feet to the same width, in which case you would still be better off with an autofeeder or a gang saw.
Room can't bee an issue because a CNC takes less space than a tablesaw (for an 8' board you need 16' minimum with a tablesaw, and can get away with 10' or less with the router table)
I have a friend with a cabinet shop and he has a sliding tablesaw, that is fully powered, Autofeed, the fences are moved to position, tilt is powered. That saw is basically used to store stuff on. He has 2 guys on the CNC feeding wood. His CNC is a large commercial unit, and most home owners don't have the room, need or budget for a machine like that. But he uses it for cutting boards out of solid wood because of it's advantages.
I just wonder what you want the tablesaw to do that a router table can't. The only thing you mention is the angle cut that most cnc's aren't capable of. There are a bunch of ways around this limitation, I saw somewhere where a builder put a manually adjustable head that rotated side to side for just that reason. But setup and tool selection can go a long way for this.
Re: Building a "smart" table saw?
Quote:
I just wonder what you want the tablesaw to do that a router table can't
There are a lot of operations where a saw is a far better option than a CNC Router. Remember that CNC Routers are just tools, and are not always the best tool for the job.
Re: Building a "smart" table saw?
The beam saw at our shop can overwhelm the two CNC easily.
For simple cabinet making you can away with only a beam saw and it is way faster than a CNC.
Got to use the right tool for the job. ;)
Re: Building a "smart" table saw?
Quote:
For simple cabinet making you can away with only a beam saw and it is way faster than a CNC.
Unless you need holes in them. :)
A beam saw is great, if you cut a LOT of identical parts, where you can cut multiple sheets at a time.
When you are cutting one sheet at a time, and need a lot of holes drilled, nested cutting on a CNC router is a better option. IF you have a high end machine.
On our Morbidelli, we average about 6 minutes/sheet. An average sheet is 8-15 parts, and ~100 drilled holes
Re: Building a "smart" table saw?
I think the best statement is the right tool for the job. Since rmitt90 didn't specify what he planed to make, his idea may be the best tool for the job.
Just building it for the fun/challenge of building it would be enough justification for the time and expense to me.
On that note I still think starting with a saw would be the way to go, but if he really wanted to build his own I would look at this design https://woodgears.ca/reader/hector/tablesaw.html. I believe it's far better thought out and engineered. And would allow adaptation to CNC better and easier than the other one.
Re: Building a "smart" table saw?
2 Attachment(s)
Re: Building a "smart" table saw?
@ttalma - I might be ok with the idea of building around an existing table saw if I find one in my area, though I do really like the idea of my own engineering going into this. I'm building a workshop with my brother, and it's kind of a weekend thing where we just keep building more projects, so the table saw was an easy fit there. The plans I have based this off of were really chosen because of their simplicity, allowing me to reshape and add to the design as needed. I've taken note of your motor recommendation and will switch to a 1.5HP belt driven arbor instead, and will update this thread once that redesign is complete.
My thought process behind implementing this into a table saw, rather than building a large CNC, is 1. the space required for the tool itself and 2. the projects I'm going to be building can be done pretty easily on a table saw. I can design what I want in sketchup, reference the design on a computer we have in the garage, and set the saw to the desired measurements one after another, thus cutting out the time spent on stopping to measure each time.
As for the second reply, yup, building for the fun of it is also a large factor. I like the design that you linked, and will reference parts of it as well, thanks!
@ger21 - There are some interesting features in the design you linked, I'll look into that as well, thanks!
Here are a couple examples of the projects we'll be making, if you're curious:
http://www.cnczone.com/forums/attach...d=375740&stc=1
http://www.cnczone.com/forums/attach...d=375742&stc=1
Re: Building a "smart" table saw?
While I can't read the original posters mine, I can relate to one use that had me thinking about an automated saw. That was for making box joints which a table saw is ideally suited for. I haven't pursued it due to other pressing needs but there are many possibilities from automating a conventional jig to a ground up design. Perhaps we will hear from the original poster what his goals are.
Like many here I'd put emphasis on a CNC router first due to the versatility. Even so that is reflecting what I would like to have right now, box joints aren't the only use for an automated saw.
Re: Building a "smart" table saw?
The saw that Ger linked to would probably be a good place to start, they have worked out a bunch of the issues you'll need too work out. Saving time for the issues you need to work on. You can probably incorporate the automated motions into the design. But personally I would still want to start from a commercial saw. I think the saw you originally linked to does not have enough reinforcement in it. and you will be chasing accuracy problems do to the lack of stiffness in the design. You'll probably have enough problems to work out!
But it sounds like an interesting concept, post all of your work here on CNC zone.
Re: Building a "smart" table saw?
Around here it's not uncommon to see table saws go for under $1 per pound on CL. I'd listed a belt driven cast iron table saw with questionable motor a few years ago for $40 or $50 and didn't even get any inquiries or offers.
The biggest issue with starting with a commercial saw (especially an older one,) is that if you want to try to use the mechanisms in the saw for blade tilt and height you might have trouble with accuracy as well as needing more torque from your motors than a scratch or semi-scratch design would need.
Even if you only reuse the frame and table (and motor if the saw has a good one,) that would be quite an effort savings.
1 Attachment(s)
Re: Building a "smart" table saw?
Thanks for the feedback! @ttalma - yeah, accuracy is a very high priority for me, as I'm very picky (even with our current $60 saw I demand accuracy to within .1mm). There will be A LOT more reinforcement as time goes on, and I did purchase the plans that ger posted (thank you for that!). At this point I plan to integrate the power box pretty much as it is in the original design, including hand control, then will leave room to upgrade it to a more automated system later (need to learn to walk before I run). I'll probably use a Wixey digital readout for my distances for now. I like the locking system/design of the rip fence on the AW design as well, so I'll put my spin on that too.
@skrubol - I'm of the type that likes to get it done perfectly first and foremost, so what I really want to do is perfect my own design and do it right, rather than build around something that will likely not suit 100% of my needs going forward.
As another example of my planned projects, here is what really started all of this:
http://www.cnczone.com/forums/attach...d=375792&stc=1
I'm building this PC case from the ground up, and require very precise cuts. My work on this extended to designing a new media center, then of course building the tools to achieve my required level of precision. If all goes will in building this case, I may start building them and selling them in a few different form factors.
