In my research of a sub $4000 CNC Router setup, I was initially dead-set on a Shapeoko Pro XXL. However the XXL has continued to be out of stock, and in the meantime I noticed a new model that looked interesting- the Onefinity Woodworker.
This machine retails for $2000, has the same carve volume of the Shapeoko Pro XXL, has ballscrews on all axis (including HDZ, a $450 Shapeoko upgrade), a built in independant controller, and touch display. There is no built plate included, but that is somewhat a minimal expense and one can use the CNC itself to carve channels for aluminum extrusions. There is also a sub <$50 joystick controller, which seems very convenient. The ballscrews are the most attractive aspect; no belt stretch, and just squirt some lube in them every so often for maintenance.
However, the Onefinity machine and company is pretty new: it does not seem to have nearly as well developed software compared to Carbide Motion and Carbide Create, nor nearly as large of a user base. The open source controller has basic integrated CAM as well as easy dropdown menus to use a PWM spindle, and loads of inputs/outputs.
This review by Myers Workshop is extremely in depth, and appears at first glance to be unbiased since both machines were provided to him for free (By Carbide3D and Onefinity).
I know there are a ton of advanced hobbyists as well as Carbide3D representatives here, has anyone tried both machines or have any additional information that could help make this decision?
Thanks. If this topic is unacceptable on this forum feel free to remove this thread- I am just looking for purchase advice and could also try on Reddit. I am not sure if I am missing something, because besides the risk of investing into a new companyās product, the hardware seems more appealing and at a lower cost. Of course I would just spend the difference in endmills
In this class of machine ballscrews usually donāt equate to more cutting force. The torque of the stepper is the limiting factor and with properly tensioned belts on the SO3 and Pro, the stepper skips before the belt. I think many people attribute any skip as the beltā¦ when in my personal experience, that is not the case. So the ballscrew on the same stepper really doesnāt present an advantage. And maintaince is more than oil. The nuts have to be checked for proper pre-load as a maintenance item and if theyāre cheap, theyāll likely have more backlash. There are already people reporting skipping on the OF, and they even had to release new movement settings to keep it from skipping steps on unloaded rapid movements. Iāve seen people replacing the steppers with higher torque ones to keep from skipping steps.
I solve the controller issue with a touch screen tabletā¦$220. Carbide Motion can use the same type of game control pad. Theyāre like $20.
The bedā¦ in my opinion that one factor is WAY more important than people seem to attribute it. It makes no difference if the rest of the machine is made of unobtanium, if itās on a weak base, itās all for naught. The base of the Pro is well engineered and would take more than MDF and t-tracks to match. You could do a full torsion box base for the OF, then add the t-tracks and spoil slats. But even then I still donāt think theyād be apples-to-apples with the SO3 Pro ahead. I have a whole extrusion bed in my Original Model XXL (not Pro) and that alone would be $500-750 to replicate on the OF as youād need more frame with no existing one for the base.
The Shapeoko controllers can use PWM for spindles.
Iām not intending to bash the OF. Just adding more perspective to the thought process.
A solid, square, rigid, non-vibrating base for the machine to support the mechanism and the workpiece, provide workholding etc. is a good deal more complex and expensive than it at first appears.
Having to engineer something from scratch and hope it matches up well with the machine is not something Iād want to do. I can see why youād do that in a product design mode to keep your retail price below a target but, no, itās not a cheap and easy job to make up for the missing parts.
I have looked at the Onefinity machine a bunch and here are my thoughts:
Hollow steel tube is not as rigid as the extrusions that are used on the standard Shapeoko or Shapeoko Pro.
I like the fact that it doesnāt have any V-wheels when compared to the standard Shapeoko.
Ball screws are overkill for this class of machine. They require more power to drive and donāt add any real value over belts when the rest of the machine isnāt rigid enough to take advantage of them. The location of them on the Onefinity will lead to a large amount of maintenance.
A good base contributes a significant amount to the rigidity of the machine. The lack of one on the Onefinity machine meant that it was something I would need to make before actually getting to use it. An MDF base does not equate to an extruded base.
While I am familiar with the people who started Onefinity because they also made the Suckit Dust Boot for the Shapeoko, I am wary of using their built in controller and software vs Carbide 3Dās because of the time they have had to work on it.
Overall, I think the Onefinity machine is an interesting first attempt at a machine that has a few flaws. I determined that I would rather have the Shapeoko Pro when I upgrade from my standard. I will be paying attention to Onefinity in the future though.
Those are interesting points, but I have some questions about the particulars and how it matches with what I read online and on Carbide3Dās site.
The point of the $450 Shapeoko HDZ upgrade is the ballscrew: āDesigned around a premium ball screw and linear rails it will allow your Z to glide up and down with ease ā even with a weighty spindle.ā
That implies that ball screws give more torque and/or allows smoother movement of the axis. It uses the same stepper motor- so per Carbideās own listings, ballscrews improve functionality and weight capacity using the same steppers. On this forum I see a reference to the HDZ having <.005mm backlash too. There is no reference for checking pre-load, and according to Carbide3D, āThe HDZ is pretty much maintenance freeā.
The bed is a significant shortcoming, I agree. But as you said, if you have some DIY or general engineering experience you can replicate it with a torsion box or even with similar aluminum extrusions for around $500 (Potentially even stronger, if you can find inexpensive extrusions locally). Or, you can just consider that as part of the table-cost itself and have a 1/2" steel top or something
I think the biggest risk is the controller and investing into a new company. No long term reviews, nor proof that they will be around in 2-4 years. While Carbide3D is here to stay, provides upgrade paths, and has a large community.
@nwallace
One thing to note is that the OF us using another companyās control system. So it is more figured out than the time OF has into it, and by a dedicated company. I also echo that Iāve had great interactions with the owners when they were Suckit. I have nothing against them and wish them success.
@Thinkiverse
Iāve actually talked with @Luke a lot about the nerdy minutiae of components, ballscrews included. Ballscrews can provide some āleverageā help, but in practice on the OF, it doesnāt seem to equate to an improvement in material removal rates from what I have read. Skipping is still an issue with incorrect feeds and speeds, and I havenāt seen any OF machines running higher material removal rates than an Original or Pro Shapeoko have been running for a while on the hands of knowledgable users. Theoretical and practice havenāt caught up on the OF. But that could also be OF users generally being newer to the machine than most SO3 owners. There is the possibility the gap is knowledge, and it will close over time.
One reason the HDZ is pretty pricy is the quality of ballscrew and nut used. In my conversations with Luke, it is much cheaper to buy nuts that require maintenance, or at least verification. Units with tighter tolerances and QC (what C3D uses on the HDZ) require much less verification. This also explains the lower backlash specs for the HDZ. Is been fun and educational nerding out with him on that, hearing about his research, and learning.
Donāt know if I would completely agree on thatā¦ consider this, whatever machine one chooses, you will need a table or surface to āplaceā the machine on. I have not seen one example of the shapeoko placed on the floor to use it. So, the workbench or table you provide for the shapeoko or any machine like this can be the work bed as well, with an added spoilboard. Only difference here is instead of placing the machine on, you would mount the machine toā¦ sort of like I did with my shapeoko, threw the workbed slats away along with the supplied MDF, mounted the machine to the custom built table and built up a work bed on top.
@Thinkiverse
Enough 20mm thick Misumi extrusion for my Original XXL bed was just under $400 with some drop in tnuts. But I had some 2020 sitting around for under the bed. Without that extra support in the opposite axis, it would have been compete crap. And that was short of the width required to reach the end supports of the OFā¦ so ~$400 gets you a ~3/4" thick bed with no support under it. A torsion box would be great, but Iām just saying you have to factor that cost in, and include your time in the equation. A properly engineered torsion box at todayās wood prices (good lord todayās wood prices!) and your hourly wage has a cost. And then is the result as good as fully machined extrusions? I canāt answer that for you, but again, a thought exercise and something to consider.
Iāve gone thru the upgrade path on an Original XXL, and now have a Pro. The differences are night and day. The fully tied-in base (on the XXL Pro, the base extrusions that go in the X are tied into the Y rails at 4 points) and linear rail system really are that well engineered. They match the whole system and uses.
Disclaimer: I started and admin the primary FB group. C3D has seen the value in that and sent me my Pro at no cost, and no obligation. The XXL I bought 4 years ago. I mention that so you can judge if I have any bias and Iām an up-front person.
You have a very valid point with the Original XXL. The Pro is a different animal base wise, and would work well on any semi level surface. You could even use a few 2x4s on the floor (arranged front to back lengthwise) without a surface and still have a base sturdy enough for the uses. But the sturdier the foundation the better, and the Pro amplifies whatever you put it on rather than being completely reliant upon it.
To interject - whilst we did send Ben a machine it was quite a long time ago. I could confidently say prior to 3-4 years worth of improvements - on a stock shapeoko - not just the pro.
The āShapeoko 3ā has been around for a long time now. Since I had my first one in 2017 I thinkā¦ itās had a huge number of changes/upgrades and enhancements. Looking at the few photos he keeps up it looks like a 2016/2017 model but I canāt be sure.
I recently assembled my Pro XL the other week. I did it in one hour 30 without instructions. I might be a little over qualified but the the base probably took 30-40 minutes, and drag chains another 30. The OF has neither. Personally a machine without a solid bed is a no-no for me. I know they all go on a table etc but I would think most will agree steel or aluminium is stronger than a MFD table, and less prone to environmental changes.
Yes Iām aware he stands on his machine and no that doesnāt teach us anything about a machine.
If you were just talking pro, you also get a free work holding kit, dust boot, bitsetter, a baseā¦
If itās ball screws you want make sure you know the quality, preload, and source of ball screws and nuts, these are huge factors in them ābeing betterā and from what I can tell they are straight out of china with ground ends. No idea if they held down with DACās. Iām happy to be corrected.
Proof really is in the pudding - look at what people are doing with the Shapokoās. Itās inspiring and ābeltsā sure aināt holding them back.
@Luke is just going to go ahead and belittle my standing on my Pro!
I actually did stand on the middle of the X rail of my Pro, but I added measurement, therefore SCIENCE! On my bench (kinda-semi-amateur torsion box and the composite Amazon bench legs) with me standing in the dead center of the X rail, thus the point most likely to flex, I measured 0.020" of flex. I got about the half that in deflection standing in the middle of the bed. I weigh around 180 lbs for reference.
The OF standing test was standing near the X rail ends and had zero measurementā¦which means no SCIENCE!
I did not know his Shapeoko 3 was that old- when I watched the video I assumed it was the most recent model. I canāt find any information available on the brand of the Onefinity ballscrews, though its not like either company provides branding of their rail systems, motors, etc. To me it just assumed that the equipment is ātypicalā for production; the least expensive available that could maintain sufficient accuracy and QC.
I am super impressed with the videos NYCCNC has put out on aluminum machining on the standard Shapeoko 3, which in itself is significantly more than I would attempt on the Pro. I am more interested in plastics and wood, and only imagine attempting aluminum for novelty sake.
This may just be me overthinking while waiting to order the XXL PRO, as I can only stalk the page so often before going off the ārailsā, so to speak . One of the most valuable parts of sticking with Carbide3D is the community and endless resources, from Carbide Create and Carbide Motion to ProvenCut (Hopefully they get Pro soon??) and Winstonās Videos/ this forum.
You have to ācatchā a Pro batch going up for sale. Hopefully that is resolved soonā¦ but right now you gotta be quick on the draw. Hit the email notification button and have your email send you a notification. Iām only being slightly facetious. As soon as a batch goes up, itās spoken for in less than an hour. I know theyāre working on production bottlenecks as hard as possible, but thatās currently the reality.
The converse is that the OF is āfor saleā but I think the current wait is 4 monthsā¦I could be wrong. But theyāre not shipping upon order.
So I think the time from today that you could get one is the same in reality, or at least not as different as it seems.
Yeah @SLCJedi I am waiting both on email notifications and checking the site probably 8x a day in case the notification is delayed. Already have a cart goingā¦
I was also weighing a Onefinity vs Shapeoko Pro as well, after having just sold my S3 XL. I ended up getting an XL Pro.
Another factor to consider is the community support, which has been fantastic for Carbide3D. It seems like every week I find another old post by @Julien and follow him down the same rabbit hole he found. Seriously, his contribution to this community alone would have been enough for me to choose the Shapeoko (unless he also has a Onefinity and is in that community as well! :p).
Fascinating threadāIāve seen the Meyer video and Iāve been considering the OneFinity. I donāt think I need to recount the claimed advantages, but looking at another thread up on the board tonight about a guy pulling his hair outāI think āDumpster Fireā was in the titleāIāve been there with my Z axis, and with tramming, and a few other minor things as well that still seem to mess up a job. The idea that a machine that runs 100% on ball screws definitely has a āCalgon, take me awayā appeal.
I had been budgeting a few upgradeāHDZ, limit switches, Bitsetterāwhen the OneFinity came along, and it seemed like I could score a slew of upgrades in one fell swoop, with a budget that compare favorably to upgrading the XXL, assuming I could get decent price for my current machine.
All that said, I donāt really even need the ShapeOko upgrades. Iām not making money with my machine, and donāt get to spend nearly the time on it that I want. I ran a fairly simply job over the weekend, and I was just glad it all went well. Iāve got some of the minor-but-frustrating things figured out over the years (Iām looking at you, pulley set screws) and maybe the answer for me is to wait. My current rig still works. Both the Pro and OF are still months away in delivery. Perhaps the prudent thing for me right now is to sit tight, enjoy the machine I have, and instead of spending my disposable income, dive into it and swim around in it like Scrooge McDuck.
@Gray, I followed that thread. He simply had a loose set screwā¦ nothing to do with a belt specifically. You could have that same issue with the coupler that attaches the steppers to the ballscrews on the OF.
I suppose the point is that itās very easy to look at the flaws of the ShapeOko (and hey, no oneās perfect) that we have come to know through extended familiarity, and look at a newcomer and say, āWow! Thatās awesome!ā I think Meyer even goofs on that meme of the guy walking with his girlfriend looking at the girl walking the other way.
I donāt have a OF machine, but the reality no doubt is that the machine comes with itās own set of flaws of quirks and foibles, some of which the team and user base have not yet had time to work out (by which I mean for ShapeOko, Will Adams).
So yes, Screws, not belts and v-wheels! All around! No computer needed! In fact, I am sure the OF is a fine machine, and has no doubt benefitted by standing on the shoulder of those who came before.
For meāthis thread has been great, because I really havenāt seen some good honest discussion of both machines by people who know them. From a decision making standpointāI can do nothing, for free, and still have a perfectly good ShapeOko XXL. If for some reason I decided I HAD to have OF (or even SO Pro), by the time all is said and done, Iām at least $1000 out of pocket after I sell my current machine, if not moreā¦ and what can I really do that I canāt do right now?
And that was short of the width required to reach the end supports of the OFā¦ so ~$400 gets you a ~3/4" thick bed with no support under it. A torsion box would be great, but Iām just saying you have to factor that cost in, and include your time in the equation. A properly engineered torsion box at todayās wood prices (good lord todayās wood prices!) and your hourly wage has a cost. And then is the result as good as fully machined extrusions? I canāt answer that for you, but again, a thought exercise and something to consider.