I have both delta & cartesian machines and I generally prefer using deltas.
I use a suspended extruder in both of my deltas and that pretty much entirely mitigates the bowden issue with tall delta machines. Suspended extruder + running 6-9k acceleration + 450mm/s+ travel moves = sub 1mm retraction and 0 blobbing.
Deltas also have the not often spoke about issue of the “moire” pattern. If you’re not running very high microsteps, your prints will develop a pattern on the walls that isn’t a big deal, but bothers some people like myself. I fixed this by moving to 256 interpolated microstep TMC2100 silent stepstick drivers.
I will definitely say that deltas require much more build precision. If you build it sloppy, you’ll forever be getting mediocre prints and trying to correct for bad hardware with software offsets.
IMO if you just want a machine that works and gets pretty good quality / easy dimensional accuracy, go with a cartesian.
If you’re big on speed/accuracy/super smooth walls / crazy high acceleration, go with a delta.
Build accuracy can be said the same about cartesian. If a cartesian is not accurately built, it won’t produce prints as good.
If we’re talking about assembly accuracy of the printer itself, you have to differentiate between “pro” and “reprap”. Pro being it’s made out of machined, cnc’d, stamped, or whatever (fit for retail).
I still don’t see how Deltas are supposed to be on a different playing field. Cartesians may be more straight forward so that a new comer doesn’t have to tilt his head sideways to figure out how it works, but then again we have auto-leveling nowadays.
Processing power of the board is irrelevant I think. We’re talking about the consumer’s approach (unless the debate is of a hobbyist’s approach). No new 3D Printer user will really care about what processor it uses… when was the last time someone had asked this? They care if the machine performs as intended. Again, differentiate wether we’re talking about a hobbyist or a regular consumer user here.
At the end of the day, I think if you’re selling a machine that works, just as good as a cartesian, then the user won’t care. It’ll just come down to price at that point as being the deciding factor.
I really fail to see how a delta or a cartesian can be better or worse. At the end of the day it depends on what company makes the machine and how good they are at making their products. This ranges from small time mom and pop shops that print out the components of their printers to more professional companies that have supply chain management and make their parts with high tolerances using material designed for retail.
Shai, build issues on a cartesian can often be calibrated out or easily adjusted. If your delta isn’t exactly 60 deg on each corner, it’ll be inconsistent across many points on the bed and software compensation just swaps leveling issues for other distortions.
Obviously any printer will benefit from being built with tighter tolerances, but with a delta it isn’t optional if you want even remotely flat movement across the bed. Definitely less of an issue with prefabricated kits/etc.
I use deltas simply because I can get away with much higher travel speeds / acceleration and still have perfect looking walls.
Moving Y causes issues with constantly cooling the part from movement, increasing Y weight as the part gets larger, and also having to move a heavy bed is often a source of slop / limits accel. CoreXY doesn’t have this problem, but definitely increases the complexity of the build.
I’ve honestly never been able to get my non coreXY cartesians to make parts that look as polished as my delta. Doesn’t mean it can’t be done, but I have yet to see a moving Y printer make better lookin stuffs.
@Shai_Schechter
My statement is that 1-1 if you give the 2 people the exact same bom (and that bom is optimized for a delta) the same build volume, print speed, and general resolution will be obtained. The delta will actually be harder to build an calibrate.
There isnt a “better” just a different approach. Personally i don’t think that deltas are worth the hassle (i cannot see a benefit to this movement structure), but i would be lying if i told you that the movements are not captivating. Deltas are art, but you pay for the form in ease of use.
One thing I have observed with my delta is that it has less overall movement from side to side on a table. It doesn’t shake the table from side to side like my Cartesian does since most of the weight moves up and down rather than side to side
@Camerin_hahn deltas do impart far less force on the frame itself in lateral directions. You can tone it down on a cartesian by reducing jerk/changing junction deviation, but @Abc_Def made a correct statement.
@John-Paul_Hopman excellent advice cartesians are the simplest to get going, but delta/corexy has some big advantages.
I don’t think it’s fair to say deltas impart less force on the frame or table. Newton’s laws still fully apply. Mass*acceleration=force, period, end of story. If anything, deltas may experience MORE total frame reaction load since the carriages see both inertial-push/pull loads to accelerate the mass AND constraint-torsion loads to keep the end-effector parallel to the build table against the moments imposed by the arms not acting through the center of mass.
To the extent deltas are USUALLY made lighter than Cartesian printers, sure, you can accelerate the delta faster. But strap a big-ass Wade’s extruder on a typical delta (if it even fits…) and it’ll wallow like a drunken hippo. It’s not quite fair to compare a heavy Y-bed i3 to a precision-fabbed ultra-light mag-joint Delta. Let’s compare apples to apples. A Bowden parallel Cartesian printer like an Ultimaker gantry can perform at just as high speeds and accelerations as an equivalent-quality delta, except the Cartesian printer doesn’t need a probe and auto-calibration to work right.
@Mike_C
Yes there do impart less lateral force, but they exert more vertical force… this converters to rotational moments, which induces lateral movement.
There is no silver bullet.
@Justin_Edwards I think that is the key, The semantics of delta Cartesian are irrelevant. A bad cartesian printer will make bad prints, a bad delta will make bad prints. Both can produce good prints if properly tuned.One is not inherently better.
@Camerin_hahn In practice, regardless of the physics at play, deltas seem to get superior repeatability – definitely visible in the walls of the parts. I haven’t really seen any non commercial cartesian 3dps that have 0 visible banding, but i have definitely seen that in DiY deltas.
Our new Tinyg board does eliminate resonance due to jerk… When it comes to market, both Cartesian and deltas will be able to eliminate resonance and any wobble. Our plus can go quite fast with no shake or wobble w the Tinyg board. The acceleration is 7th order or some such voodoo…, it works. And ooze is almost completely eliminated due to true acceleration in the extruder (no retracts). Maybe this will level the playing field a bit. It’s an ARM processor. Although I don’t know how high of a priority delta is w alden, the community will figure it out if nothing else. Then we should have apples to apples for comparison. I totally agree with the comment: proof is in the results!
I have two and half delta printers and Iove just the way they move. It prints well for my tastes but Deltas are really all I know. I would like to get a cart. printer sometime in the near future. From what I see you can get either printer to print anything with the right mods. With a 1000 budget I’d get both kit printers and have fun with them.
cartesians are the simplest to get going, but delta/corexy has some big advantages.