Current RepRaper looking for new build

Hey guys,

I'm Rob, first time poster and current RepRap user. I've owned my MakerFarm 10" i3v for nearly a year now, but have been going at it alone and haven't had much interaction with the RepRap community up til now, so please excuse any ignorance or misinformation I may display. My wife and I are looking to build two more machines, but we're pretty overwhelmed by all of the different builds, styles etc out there. I was considering buying two more i3v 12"s from Colin, but honestly I haven't had the best interactions with him and would like to see what else is out there, and the recommendations of those here with a greater wealth of knowledge beyond my own.

Essentialy, I'm looking to build a solid cube-shaped model similar to the Haeckel, only with a very large print volume (around 400 x 400 x 400), with maximum rigidity and low vibration, high resolution and minimal maintenance. I'm not particular in terms of axis-style, be it CoreXY or XYZ, as long as it gets the job done and has a large print volume, since the wife and I print a lot of movie and cosplay props. Here's a rough list of what we're looking for:

* All-aluminum cube/rectangular V-extrusion design
* Heated bed with a volume of 400mm³ (+/- 50mm)
* Bulldog XL or belt-drive single extruder, classic style (non-Bowden), upgradeable to dual later, using 40mm hex hot end, 3mm nozzle
* Resolution of at least 40-50 microns
* Minimal printed parts
* Leadscrew design, or whatever provides the highest performance

Again, please excuse any ignorance I may have displayed above in my requisites. I'm open to any suggestions, but ultimately I want a very sturdy, rugged box-shaped design with a very large envelope. I've seen a lot of great designs all over the Web, but almost none of them have a complete parts list/BOM or meet the build volume specs we want. Any help would be greatly appreciated!

Having built both a Cartesian printer (Ormerod) and a delta (Mini Kossel), I'm very impressed by the speed, stability and lack of maintenance (no leadscrews etc. to lubricate) of the delta. In particular, on the Cartesian printer I have to re-level the bed from time to time, and when I move it. Not so with the delta, which has a fixed bed.

Another thing to bear in mind is that a Cartesian printer that moves the bed in the X and Y direction takes up a lot of bench space - especially with a 400mm bed. The delta needs additional height, but that's not a problem unless you want to put it under a cupboard or similar.

However, I don't think a non-Bowden extruder is practical on a delta because of the weight. Was there a particular reason for your choice of extruder?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

The materials for the raw materials and hardware would be about $1,000
Linear rails and high quality bearings.
Ball screws

The machine work to make the aluminum parts could easily run another $1,000

Bigger motors and drivers for the x & y. $200 z axis $200

Electronics $200

Total cost for "kit" $2,500 - $3,000.

Still interested?

I'd love to know how you came up with that crazy price lol. $200 for the motors? How do you figure, when NEMA17s are $15. The aluminum parts are easy, as I work for a machine shop. $100 tops. The hardware, I can't imagine going over a few hundred from OpenBuilds. Electronics seems about right.

All in all, I think your estimate of $3000 to be a "bit" off.

This is an interesting thread, and I'm sure you're going to get a wide spectrum of responses. I have built 3.5 cartesian printers, but I cannot comment on the delta bots since I have yet to build one. My first printer was a Prusa i2 that I also purchased from Colin at Makerfarm. I had a great experience in dealing with him, but it's also been a few years. Second, I self-sourced and built a Lulzbot Taz clone with a 300mm² heated bed. Third was a Smartrap with bowden extruder and no heated bed, and my current project is a Mendel Max 1.5 which also uses a 300mm² heated bed.

In my experience, I have found that the box style frame for a printer with such a large build volume is pretty flimsy when built from the standard 20x20mm extrusions. As a consequence, I print at 30mm/s with the Taz clone, but nearly double that with the Prusa i2. That lack of rigidity also caused me to revert to the traingular design of the Mendel Max 1.5 for my current project. With that, I would have a hard time recommending a box frame printer so large that moves the part in anything other than the Z-axis. The Core XY designs with the printbed that lowers throughout the print look very promising to me.

For my next project, which will likely be a Core XY design, I have been looking toward the Open Builds version of a Makerslide-type system and likely the 20x40mm or 20x60mm extrusions to make sure that it will have a rigid structure. I have also been looking to move away from RAMPS 1.4 to something built on a 32-bit architecture. This is probably even more important to consider if you end up building a delta, because it will require many more computations per second.

Your desired level of accuracy is certainly doable, but at what scale and what level of detail? Anything smaller than a 0.5mm nozzle will be painfully slow if you're printing 400mm³ objects. Of course you can get dimensionally accurate large parts from any size nozzle but, if you're looking to make huge parts that still offer .05mm resulotion on fine details, you may find that you're working against yourself unless you've got days or weeks to print each part.

Whatever you choose, it sounds like a great project. Please keep us up to date with your decisions and the build process!

Thanks so much for the reply. The build size in relation to stability issues from such a large volume is what I was concerned with, and you've given me a bit to think about. As someone mentioned above, I like the look/feel of the Deltas, but not sure if it would be the right fit for what we're trying to do, i.e. printing helmets, etc. The reason I was thinking classic as opposed to Bowden is due to my own experience with classic extruders, and having very few issues. What research I've done into Bowdens is limited, but not very positive from the feedback I've seen from others. I may be unfairly biased there lol. A coworker of mine is designing a very interesting CoreXY setup unlike any I've researched, but it's a ways off from v1.0. As far as print time is concerned, I'm used to 24-72 hour print runs for single items. I'm in no hurry lol.

Delta printers generally have circular print beds, so I would have thought they were ideal for printing fairly round objects like helmets. Less suitable for printing large square objects though.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

You said you wanted a more robust machine. You will have to increase the weight of the table by using heavier material, Bigger rails and bearings. That being the case, you will have to increase the size of the motor to a 23. The cost of a 23 and a driver is upwards of $100. I added a $1,000 for machine work not knowing that the machine work for you is "free". The aluminum 8020 extrusions are going to run $100 - $200. That doesn't include any brackets or hardware.

Here is a model of a frame and axis that I had in mind when I made the above posting. I am using 4 motors on the table. By adding a toggle switch in series to the z limit switch circuit I can deactivate the z axis. I can then run the leveling routine and adjust the motors independently to level the table. This eliminates the need for any leveling hardware, springs, etc.

I frankly desagree with the use of Nema 23 instead of Nema 17. The Nema 23 vibrates much more at low speed thant Nema 17. The Nema 17 will be pretty enough whatever the size, especialy with ballscrews. I'll be affraid to build a delta with custom size, because of the maths and calibration. the Reprap Lautr3k uses balscrew only. But I dunno if it's scalable to the required build area.

If I would build a printer with your needs, I'll go for a CoreXY printer. To me it ain't necessary to use ballscrews. You can reach the required resolution with belt driven 0.9° Nema 17 stepper motors. To get high precision ball screws in your dimensions, it will be very expensive. Through, ball screws are slow compared to belt drive. Ball screws are interesting when you need to hold a lot of torque, like on a CNC mill/lathe. With lasers or FDM printers, you definitely don't need it. Otherhand, ballscrews needs to be aligned with a lot of precision, or you'll get important wear soon. So if you realy want to use ballscrews, you should invest on expensive machinist measuring tools. A belt system is inexpensive and pretty easy to setup, especialy for the hobbyist. To be confident, invest into quality belts and pulleys. It'll be ways cheaper than 500mm precision ballscrews (>$400 each !). You could buy chinese ballscrews, but @ $120 each, you'll get only a comparable precision as a blet transmission.

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)