Diamond Hot End on a Delta

I've managed to graft a Diamond hot end on to my RepRapPro Mendel 3 and got it working quite well. I've been using colour mixing to create some quite nice looking vases but I'm limited to about 140mm in height. I know next to nothing about Delta printers and was kind of wondering about the practicalities of grafting a Diamond hot end onto one. It's quite a heavy beast as it takes 3 E3D heat sinks and needs a high efficiency 50mm fan as well. Then of course there would be 3 Bowden tubes trying to pull it in all directions as well as the practicalities of somehow mounting 3 extruders. Does all this mean that as an idea, it's a non starter?

It's common practice to fit the extruder drives on the towers, so you could fit one on each tower. With any luck, the sideways forces exerted by the Bowden tubes will mostly cancel out.

To handle the extra mass of the hot end, you might need to use thicker than normal carbon fibre tubes. 6mm tube is commonly used, and I use 7mm tube on my delta.

How is the Diamond normally mounted to the carriage assembly? It's hard to tell from the images on the Wiki, but it looks to me that you might be able to sandwich the effector between the fan and the fan duct, with a hole cut in the effector for the airflow.

I am interested in fitting a Diamond hot end to my delta too,.

---

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 question is what is the weight induced ?
The initial Kossel had an effector with J-Head weighting 40~50g all included (no fan), super light, which allow it to work with a relatively weak structure.
The Fisher effector weight is ~ 85g with one radial fan, with a relatively small machine, so average stiffness.
Larger deltas with multiples fans and magnets (which are quite heavy) may have effectors weighting ~200g

Heavier mobile mass need more stiff overall structure
With larger mass, you also need to reduce acceleration and you lose some of the advantages of a delta.
You may also experience structural resonances, which are quite difficult to chase.

Arms are easy to stiffen longitudinally without a lot of weight increase.
More stiff structure and carriage may be a bit more complex. Often using aluminum parts increase stiffness, but it also add mass, so, you are running on a bad path, but something like the Berrybot may be workable [www.youtube.com].

As another reference, you may have a look at Ultibot Kossel derivative, as they have the extruder on the effector, so an overall effector mass higher than other deltas.

I think this might be workable, but you shall choose or design a very stiff model, so, not a standard Kossel, and sure not a Rostock. A solution may be to build a Kossel with Aluminium corners and 20x20 extrusion, of standard size (not a XL), as the stiffness reduce a lot while you lengthen the extrusions (for a given load, beam bending displacement goes with the cube of the beam length)

A shortened Rostock max may also be an option.

As for the traction of the Bowden/cables, it is quite significant, with something like 500~600g vertically while effector is at most distant point from bowden start point (I know because it was triggering my sensor on the D-Box).
With 3 Bowden tubes, you may be capable to eliminate a big part of the horizontal load, which is the most important.

---

Pierre

- Safety [reprap.org]
- Embedded help system for Duet and RepRap Firmware [forums.reprap.org]
- Enclosed delta printers Lily [rouzeau.net] and Lily Big [rouzeau.net]
- OpenScad delta printer simulator [github.com]
- 3D printing on my site [www.rouzeau.net]

Quote
dc42
It's common practice to fit the extruder drives on the towers, so you could fit one on each tower. With any luck, the sideways forces exerted by the Bowden tubes will mostly cancel out.

To handle the extra mass of the hot end, you might need to use thicker than normal carbon fibre tubes. 6mm tube is commonly used, and I use 7mm tube on my delta.

How is the Diamond normally mounted to the carriage assembly? It's hard to tell from the images on the Wiki, but it looks to me that you might be able to sandwich the effector between the fan and the fan duct, with a hole cut in the effector for the airflow.

I am interested in fitting a Diamond hot end to my delta too,.

Hi Dave,

I've attached the diamond hot end assembly.scad file which you can also get from RepRap.Me's web site. The mount is basically a fan duct. You, screw the 3 E3D heat sinks into the brass cone and these kind of snap into the duct and are secured with cable ties. To actually attach it to the carriage is just a question of adding whatever flange arrangement is needed to the basic fan duct. I had a job understanding how they had written the open scad code because they seem to have used some strange (at least to me) origin and I'm not all that proficient with OpenScad. I just kept adding bits on until it looked about right and got there in the end by trial and error. The hardest part for me was adding a bracket to fit RepRapPro's bed probe which was fitted to the original cooling fan spacer. As I am using it for the X end stop, it needed to be in the correct position relative to the nozzle and the "tab" on the X carriage mount. In hind site, I should just have used a different X end stop and/or used your bed probe (had I known about it at the time). I guess if it's possible to have a 51mm square hole in middle of the effector, then mounting it should be a piece of cake with a simple flange added. Alternatively a 50 mm round hole and sandwich the effector between the fan and duct with long screws as you said.

It does need high airflow to cool it so if a 50mm hole isn't possible then I guess that makes it a no no.

Ian

Update - from the Wiki,

Quote..............

Weight of assembled hotend: 250 grams (nozzle, HeatSinks, cooling shield, fan, heater cartridge & thermistor)

...........End of Quote

So you may end up with an effector 350~400g depending arm attach type and fan installation, which is quite on the heavy side.
The Ultibot extruder weight ~190g and I have no data for their overall effector weight, so we are somewhere above that (don't forget bowden tubes do also have a weight). Workable, but with a fairly strong machine.

Quick and dirty trial, something like that (shall have bolted cups, to be printable without supports).
Top ring shall be reinforced to cope with moment created by tensioning wire (which is quite important), or set bottom bracing (with 3mm rods ?).

Offset 30mm Arm space = 84mm, balls 8mm (may be increased to 10mm) , tensioning wires (This is the data of my D-Box)
No magnets -> low weight

---

Pierre

- Safety [reprap.org]
- Embedded help system for Duet and RepRap Firmware [forums.reprap.org]
- Enclosed delta printers Lily [rouzeau.net] and Lily Big [rouzeau.net]
- OpenScad delta printer simulator [github.com]
- 3D printing on my site [www.rouzeau.net]

Found a video with A diamond on a Rostock max :
[edit] and flying extruders !

[www.youtube.com]

Edited 1 time(s). Last edit at 04/20/2016 12:38PM by PRZ.

Quote
PRZ
Found a video with A diamond on a Rostock max :
[edit] and flying extruders !

[www.youtube.com]

Good find! Looks like it might be doable then.

I've been thinking about this some more. My main requirement is for extra height over the Mendel design that I have now. One of the issues of course, is that I need to use Bowden type extruders with the Diamond hot end, so a tall Mendel would mean really long Bowden Tubes. As I understand it a Delta design gives extra height and also faster speed, and with flying extruders the Bowden tubes could be kept reasonably short (if I could live with these things flapping about like some prehistoric pterodactyl). However, much of the potential speed increase would be limited due to the mass of the Diamond hot end. So I'm thinking that a tall corexy design might be a better proposition. From the limited research I've done, most of the designs seem to be cubes but I can't see any reason why they could not be extended in the Z direction (with the usual caveats about stiffness and rigidity etc). A (tall) corexy design would give me the height I need and potentially some of the speed benefits of a Delta but because the hot end is static in the Z direction, it makes mounting extruders much easier and I could probably have shorter Bowden tubes than I have now. Does this seem reasonable or have I missed something?

Delta design and CoreXY share the objective to minimize mobile mass. Installing heavy head defeat a bit this purpose, which doesn't mean that will not work, but that you will loose one of the advantage of the design, so reducing speed/acceleration ability.
I am a partisan of structural boxes, which makes very stiff structures, either for Delta or CoreXY, but stiffness shall also be improved in the movement parts and carriage, which is not easy, particularly is using ABS, which is much more flexible than PLA.

The main point here I think is that as the hotend is moving in a plane, the Bowden tube of a CoreXY will be shorter than on a Delta.
The shortest possible Bowden on a Delta is for a 200mm diam bed approximately 480 mm and for a bed diam 300mm approximately 650mm. I have not designed Bowden on a coreXY, but a simple 2D drawing may help . On A coreXY, shortest Bowden will be with the Extruder(s) installed at the middle of the long side, with the start and end of the Bowden in the same plane.
With that, Bowden length may be near that : sqrt ((travel width)^2+(travel length/2)^2) * pi/2 (half circle on the diagonal between middle of the length and opposite corner)

As you have three Bowden, minimizing their length may be important, also I assume longer Bowden may give more trouble for mixing filaments.

---

Pierre

- Safety [reprap.org]
- Embedded help system for Duet and RepRap Firmware [forums.reprap.org]
- Enclosed delta printers Lily [rouzeau.net] and Lily Big [rouzeau.net]
- OpenScad delta printer simulator [github.com]
- 3D printing on my site [www.rouzeau.net]