New Cyclops nozzles > Old Cyclops nozzles

[626Pilot]

I had to order a new E3D Cyclops nozzle because my old one got hammered into the glass one too many times. I noticed that whereas the old nozzle flange screwed in flush with the heater block, the new nozzle flange stands off by about 1-2mm. Indeed, the threaded section is longer. It's more of a gap than you'd want to use on a v6 hot end by a good margin. At first, I was a little worried, but this new nozzle flows noticeably smoother! It seems to require significantly less extrusion force. I think this is E3D's "secret sauce" of keeping the nozzle flange separate from the heater block.

PLA flows through this hot end just like butter now. There is so little resistance that I think you could easily run them WITHOUT geared extruders. It's about the same back-force as a v6 nozzle, and perhaps even a little bit less!

[bubbasnow]

did you get from filastruder?

[626Pilot]

Yeah. They don't have them on the order form, so you have to send them an e-mail.

I'm also noticing that the extrusion temperature for PLA has gone down from 220 to 210.

[mhackney]

This could be a design tweak to lengthen the effective melt zone. Your extrusion temperature results appear to support that. I need a new nozzle too so thanks for the heads up. I had been emailing E3D back and forth since Jan about needing a replacement.

[bubbasnow]

can you snap a photo right quick?

[626Pilot]

Nozzles.jpg (19.4 KiB) Viewed 11990 times

[Eaglezsoar]

Thanks for posting this. I did not know that filastruder could sell the nozzle for the Cyclops. Thanks!

[mhackney]

I ordered 2 last night. They are only available in .4mm like the original. I'm looking forward to trying this.

[elmoret]

Tim here, from Filastruder.

I went ahead and added them to the site:

https://www.filastruder.com/products/chimera-cyclops" onclick="window.open(this.href);return false;

there’s now a “Cyclops Spare Nozzle” variant available. E3D only makes them in 0.4mm, though the request is in to have more sizes available.

[626Pilot]

Tim here, from Filastruder.

I went ahead and added them to the site:

https://www.filastruder.com/products/chimera-cyclops" onclick="window.open(this.href);return false;

there’s now a “Cyclops Spare Nozzle” variant available. E3D only makes them in 0.4mm, though the request is in to have more sizes available.

Yay!

Can you also get them to ship you some of these, but without the useless brass parts? The plastic grippy things wear out, but I have no idea what the brass fitting is good for if you already have a hot end.

[SanjayM]

Hey 626Pilot

There has indeed been a super slight revision! We're impressed that you even noticed.

With a few of the first Cyclopes out the door we had a couple of people that managed to strip the very short length of thread on the nozzle. At that time (around a year ago now! hardly feels it) we had also just found in research that using a 60° drill cone vs a 118° cone results in a significant drop in backpressure and generally improves printing. So we were transitioning to 60° drill cones on all the v6 stuff.

We added some additional threaded length to the Cyclops nozzles to address the stripping issues heard from testers and early customers. We also increased the depth of the hole in the Cyclops blocks to accommodate for this, but the difference is so slight that it doesn't really have any negative impact when the newer, longer nozzles are used in the older blocks - as you are experiencing! So there wasn't really a need to explicitly have a new revision.

While we were at the task of adding the length it we folded in the 60° drill cone from v6 over to the Cyclops nozzles. This is what I would ascribe your observed improvements in flow/pressure to, and we also saw a good boost in tested performance, it seems like the 60° taper makes a more significant difference in Cyclops than elsewhere, almost certainly because pressure plays a much more significant role in Cyclops than other hotends. I wouldn't say that adding the length has really made any effective change in "melt zone" as the difference is really very small. Also, Cyclops has miles of length for the melt to get up to temperature before it gets near the nozzle, and it's not just length - the filament channels in Cyclops being smaller in diameter than a v6 etc results in much more effective heat transfer to/through the melt.

We have both smaller and larger diameter Cyclops nozzles on hand, but they are not yet released as products as they remain untested and I'm quite keen that we give them a proper testing before release as Cyclops is a funny beast and playing around with the pressure changes involved with different nozzle diameters might have some unintended consequences. However if anyone wants to play with some other diameters we'd be happy to throw some into an order free of charge on a "no promises" basis and hopefully you would be able to provide some feedback. Just get in touch through the contact form and we'll add them to any order you might be placing.

Cheers,
Sanjay

[Renha]

Hi SanjayM! Are you planning to release some kind of Diamond Hotend competitor? I.e. ~4 in -> 1 out?
I have bought Kraken, and i never was able to align nozzles correctly (delta imperfectness). Love water cooling.

[bot]

SanjayM, does that mean that new v6 heatbreaks will be drilled with a 60 degree point, or will that still be 112 degrees?

[626Pilot]

The ultimate hot end, I think, would support up to five filament inputs with a nozzle dead center, and come with set screws you could put into any filament channels you aren't using. This would support true full-color (WCMYK) printing. This guy is doing pseudo-full-color printing with a Diamond, but without white and black, there are many colors that can't be achieved.

A second, offset v6 heater would be used for support material, which often has to be at a different temperature. They could all be cooled using the same water jacket. A water-cooled 5-in, 1-out + 1-in, 1-out would look sort of weird and futuristic. There are some drawings of a 6-in, 1-out extruder somewhere on the RepRap Wiki, from 2012 or earlier I believe, and if no one patented it by then, they can't.

[gestalt73]

Sweet! Just ordered from filastruder. I haven't hooked my Cyclops back up since I moved everything around.

I'm *very* interested in the observations on extrusion force.
(dreams of effortless prints with breakaway support material...)

[Xenocrates]

In all honesty pilot, I think a 6 in 1 out nozzle, plus a pair of 1-1 nozzles would be better, as it would let you potentially add clear/translucent areas (Although the chemistry of some colorants may not be good for that), as well as having a pair of support/engineering filaments, say one for HIPS and one for ninjaflex/carbon/ or other filaments with very different mechanical properties that you would want to avoid having bleed into them or waste of them. Arrange it such that you have the heater cartridges in between pairs of filament channels (3 cartridges in total), with potentially 4(or 7) points for thermistors/thermocouples/pt100s, one above the mixing chamber (As close as practical), and one each between the filament channels as well (If you use thermistors or other sensors threaded the same as your grub screws for the heaters, you could use those holes and have the heater grub screws be short, far in, and protected from crud that way, otherwise you may want them in pockets inboard of the filament channels (With the heaters themselves outboard), or in the sections without heaters (Although you would lose the ability to potentially adjust color pairs semi-independently from the rest of the hotend).) The set screws to close off channels you aren't using would be much harder though. Most of those I am familiar with are not designed to be liquid tight (And E3D doesn't depend on them for seal, they use ball bearings pressed into place with the set screws, which would be hard to undo). They would also foul very rapidly. I have issues with threads on the nozzle and the two set screws fouling, and they don't have filament passing directly through. There aren't a lot of good ways of sealing that that I can think of. Ball valves would not like the heat much, would be bulky, and would potentially jam over time. Butterfly valves would probably be too much of an impediment to standard filament feed, and might have issues with the temperature making most sealing materials unsuitable. A small knife valve would be ideal, but would add a great deal of complexity and bulk to the nozzle. Honestly, I think it's a nice idea that would be next to impossible to implement well. The best you could really hope for would be a blanking plug that goes where the heatbreak would go (That may have been what you meant, but it would require dissassembly/reassembly of the hotend to change the number of filaments you're using, which is something I do not want. Easier to just stick a piece of filament in, possibly of a higher temperature band, and then tape it in place if you lack an extruder to hold it in place (And considering how much of this would have to be custom made, and the costs involved, why would you not have at least a basic extruder for that?)

Worst to me, is that if you wanted to make these, it would likely take 5 axis CNC's (Or some very complex fixturing), which means that the cost of having it made would be very high (Few machines we could get access to are set up to do routine 5 axis work, and full 5 axis CAM is not cheap (Admittedly, it might not take full 5 axis CAM if you were willing to start and finish it on a lathe, and not trim absolutely every gram of thermal mass you didn't need). But the operations involved would be rather complex.

1) machine blank from stock. Requires a nub for the machines to hold onto, or a tapped hole.

2) Machine to basic shape (Conical, would likely be best on a lathe). on a lathe it would be sub-divided somewhat
A) face end of blank to length
B) Rough profile
C) Semi-finish profile
D) bore central hole for nozzle
E) thread central nozzle hole

3) Chuck part into 5 axis fixturing using the tapped hole from step 1.

4) Drill holes for filament feed, heaters, and PT100 type sensors ( These would all be parallel to the surface)

5) Drill holes for grub screws, set screws, and other sensors

6) Machine to finished surface (You may exclude much of this step if you don't mind having a giant cone of a heater block, and the extra mass that goes with it, but for a good product, I would cut the mass as much as practical, since that improves heating speed, acceleration, and cornering)

7) Ream filament feeds
A) If designing for use with abrasive filament, possibly press in a liner for the channels to minimize erosion, which then needs to be reamed itself
8) tap holes for set screws and other sensors.

9) drill seating locations for heatbreaks into filament feeds.

10) Tap seating locations for heatbreaks.

11) Remove part and re-fixture to clean up back side

12) remove previous locating hole and threads

13) Drill mounting hole for sensor above mixing chamber, such that it encounters the set screw hole drilled earlier, if you want to avoid having to drill a new one and tap it

While it seems like a short list of operations, many would require one to four tools unique to that operation, and at least one requires that the part be continuously re-positioned in all five axis's (Machining the final profile), while most require it to move in at least 3 between cuts. You would also have to have a fixture that can grab onto the front of the heater somewhat (With sprung pins to hit the chamfers (What do you mean you didn't chamfer it? You want every hole meant to have assembly/disassembly involved with it chamfered, even if slightly, so you can actually do it in reasonable time) on the set screw holes, and a threaded piece to lock it in by connecting where the nozzle goes.)

If you already own a CNC, you can make much of the fixturing (Other than potentially hardening it), but you'll need the rotary axis's. There's a few grand to put it on a machine (24 grand at least to get one new from Haas, unless you plan on trying to put this in a collet somehow, 8K the lowest I would trust off Ebay (Other machine tool manufacturer's are likely cheaper, but it gives you an idea of the cost), and the cost to get 5 axis CAM ( Mastercam is something like 4K a seat at the low end, and 40K at the high end) And no, 3+2 is unlikely to cut it, even though auto-desk is giving it way next to free. Tooling would also be somewhat expensive if you wanted to make it fast, as you would want to go carbide. So the start up costs would be huge. A kickstarter or Indiegogo with a realistic goal on it would need either very expensive hotends, or to have so many that you would be absolutely swamped dealing with the orders from it. So unless someone already owns a machine shop with 5 axis capacity, and wants to make/sell hotends for some reason, rather than potentially taking better paying work, we are unlikely to get a something like that. I would love to be wrong of course. But realistically, it's unlikely to happen (And the machining steps mentioned are just for the heater block, not the nozzles, heat breaks, or the heat exchanger) anytime soon.

[elmoret]

Can you also get them to ship you some of these, but without the useless brass parts? The plastic grippy things wear out, but I have no idea what the brass fitting is good for if you already have a hot end.

E3D doesn't sell the collets separately, to me or through their website. I'm told that they can't buy them separately from their supplier, so if they sold collets only then they'd just end up with a pile of brass rings themselves.

However - a bulk ordering customer recently purchased 1.75mm hotends but didn't need the collets, so I do happen to have some collets on hand. I've added the option to the site, it only exists in 1.75mm. We do not have any 3.0mm collets available. One the collet surplus is sold out, it will be back to whole couplings so get 'em while they're hot.

http://www.filastruder.com/products/emb ... encoupling" onclick="window.open(this.href);return false;