Wisdom needed on Tip of Hot end detail...

I have been studying the various hotend designs and wondered if there was a consensus on the detail of the very bitter end of the hot end tip. As best I can tell, some designs end in what I will call a truncated cone of about 118 degrees included angle (the cone) and the truncation ("flat spot") on the tip maybe about 1mm in diameter.

Others seem to be better described as a truncated cone but with a small, final projecting cylindrical tip, the tip maybe 1 mm in diameter and maybe about 0.5 to 1 mm tall.

Any comments on cone angle, how large a "flat spot" on the end or whether have a small, final cylindrical pip at the end and its impact on print quality or other performance...how forgiving are these details?...for an ABS machine with a 0.5 mm orifice.

Thanks in advance.

My comment, avoid the tips with nipples.

It's a tradeoff like most things are. Larger flat area around the orifice can produce smoother output but can also cause filament drive problems whenever there is too little space for the extruded filament (starting even a little too close to the print bed, printing in a spot that has ended up overfilled for some reason). A sharp tip with very little flat area is more tolerant with these issues but can produce more uneven surfaces.

I think the cone angle is usually chosen to match the drill cone inside, or be shallower. That allows a short exit hole.

The tip needs to be at least a little wider than the widest filament you want to extrude.

---

[www.hydraraptor.blogspot.com]

I never liked the nipple until I started using a hotend with one and I am sold now. Why? Because it allows the filament to be molten in the melt chamber which helps to keep extrusion pressure down while allowing the tip to be cooled a little which helps reduce ooze and makes the prints solidify faster giving them better resolution.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Your experience with a nozzle nipple requiring less extrusion pressure is the opposite of my experience. If I am targeting a less than 2mm exit length then a nipple ends up being a significant fraction of that.

To make good use of the nipple, you must have a level bed, heated bed, and ABS dialed in at the right temperature. Printing on glass even better. PLA is too gooey for the nipples, IMO.

billyzelsnack Wrote:
-------------------------------------------------------
> Your experience with a nozzle nipple requiring
> less extrusion pressure is the opposite of my
> experience. If I am targeting a less than 2mm exit
> length then a nipple ends up being a significant
> fraction of that.

I have never seen a hotend with an exit hole longer than 1mm. The J-heads Tantillus uses have a nipple but the exit hole is under 1mm in length.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

jcabrer Wrote:
-------------------------------------------------------
> To make good use of the nipple, you must have a
> level bed, heated bed, and ABS dialed in at the
> right temperature. Printing on glass even better.
> PLA is too gooey for the nipples, IMO.

Only ever printed with PLA and at high resolutions. But I am a true believer that lots of air flow is the key to a good print (at least with PLA).

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

I don't like nozzles with nipples because it is hard to keep them clean when using PLA. A simple cone can be wiped when hot, but plastic gets trapped in the crease around a nipple. ABS can be peeled off when cold but PLA sticks too well.

I would be surprised if it affects the temperature of the plastic exiting. Plastic has such a high specific heat capacity, low conductivity and is moving so fast at the nozzle exit that I doubt it notices the nipple is a slightly lower temperate. And the nipple is so short compared to its width I doubt it is much cooler at the end than a cone.

I think the amount of ooze is affected by the length of the exit hole, but whether it is in a cone or a nipple shouldn't have much affect. I have never had an issue with ooze when printing, only during warm up and that is solved by warming up with the nozzle on the bed.

Edited 1 time(s). Last edit at 01/31/2013 07:14AM by nophead.

---

[www.hydraraptor.blogspot.com]

Thanks for the all the thoughtful responses to my request. I think I will start with no "nipple" for the first cut and see how it goes.

The reason I asked is that I am designing a hot end that is based on using hard temper stainless precision tubing (basing large bore hypodermic needle tubing) as the barrel (1.75 mm filament...tubing has a 0.01" thick wall) and providing the structural support to the heater block/tip using two steatite L5 ceramic standoffs.

In design terms, I am separating the mechanical structure/strength elements from the melt/feed elements as opposed to most designs I have studied which combine them. I am aware of the tradeoffs...higher mass, heat loss through the ceramic standoffs (though they do have a surprisingly low thermal conductivity), more complex building, more parts, etc...

I am brazing the stainless tubing into the brass tip with the goal of making the melt chamber/transistion zone/feed barrel seamless while minimizing the thermal transfer of the heater block into the barrel (transistion zone). Does the world need another hot end design? No, probably not...doing it for the fun of exploring the design space and learning about materials.

An added question to this: I have noticed that some hot ends put a chamfer on the exit hole of the nozzle. I understand this is probably how they drill the hole, but don't you guys think that is a bad idea/defeats the purpose of a small exit hole?

Makergear hot ends have this, and on the Ultimaker website it looks like their new hot ends have this too (but not the old ones! I have two and they're both fine).

That said I have been printing using the makergear hot end with this chamfer and i've noticed that I can't stretch the extrudate as much as I can with the old ultimaker hot end (sans chamfer). Otherwise, the quality seems pretty similar.

Any thoughts on this?

in my experiments with hotends, I've angled the brass to about a .5mm land near the .4mm dia hole. Easy to clean as mentioned above.
No specific angle for me. I've cut back nipple features as well. I see no cooling issues with a fan blowing and +1 Nopheads comment.
It's all hot and kept hot by the heater and plastic leaving the hole.
I worry about bed collisions with nipples, so I avoid them now. I've deformed one and wasted some time remachining brass.

What I'm seeing are two styles of printing: one that uses the tip to smear around the plastic while depositing and one that lays down discrete roads. The smear style is a result of large land areas,small layer heights, extruder gain or other settings. I can make some void free and translucent PLA parts using smear printing. The discrete roads are produced by layer heights around 50% of the nozzle dia and up and settings specific to produce deposits stacked and orderly. These parts are not translucent due to all the voids and internal round surfaces.

I've made many hotends and drive systems now, and used the hypodermic tubing (.095/.077 size) from the drive to the very tip, or between the drive and the top of a Jhead (I use Jheads because I'm losing too many hours making franken-hotends). I press fit the tubing deep into the brass and avoid the brazing, which I could never get to work with my plumbers flux and solder. I flare the tubing inlet by rotating a sharp hard pin at an angle, quick and easy.


I've realized that jamming and cold slugs are common due to long melt zones, so pay attention to reducing the distance between the hot and cold zones. Heat sink and actively cool the tubing just above the melt zone to narrow the distance as the filament softens along its length readily if the melt zone has a gradual temp reduction along its length. For me, jamming from long cold slugs was common with long hypodermic tubes. I've never built a long tube system with active cooling to narrow the melt zone, but I came to this unproven conclusion.

I also had tube jams from non-linear Ebay PLA, which I can see if I unspooled some length. Lots of slight kinks and sometimes, a kink will be enough to seize up in the tubing due to friction, and my drive system will strip out the filament and the print is dead. I would suggest an ID larger than .077inch, which is the diameter Stratasys uses. (note: they have very straight filament in every roll). As you increase the tube ID, you can get backflow/cold slug up above the melt zone and I don't have a good suggestion for an ideal ID. Maybe monitor the vendor quality for straightness?? I recently removed the hypodermic tubing between the drive and Jhead and replaced it with a teflon tube, so no more ss-tube for me. No more jams now, and I am back in tuning mode and not designing/fabricating mode.

davew tx,

Great info. You addressed one of the issues that I have been wondering about...the best diameter for the stainless steel tube. Question: are your comments based on PLA exclusively? Would you expect an easier time if I just shoot ABS?

As far as the press fit idea, very nice. Also, I like the idea on how to flare the input end. I am using brazing...."easy" silver solder (jewelry stuff...56% silver content...melts at something like 600+ degrees F as I recall) and special stainless steel brazing flux. Makes for toxic flux fumes and beautiful fillets.

I am using active cooling to minimize the transition zone. I also plan to try to minimize the tubing length to just what is needed to transition the temperature and then transition to PTFE tubing up to the drive hub...so metal only where the plastic is soft.

I hope to be able to put together a few pictures soon to post.

While I have high hopes for this design, I won't be surprised if I find my way back to a full PTFE liner and ditching the hypodermic tubing...but not until I try it.

I did get some ABS to work with the tube on top of a Jhead, but mostly I am using PLA for now.
I have P400 Stratasys black, P430 ivory, Makerbot nuclear green and an Octave purple.
All of these are very straight filaments and I have many test prints that worked, but with my Jhead and not a custom head.

To be honest, all my efforts to make an all metal hotend fell short, but I understand what's happening now.
I learned with every new version, but gave up because I am under a deadline for a large print.
I spent about 10 hours making 2 versions or modifying to salvage a version.

I also just read about an Aussy (none other than TheJollyGrimReaper) [forums.reprap.org] and seeing what he's doing fits my conclusions.
I should have read up before tackling this. It seems the all metal hotend is a rare item still. Prusa has one out now which i see looks nice.

Note: on one version I had the brass hotend mounted in an aluminum/phenolic part, and extended the ss-tube up about 10mm, but did not cool it. Then I put a 1.5in length of teflon tube over that up to the drive. ( pushed over the tube flare so there was a full 10mm overlap)
This did not work because the ss- tube was 200C just like the brass block it stuck out of.
That caused my melt zone to be above the ss-tube in the PTFE section, with a 3mm ID, so I noticed the large melt puddle building up in the teflon section, and I kept feeding it, just for fun at this point, I quit just before it reached the drive wheels.
Somewhere in there the built up pressures caused leakage in that overlap tube joint area also.

Good info on the brazing. I had one version where the ss-tube was 1" deep inside the brass but not press fit. The drive system drove the tube right through my brass tip, just punching clean through. I inspected the brass tip and did have the hole deep and it left that area thin, but I was hopping out of my chair to hit the power when I realized the tube was dragging around on my glass. Then I made a press fit brass tip and that area was leak free. I have a 6mm cartridge heater running parallel to the 3/8 brass part/tip, and both those in a 3/4 inch diameter aluminum extension, which is press fit/pinned into a phenolic isolation.

I know a picture is worth a ton right here, so I will snap a pic of this whole bad idea tonight if I can.
There are several "things to avoid" in the last version.

Edited 1 time(s). Last edit at 02/01/2013 04:47PM by davew_tx.

I'm reading and JollyGrimReaper says something worth noting. if you use too high of temperature the PLA swells and jams up.
This probably was a major contributor to all of my problems. I just chose a temperature randomly, like 200 or 210 or higher if I get jams.
I was thinking I could melt the jams and push them through, but I was likely just softening my PLA up the tube and creating a longer jam.

I feel like I wasted my time a bit, by not researching first.

Besides Jolly's, are there any good reading blogs about this?

in hindsight the it's quite suprising how effective a stainless thermal barrier can be made,

click here

thats the 3mm version of the one in my newer hotends,

its essentially the volume of stainless steel that the heat energy has to get across that is where a good thermal barrier effect can be achieved by reducing the volume you can decrease the amount of heat going across it dramatically,

what i tend to find now is that the abs and pla is now solid right down to the point where the threads start again, and because the meltzone is so short it's very difficult to jam up if not almost impossible because there is very little melted plastic inside the meltzone so when you pull it back again (like what happens during retraction) there isn't as much melted plastic to cool down and cause a jam,

in saying that, my latest tests with my "retractionless hotend" now have more heat going up the filament than up the thermal barrier, and because of the very different behaviour of the hotend i've had to pretty much re tune the printer completely,

with the nipple vs pointy orifice, it's more of a manufacturing thing, drilling a nozzle and making a "pointy" hole at the end is actually quite difficult and can go lopsided quite quickly and the reject pile can get significant, so with a flat or nipple on it you can reduce the reject pile and get more viable nozzles per run, plus you can drill them from the front and see what you are doing,

i think what we're going to see in the comming years (providing we don't have a viable alternative to filament) is people actually having more than one hot end and swapping them out depending on what is to be printed and the material used, rather than one general use hotend for everything,

Edited 1 time(s). Last edit at 02/01/2013 07:44PM by thejollygrimreaper.

---

-=( blog )=- -=( thingiverse )=- -=( 3Dindustries )=- -=( Aluhotend - mostly metal hotend)=--=( Facebook )=-

It's just like a cnc mill/router and toolheads. One day the technology will permit automatic changing of nozzles; Even in the middle of a print.

JGR: I assumed nipple was a cylinder extension and the bevel was just to the flat land. the tip pic shows what i call a nipple in the right brass part and the left one is beveled to a point and also the mix shows a bevel in the assembled.

the aluminum hotend was a bad idea because those two large members that go up to the phenolic carry all the heat and shroud the PTFE tube and make an oven for it. blowing air across it did not help much.

that white plastic is fluorosint 500, a mineral filled PTFE.
Great stuff, about $3/inch for 1" diameter though. machines and feels like a hard soap bar. No strength to itself, but handles high temp.

In the mix pic you can see a ss-tube above a Jhead (hotends.com!). This jammed when kinky PLA was used.

Attached .jpg showes progress to date on the stainless hypodermic tubing hot end.

The structural supports are 1/4 inch diameter ceramic (steatite L5 grad) threaded standoffs.

The tubing is brazed with jewelry "easy" silver solder (56% silver). The tip is brass and is held in the aluminum heater block with a compression "clamp" arrangement (not threaded into the block).

The heat sink is a custom part I fabbed (as is pretty much everything). It clamps unto the hypo tubing. I am not totally pleased wit the heat sink...I think it should have bigger air slots.

What is not shown is the ceramic tape insulation that I will be covering the heat block. The funny angles of the block and the heat sink are a long story...I designed the exturder for a different custom hot end and I needed to play funny games with the angles to salvage the machining I had for the old design.

The hypo tubing input end mates with some PTFE tubing, the output end of which is slightly bored to mate with the hypo tubing. I plan of force air (fan) cooling of the heat sink. The heat block accepts a 40 watt/12 cartridge heater.

Wish me luck.

nice. what are the ID/OD you selected on the tube?

I found Amazon a good source for hypodermic tube of all size.

OD/ID=0.095/0.071 inches...and yes, anticipating a question based on your previous post...it may be too tight. I plan to try it and it needed, up the bore size. I did a rough calculation of the thermal expansion of ABS and measured the roll of filament I own, there is some margin...but, over the full range of variability of diameter of ABS filament...it could well be a problem. I may go up to 0.077 inch ID...

Good tip on Amazon for tubing...I bought mine from Mcmaster's for not alot of money (something like $8 for a foot if I recall).

Geez.

I finally got around to reading Nopheads blog [www.hydraraptor.blogspot.com] .

Really worth reading, as he's see it all and done it all.
Really a cornerstone.

Props.

Hello there.
I made a very simple hotend taking inspiration both from Watson and Direct hotend.
It is made only from 3 pieces, 1 piece of PTFE with 25% glass content, aluminum heater block and aluminium nozzle.
External isolated with ceramic tape.

It doesn't jam at all, melting length is really only tip but I cannot get the PLA to flow straight down, it just curls around the tip of the hotend whatever.

Do you know what explains this? Is it the construction of the tip that determines this curl or what may be the cause?

the curling should not matter. if it is placed to a surface it will extrude properly. likely the curling is from uneven heating and cooling. raise the temp 5 deg or more and curling should start to disappear at lower flow rates.

Sadly it will not stick to bed at all.
I don't have a heated bed, I tried with kapton tape but it doesn't stick at all.
The only way I could make something stick was to melt some PLA with the solder station and spread it on the kapton to form an 1 mm bed of PLA. Then it would kind of stick but the extruded filament is not straight it still curls around so what I got was an ugly blob instead of a cube. So really I can't get some straight lines of extruded PLA tk stick to each other, seems like the tip is attracting them like a magnet.
I will try to make some pictures and videos.

Some pictures and videos:







[www.youtube.com]