E3D V4 All metal hotend

[mhackney]

3 parts down - all perfect, no starvation all at 185°C.

Polygonhell, not sure what you mean by KISS having particularly bad tool paths? I've been very happy with KISS' tool paths compared to other slicers. However, that round infill is a stressed.

I suspect that the nozzle bore length might work for many users - raise the temp a bit, no "wild" infills, reasonable print speeds, sporadic printing, etc. But I print a lot (production parts) and am very demanding on quality and speed. I've emailed Josh and Sanjay and explained my findings and sent some photos, etc. Pointed them here too. I also asked if they have measured this bore length and if they think there might be variability. I got my E3D months ago and it was right at 2mm as were the 4 with Kraken.

[Polygonhell]

My comment about KissSlicer was more it seems to occasionally generate tool paths with seemingly maniacal retractions, usually when it's doing the fill the unspdersized gaps thing, or filling out the perimeters to the specified thickness.
In general I think it's still the best of the slicers with some caveats, Cura is very good right now, generates less crossing of edges and less retracts in general, and it actually bridges. I've been using that for my mechanical parts for the last couple of weeks.

I would be surprised if there were significant disparity in the nozzle lengths, these things have to be CNC machined in a jig, wondering if we can find the difference between the working nozzles and those that do not. I guess the higher extrusion pressure coupled with other factors like the overall finish of the stainless piece could be the answer.

[Nylocke]

Speaking of Cura, what are you doing for bridging Poly? for some reason Cura has been trying to bridge diagonally it seems, instead of crossing straight across from one side to the other, it prints like there isn't an overhang. Don't quite know how to explain it, but its been acting kinda weird for me. Also I've been having troubles with retracts and ABS, though that may be because my extruder split almost in half :/ (also might be due to El cheap sainsmart ABS, I haven't printed with my IC3D stuff for a while...)

[Tinyhead]

I tried the speed/temperature method on a print last night. In Cura, it was set for 30mm/s speed, 60% infill, 0.1 layer height with supports and would take 24hours. Once it got the bottom layer (20mm/s) down, I ramped up the speed using the LCD to 185% (55.5mm/s), and adjusted the temp to 230C and was using black PLA. With the stock hot end, I can run this black at about 190-195C, so 230C was putting it up there. The print ended up taking about 13hrs.

This ended up not only working for the entire duration of the print with no starving of filament observed anywhere, but it also never plugged due to retracts (another problem I've had), and it produced some of the smoothest walls I have seen on my prints to date.

3/4 done.

Complete.

I think the temperature could have come down as there was some decent blobbing on a portion of the backside of the supports, but it was just on the supports, so who cares. I thought the supports were going to be murder to remove due to the high temp of the PLA, but it snapped off nice and easy. Very surprised.

I'm very much liking the news regarding the drilling of the nozzle by mhackney. It would be nice to hear that it might allow printing at regular settings. While printing this time at high speeds was awesome and may work great on prints with a large layer surface area, small prints would stay too hot and sag without some serious cooling.

[mhackney]

Well so far so good today. As I ramped the speed up to >50 mm/s I did have to bump the temp up to 190°C but that is typical with other hot ends. I am still guardedly optimistic about this insight. I'm looking forward to hearing from E3D-systems and printing many, many more parts successfully.

That is a great looking and large party Tinyhead. It is good feedback to see that ramping the temp up to an extreme setting also worked for you too. I do still think that the nozzle might not have as good heat transfer as other hot ends since 1) the tip is not tightened against the heater block, a gap is prescribed to allow tightening the nozzle to the heat break, 2) the back end of the nozzle is tightened against less thermally conducting SS heat break and 3) the primary heat transfer path is through the threaded section and there is not that much surface area/threads to conduct the heat. I think this combined with the long bore result in two parameters that play against each other.

[Generic Default]

When you say Bore Length are you talking about the length of the tiny orifice drilled hole? I started a thread in the louge a few days ago about nozzle design, and this E3D thread just happened to turn into a conversation about nozzles. Maybe take pictures/draw sketches of your modifications and put them on this thread or the nozzle design thread?

[mhackney]

Yes, I am talking about the .40mm bore's length. There are photos I posted this weekend showing that the E3D bore is 2mm long - much longer than other nozzles. I simply drilled the 2mm hole that leads up to the .40mm bore a bit deeper with pin vise and 2mm drill bit. I'll try to draw something tomorrow.

[McSlappy]

I haven't commented much on this lately, just waiting for my new head (which arrived, but I've been now too busy to install >_< ).

I've watched this thread lately with great interest and I'm pleased that someone else was able to replicate and delve into the issues with PLA that I had. I'm curious now to see what depth my nozzles have. Even without knowing much about nozzle geometry I would have thought that 2mm was waaay to deep for a nozzle opening.

Michael keep us posted on the progress (as if you wouldn't) I appreciate your methodical problem solving

I'm also very interested in trying out a Prometheus and see whether the single nozzle/heatbreak combo will do wonders.

[JohnStack]

Yep, this thread is so long that it should have been broken into it's own sub-forum.

On my end, after reliable multiple shipments from Filastruder, I can't seem to get a real tight assembly without bending the heat break shaft at it's weakest point.

In the photo below, the two shafts on the left are bent and they caused (I think) two more hot end blowouts. On the right is another shaft - which is straight - but needs cleaning out as the result of a blown thermistor a while back.

All in all, I'm frustrated with these - the performance is great - as long as you don't have any mechanical or electrical screw ups. I've had and done both.

Here's why I'm frustrated:

1. Polygonhell's PLA problems.
2. You need to build a separate fixture to heat the assembly in order to avoid burning yourself in the process.
3. If you're doing this without, you're risking a bend to the weakest part of the heat break shaft.

These are all known problems - I guess I've reached my proverbial limit with hot ends. Surface adhesion and hot ends! Damn. Going back to SeeMeCNC's 99% solution - their hot end.

[Broose]

I am wondering if, as Poly said, there may be some manufacturing variation in the nozzle bore length. I'd love to know what the bore length specification is. I have unused .4mm and a .6mm nozzles and the bore length I measured was just about 1mm for each. I measured a 2mm drill bit bottomed out in the nozzle then subtracted the length of the 2mm bit to get the length of the hole. The difference was about 1mm in both cases.

[Polygonhell]

Using the same measure the drill concept yesterday, I measured the orifice on the first one I had at over 3mm, but that's the older nozzle design, the newer one I checked was closer to 2mm, but it's hard to get an accurate reading without knowing the angle of the original drill, since there is no way to know if the drill is seating at the bottom on the hot end. Although it gives you a maximum. I drilled one of mine out to a 0.4mm orifice, though I'm not printing PLA right now.

MHackney's measurement solution seems like it ought to be accurate, and that seems supported by the fact that he measures the JHead at 0.5mm. If your seeing a worst case of 1mm and he's seeing 2mm's then it would seem there is variation, I'm at a loss as to how this happens. I would assume that the nozzles are made from hexagonal stock on a CNC lathe, everything would be relative to the facing operation, and unless the actual code is changing between runs or the drill's offset is not correctly set, they should be identical to within 1/10ths of a mm.

[joecnc2006]

where do you get the tiny bits?

[mhackney]

The "frozen plug" method is the most reliable measurement since it captures the actual bore length, cone leading up to it and hot zone. It's easy to do. I measured the J-Head with the frozen plug technique.

[mhackney]

MSC, McMaster Carr and even Amazon carry the tiny metric bits.

[ceramichammer]

Well so far so good today. As I ramped the speed up to >50 mm/s I did have to bump the temp up to 190°C but that is typical with other hot ends. I am still guardedly optimistic about this insight. I'm looking forward to hearing from E3D-systems and printing many, many more parts successfully.

That is a great looking and large party Tinyhead. It is good feedback to see that ramping the temp up to an extreme setting also worked for you too. I do still think that the nozzle might not have as good heat transfer as other hot ends since 1) the tip is not tightened against the heater block, a gap is prescribed to allow tightening the nozzle to the heat break, 2) the back end of the nozzle is tightened against less thermally conducting SS heat break and 3) the primary heat transfer path is through the threaded section and there is not that much surface area/threads to conduct the heat. I think this combined with the long bore result in two parameters that play against each other.

Mhackeny,
I was out of the country all weekend but whenever I found wifi, I compulsively checked this thread. I'd never heard of freezing the filament out and some of those results are fascinating. That ~2mm bore length looks way too long. It may also explain why I had good luck with the 0.6mm nozzle. Even if it still had a 2mm bore length, there would be much less resistance and possible plugging and jamming. I have also been skeptical of the thermal stability and heat transfer to the nozzle itself. I installed a much larger heater block in hopes of making the temperature more stable. Even if the heater block temp is stable, the nozzle itself may be fluctuating wildly or even at steady state but at a much lower temperature. These "dinky" nozzles have next to no thermal mass and very poor conductive heat path because they aren't seated flush to the bottom of the heater block. Compare the thermal mass of a stock Ultimaker nozzle vs an E3D hotend.

[img]http://i.imgur.com/SDSpqvUl.jpg[/img]

[McSlappy]

Yep, this thread is so long that it should have been broken into it's own sub-forum.

I suggested this when it was only 40 pages long.... Ahhh those were the days...

[Broose]

where do you get the tiny bits?

I got mine through Amazon http://www.amazon.com/SE-Drill-Mini-Siz ... _ac_text_y

[mhackney]

ceramichammer, your photo clearly shows how little contact surface the E3D nozzle has! The base of the nozzle itself does not snug up against the heated block by design and on top of that, there is an untreated section right above that so heat isn't able to transfer conductively until the first thread (the lowest in your photo). Then, there are only 5 threads total. That is not a lot of conductive pathway!

A quick update: printed all day at 185-195°C with NO filament starving on parts that are usually problematic. On top of that, I started ramming the speed up and was successfully printing at 60mm/s the tenkara line holders I manufacture - again, with no filament starving.

[Eaglezsoar]

ceramichammer, your photo clearly shows how little contact surface the E3D nozzle has! The base of the nozzle itself does not snug up against the heated block by design and on top of that, there is an untreated section right above that so heat isn't able to transfer conductively until the first thread (the lowest in your photo). Then, there are only 5 threads total. That is not a lot of conductive pathway!

A quick update: printed all day at 185-195°C with NO filament starving on parts that are usually problematic. On top of that, I started ramming the speed up and was successfully printing at 60mm/s the tenkara line holders I manufacture - again, with no filament starving.

Have you recieved any response from Josh or Sanjay about this?

[mhackney]

No, not yet.

[ceramichammer]

ceramichammer, your photo clearly shows how little contact surface the E3D nozzle has! The base of the nozzle itself does not snug up against the heated block by design and on top of that, there is an untreated section right above that so heat isn't able to transfer conductively until the first thread (the lowest in your photo). Then, there are only 5 threads total. That is not a lot of conductive pathway!

A quick update: printed all day at 185-195°C with NO filament starving on parts that are usually problematic. On top of that, I started ramming the speed up and was successfully printing at 60mm/s the tenkara line holders I manufacture - again, with no filament starving.

Because I have 3mm filament, I believe I'm at a geometric disadvantage in how much I can shorten the barrel length. Because the tip of the nozzle is tapered, my 1/8" drill bit will leave me with a much thinner cross-section and I'm more likely to blow the entire bottom out. Here's my next test with the flange of the nozzle snugged up against the heater block to promote heat transfer. I'll have to wait on my machinist friend to attempt to drill it out.
[img]http://i.imgur.com/EEnY6bql.jpg[/img]

[bvandiepenbos]

to get most thermal transfer to nozzle why not screw the nozzle into heater block FIRST until flat seats against block, then thread heat break in from other side until it bottoms against nozzle, then lastly screw heat break into the cooling fin body?? I have not done this yet, but am planning on assembling my E3D that way.
why leave the gap?
am I missing something here?

[mhackney]

Because that's not how E3D says to do it - most likely to prevent breaking the heat break at the thin section. You want to tighten the nozzle agains the break, not vice versa.

[ceramichammer]

to get most thermal transfer to nozzle why not screw the nozzle into heater block FIRST until flat seats against block, then thread heat break in from other side until it bottoms against nozzle, then lastly screw heat break into the cooling fin body?? I have not done this yet, but am planning on assembling my E3D that way.
why leave the gap?
am I missing something here?

That is what I did in the photograph above. I agree it sounds dumb but the E3D assembly instructions say specifically to leave a gap. http://files.e3d-online.com/Drawings/E3 ... Manual.pdf
My best guess is having that gap there is confirmation that the nozzle is seating against the heat break and not accidentally the heater block. Either way it seems silly.

[bvandiepenbos]

Because that's not how E3D says to do it - most likely to prevent breaking the heat break at the thin section. You want to tighten the nozzle agains the break, not vice versa.

so don't use your big pipe wrench