Update on the E3D All Metal Hotend (Now finally shipping!) [Lots of pictures]

I just got my E3D - thanks guys! It's been printing for a few hours now and is looking good. I was wondering about the practically of larger nozzle size. I am using the 0.6mm nozzle at the moment. Would 1mm be pratical? If so is there a chance of getting one?

What temp are you printing at, and what material?

Chinese PLA 220C

I fitted one of these to the hotend - much quieter :-)
[dx.com]

Here is my 40mm shroud mounted with the fan on the x-carriage: [www.thingiverse.com]

Hi Guys,

I'm facing an issue with my 1.75mm E3D hotend with 0.4mm nozzle.

Looks like the extrusion width is not constant. When I extrude outside the bed (I mean, in the air), the extruded filament looks like pretty regular!

But check what happens on prints (good ABS at 220C):



Any clue??

Cheers

Alex.

turn up to 230 for a more fluid extrusion?

I've tried from 200C to 240C... same results :-(

is there some kind of blockage in the nozzle? some stories about impurities in chinese plastic etc like metal fragments..

Another possibility is that you hotend is not secure and "wobbles". How are you mounting it and can you move the tip around easily?

I have just replaced the nozzle - I had a 0.25mm spare one. Same issue

About the wobble, I don't think so.. the hotend is well fixed and even on Y axis, when the hotend is not moving, I've got the issue.

My next approach will be replacing the motor.. before replacing the hotend.

Cheers.

Alex.

if you look at the image there is actually a repeating pattern to the lines, have u tried tightening ur timing belts? also gt2 belt is better than the other one in terms of getting smooth movement.

Could be an eccentric hobbed bolt or similar.

---

[www.hydraraptor.blogspot.com]

@mrPrik, I'm using GT2... it is not that tight. I cannot see a relation between belt tight and such issue, but I will give it a try and let you know.

@nophead, I've already replaced the hobbed bolt and extruder. Former I was using a regular extruder with a hobbed bolt and know I'm trying a direct extruder with a pinion gear... same issue.

Cheers.

Alex.

How many E steps per mm do you have?

---

[www.hydraraptor.blogspot.com]

I have tried such combinations:

- 0.4 nozzle, 1/32th DRV8825, geared extruder (1.5:1) and 8mm hobbed bolt: 470 steps/mm;
- 0.4 nozzle, 1/32th DRV8825, direct extruder (11mm pinion): 195 steps/mm;
- 0.4 nozzle, 1/16th A4988, direct extruder (11mm pinion): 98 steps/mm;
- 0.25 nozzle, 1/16th A4988, direct extruder (11mm pinion): 100 steps/mm;

I'm changing the extruder motor right know.. let see what happens.

Cheers,

Alex.

I doubt DRV8825 can actually do 1/32 microstepping with low voltage motors. And A4988 will only achieve 1/16 if it has ROSC shorted, see [hydraraptor.blogspot.co.uk].

Taking that into account your values are about 1/7th of a typical Wade's with a hobbed bolt. I wonder if it is simply the steps you can see.

---

[www.hydraraptor.blogspot.com]

the pattern makes me think of some kind of dragging moving part that is catching and releasing in a cycle, worth checking everything imbetween the filament going in and the build platform in terms of the extrusion process. interesting u'v tried diff extruders with same results, and that extruding from a height gives a uniform shape.

looks like either temp fluctuation or mechanical fault anyhow. best elimination would be to change hot end to a j-head or similar and prove/disprove normal operation. good luck

@nophead, I ran the motor holding the axis with my hands and I can "feel" the steps.. I mean, the rotation is not fluid, it is more like a pulsation.. Since this variation in width is uniform, I could wonder whether this is a result of a pulsed extrusion?!?!? With A4988 ROSC shorted to ground, does the rotation become more fluid??

Maybe the solution will be using a geared extruder 4:1 or more...

Quote
nophead
I doubt DRV8825 can actually do 1/32 microstepping with low voltage motors.

What does this mean? and what do do consider a low voltage motor? I have all 3 steppes on my delta set to 1/32 (200 steps per mm). Prints fine. Yes I have read your step stuck fast decay for the 4988, and the same for the DRV 8825 but I have not seen the same issue. I use a 4988 on my 5.1:1 extruder at 1/16 with no visible artifacts like Alex shows.

What I have seen with the E3D is I cannot go slower than 20mm/s or the PLA will cool in the nozzle (temp set at 227C for PLA) and if I add an external fan to cool the print, I have to raise the temp. This is based on my experience.

As far as I can see the DRV8825 can't get the mark space ratio small enough to do 1/32 steps unless the motor voltage is similar to the supply voltage, which is not typical for Repraps.

The rotation is only fluid when moving relatively quickly. If you slow it down enough you will always feel the discrete microsteps. When ROSC is not shorted on the A4988 it doesn't move for some steps and then moves twice as far on the next step to catch up.

It is easy to test, just extrude so slowly you can hear it tick and see if it sounds regular or syncopated.

---

[www.hydraraptor.blogspot.com]

It sounds like you are feeling and witnessing a "pulsing" when the printer moves on a single axis, but as far as I can tell from the posts it's not yet clear to you if the extruder and or hotend are partly to blame. What about trying this to isolate the varying trace widths from the extruder/hotend:

-- Run off a few traces using the hotend and extruder and steps per mm of extrusion that give you the worst result in terms of varying width of extrusion.
-- Change the steps per mm of extrusion by 10% or 20% and keep all other settings the same and run off a few traces of material again.

If you see the pulsing width of the traces still happens on the same length scale even when the steps per mm of extrusion are changed, that suggests that the linear movement of the printer is the culprit. If the movement along the X and Y axes is not causing the issue but the problem lies in the extruder or hotend, changing the steps per mm of extrusion should cause the pattern of the varying widths of the traces to either stretch out or compress, since the extruder would be moving faster or slower compared to the linear movement of the machine.

You may have done something similar above, so sorry if this has already been tried.

Best regards,
Daniel

Daniel, you had a great idea to isolate the issue. I will try it as soon as I get home! And let you guys know!

Thanks
Alex.

Sanjay from E3D here again!

So much going on here, lots of questions and lots of answers. Not a lot of time. I will come back and answer everything once I have a moment. Just moved house, quit my job, gone full E3D, its been a crazy time. Thank you all for your patience, and I'm sorry if some people have been waiting for customer service a little longer than usual. It's been hard!

Just wanted to let you all know that we are doing a 20% off everything sale - so hotends are only £35.20 right now, along with all sorts of other cool bits. However I'm probably going to have to close the sale soon before we bankrupt ourselves and have no stock, so do get in now if you want in. The amount of orders the sale has produced has been beyond expectation. So was just going to jump on here and let you know, because you guys have been the most informative and valuable resource to us at E3D - I don't want you to miss out on getting a good deal when you of all people deserve it.

[e3d-online.com]

Back tomorrow to comment on the other matters raised!

Sanay,

I saw that you were having the sale--2 days after I had placed my order!

No worries, I wanted to comment here that the package arrived very quickly. I placed my order from here in the US on Sunday the 8th, it shipped on the Monday the 9th and FedEx made the first delivery attempt on Wednesday the 11th.

I ordered two hotends that I am using on a new machine build. I haven't been able to try them out yet because I am waiting for electronics and extruder parts to arrive. Looking forward to seeing what they will do though!

Daniel

can these hotends go up to 300 without melting the ABS fan shield? wanting to be ready for nylon and polycarbonate printing

Daniel, you were right:

195 steps/mm: 6mm between "pulses"
96 steps/mm: 12mm between "pulses"

I will print new gears to get a 4.7:1 extruder.

But I wonder if I'm the first person to realize it ? Nobody else has seen such behavior with direct drive and 1.75 hotends?

You are not the only one to realize the bad quality aspects of direct driving filament. Using 1.75 mm filament (vs. 3 mm) helps a bit, also using a 0.9 degree stepper helps. You know, steppers are only truly accurate to half step. Anything further than that cannot be considered as added resolution, but rather added smoothness of motor rotation.

Microsteps do give more resolution, they just don't give more accuracy because they interpolate between the full step positions so can never have more accuracy than they have. To get proper micro stepping the driver needs to be configured correctly for the motor and the supply voltage. It is very easy not to get 16 steps when the supply is too high relative to the motor voltage with constant off time drivers.

---

[www.hydraraptor.blogspot.com]

Quote
jkoljo
You are not the only one to realize the bad quality aspects of direct driving filament. Using 1.75 mm filament (vs. 3 mm) helps a bit, also using a 0.9 degree stepper helps. You know, steppers are only truly accurate to half step. Anything further than that cannot be considered as added resolution, but rather added smoothness of motor rotation.

it helps a lot especially in direct drive to run you steper drivers up near their rated current, the pololus with the a4988 aparently only really deliver about 1amp and most of is are running motors that need 1.7amps for full torque output, if you swap drivers or just the extruder driver to a drv8825 you are able to get the current up higher and you'll find this issue for the most part will dissapear

---

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