E3D V4 All metal hotend

[bubbasnow]

got an update.. usps said it left LA this morning.. i should have it Saturday....i better have it Saturday.....if not i will punch my postman in the face (he has been pissing me off as of late anyways!)

got it today... you were lucky usps... kthanks!

[626Pilot]

Couple easy and highly useful mods I figured out for the E3D.

First, consider potting your heating resistor with f̶i̶r̶e̶ ̶c̶e̶m̶e̶n̶t̶ RTV. Put a thin layer on the outside of it, push it in one side of the heater block with a gentle twisting motion, then do the same from the other side of the heater block to get a good coat on the inside of the block before you press it in. Then, seal up both ends (making sure to coat all the metal of the resistor casing) with some more RTV. If you do this, you will notice a LOT more thermal stability in the temperature curve screen. Much less over- and under-shooting, and I think it heats up a little faster. (Not that the E3D needed much help!) Because the hot end can run more efficiently, it should sink less power and last longer. Relying only on the grub screw to secure it lets a bunch of air get between the resistor and the heater block, and this fixes it.

Second is a preemptive fix for two things: PLA jams, and buzzing noises as the shroud vibrates against the platform. It's easy. All you need is about 2" of half-inch-wide Kapton tape. Cut this into three strips. Line up the bottom of the fan shroud with the bottom of the lowest cooling fin, so they form a plane together. Use the Kapton to seal the seams. This will ensure that the shroud is always properly lined up (which it needs to be to do its job), that no air leaks down (which will drive up the duty cycle & encourage melt creep), and that it can't "buzz." This can also revive a fan shroud that's warped because of the cooling fan being left off - it's the first thing to get ruined, and it likes to let go of the heat sink and flare out so that air can escape out the bottom.

[bubbasnow]

Couple easy and highly useful mods I figured out for the E3D.

First, consider potting your heating resistor with fire cement. Put a thin layer on the outside of it, push it in one side of the heater block with a gentle twisting motion, then do the same from the other side of the heater block to get a good coat on the inside of the block before you press it in. Then, seal up both ends (making sure to coat all the metal of the resistor casing) with some more fire cement. If you do this, you will notice a LOT more thermal stability in the temperature curve screen. Much less over- and under-shooting, and I think it heats up a little faster. (Not that the E3D needed much help!) Because the hot end can run more efficiently, it should sink less power and last longer. Relying only on the grub screw to secure it lets a bunch of air get between the resistor and the heater block, and this fixes it.

Second is a preemptive fix for two things: PLA jams, and buzzing noises as the shroud vibrates against the platform. It's easy. All you need is about 2" of half-inch-wide Kapton tape. Cut this into three strips. Line up the bottom of the fan shroud with the bottom of the lowest cooling fin, so they form a plane together. Use the Kapton to seal the seams. This will ensure that the shroud is always properly lined up (which it needs to be to do its job), that no air leaks down (which will drive up the duty cycle & encourage melt creep), and that it can't "buzz." This can also revive a fan shroud that's warped because of the cooling fan being left off - it's the first thing to get ruined, and it likes to let go of the heat sink and flare out so that air can escape out the bottom.

Can you snap a pic of your tape job?

[daftscience]

I solved PLA jamming with some canola oil. The only time I've jammed so far is when I forgot to connect the fan. Even that was easier to clean out.

But I thought I would share this.

[youtube]http://www.youtube.com/watch?v=EgE7_cTqKoU[/youtube]

[626Pilot]

I solved PLA jamming with some canola oil. The only time I've jammed so far is when I forgot to connect the fan. Even that was easier to clean out.

But I thought I would share this.

[youtube]http://www.youtube.com/watch?v=EgE7_cTqKoU[/youtube]

What did you do with canola oil? Also, dang, I want one of those.

Can you snap a pic of your tape job?

Sure

[dpmacri]

Couple easy and highly useful mods I figured out for the E3D.

First, consider potting your heating resistor with fire cement. Put a thin layer on the outside of it, push it in one side of the heater block with a gentle twisting motion, then do the same from the other side of the heater block to get a good coat on the inside of the block before you press it in. Then, seal up both ends (making sure to coat all the metal of the resistor casing) with some more fire cement. If you do this, you will notice a LOT more thermal stability in the temperature curve screen. Much less over- and under-shooting, and I think it heats up a little faster. (Not that the E3D needed much help!) Because the hot end can run more efficiently, it should sink less power and last longer. Relying only on the grub screw to secure it lets a bunch of air get between the resistor and the heater block, and this fixes it.

You call it "fire cement", but your picture looks like RTV. Did you use RTV just on the outside or in the inside as well? If not, do you have a link to the "fire cement" you used?

[626Pilot]

You call it "fire cement", but your picture looks like RTV. Did you use RTV just on the outside or in the inside as well? If not, do you have a link to the "fire cement" you used?

It's RTV.

[dpmacri]

You call it "fire cement", but your picture looks like RTV. Did you use RTV just on the outside or in the inside as well? If not, do you have a link to the "fire cement" you used?

It's RTV.

Thanks! I hadn't finished setting up my E3D, yet, so I just took out the heater cartridge and followed your advice

[daftscience]

What did you do with canola oil? Also, dang, I want one of those.

I took a filament cleaner, something like this, and cover the sponge, microfiber cloth in my case, with canola oil. Obviously, not so much that it's dripping but enough that it will give an even coat to the filament.

I think I found this tip on reprap forums.

[Av8r RC]

I ordered the E3D V5 hotend on Tues. and got it Fri. Super fast shipping thru FedEx.

Spent last night getting it all assembled and printing out 626's mount (very nice design, ) with my Budaschnozzle. I didn't use any RTV on the resistor, but I did wrap the heater block in approx. 10 layers of kapton.

Wow this thing heats up fast! Used to take my Buda around 10min to get to 230, with the fan on. Even with the fan on, the E3D gets there in about 2. Very nice. One of the main reasons I got the E3D was the limitations of my Buda. My Buda couldn't get past 230 with the fan on, and I couldn't get past 75~100 mm/s, even with PLA, because the plastic was flowing so fast thru it it couldn't keep the temp up. Can't wait to try some fast prints with the E3D.

What kind of settings are you all using on KISS with this hot end? With an extrusion width of 0.4 and flow rate of 1.0 I'm getting a single wall thickness between 0.4 and 0.425. When I tried getting retract settings I found 6mm retract to look good as far as stringing goes, but I'm getting some weird blobs at every layer. What kind of setting do I need to look at.

Here in this pic the in focus tower is 6mm retract, the slightly out of focus tower is 8mm. They look exactly the same. I thought this setting would have helped it.
[img]http://farm4.staticflickr.com/3758/1090 ... 674f_k.jpg[/img]


EDIT: Not letting me upload a pic. Just getting a white screen. Any ideas?
EDIT: Put the pic on flicker.

[626Pilot]

There are some .ini files for settings in the KISSlicer section on these forums. I've found that a retract of 4mm and a wipe of 0 is best. Keep the speed up too. I like at least 150mm/sec in the retraction settings. Wipe generally seems like a good idea but it can mess things up, and you should never use it on a piece with any sort of narrow features (my E3D top mount spacer and Hall-O probe on Thingiverse both print a lot better with it turned off.)

With retracts of 4mm and wipe turned off, I can print things like this featherboard with the barest hint of spiderwebs in some places where it retracts. That's as far as stringing goes. Blobs can be addressed by printing slower. Unfortunately there is too much hysteresis in the Bowden setup to do much about that - the faster the filament is moving before retract, the more it's going to blob out. To make it worse, slow print speeds can have a negative effect on sharp corners, so you have to consider the likely sweet spot for each object you print.

[Av8r RC]

Thanks for the suggestions. I did only have a retract speed of 80. I tried all the way up to 500 and it had basically zero affect with the blobs. After watching about 8 prints today. It's happening at the end of the perimeters at the retract. And I tried anywhere between 4 and 8, for retract, and still had the same size blobs.

[626Pilot]

Retract speed helps but can't completely compensate for it. Try lower print speeds or thinner layers. The running average of filament volume through the nozzle is what makes this happen.

I guess you could also reduce the acceleration and jerk settings. That would give the filament extra time to slow down before retraction.

[bubbasnow]

I upgraded my E3D Bowden mount to have better cable management (it now has strain relief) and by making the printable spacer 2mm taller. That should help wash the bottom cooling fin with more air.

BTW, I ordered some more nozzles. The 0.25mm nozzle is GREAT! I highly recommend it. I can feed PLA through it at 200mm/min and it won't stall, and with a line width of 0.3 the stuff I can print comes out looking nicer around any fine details. Better ceilings too.

with this taller spacer, i cant find a fan mount for a 40mm that really fits well. Do you have one as well?

[Av8r RC]

I tried some crazy retarded retractions speeds. 500mm/s w/ 3000 accel. Also noticed the bowden tube was moving some during retracts, so I took the play out.

Tried printing out the hollow cube. One at 50mm/s and the other at 15mm/s.

Can you tell which is which? Very disappointed that these blobs are still there. Even at crawling slow (for me) speeds.

[img]http://farm8.staticflickr.com/7362/1093 ... a705_o.jpg[/img]

[626Pilot]

with this taller spacer, i cant find a fan mount for a 40mm that really fits well. Do you have one as well?

The stock 30mm fan or whatever it is keeps the bottom cooling fin barely above room temperature, so I never saw a need to use a 40mm fan. If your fan duct is set up correctly you should not need a 40mm fan unless you're running your printer inside an oven.

Av8r RC, did you try lowering the jerk/accel settings? It might help to let the filament slow down near corners. Also, what's your nozzle diameter and extrusion width? Have you done the extruder calibration? Did you measure the filament diameter ten times and put the average into your slicer? Is your filament multiplier 1.0 or something else?

[bubbasnow]

with this taller spacer, i cant find a fan mount for a 40mm that really fits well. Do you have one as well?

The stock 30mm fan or whatever it is keeps the bottom cooling fin barely above room temperature, so I never saw a need to use a 40mm fan. If your fan duct is set up correctly you should not need a 40mm fan unless you're running your printer inside an oven.

Av8r RC, did you try lowering the jerk/accel settings? It might help to let the filament slow down near corners. Also, what's your nozzle diameter and extrusion width? Have you done the extruder calibration? Did you measure the filament diameter ten times and put the average into your slicer? Is your filament multiplier 1.0 or something else?

sorry i didn't say it clearly, i was talking about a fan for the layers not my cooling fins on the actual hotend. the spacer pulls the nozzle up closer to the hotend platform so my old fans dont have enough clearance between it and the print anymore.

[626Pilot]

sorry i didn't say it clearly, i was talking about a fan for the layers not my cooling fins on the actual hotend. the spacer pulls the nozzle up closer to the hotend platform so my old fans dont have enough clearance between it and the print anymore.

I don't. The solution I intend to use is this: http://www.thingiverse.com/thing:170317

It's designed to mount under the platform, so no spacer. I'm going to use that when I start working on my direct extruder mount.

[bubbasnow]

I don't. The solution I intend to use is this: http://www.thingiverse.com/thing:170317

It's designed to mount under the platform, so no spacer. I'm going to use that when I start working on my direct extruder mount.

o i c, that looks interesting

[Av8r RC]

626,

1st off. Yes I did do my extruder calibration. Ez struder, 96 steps per mm (extruded 100mm, measured 100mm with digi calipers), measured filament multiple times (avg. 1.71mm), filament multiplier 1.0, nozzle dia. 0.4mm, KISS set to 0.4 extrusion width, thin wall extrusion printed at 0.4 to 0.42

I tried messing around with jerk settings first. From the bottom up 25, 15, 10, 5 & 1.
[img]http://farm4.staticflickr.com/3772/1095 ... d522_o.jpg[/img]



Then left it set at 5. Then tried accel settings. From the bottom up 800, 400, 200 & 100.
[img]http://farm8.staticflickr.com/7390/1095 ... 0442_o.jpg[/img]



The 100 looks pretty good as far as the blobs go. But slowing the accel. down had a very adverse affect. These are the same parts from the accel. tests taken from above. Very overfilled corners. From left to right 800, 400, 200 & 100.
[img]http://farm3.staticflickr.com/2814/1095 ... 7174_o.jpg[/img]

I'll try some different combos tomorrow. Any suggestions?

[626Pilot]

Yes. Take your extrusion width up to 0.45. Never leave it at the measured diameter. You have to have some margin for the nozzle end to trowel the filament flat and ensure that minor fluxuations in filament width don't result in less than the whole nozzle diameter discharging at the correct rate. A little too much is OK but not enough can potentially make things uneven.

Also, try printing at a thinner layer height. I'm going at 0.08 right now. There are blobs but they are so teeny tiny you'd never notice them unless you were looking for them, and you'd have to look CLOSE. The filament goes slower the thinner your layers are, and the running average speed should be as low as possible for eliminating blobs.

I'm going to do a direct EZStruder mount for the E3D at some point, hopefully soon. That's probably the best way to get rid of blobs, because the Bowden tube hysteresis is 100% gone. Someone else mounted a QUBD extruder on his Rostock and he said that gave him the best print he'd ever seen. (And he hates Rostocks!)

[Eaglezsoar]

Yes. Take your extrusion width up to 0.45. Never leave it at the measured diameter. You have to have some margin for the nozzle end to trowel the filament flat and ensure that minor fluxuations in filament width don't result in less than the whole nozzle diameter discharging at the correct rate. A little too much is OK but not enough can potentially make things uneven.

Also, try printing at a thinner layer height. I'm going at 0.08 right now. There are blobs but they are so teeny tiny you'd never notice them unless you were looking for them, and you'd have to look CLOSE. The filament goes slower the thinner your layers are, and the running average speed should be as low as possible for eliminating blobs.

I'm going to do a direct EZStruder mount for the E3D at some point, hopefully soon. That's probably the best way to get rid of blobs, because the Bowden tube hysteresis is 100% gone. Someone else mounted a QUBD extruder on his Rostock and he said that gave him the best print he'd ever seen. (And he hates Rostocks!)

We've had many threads in the past about placing a direct drive extruder on the effector and I seem to remember that everyone decided that the solutions were too heavy to
work well. In your experience have you seen this working well enough to be a permanent solution? Which would you recommend, the QUBD or the EZStruder? The QUBD
would seem like it would have the lowest profile. The problem would be interfacing the QUBD to the E3D. Most of the ones I have seen on Thingiverse are designed to work
with the QUBD nozzles. Any thoughts from you would be appreciated.

[Cleveralias]

We've had many threads in the past about placing a direct drive extruder on the effector and I seem to remember that everyone decided that the solutions were too heavy to
work well. In your experience have you seen this working well enough to be a permanent solution? Which would you recommend, the QUBD or the EZStruder? The QUBD
would seem like it would have the lowest profile. The problem would be interfacing the QUBD to the E3D. Most of the ones I have seen on Thingiverse are designed to work
with the QUBD nozzles. Any thoughts from you would be appreciated.

As to the weight, I've also wondered whether the added weight of a direct drive extruder could be offset by negating the whole effector/extruder weight with counterweights attached to the cheapskates. It seems like it wouldn't be so difficult to implement since we could add a pulley to the top of the towers relatively easily. Am I off in my thinking or could this work?

[cope413]

When you add mass to the effector, you're increasing its momentum at any given print speed. That means it requires much more force to stop the effector or change direction. Counterweights don't change that. The momentum is still higher, and therefore your print speed potential while still maintaining precision will be lower.

So if you imagine you're printing a cube. As the head is traveling toward a corner, it will require much more force to make the 90* turn properly. If you were to keep all of your settings with a 100g effector, and print with a 500g effector, chances are that you would get overshoot.

Adding mass to the effector may increase print quality, but it will certainly limit the max potential speed of the machine, and it may also limit or lower the precision.

It's certainly do-able, that's never been in question. The question is why?

If you can't produce the results you want with a bowden, then that would seem to be a pretty good reason to try it.

[0110-m-p]

Couldn't have said it better myself.