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

This post blurs the line a little between information and advertisement - I really don't want to spam, but I genuinely think this is of interest to most of you here and I will be posting information about design/engineering decisions and rationale for critique and for the benefit of others. Along with info and discussion about printing Polycarbonate and other high temperature plastics.

After much ado all our parts have arrived from the machinists. They look so much better than we could ever have imagined, worth the wait - check out the images below.





We have two kinks in our supply chain - 30mm fans have not arrived in quantity and we have had to make a last minute amendment to the printed fan-mount which we are printing as many as we can as fast as we can - but still can't keep up with demand, which has taken us by surprise, we really didn't expect this many preorders.

So, we can ship you a hotend tomorrow, but only if you supply your own fan and a way to point said fan at the heatsink. Cooling is required for this hotend. £3 discount for this option!

To all our early adopters - THANKYOU. Your orders are our absolute priority, Dave and I have been working literally from sunrise to sunset to get these to you ASAP. All our early adopters (except a handful placed in the last few days) who have placed an order will have their items dispatched in the next 2-3 days maximum - most going out tomorrow. Those who have ordered in the last few days will ship by Friday when the next batch of fans arrive.



It's crazy how much time packaging things properly takes!


Lots of heater cartridges....


Cutting and packing little lengths of high temp heat shrink, and special high temp wire takes a loooong time...

The parts look and perform really really nicely, extrusion is smooth and easy as well as precise and immediate. Cooling is also really good, with the extruder running at 230 printing ABS over a 110C heatbead I can hold my finger on the heatsink which is only 2.3mm away from the heater-block and it feels barely above ambient. I have taken the hotend to a temperature higher than I can measure, at this temperature I can freely feed in PLA, but it exits the nozzle in a powerful jet of vapour instead of liquid plastic. I can still hold the heatsink in my hands while doing this.

Videos going up of some printing tonight with ABS and PLA - just a teaser, tomorrow I will show off printing Polycarbonate, Nylon, PVA. I am trying to find some PEEK and FEP that I can extrude - Maybe I can print a J-Head?

More updates to follow.

[e3d-online.com]

Sanjay

If I buy it without the fan mount can I get the stl to print my own?

Edited 6 time(s). Last edit at 04/05/2013 11:11PM by hendo420.

Yep, its going up now on the downloads section of our site and thingiverse.

Looks great! I placed an order just a couple of days ago. Can't wait to get it!

---

Help improve the RepRap wiki! Just click "Edit" in the top-right corner of the page and start typing. Anyone can edit the wiki!

How does this mount? My J-head mounts with a hollow threaded bolt, 6mm OD 2mm ID can I use this same bolt to mount your hotend?

My extruder just has a hole to mount my hotend. Will I have to make a custom bracket?

Active cooling is a dealbreaker for me. I already have a head fan to cool the object being printed, the last thing I want is a hotend that needs an additional fan blowing more air around in order to work properly.

With many many hotends available that do not require active cooling, what is the advantage to making a hotend that *does* require it? Can you print faster or something?

EDIT

Just saw your post on reddit explaining the thermal transition advantage. I understand the theory behind this, but I have yet to see it make a meaningful difference in print quality. For example, I've seen Jo Prusa churning out prints on his i3 with one of his prototype nozzles and it looks flawless.

Have you by any chance done empirical testing with existing hotend designs, and different configurations of yours, that quantitatively show an improvement in print quality when using active cooling? This would convince me that maybe there's something to this approach.

Edited 1 time(s). Last edit at 04/06/2013 01:42AM by crispy1.

Just placed my order

Nice to see a uk hot end

Printing the peek tube for a J-head would be quite a sacarstic jab

What is the nozzle size. I have looked on the web site but while I can see selection of filament size, but not the nozzle size, nor any definition of what is normally shipped?

David

Nozzle is 0.4mm if you check the drawings on the site.

i have a prusa 2, will this hot end fit between the steel rods?
and is there a duct fan design that works on this to cool the plastic coming out (i need one for bridging stuff)?

i'm still waiting for prusa his nozzle to come out, so i'm willing to test this one in the meanwhile.
it looks awesome.

How do I order? there is no order form or contact details on your website.

TC

TCase Wrote:
-------------------------------------------------------
> How do I order? there is no order form or contact
> details on your website.
>
> TC

Click here: [e3d-online.com]

---

Help improve the RepRap wiki! Just click "Edit" in the top-right corner of the page and start typing. Anyone can edit the wiki!

Hendo420 -
It uses a groovemount system that is nearly identical to a J-head (with the tiniest smidge more space for the screws as aluminium doesn't deform as much as PEEK in a JHead). You should be able to use any J-Head groovemount type extruder. What printer are you using?

crispy1
There are a few reasons we went for a short thermal break -

1) First is that it confines the melt point/transition to a very specific place, which gives consistency in many aspects of extrusion. This is borne out in personal experience, but you are right - some sort of more quantitative and empirical measure of this is really needed for good comparision. The theory is certainly there, and stratasys detail this very very exhaustively and go to HUGE lengths to ensure a consistent and 'locked' melt point on their FDM machines.

I am unsure of a good way to measure this - high framerate slow motion video of extrudate exiting the nozzle with a view to observing a sharp start and end to extrusion would be really nice.

Perhaps I will retrofit an extruder with a long break and try to do some comparison prints using the same GCODE on challenging items such as many towers where stringing may occurr. Suggestions welcome.

2) Extrusion force - As filament goes above its glass transition temperature it becomes rubbery, buckles somewhat, and begins to adhere to the walls of the tube it is being forced down. Once filament liquifies it no longer impedes the travel of filament behind it (setting aside backpressure etc). The longer your transition the longer your amount of rubbery filament, more rubbery filament - more adhesion, more adhesion means more force. This we have absolutely experienced this too - but if I get an extruder with a long break rigged up I will try and measure it.

3) Manufacturability - You will notice how long it has taken Prusa to get to market, drilling these long holes in 300 series stainless steel is not easy, or cheap. Then talk about reaming/polishing the inside too. Prusa seems to think that a 1.75mm version of his design might not be possible, and I would agree with him, the hole is too deep and too thin.

The reason we can sell these for the price that we are is because this whole setup was designed with a specific lathe and specific tools on that lathe which makes manufacture fast and easy, therefore cheap. It seems like other hotend designers have come up with thermally fantastic designs but have utterly disregarded the fact that someone actually has to make it!

alj_rprp

In all seriousness though, I genuinely think PEEK is a possibility - our efforts to get some made for us have so far been in vain. However we are looking into self-extrusion of granules to get around this. I would also love to print FEP,
(A fluoropolymer like PTFE in many ways, but can be melt processed) would be amazing for bearings and such, but I imagine it won't stick to anything... ideas? Perfboard?

davidgoodenough
Yeah, that is a stupid omission, we have update our site to include this prominently. We are moving to an entirely new and improved website/hosting soon so watch this space.

possenier
It's tight between those rods with them being 50mm apart and the hotend being 30mm at its widest, but it does fit! (Just) Depends a bit on the XCarriage - Kuehling style ones fit beautifully. Gregs style ones are a bit small between the bearing holders, and you would need to nibble away about 1mm of plastic on the bearing holders to fit. Alternatively you can duct/mount the fan elsewhere or integrate a fan mount into an XCarriage.

TCase
The link that Cameron posted should take you right there - the website isn't great, and we are trying to change it rapidly. If you are having any problems I apologise. You can always PM me here on the forums to order.

SanjayM Wrote:
-------------------------------------------------------
> Hendo420 -
> It uses a groovemount system that is nearly
> identical to a J-head (with the tiniest smidge
> more space for the screws as aluminium doesn't
> deform as much as PEEK in a JHead). You should be
> able to use any J-Head groovemount type extruder.
> What printer are you using?
>
Rapidbot 3.0
[makemendel.com]

It uses a hollow m6 bolt that attaches to my 1.75mm J-head where the set screw would go. I have a 3mm J-head and it will attach with a hollow m8 bolt(I'm still looking for one)

I've been looking for a way to convert my extruder to groove mount but am not having alot of luck designing my own. I'm trying to avoid putting more holes into the acrylic plate that is the base of my extruder.

Edited 1 time(s). Last edit at 04/06/2013 03:27PM by hendo420.

> In all seriousness though, I genuinely think PEEK
> is a possibility - our efforts to get some made
> for us have so far been in vain. However we are
> looking into self-extrusion of granules to get
> around this. I would also love to print FEP,
> (A fluoropolymer like PTFE in many ways, but can
> be melt processed) would be amazing for bearings
> and such, but I imagine it won't stick to
> anything... ideas? Perfboard?

An important point is that FEP is relatively safe to melt when PTFE outgass some nasty stuff and also deteriorate very near its melt point (320°C for pure PTFE). Melt point is lower too (280°C ???)

I know of 2 standard methods for getting teflon deep coatings sticks to other stuff. I suppose they should work for FEP too :

- Diamond dust pyramids embedded by pressure in the interface layer. not really accessible for DIY 3D printing.
- Use an intermediate layer of nylon with very short and fine glass fibers in it. This method works and is quite cheap but reduce a bit the thermal capability. It is used though to keep in place PTFE seals in steam valves that tend to creep. Problem is that one the PTFE gripped the fibbers, dont even think to separate it.

What could work but need to bit tested is a porous ceramic tile. I imagine the tile would have a limited life (saturation).

There is probably other but I'm not a chemist.

As for finding peek, you are in UK no ? so, one of the main manufacturer of PEEK is english (VICTREX)

. I dont know if they sell retail and you have probably to use either powder or pellets and make your own filament, though.

Can this hotend be used with a bowden setup such as on the reprappro mendel?

Thanks.

Ben

Very very nice Job Sanjay and Dave!! The Berrybot3D would be proud to have a hotend like yours!! Also very very nice job on the web site!!
I think this hot end will make a lot of printers very happy!

Werner

Just placed an order. Great job on the hotend, hope it performs as good as it looks.

Well, its here. Looks good. Wont fit a Sells carriage, but I can work around that.

The assembly instructions are not yet on the website, most importantly theres no information on the thermistor.

Hi All,

Been working flat out today getting these shipped and need to go to bed.
Will reply to all comments/questions as soon as I can.

Assembly manual is going up tomorrow.

Thermistor is a pretty standard 100K EPCOS B57560G104F job.

Demand has been way higher than anticipated, and we are a little swamped, but shipping everything very rapidly now. We are just waiting on fans which arrive on Thursday/Friday. You can order one without fan with next day dispatch in the meantime for a discount.

It *should* work with any J-Head/groovemount printed bowden adaptor type setup BUT we are working on getting a really nice neat integrated bowden solution sorted out.

Sanjay

Nice to finally see this. I seem to recall that you draw inspiration from the Emaker hotend, which was very good.

I think you've taken the idea to greater heights.

Sanjay, do you know the exact weight of the complete hot end (sans fan and shroud)?

Everyone else, do you know the weight of the different hot end options, or possibly a link to a thread where hot end weights have already been discussed? And any thoughts of possibly doing a version that has thinner machine heat sink fins? Shaving it down could same some precious weight for those who may mount 2 or 3 of these on a carriage.

Edited 1 time(s). Last edit at 04/09/2013 11:26AM by maddox.

Very interested in this head but the website seems to be down / empty!

---

---

PrintB.it

just a question i'm intending to get one of these for my build so if i get a piece of aluminium cut for the groove mount no heat will get transferred to that?

I ordered one without a fan.

Can I print the fan mount in PLA or do I have to print it out in ABS?

Also, what are your expected shipping times to the US?

Hi all,

The full assembly manual is located here: [www.scribd.com]
Lots of lovely photos etc, but if there is anything you are unclear of just let me know and I can update/revise the manual to make things clearer.

Lots of questions about the weight:
The weight of all the metal parts is 45g combined.
If you weigh everything at once including electronics, fan etc, it comes to 96g, but this includes 2 whole meters of very heavy gauge heavily sillicone insulated wire from the heater cartridge, most of this wire will obviously be off-carriage/non moving mass. The whole heater cartridge incl wires weighs 25g alone, so your on-carriage weight is looking at about 75g including fan etc.

These weights really do include everything, and we haven't scrimped on good quality heavy gauge components, like including good quality sillicone thermocouple heat-resistant wire, as well as including high temp PTFE heatshrink for joining to the thermistor, using high quality hex-socket hardware all over etc, you even get a patch of Kapton for insulating the really hot areas. Check the assembly manual to see what I mean.

Weights breakdown!:
Sillicone thermistor wire + PTFE heatshrink for thermistor = 3g
Thermistor : <1g
Heater-Block: 7g
Nozzle: 3g
Stainless Heat-Break: 2g
Heatsink: 35g
Fan: 10g
Heater Cartridge: 25g
4x Hex Machine Screw + 1x Knurled Cup Faced Grub Set Screw: 4g
Printed Fan-Mount: 7g

Some very minor weight differences between 3mm and 1.75mm, but this is below 3g in total change.

maddox
We are very much looking to reduce weight, and the sink is the obvious target here. Thinner fins is on the cards. We also will look to be integrating a bowden coupling system that is totally integrated and very easy to setup.

printb.it:
Website seems fine for me? Don't know what's up. We don't like the current flash/moonfruit based site, it will be gone and changed for a more superior site in a few days.

chris33
Yep, absolutely no heat will be transfered from the mounting area to the carriage/mount. The heat is totally eliminated by that point.

hendo420
I have looked all over for the dimensions/specs on this J-Head you are using, and can't find data about it to answer your questions about mounting. Sorry about that. We have used PLA fan-mounts quite happily for long periods. Try and orient the block so it is away from the mount to reduce radiation. Do remember it does require a fan on the heatsink.

Cheers,
Sanjay

Sanjay, thanks for the reply. Yeah, thin out those heatsinks by like 50%, and you can consider me sold on at least 3 (as well as a consideration of having you supply many more, if I decide to market the machine that I'm building).

Maddox:

We have a bowden-multi-struder in mind that uses a common heatsink, and could be very compact even with 4x nozzles. We can probably keep a full Quadstrusion setup well below £100 all in, including bowden tubes, electronics, couplers, basically everything you need. Watch this space.

@Sanjay Thanks for the response. It all seems to be working now. I was just getting a large grey box with a moonfruit link at the bottom.

---

---

PrintB.it