New all metal hot end modification

[ayahoo 0]

Hi All, I started to think about redesigning my hotend, the main Idea was easy to replace V2 hot end, I wanted to stay as close as I can to the current design, without modification to the head \ nozzle, and to try and make "simple" parts to manufacture to lower the cost of the design.

 

My Idea was using Titanium "Core" to replace the current brass connector & the peek, and to use aluminum hotsink to cool it.

 

I must say the according to the computer the results are very good, about 100deg C less than the peek at the end of the hotend.

 

What do you think about the design?

 

[snowygrouch 1]

Since the pics are not labelled, im just guessing here from looks- but the RH one is the standard V2 hotend and the

LH one yours?

If so, the design on the left looks worse than the V2 hotend from your pic. The hotzone is extending further up

than the V2 hotend.

The design on the left is (if I am interpreting correctly), showing upper side peak temp of 179 Deg C vs 25 Deg C for the V2 hotend. Which means heat is getting higher up, which is bad.

Also stainless steel would be better (16 vs 21 W/m/Deg K) than titanium from conductivity perspective. As well as cheaper.

Unless I am missing something with your pictures you attached.

 

[ayahoo 0]

Hi snowygrouch, You are right, I played with so many simulations that I uploaded the wrong one :shock:.

Regarding your comment on stainless steel vs titanium, if I use 6Al-4V, it is 6.7 W/m/Deg K, b which is much better than the 303 stainless steel.

Although you can see that the temp in my design is higher than the peek (70 vs 28) the heat goes up gradually down the tube vs the design in the peek which is jumping from 28 to 200 when the material goes from the peek to the brass connector, and I think 70 C should be OK to work with, with no problem of peek melting at 320 deg (and the simulation is on a "stationary" state, when there is no air flowing, while using the heatsink you get much bigger surface area so when the head is moving i should get better results).

what do you think?

 

[nick-foley 5]

I wonder if a stronger design from a cost+effort perspective might be to ditch the titanium for SS and instead just add a small fan to the heatsink? I'd be interested to see that simulation.

 

[snowygrouch 1]

Hi,

I would only comment further about the contact area between block and upper hotend. See pic attached.

Also its true it will work alot better when its moving with the airflow...however alot of plugging problems

stem from when people have the thing sittting idle but heated up. So its still worth considering what

happens in that situation.

C.

 

 

[ayahoo 0]

I wonder if a stronger design from a cost+effort perspective might be to ditch the titanium for SS and instead just add a small fan to the heatsink? I'd be interested to see that simulation.

 

Using a fan will help alot, but I'm looking for a design that will be problems proof, and fan is alway a point for problems.

Anyway i managed to burn the peek... so I will make the hotend today or tomorrow, I desided to change the desing, go with SS 303 and make it much longer, and by this desing I solve 3 problems, first. All metal no problem of burning plastics, second is the bowden tube, I make the hotend go all the way up to the top of the head, 90mm so I can connect an industrial grade connector to the end of it and never have the problem of disconneced tube. And last is the leakage at the top of the aluminium block. this is the answer to snowygrouch.

I'll post a pic of the simulation & the real part tomorrow :-)

[ayahoo 0]

This is the new design, I've simulated it also with 400C with no problem, adding a fan will also help to minimize the hot zone, but the overall looks very good to me, it looks like it would be very easy to handle & fix problems (since the tube is connected directly to the hotend) & no modification is needed to the head of the ultimaker.

 

[baasb 6]

Wow, that looks great. Also, nice drawing/technical skills :wink:

Are you able to manufacture this hotend?

 

[ayahoo 0]

Actually yes, and it is working but not 100%, I think I need to polish the hole so the material won't stuck inside the "barrel", but I already have a better design than this .

Trying to find the right material to use, I'll update soon.

P.s.

Thanks for the complements

 

[ayahoo 0]

Ok, I would like to hear you thought about this design, I'm thinking about using calcium silicate as a connector between the hot end and the tube, according to the simulation and the datasheet (0.058 W/m/Deg K) it should be great material with melting point over 1000 deg C.

the material is cheap and I need to get a sample to check if I can work with the lathe on it.

any thoughts?

 

[gluetools 0]

Which software package are you using to model this? (Solidworks Pro?)

bob.

 

[Daid 306]

Not to think less of your efforts. But remember there is more in play here then meets the eye and the simulations. You won't know how your hotend really works with PLA in it, and how it responds to retractions and backflow, until you test it.

 

[thingster 0]

AYahoo, may I ask you what program do you use for simulating the heat flow?

[ayahoo 0]

I'm using Solidworks to model & simulate it.

Not to think less of your efforts. But remember there is more in play here then meets the eye and the simulations. You won't know how your hotend really works with PLA in it, and how it responds to retractions and backflow, until you test it.

I know, but like anything else in life, you try to think about the problems before making the final product, and simulating & thinking about potential problems is better than just making the part & than thinking about the solution .

 

[Daid 306]

I'm using Solidworks to model & simulate it.

I know, but like anything else in life, you try to think about the problems before making the final product, and simulating & thinking about potential problems is better than just making the part & than thinking about the solution .

Oh, but my comment doesn't come without background. I've seen a small stack of well thought-out but not working properly hotends. (Sadly, that's all the information I can share)

 

[snowygrouch 1]

I dont think Calcium Silicate has useful mechanical properties, and I would be very surprise if you could clamp into lathe jaws without crushing it like chalk (EDIT, yes Calcium Silicate has a compressive strength of about 600kPa....so thats about 100x less than ABS plastic and in fact slightly LOWER than chalk)

MACOR (developed as a machinable ceramic) is a much better material if you want a high temp insulator, its a little more conductive - but

its not really expensive, its machinable and quite strong. Its also stable to extreme temperatures

far higher than you would ever need to print any thermoplastic.

C.

 

[ayahoo 0]

I dont think Calcium Silicate has useful mechanical properties, and I would be very surprise if you could clamp into lathe jaws without crushing it like chalk (EDIT, yes Calcium Silicate has a compressive strength of about 600kPa....so thats about 100x less than ABS plastic and in fact slightly LOWER than chalk)

MACOR (developed as a machinable ceramic) is a much better material if you want a high temp insulator, its a little more conductive - but

its not really expensive, its machinable and quite strong. Its also stable to extreme temperatures

far higher than you would ever need to print any thermoplastic.

C.

 

Thanks for the input, you are right, I went and check and this material is not suitable for hot end, and after many simulations & designs, I manage to choose the one I think will be simple & effective.

I'm using aircraft grade titanium (grade 5) M6 bolt, it has very low heat conductivity, and should be easy to use & make, I believe the whole cost will be lower than 50$ material & manufacturing, so I've ordered the bolts, and I'm waiting for them to arrive to check.

thanks for all the inputs

 

[ayahoo 0]

Ok, So I promised to update, so it took some time for the items to arrive, but now that I finished making the hotend, I must say it is working GREAT!, I'm uploading a video & a photo, that shows the hotend, I'm pushing the filament with my hand with very little force, and the top side of the hotend is cold... no modification is needed for the ultimaker, not even to recalibrate the platform, total assemble time is about 5 min

so bye bye plastic parts!!

 

 

 

[ian 32]

looks very cool. are you going to do some test prints, would love to see how it works with lots of retraction... :-)

Ian

 

[Daid 306]

looks very cool. are you going to do some test prints, would love to see how it works with lots of retraction... :smile:

Ian

 

Yes. Retraction. Print this one:

http://www.thingiverse.com/thing:34942

At "full machine scale" with retraction enabled. That's how I've managed to kill every "all metal" hotend so far. (works with the stock hotend)

 

[ayahoo 0]

I will try as soon as I get the direct drive mod, since I managed to tear one short belt :(

Just excuse my ignorance in 3d printing... what is the benefit of using Retraction?

 

[SittHein 0]

Ayahoo, I am beginner to SW stimulation and trying to test some thermal analysis. Let's say for aluminium, I only know Thermal Conductivity (W/mK) of ~200. How can I convert it into Convection Coefficient (W/m2K) which is required in SW stimulation? Any maths or assumption behind?