Water cooling questions

A place to talk about user-made mods and upgrades to their machines
BenoitD
Plasticator
Posts: 8
Joined: Thu Jul 06, 2017 3:58 pm
Location: Montreal, Canada

Water cooling questions

Postby BenoitD » Thu Aug 10, 2017 10:51 am

Hi,

I just want to know if anyone can tell me why (except price) we don't see more water cooled hot end. I would think for delta 3D printer, it would be great to have a zesty nimber extruder with a water cooled hot end. E3D seems to say that it is coming soon, but when.

Regards,
Benoit Despatis, P.Eng.
Chief Engineer at Sanuvox

User avatar
Jimustanguitar
ULTIMATE 3D JEDDI
Posts: 2618
Joined: Sun Mar 31, 2013 1:35 am
Location: Notre Dame area
Contact:

Re: Water cooling questions

Postby Jimustanguitar » Thu Aug 10, 2017 1:55 pm

It's a lot of cost (like you mention) and an increase in parts count and complexity to address an issue that very few people actually have. Air cooling works just fine except in heated chambers or at super high printing temps. Plus, the cooling block and plumbing (and the coolant) would add inertia to the movement of the effector, which is in opposition of what you're likely trying to get out of the Zesty extruder...

I'm no expert on that sort of a setup, though. What problem are you experiencing or trying to find a better solution for?

User avatar
Xenocrates
ULTIMATE 3D JEDDI
Posts: 1444
Joined: Wed Sep 23, 2015 2:55 pm

Re: Water cooling questions

Postby Xenocrates » Thu Aug 10, 2017 2:51 pm

Since I've been looking into it, I'll tell you about some of the problems I see with it (Since I have most of the parts already anyways).

Firstly, if you print with moisture sensitive materials such as nylon, you end up needing to go closed loop on the cooling, increasing cost, as you need either a much larger reservoir or a radiator (And it needs to match your other in loop materials, you don't want to mix copper and aluminum while dumping a lot of heat into the system). And while most computer radiators are overkill (An I3-6100, which is a relatively low end desktop CPU has a 51W thermal design power, which is meant to be kept at a much lower ΔT than a hotend), you still want a reasonable size (120MM), to keep the fan speed down and thus noise, as well as to allow you to have a lower circulation rate to reduce pump noise.

Secondly, most hotend manufacturers don't make them available as a default. Meaning if you want a watercooled E3D setup, you need to either pay for a custom machined part (Quotes ranged from setup cost of 150+20 per part, to 45 per part without setup, MOQ of 3 from the various places I checked with), or make it yourself (Which takes some semi-specialized tooling, as you really need the right diameter of stub endmill with the quality of most of the open use or hobby lathes, and a metalworking lathe is rare in it's own right).

Thirdly, if the chamber is being heated, which as Jim noted is one of the few reasons to do watercooling (Other than high density stuff like the Kraken), then your routing of tubing must either avoid the heaters, or be capable of withstanding direct contact, as well as potentially the temperature rise should the pump fail (as having it dump water over your power supply, electronics, and a bunch of other equipment is very bad). This means that to get acceptable flexibility and temperature resistance, it needs to be specialized tubing, and somewhat low diameter to play nice with fittings. This stuff isn't particularly cheap, at around 30$ for the low diameter connections I planned for the extruder to hotend and hotend to pump runs, and another 20$ for the shorter, higher diameter stuff to work with the other runs and their fitting sizes.

Fourthly, as is the big cost factor in any water cooling setup, fittings. If you're using aluminum, it has to be all aluminum, or if you're using copper, it should all be copper. Meaning you need to use either plastic or matched fittings, instead of the cheaper brass stuff, especially if you end up needing an anti-freeze agent in the coolant like I do (And that stuff isn't cheap if you don't want garbage premixed stuff without the corrosion inhibitors and no way of making sure you have the correct ratio), I found that using aluminum fittings, as well as the disconnects I chose (All high temp plastic, no metal contact in the flow path, self sealing on both sides) added about 50$ to the base cost, as well as 30$ for any hotend that needed to be equipped with them (Plus another 10$ in tubing mind you, and the water jacket and heat exchanger themselves), making each hotend conversion something like 100-120 $ in parts if you want to have the ability to swap them out without dumping water everywhere.

Fifthly, longevity. Any mixtures of metal types invites bimetallic corrosion. Aluminum which is not anodized can see that occur somewhat quickly, and the thin nature of radiators, as well as the usual lack of internal anodizing means that if you aren't very careful, it will develop leaks. The seals also need attention paid to them, especially on the heat exchanger and water jacket, as due to it being mobile, having the highest temperature gradients, potentially being exposed to acetone or other chemicals as you clean the bed, and having the highest pressure in my setup as it's the lowest point, any seals not designed for these sorts of conditions could degrade rapidly.

Sixthly, as Jim again notes, the additional mass from the waterlines (As well as their resistance to flex), the water in the block itself, etc, may add more mass than it takes away. In my case, the ability to run the chamber at higher temperatures, as well as the elimination of the barrel fan in the chamber are a large part of my motivation. It's expensive, technically complex, and requires a large amount of specialized components, some of which need design and fabrication. Unless you're working in the upper end of the machine capacity, you won't ever notice it (I intend to work with MG-94 ABS and PMMA at a nozzle bore of .15 mm, using relatively high chamber temps and other exotic idiocy to produce models and push the limits, since anything I work out can potentially turn into job related experience later, thus validating the effort, if not the cost)

In short, I would recommend against a retrofit of an E3D or other existing hotend, unless you have someone willing to sell you an effectively tested solution. But if you plan on any of the following, I would look into it.
Chamber temperatures above 40C
Hotend temperatures nearing 300C or greater, or heater cartridges above 50Ws
Nozzle apertures at or below .2 mm
Work with materials such as PEEK, PEI, or other engineering plastic filaments, where the cost of a failed print is high, and post processing the exterior is insufficient remediation.
You have a research budget or grant, or need a project for engineering students or machinists.
You can't leave well enough alone, and need the craziest piece of hardware on the block (This is honestly pretty close to my motives)
Machines:
Rostock Max V2, 760W corsair modular PSU, PT100 enabled E3D V6 and volcano, self adjusting carriages, Raymond style enclosure
Automation Technology 60W laser cutter/engraver

Sic Transit Gloria Mundi
01-10011-11111100001

User avatar
Jimustanguitar
ULTIMATE 3D JEDDI
Posts: 2618
Joined: Sun Mar 31, 2013 1:35 am
Location: Notre Dame area
Contact:

Re: Water cooling questions

Postby Jimustanguitar » Thu Aug 10, 2017 3:07 pm

Xenocrates wrote:Nozzle apertures at or below .2 mm

Work with materials such as PEEK, PEI, or other engineering plastic filaments, where the cost of a failed print is high, and post processing the exterior is insufficient remediation.


Can you expand on those? Why does water cooling aide smaller nozzles or improve production reliability?

User avatar
Xenocrates
ULTIMATE 3D JEDDI
Posts: 1444
Joined: Wed Sep 23, 2015 2:55 pm

Re: Water cooling questions

Postby Xenocrates » Thu Aug 10, 2017 4:45 pm

I have had issues with heat creep, which is partially due to the chamber, but also due to the low flow rate of the filament, which has caused jams at times. Watercooling removes the heat creep, as it is not heating with the chamber, but also works with a lower ΔT, meaning cooling is more efficient.

As for the second one, the first two listed are high printing temp (340 to 370 for PEI, 370-390 for the PEEK), and generally want a high chamber temperature, while to mimic the Stratsys machines, even ABS would be printed at a relatively high (70-90C) chamber temp, which for me tends to cause both heat creep issues, and a higher chamber temperature helps with interlayer adhesion, as well as reducing warping in 100% infill parts.

This is anecdote, and applying common sense to the physics involved (And limited thermal simulations using Fusion 360), meaning that it may very well not be repeatable, may be totally off base, and definitely isn't everything involved. For another variable, my setup also cools the extruder motor, to keep the drive gear temperature down, and avoids softening it too much there.
Machines:
Rostock Max V2, 760W corsair modular PSU, PT100 enabled E3D V6 and volcano, self adjusting carriages, Raymond style enclosure
Automation Technology 60W laser cutter/engraver

Sic Transit Gloria Mundi
01-10011-11111100001

User avatar
Jimustanguitar
ULTIMATE 3D JEDDI
Posts: 2618
Joined: Sun Mar 31, 2013 1:35 am
Location: Notre Dame area
Contact:

Re: Water cooling questions

Postby Jimustanguitar » Fri Aug 11, 2017 8:02 am

[img]picture_of_thumbs_up_that_i_was_too_lazy_to_go_find[/img]

BenoitD
Plasticator
Posts: 8
Joined: Thu Jul 06, 2017 3:58 pm
Location: Montreal, Canada

Re: Water cooling questions

Postby BenoitD » Fri Aug 11, 2017 8:58 am

Thanks.

My goal is more to have the better setup and learn from the best. I was able to print 1 Kg of PLA+ in less than a week because of all your knowledge and tips. My needs are more on the "print fast" and test now our designs. I think your opinions will help slow down my desire for water cooling for the moment. I will start my other type of prints with Polysmooth and Isopropyl Alcohol.
Benoit Despatis, P.Eng.
Chief Engineer at Sanuvox


Return to “Mods and Upgrades”

Who is online

Users browsing this forum: No registered users and 2 guests