Water cooling + 230V heated bed

Hello,


I'm upgrading the extruder to be water cooled since the temperature in the heated chamber is causing clogs.

I'm worried about safety (electrocution, fire hazard). The bed runs on 230V and .. well it's right below where the water flows.
If it happens to leak somehow, maybe things could catch on fire, or if I'm operating the printer when there's a leak, I might get electrocuted.

A third possibility is that the pump stops working and the cold end overheats -> also fire hazard.



What can I do to make this setup safer?


Some things I've thought of:
- Temperature sensor on cold end, stop printing when it becomes too hot. (however, I have no thermistor connections left on the ramps board)
- some kind of water leakage sensor? Checking the level of the water tank
- Maybe a big water sensor in the bottom of the printer which cuts the mains if it gets wet.
- ?

Edited 1 time(s). Last edit at 11/20/2016 06:19PM by FalloutBe.

What temp inside chamber? are you sure this is the cause of the clogs?
I bought a water cooler for the PC, never had the courage to fit it.
I'd be just as uncomfortable on the printer, but it has been done,

Edited 1 time(s). Last edit at 11/20/2016 07:20PM by MechaBits.

Not really condoning what you are doing but.... They make non conductive coolant for computers so they would not short out the high voltage if spilled.

Yeah I've seen whole boards submerged in some liquid, while it still functions, weird...cool... but still I just not ready to go there...bit like walking on glass bridges

Quote
FalloutBe
I'm upgrading the extruder to be water cooled since the temperature in the heated chamber is causing clogs.
...snip....
What can I do to make this setup safer?- ?

The obvious thing to do is to turn the enclosure temperature down and forget about water cooling. If you aren't controlling the enclosure temperature, start controlling it. If you're printing PLA, don't close the printer.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

How are you incorporating the water-cooled hotend? Utilizing properly sized barbed fittings and clamps is necessary. You don't need to fear water cooling if you build it right. First, if you have a heated chamber, I'm assuming your heated bed is insulated on the bottom as well. Heated Chamber or not, your electrical contact points should not be exposed. It should be appropriately sealed with or without water cooling.

A Failed water cooling pump or leaking is not going to start a fire. And if your pump fails, just as a fan can (which actually would likely fail first due to the heat in the chamber shortening its life) Your cold end will heat up and your hotend will clog, but if your thermistor is operating normally, it would keep your hotend regulated, which may degrade the filament in your nozzle, but it's not going to combust.

If you have your power and electrical board underneath the bed, then they should be isolated in their own box, such as the SSR (which should be anyways since its main voltage)

For example, I have a Kraken + 4 water blocks, with zero chance of electrocution if water broke loose onto my bed in a case of ultimate catastrophe. Even if I ripped the tubes off due to a massive crash causing massive water leakage (which wouldn't happen because I have quick disconnects on my Kraken). The thermistor connections and heater cartridges are also isolated from the pump lines. I have a Delta with a 5/16" aluminum plate surrounded by isolated 3/4" insulation, and then sealed on the bottom of the heater with additional 3/4" insulation. I have a 450W 110V silicone bed heater where the connections are high amperage XT60 connectors. Then the secondary wires extend far into a SSR and Power supply that is isolated and the wires go through small rubber grommets.

I don't even think its possible for water droplets to even enter the box.

I have a reservoir that has a temp gauge on the front just so that I can see if the temperature rises too much.

You do need to consider what you're using your heated chamber for. If its for ABS, you can get away with not using watercooling if you don't mind slightly weaker layer bond adhesion. You'll need external air funneled in or watercooling to properly print PEEK, PC , certain kinds of PP at any kind of large scale though.

---

M8K-X Quintuple Delta Printer (as in Make X)
E3D V6 Kraken + Syringe Extruder
Automatic Tool Change Mechanism (Kelvin Kinematic Mount) (WIP)
Auto-Filament Changer
Heated Chamber
Watercooled Components
380mm Printbite surface
Duet V.8.5. + DueX4
Recirculating Air Filtration
Sealed Filament Storage

Follow along with my trial and tribulations
www.m8kx.wordpress.com

I don't like the idea of watercooling for it's drawbacks regarding safety, and mass. To achieve good cooling in a heated chamber i switched from fans to a membrane pump (aquarium pump) that uses a hose to deliver the air to the hotend.
This needed a redesing of the direct extruder i use but turned out to be working nicely. The pumps are cheap and very reliable and over all the solution weighs even less than a fan would. Also the cooling effect is better than it was with a fan, i can push my PTFE/PEEK based Merlin to 260°C without problems now.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

The chamber will be heated up to about 50C and I'm willing to experiment with 70C, for printing big pieces with ABS.
Is it correct that regular destilled water works fine too instead of liquid PC coolants?

This is what I have for now:


E3D heatsink with 3D pen art and some epoxy.
ok, it's very ugly, but it's water tight and pretty light weight.
I'm just concerned if the two connections brake off somehow.

The bed heater is a silicone heating mat that's sticked to an aluminium platform. The wires are covered in siicone where they enter, but they only have a fabric sleeve.
It's powered by a SSR.

"A Failed water cooling pump or leaking is not going to start a fire"
No, but the hotend is mounted in a printed part, so it'll melt :/ and then the whole hotend might melt out of there, falling inside the printer.

Thanks for all the replies!

Edited 2 time(s). Last edit at 11/21/2016 07:10AM by FalloutBe.

Quote
FalloutBe
"A Failed water cooling pump or leaking is not going to start a fire"
No, but the hotend is mounted in a printed part, so it'll melt :/ and then the whole hotend might melt out of there, falling inside the printer.

No that bit isn't correct as was pointed out above, the thermistor will still be attached, the temperature will still be regulated, the heatsink fins water cooling block will still sink heat away passively even without a fan, it might even convect around the circuit. You might be pushing it a bit if you're using very high temperatures in the hot end like 260+ deg C. A fan/watercooler block, is to create a sharp transition from cold to hot, so that filament only melts/softens where it's meant to, its not a hot end cooling system. Heavily insulate your heater block with silicone/ceramic wool/kapton, this will improve heating times, use less power and allow more stable temperature regulation in use.

I have water cooled PC's, do it properly and it works, but I agree with the reservations about mass, is pumped air not possible? Maybe try a teflon coated heatbreak, they are not evil and are much easier to use than all metal. I think "all-metal" hotends are a bit of marketing spin, which unless you print at very high temperatures probably cause more problems than they solve.

I have a chamber which gets to 65 deg C printing ABS passively and no jams. With PETG it only gets to 40 deg C as the bed is cooler, with PLA I take the lid off.

Edited 2 time(s). Last edit at 11/21/2016 12:27PM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

At those temperatures, you might as well get external air funneled in, i mean that is how Stratasys does it. Also like DjDemonD repeated, temperature on the hotend is regulated, and it is still a heatsink and it will not get hot enough to melt it. Also if you were to use a fan internally such as on an E3D, your fan has a much much higher chance of failing than an external pump.

Also, no I do not trust that setup. I figured you would use a RC boat metal watercooling jacket. Your barbs are way too short. Furthermore, is the silicone heater completed covered on its underside and the silicone sleeve containing the power lines, its that covered as well? Or is it just feeding to your SSR ?

---

M8K-X Quintuple Delta Printer (as in Make X)
E3D V6 Kraken + Syringe Extruder
Automatic Tool Change Mechanism (Kelvin Kinematic Mount) (WIP)
Auto-Filament Changer
Heated Chamber
Watercooled Components
380mm Printbite surface
Duet V.8.5. + DueX4
Recirculating Air Filtration
Sealed Filament Storage

Follow along with my trial and tribulations
www.m8kx.wordpress.com

The barbs for the hoses are indeed quite crappy haha. However, it's a proof of concept for now. It'll have to change in the future for sure.
If I'll be using the water cooling anyways because I didn't know pumping air would be easier, so I'll probably go with that.

What kind of pumps do you suggest? What tube diameter?
And won't the air pump cool down the heated chamber? And if air goes in, air needs to go out, so the heated chamber won't be very efficient when heat is leaking.



EDIT: I'm not sure which of the two it'd be like, do you mean like a fan with a big tube attached to it, or more like a compressor with a small tube?



For the silicone heater, it's easier to just show you a picture:


Edited 2 time(s). Last edit at 11/21/2016 03:59PM by FalloutBe.

Since you kept the original heatsink, you can use an aquarium air compressor on the same barbs as the waterpump. At least, that's what I read from the posts above.
If you also use the outlet barb to guide the warm air out of the chamber, your temps will be stable. But I'm sure, the aquarium airpump won't feed that much air...
The only drawback is the noise of the pump..

I cooled my CPU's with demineralized water too and spilled it into the cpu socket more than once. The connections were getting green from moss, but one of the PC's is still running 20 years later

Oh yes of course! thanks
I can indeed just add another tube behind the heatsink to direct it back outside of the heated chamber.

So the aquarium pump is noisy? Then how do aquarium people live with that Aren't there silent ones, just like the silent aquarium water pumps?

They make some noise, but it is very bearable. If you get good quality pumps it is a soft hum in the background, not nearly as disturbing as small fans.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

"not nearly as disturbing as small fans"

Oh, then it'll be perfect! That was the goal as well actually to make something quieter than the small fan.

Keep in mind that the pumps can't deliver the same volume of air than fans. Their advantage is the much higher pressure. To make the cooling effective you have to guide the air tightly around the cooling fins. For me this was a big advantage with the Merlin hotend, since it has a very compact design. If you try to use a cooler that is designed for a fan you might not get the efficiency you need.
Edit: Don't worry about guiding the air out again, it is very likely hotter than the chamber air and much to little to make any noticable difference in the chamber.

Edited 1 time(s). Last edit at 11/22/2016 02:38PM by Srek.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

How many l/min would I need to effectively cool the hotend if I don't make any changes to it?

And is an aquarium air pump a good choise?
I think they are made more for higher pressures since they need to be submersed in half a meter of water.
Are there other pumps that are more designed for higher volume rates, but can still use quite thin tubes? (thin meaning smaller than 8mm)

Edited 1 time(s). Last edit at 11/22/2016 03:55PM by FalloutBe.

I use these
[www.amazon.de]
with 215 l/min
For the e3d i would print a new airguide, but the cooler is huge in comparison to what i use and compressed air cooling works best if you have a fast flow over the cooler, which will be hard to achieve with the original cooler, especially since it is thickest at the lower end. This makes sense if you want to spread the heat over as much of it's surface area as possible, but makes it hard to cool it with only the lower cooling rips, which is what i would try to do.
Even a much larger pump will be hard pressed to produce the volume you need to use the cooler as is. Keep in mind that the membrane pump i use has 215 l/min, while a good 30mm fan can produce 4000 l/min.
If you don't want to change the hotend at all maybe don't use compressed air but a large diameter (15-20mm) tube/hose with a radial fan at the end that pushes air through. Radial fans have a good deal more pressure than axial so you can try to mount it outside of the heated chamber.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Thanks a lot! I could only find aquarium air pumps of about 60l/min max. That one seems a lot better!

Isn't it possible to have a tube from this pump expand to the size of the heatsink of the E3D at the end? or won't that work?



Do you think a radial fan will be quieter than this pump?

If you increase the diameter the speed of the airflow and the pressure will be greatly reduced. At the end you have a lot less air than with any fan. The key advantage of compressed air is the pressure, which allows you to get a high cooling effect in a small space. You have to have a small cooling area though, otherwise it uust doesn't work.
Quiet or not only depends on the specific pump or fan. The pumps i use ate not super silent, but the hum they produce is much less annoying than a fan to me.
For me the key reason for compressed air cooling was the higher temperatures i can reach, i can now print PC with a hotend designed for ABS. The lower moving mass is the second reason.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

When using compressed air, it cools better because the air density is higher?

I'd like to use this method but I'm not sure how to modify my E3D. Do I need the fins to be thinner too then? More thinner fins in a more compact space?

You want a large surface area in a small volume, so finer fins are a good idea. Thy should not need to be huge though. With compressed air cooling you are heating up a small amount of air a lot, while with a fan setup you are heating up a lot of air a little bit. If you take a look at the Merlin hotend, the M10 thread on the PEEK is sufficient for cooling. In my setup i put a tube around the thread that is only slightly larger in diameter and let the air exit through a thin gap at the bottom towards the heatblock. Since the heatblock has a silicone cover this does not cool the heater. This design cools the lower part of the heat barrier the best, which is what you want, the gradient from hot to cold should be as steep as possible. I designed the tube around the hotend into the extruderblock, this way it is sealed towards the top and i don't need any mounting parts for the cooling. The air hose connector is integrated as well.

The part with the gradient is the M10 threaded part of the Merlin hotend, you can see that the air exit to the left is just a small gap. I left a bit of volume around the upper part (to the right) so the incoming air can be distributed more evenly around the hotend. The smaller the gap the closer the air flows along the hot surface and the more heat will be transfered out. If the gap size increases the cooling efficiency goes down very fast. If it is to small you need a lot of pressure to work against it. Finding the balance for this is a bit tricky.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Thank you for sharing
It looks like I'll need to redesign the heatsink.

Too bad I don't have any metalworking equipment :/ A lathe would be great haha


Or would there be any part available where the E3D's heat break part fits into?



Edited 3 time(s). Last edit at 11/23/2016 08:57AM by FalloutBe.

Hm, maybe you could do it just like i did with the Merlin. Use a 10mm Aluminum rod, put a M10 thread on it and on the inside it gets the thread to screw in the nozzle. You might not even need serious metal working equipment, a threading tool and a drill press might be sufficient for this. On the downside, this would give you a M10 thread for mounting, so you would need to modify the exturder for this. But then, since you have to seal of the top of the hotend anyway this is not neccesarily a big disadvantage.

Edited 1 time(s). Last edit at 11/23/2016 09:12AM by Srek.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

"and on the inside it gets the thread to screw in the nozzle"

Did you mean the thread to screw in the heatbreak?


I have threading tools for threading a hole, but not for the outside.
What if I just take an M10 bolt like this

Yes, i meant the thread for the heatbrake.
Using a M10 screw might be even better, the aluminum might be to good a heat conductor.

Edited 1 time(s). Last edit at 11/23/2016 09:41AM by Srek.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

to good? oh yes you want to keep the heat only at the bottom ; )
I'll give this a try!

If you want to see the cooling in action, this is a part that is currently printing using two compressed air cooled extruders with Merlin hotends
[youtu.be]
In the sound the pumps sound louder than they are, since the printer itself is pretty silent.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Thanks a lot for the video!
Sounds like a pretty quiet printer

So this is the pump from Amazon which you linked me, right?

Yes, two of them, one for each extruder

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]