Watercooling the Printhead

[Simon2.4]

21 hours ago, Todd said:

I didn't think it was possible but I received a quote more absurd than the one you shared with us. Check this one from Fathom out...

What material/process are you getting this done with? FYI copper was around 270$ for my part but I'd rather have Aluminium for the weight anyway. What feedback did they give you after looking at your part? I don't know if you checked out the guide I posted above for part design:  and SLM printing: https://www.materflow.com/en/design-guidelines-for-metal-printing/

Given the internal structure of your part, that may explain the obscene quote. If you intend to use https://craftcloud3d.com/ and get a decent quote, you are probably going to have to respect these guidelines.

In my part there:

-are no walls with an angle below 50 degrees from horizontal except the crown thing on top which will require external print support that is easily removed.

-is no structure thinner than 1 mm, though the process can do better, this will produce better results. The automated review process rejects parts at 0.8mm and below I think

[Todd]

The Craftcloud3d.com quote for aluminum was $40 all in for the same file. I did modify the water tube connections, but that price was even after the mods. The piece meets all the restrictions. It's clear that the >$2k quote was a "don't want to be bothered with this" price.

[Todd]

Try to stay away from band-aid solutions like silicone inside joints. It WILL fail before you expect it to. The rings look like they will serve nicely, but you may need to experiment with i.d. if you see any seepage or if it becomes too difficult to set in place.

[Simon2.4]

3 minutes ago, Todd said:

Try to stay away from band-aid solutions like silicone inside joints. It WILL fail before you expect it to. The rings look like they will serve nicely, but you may need to experiment with i.d. if you see any seepage or if it becomes too difficult to set in place.

As you can see, "band-aid solutions" is not my style. Every part has been pressure tested as it was designed and prototyped I can print those rings in 5 minutes so iterative design is no issue. I'm just waiting on FedEx to deliver the printed part to put it all together. Not looking forward to accessing the bottom. That's an adventure every time; It weighs about 100 lbs:
 

[Simon2.4]

I got the printed aluminium part yesterday; it looks gorgeous and the walls are airtight. I do have a major issue though, the base material powder that expectedly fills the whole inside is quite cohesive and does NOT just pour out. The powder is extremely fine and has the aspect of icing sugar.

Initially, I used a drill bit to dig out the two tubes but that's all I could get out by pure mechanical means.

I then put 100% Isopropyl alcool down the tubes and that was able to put some of the powder in a suspension that I poured out; it worked well for the lower half of the internal volume, but very little came out the top half.

Next I added refrigerant from a 'dust off' can: its super low viscosity so it penetrates the powder cake and expands while evaporating so I was hoping that would loosen up the powder. It didn't..

So now there is a tiny passage that lets pressure through but it's still essentially blocked and I've almost run out of ideas to get the powder out. Here's 2 ideas of at more 'extreme' part of the spectrum:

1 - put water in it and heat it with a torch to try and use steam to loosen the powder

2 - use a pressure washer to force a 1000psi driven jet into one side, hopefully eroding its way to the other... This would need some sort of vice jig to hold it in place..

Last resort will be to drill through the bulkhead and shorten the coolant path:

@Todd how's your adventure going? did you start work on the peripheral stuff?

[Simon2.4]

Version 5...

Single chamber with bigger volume this time; the alcohol trick should work ok with this design:

And some more design guidelines for SLM printing...

https://www.sirris.be/removing-raw-material-when-designing-additive-manufacturing-0

 

 

[Todd]

Another idea you may want to try…put the part in a pot of water on the stove and bring to a slow gentle boil with a little olive or vegetable oil in it for an hour or so. The idea is to simulate a cutting oil like would be used when machining aluminum. Another approach would to use an ultrasonic vibrating jewelry cleaner with finely ground media in it.

 

as for the adventure, I am nearly finished in Paraguay, then off to Santo Domingo for a couple of weeks before finally getting back home for a week. Then, Uruguay for a couple of weeks. Can’t wait to get a break and put the ideas into action.

[Simon2.4]

If someone else want to get fancy 3d printed parts: https://craftcloud3d.com/account/referral

Referal code:  REFHG992QMS used jan23

 

[Todd]

Thanks for the code!

[Todd]

@Simon, after hearing of your issues with residual powder inside your heat exchanger I'm interested in how you solved the dilemma.

After hearing about that, I did a re-design and eliminated the plates. They were probably overkill to begin with. The updated design has a single divider wall between the entering and leaving water with a 5mm gap at the bottom that serves as a return path from bottom to top. The images don't capture exactly what's going on but suffice it to say it's a single pass with very little restriction.

Thank you for sharing the residual issues!

[Todd]

And only USD $38 in Aluminum through Craftcloud3D.

Edited September 17, 2022 by Todd

[Simon2.4]

15 hours ago, Todd said:

how you solved the dilemma

I didn't.. I now believe that the part was poorly printed; the reason I know this now is because I ordered a stainless one too from JLCPCB and that one came perfect and cleaned inside-out, from the same STL file. I sent an E-mail asking for a refund and should get an answer tomorrow. Also, I did order my "v5" last week.

When I get an answer, I'll use a grinder to cut it open and see what went wrong during the print.

15 hours ago, Todd said:

Aluminum through Craftcloud3D

You are obviously a pretty advanced 3D printing geek if you're into designing mods; I'd love to know a bit more about you; may I invite you to post an introduction and tell us a bit about what brought you here.

https://www.teamfdm.com/forums/forum/39-introductions/

If you are building a Voron or even if you're finished, you could post your progress in the builds diary section too.

https://www.teamfdm.com/forums/forum/27-build-diaries/

[Todd]

Thank you @Simon. Just posted a "welcome Bio" using the link you provided. Happy to be invited. Hopefully I gave everyone an idea of what can come from coastal SC, USA.

 

Cheers

[Simon2.4]

I received my second part yesterday: "version 5" from above. It is also completely blocked and impossible to clean out. This model was designed to make powder removal easy so there is no reason for this to happen. I followed all design guidelines religiously to avoid this exact problem...

So I decided to investigate the reason this is happening by cutting open the first part:

The base powder is sintered inside the part! There is nothing to remove that. It's hard like concrete. That is caused by poor print settings and is a clear indication that there is no QC done by the seller...

Do not use 3DPnxt on Craftcloud3D

I'm requesting a refund for both parts; now, we'll see if Craftcloud3D's customer service is any good...

 

[Todd]

That looks like a complete garbage print job. Nice design that should have been easy to clean out any residual powder, but it indeed looks like they have no QC/QA at all. The size of the walls and the simplicity of the internals should have allowed for a smooth print without the fused powder inside. Keep us informed on how they are on the returns.

 

[Demosth]

This is such a neat thread to follow - any updates?

[Todd]

We haven’t updated much here lately. After @Simon’s heat exchanger cam back with poor results things have cooled off a bit.

For my project, I have the other pieces ready (radiator, fan, pump, etc.) but am looking for a reliable (SLM?) service with reasonable aluminum pricing and quality work.

Edited November 19, 2022 by Todd

[Simon2.4]

@Demosth Yah, I'm hard at work for the autumn rush.

Like @Todd said, getting the aluminum part printed properly has proven challenging with Craftcloud3D. I have a full month of conversation with them and my conclusion was that they need to add print supports to connect the build plate to the part and conduct the heat away, as per standard guidelines. I asked them that specific question and they said no.

As a service, they are adequate but it's the material that's challenging. Since the fusion temperature of the aluminium powder is so low. Titanium would probably work ok but it has bad heat conductivity. Copper is heavy AF

The Stainless steel one I got from JLC PCB is flawless and came out perfect so I may still try this but the added weight is really a let down.

Anyway I now know alot about SLM printing so it was not all a loss.