Case Design for Liquid Cooling

[mlwood37]

excellent design and yes it could be produced .... depends who he goes to and what his cut would be.

[Mick64]

Decided to go public with my case design SketchUp models. Feel free to download and build, if you have the resources. Be warned, these files are very large! Download from here. Enjoy!

[Mick64]

Bump, I expected more reaction to this post. Large cases must be going out of fashion?

[h20xide]

Some may have gotten over the hype this case received? Not sure, but I for one am excited that you've decided to release the design. Someone make this!

[UrbanSmooth]

This is still one of the best-looking, highly-accommodating PC cases that I have seen.

[Sadasius]

Don't worry Mick. One of these days someone is going to build this. This is the best case I have ever seen for watercoolers period.

[individual]

I posted in here about 2 years or so ago, and I loved this case.

and I still do.

I'm in my 3rd year as a fitter and turner now, and i especially love working with sheet metal

you can see where this is going...

I'm going to attempt to make this case

First one will be out of steel, simply because....errors when making are easier to fix. given how long it will take me to transpose it to inventor and then into sheet metal format for the press brakes, amidst my regular work, I can't give a definite answer on when I'll start BUILDING.
but right now I'm transposing the plans into inventor

I'll update the parts and upload the files to STEP format piece by piece.

**Forgot I had to re-install inventor, fresh format. Is it sad that I've installed it that many times, I know the product key off by heart?**

[individual]

I'm studying the outer shell, the main piece, thinking it's going to have to be slightly modified, because it wraps around, its overall length is =>2400mm, which is a decent sheet of steel/ally. but then again I may need a 2.4x1.2 piece, maybe even bigger, to cut out all the necessary components, it's difficult to estimate.

Also depends on what grade of aluminium you want, or if you want an alloy (recommended) you want one with a decent tensile strength, low thermal co-ef so you're case doesn't become a heatsink, and a low thermal expansion co-efficient too.

so much planning.
Also getting a list of rads that will\wont fit in combo with standard 20mm depth fans, and ones that are a tad too wide too.

I was thinking carefully about the PSU hole, and doing a basic flow test on inventor on a quick draft, it works better having the fan facing into the case, you were right.
WITHOUT a graphics card, the case would benefit from having the PSU fan hole on the underside (which isn't going to happen)
WITH the graphics card, the flow is essentially split, top half goes straight out the back, bottom half goes under and out (little bit of a vortex effect though)
OPTIMALLY, if the motherboard was placed on the opposite side of the case, therefore flipped (CPU on top, graphics Beneath, like in a few of the different versions you posted)
WHICH is easy to do, the bracket that supports the motherboard tray could easily be made to go on either side, which it looks like was originally intended (or i may be looking at an older file of the case, i downloaded every file on the skydrive)

ALSO I thank you that it is in metric and not imperial, apart from the sheet thickness. 1/8th is overkill for this case. (3.175mm for metric users) I'll redesign in 2mm

LAST thing (promise) with the motherboard tray holder, difficult to produce those "snake" clips



so i would suggest something like this..




Sorry to be fussy, I'm used to working with micron spacing and tolerances so I'm very picky haha.

EDIT** also, you have some awesome skills with sketchup, i can't stand polygon modelling haha.

[individual]

OK so I've begun designing the outside casing, and there's an issue the radial bends in the main shell, the 27mm ones that link front>top>back>bottom, you can't flange the edge of them. essentially you can't flange a radial wall around a full-round corner, if you understand what I mean

The only way would be to turn the radial edges, into chamfers. Which in all honestly wouldn't look too bad, you'd just chamfer the other round cut features like the fan holes etc and it would look right as rain.



Also, Looking at all the parts in the case, it needs some re-enforcing to stop it from skewing backwards and forwards. the U-shaped centre section doesnt help it, the only thing that does as a permanent fixture is the sheet metal peice that the motherboard tray latches into (previous posts) but it needs something on the other side.

***EDIT***

Tried it out with a chamfer of 45deg, 3mm rad joins, still a little bit gappy.


There's no easy way around this.

[Mick64]

How about?

OK so I've begun designing the outside casing, and there's an issue the radial bends in the main shell, the 27mm ones that link front>top>back>bottom, you can't flange the edge of them. essentially you can't flange a radial wall around a full-round corner, if you understand what I mean.

By the way, this can be done in a press. It's not cheap, but it's not impossible. If you were going to mass produce the case, the dies would pay for themselves, but a one off is a little different.

[Yaka]

wish i had both the tools and the time to make my self one.:/

[individual]

Yeah it would work with some dies, but they're still beyond my skill level honestly, even with NC machines, and especially with a piece this large.the most feesible way, and to keep the round corners, would be to use a support bracket like you posted, with a round edge that sits flush with the case. ill show yo what i mean. even then it'll have to be bent and then the radius machined on an nc.

[individual]

everything else looks pretty damn good I'm guessing you've used sheet metal before yourself?

With the side panels, because they are two sheets, if you were building it would you just glue it together?

[individual]

an Idea to support the skew, would be to simply remove the hems of the main 5.25" cage at the front, and extend them out to the side as extra flanges and then they will get riveted into the main case.....Thoughts?


EDIT only just discovered you can change the accuracy of the measuring tool. I was sick of seeing ~100.1...now its 100.115 makes life so much easier.

2ND EDIT. the more I look at this case, the more impressed I am. You've done an amazing job, this tops my case designs for simplicity and efficiency by far (mine have a tendency of been....complex lol)

[Mick64]

an Idea to support the skew, would be to simply remove the hems of the main 5.25" cage at the front, and extend them out to the side as extra flanges and then they will get riveted into the main case.....Thoughts?

This is a possibility, but this is where the inner 'U' section is attached to the outside of the case, so there is already two layers. This may cause an issue when attaching the side covers?

the more I look at this case, the more impressed I am. You've done an amazing job, this tops my case designs for simplicity and efficiency by far (mine have a tendency of been....complex lol)

Thanks for your kind words. I have tried to keep it simple to aid with the manufacturing process, but also ensure that it is aesthetically pleasing and practical.

[Mick64]

everything else looks pretty damn good I'm guessing you've used sheet metal before yourself?

I've never worked with sheet metal, but have a good understanding of the principals.

With the side panels, because they are two sheets, if you were building it would you just glue it together?

Yes, there are many types of glue that could be used for bonding the sheets. This also makes the side covers very strong and rigid.

[individual]

You've got a very good understanding for someone who hasn't worked with sheet metal before, very impressive

I've been talking to some of colleagues today and we've come to an agreement that the prototypes will have to be lazer-cut, then bent and folded, to save the manufacture of punches. Even with lazer cutting, it will be costly. most likely around 300-400 alone for just the lazer cutting (I know from previous work I've had done). the only issue i can see, is bending the side panel latches, more than likely will have to be done in a special hand press.

Thinking more and more I may even do the prototype in aluminium, and do most of the bending and folding by hand.

With the side panel issue when using the 5.25 cage for structural support, that could easily be fixed by simply making the catching flange on the side panel slightly higher.

I will have to modify some things on the case very slightly to get it produced, but it is do-able, and it will be done

[individual]

SO I tried test bending some aluminium 6061 sheets today, 2mm thick, has a very large tendency to crack when bending, so I'll have to use steel, which is going to make this f***er heavy, at least 10 kilos.
I've modeled the 5.25 cage alone and in steel the mass is 1.2kg :O



Also, with the corners, this is what i was talking about.



Then simply have your corner support idea behind that, riveted in.

[individual]

OK I'm doing the motherboard tray, which is BY FAR the most complicated part of the design.
A few things I've discovered, the rear IO mount (where the I/O shield clips into) can be a maximum of 1.32mm thick and a minimum of .93mm thick. So I'll have to change that to 1mm material.

I have to ask, I can find all the details I need for the HTPX form factor mounting holes and I/O shield, but I can't find any details on PCI mounting :/

I'm after the PCI tabs that slot into the mobo tray, their placement in relation to the motherboard mounting holes, and of course the dimensions of the PCI slots, and placement of the screw holes. basically everything PCI I still need. Most of your dimensions are solid, but I want to double check everything especially when things start to get made.

[.::NeO::.]

OK I'm doing the motherboard tray, which is BY FAR the most complicated part of the design.
A few things I've discovered, the rear IO mount (where the I/O shield clips into) can be a maximum of 1.32mm thick and a minimum of .93mm thick. So I'll have to change that to 1mm material.

I have to ask, I can find all the details I need for the HTPX form factor mounting holes and I/O shield, but I can't find any details on PCI mounting :/

I'm after the PCI tabs that slot into the mobo tray, their placement in relation to the motherboard mounting holes, and of course the dimensions of the PCI slots, and placement of the screw holes. basically everything PCI I still need. Most of your dimensions are solid, but I want to double check everything especially when things start to get made.

Well if you have the exact dimensions of a 10 slot I/O shield you should be fine. Fix the I/O shield to the rear of your case and then start to place the motherboard holes from there (if you know the motherobard I/O measures. If you do not have the complete measures of a 10 slot I/O shield I think I could give you a help. I think I have them in SolidWorks and I'm also waiting for a 10 slot I/O shield from danger den which should be ideal for your needs, I could measure it for you.

Also, check out www.grabcad.com and I'm sure you can find a correct I/O shield there

I'll try to send you a couple of measures as soon as I get home. Can I ask you if you'll be so kind to share HPTX form factor mounting holes here or via PM? I really need them

Thank-you

[JoanchimMartin]

SO I tried test bending some aluminium 6061 sheets today, 2mm thick, has a very large tendency to crack when bending, so I'll have to use steel, which is going to make this f***er heavy, at least 10 kilos.
I've modeled the 5.25 cage alone and in steel the mass is 1.2kg :O

Use aluminum 5052-H32 (easy to bend): 1mm or 2mm is up to you.

[individual]

Use aluminum 5052-H32 (easy to bend): 1mm or 2mm is up to you.

Awesome ill look into it ill use steel for the prototype still just because its cheaper but if the case takes off ill switch to 5052

[Gimmpy224]

Luckily i now am the system administrator for a custom aluminum siding manufacturing company.

I will hopefully see them sometime this month and will have one of the engineers take a look at this for me and quote me a cost to make it.

Ive been eyeballing this thing since you first drew it up. Maybe ill be the first to build it

[individual]

Luckily i now am the system administrator for a custom aluminum siding manufacturing company.

I will hopefully see them sometime this month and will have one of the engineers take a look at this for me and quote me a cost to make it.

Ive been eyeballing this thing since you first drew it up. Maybe ill be the first to build it

It's a competition!! Lol if you can build it, awesome, will get done faster that way. I'm estimating as a prototype around 600-700, if you get punches made about 1k. I'm transferring all the components to inventor as final measurements, so if you get a quote and you want to build it PM me for the drawings

[individual]

Can I ask you if you'll be so kind to share HPTX form factor mounting holes here or via PM? I really need them

Thank-you

As soon as I get home ill upload them for you to this thread