DAN C4-SFX: The smallest water cooling case in the world

Obsidian 500d brought corsair back into the fold for me. Very modern designs they go for now and a liltte too much RGB.

Dan Personally I think the case looks fantastic. All I would want is enough rad space for a 240 for my cpu and a 240 for my GPU. Im a little excessive on cooling.
SFF is not for you bro.
 
AlMiDaX The rubber layer must be glued and can be compressed (on the short part) while attaching it. I think a better solution is to add a gab like 0.1-0.2mm on both sides between the short bends and the inner frame. Then I have to use a 0.1-0.2 thicker panel on both sides to make sure they flush.

Btw. What tool do you use for animating stuff like this?

@Everyone: Please give me some days I am working on two idears that I like to show, but I need time for visualization.
 
Btw: What do you think about a outer design that looks like this without the gab under the window panel:

Case-Jonsbo-UMX3-W-3.jpg
 
Btw: What do you think about a outer design that looks like this without the gab under the window panel:

View attachment 85101
Generic and bland imo. Personally I think you are getting further away from the original design that people liked. Is the complete overhaul a necessity now with the issues you described or are you just spitballing different ideas?
 
Here is a direct comparison between UMX and RM3. The UMX looks like the unibody design while the RM looks like the C4 design.

Mini-ITX-Cooltek-Case-Packs-Full-Size-Desktop-Benefits-467462-2.jpg


1113094_1__8731045-1.jpg
 
I just don't like rounded corners in general.
This looks too simple. It's just a rounded box with a glass side. It's not ugly but it doesn't have any interesting lines and details to catch the eye. If you want a box that melts into the background than I guess this would work. But when you want something beautiful that you can just look at for a while, a design object by itself, then nope.

That's what I loved about the prototype. It had those angles with the feet and especially how the side panel had that cut out which integrates the IO. The front panel was separate which creates a bit of a division to help the viewer take in part by part, unlike this unending loop of a unibody design. This feels..."slippery" It doesn't give my eyes anything to latch onto if you know what I mean. While the prototype guides your eyes along the rims ontho the details, this just guides your eyes in a loop around the glass.
I do like this kind of very dark glass tint though. It hides cable routing crimes and highlights inside lighting which are on the main components.
 
AlMiDaX The rubber layer must be glued and can be compressed (on the short part) while attaching it. I think a better solution is to add a gab like 0.1-0.2mm on both sides between the short bends and the inner frame. Then I have to use a 0.1-0.2 thicker panel on both sides to make sure they flush.

What if the padding is directly attached to the frame instead of the outer panels? It should be less of a problem, right?
Alternatively, You can use Foam instead of Rubber.

ezgif.com-resize-2j4sl9.gif



Btw. What tool do you use for animating stuff like this?
I used Affinity Designer for the Illustration and Flinto (for Mac) for the animation. ;)
 
dondan if you opt for unibody is it possible to make it a bit less curvy? Also, I'd like to maintain some of the original features.

unibody-fixvhskd.jpg


In any case, I think the C3 Plus, VR2, and U4, look better than the UMX4 (all made by Jonsbo).
But that is just my personal opinion. :)
 
Hi AlMiDaX like everytime your posts are very helpful and great visulized.

Maybe I found a solution today how it can be done without the need of that rubber. Please take a look at the Raijintek Metis. Furthermore this construction has a cost advantage because i don't need a inner frame.

I already in the progress of drawing it in CAD.
 
Hi AlMiDaX like everytime your posts are very helpful and great visulized.

Maybe I found a solution today how it can be done without the need of that rubber. Please take a look at the Raijintek Metis. Furthermore this construction has a cost advantage because i don't need a inner frame.

I already in the progress of drawing it in CAD.
Well, it's in my best interest to see best possible C4-SFX.

The Raijintek Metis looks interesting and quite spacious. I don't know how the 180° flip will function, it seems like you'll have to disassemble the whole case.

In that case, it may be even possible to consider flat packaging (like on the Ghost S1).

Looking forward to the renders :)
 
Rotating will be possible if i use vent holes on top and bottom side. You can use the holes to attach the rubber feets that you know from the A4 or a higher version.

Challenging is how to mount the radiator.
 
  • Like
Reactions: Jspr
like this
Back w/ Daily suggestion:

View attachment 85356

dondan is right tho: this forum needs to be a little more active. I guess is time to make an SFF account :whistle:

Problem is that we are mostly waiting for news and updates. Dondan has mentioned some issues, but has still to post any pictures or detailed explanation of the problems. That makes them difficult to discuss. If we go by your animations, it seems that the frame is scratched where the top and bottom parts wrap around the frame. This makes sense. But at the same time, if fairly irrelevant. It will never be visible on a finished system, heck not even when removing the side panels.

To do a complete redesign because of this just seems unecessary and excessive. Now there is talk of a completely different design that we have yet to see. So even less to relate to.

Now, as for the unibody. I just find it obsolete as mentioned earlier. But it's also impractical. Especially when dealing with tiny cases, having the top, bottom and sometimes the front, permanently fixed, just makes the building process a lot harder. I have made a build in the original C4 visually in my mind a lot of times now. I know exactly how to build it, in what order, and what extra things I need to buy and how to do cable management, etc. Having just the frame, makes everything simpler. Know, I know the A4 can remove the curved panel, but it's not a unibody design as such.

I know Dondan worries about things that can result in an RMA. That is a major risk for him, so he needs to take it serious. But honestly, there are easy ways to get around that. He can simply state/show the issue on the kickstarter/store page and say it's not to be considered a defect. It's just a simple disclaimer and something to educate the potential customer of the design.

I think some people are a little annoyed/frustrated as they fell in love with the original concept and now it's basically scrapped. Furthermore the argumentations often don't hold up. Like the issue with USB-C on the side. Sure the frontpanel header is still rare on ITX boards, but they will be less so for each year. Furthermore, there is more than enough room to push the sata SSD up in the case, or just put the USB-C connector further down, switching place with the USB A connector, if the C connector is deeper. So often, it doesn't make sense.

I think Dan hit a homerun with the original design. Some improvements got made in the iterative process, but lately it has been less iteration and more starting from scratch.
 
Now, as for the unibody. I just find it obsolete as mentioned earlier. But it's also impractical. Especially when dealing with tiny cases, having the top, bottom and sometimes the front, permanently fixed, just makes the building process a lot harder. I have made a build in the original C4 visually in my mind a lot of times now. I know exactly how to build it, in what order, and what extra things I need to buy and how to do cable management, etc. Having just the frame, makes everything simpler. Know, I know the A4 can remove the curved panel, but it's not a unibody design as such.

This is exactly my thoughts on the unibody. I had a Jonsbo U2 before and I will never own another unibody case. Doing the build in that thing was next to impossible. I don't mind SFF cases, if I can remove most of the panels and actually have access to the inside of the case.
 
This is exactly my thoughts on the unibody. I had a Jonsbo U2 before and I will never own another unibody case. Doing the build in that thing was next to impossible. I don't mind SFF cases, if I can remove most of the panels and actually have access to the inside of the case.
Indeed. And I have the Lian Li v700 unibody case now. At least it's ATX, but still not very practical. When going to ITX, everything just gets a lot harder.

Unibodies are cheaper to make and simpler. So I understand why Don want to do it. Risk management is important. And the idea in the beginning with the stiff PCI thing, was to make it cheaper than A4. But in this market, I think any and all alu cases will be expensive, and people actually going high end alu ITX is more focused on design/quality than saving 50€.

Then again, lower prices opens up the market. But I think the PCI bridge already does that.
 
Problem is that we are mostly waiting for news and updates. Dondan has mentioned some issues, but has still to post any pictures or detailed explanation of the problems. That makes them difficult to discuss. If we go by your animations, it seems that the frame is scratched where the top and bottom parts wrap around the frame. This makes sense. But at the same time, if fairly irrelevant. It will never be visible on a finished system, heck not even when removing the side panels.

To do a complete redesign because of this just seems unecessary and excessive. Now there is talk of a completely different design that we have yet to see. So even less to relate to.

Now, as for the unibody. I just find it obsolete as mentioned earlier. But it's also impractical. Especially when dealing with tiny cases, having the top, bottom and sometimes the front, permanently fixed, just makes the building process a lot harder. I have made a build in the original C4 visually in my mind a lot of times now. I know exactly how to build it, in what order, and what extra things I need to buy and how to do cable management, etc. Having just the frame, makes everything simpler. Know, I know the A4 can remove the curved panel, but it's not a unibody design as such.

I know Dondan worries about things that can result in an RMA. That is a major risk for him, so he needs to take it serious. But honestly, there are easy ways to get around that. He can simply state/show the issue on the kickstarter/store page and say it's not to be considered a defect. It's just a simple disclaimer and something to educate the potential customer of the design.

I think some people are a little annoyed/frustrated as they fell in love with the original concept and now it's basically scrapped. Furthermore the argumentations often don't hold up. Like the issue with USB-C on the side. Sure the frontpanel header is still rare on ITX boards, but they will be less so for each year. Furthermore, there is more than enough room to push the sata SSD up in the case, or just put the USB-C connector further down, switching place with the USB A connector, if the C connector is deeper. So often, it doesn't make sense.

I think Dan hit a homerun with the original design. Some improvements got made in the iterative process, but lately it has been less iteration and more starting from scratch.
While I can't disagree with you on most points, I think we have to try to be patient. He is doing all the work by himself, I don't think he is happy to scrap the original design, only he knows how many hours he spent making the original CAD drawings.

These are the actual challenges afaik:
1) top and bottom sides of the case frame are prone to scratches
2) the original design is expensive to manufacture (the project needs to be sustainable, he has to earn at least something)
3) need to find a way to hide screws on side panels (there’s no room for clips or slide mechanism)
4) increase space for cable management.

Those are the ones that I remember, so if you guys have any ideas, let us try to find possible solutions and give some constructive feedback.
 
While I can't disagree with you on most points, I think we have to try to be patient. He is doing all the work by himself, I don't think he is happy to scrap the original design, only he knows how many hours he spent making the original CAD drawings.

These are the actual challenges afaik:
1) top and bottom sides of the case frame are prone to scratches
2) the original design is expensive to manufacture (the project needs to be sustainable, he has to earn at least something)
3) need to find a way to hide screws on side panels (there’s no room for clips or slide mechanism)
4) increase space for cable management.

Those are the ones that I remember, so if you guys have any ideas, let us try to find possible solutions and give some constructive feedback.

Yeah. I won't make a judgement until the final product is up. I really want a sweet small CLC based computer. Right now the C4 is the best option. However:

  1. Afaik it's where the top and bottom panels wrap around the frame on the sides that gets scratched. I don't care about that. You won't see it even with the sides removed.
  2. Agreed.
  3. Hoesntly, I don't mind the screws. People could always gets torx screws or other mods to upgrade the look. I'm going with glass on one side anyways, so screws is necessary on my build either way.
  4. Obviously, I haven't tried building in this case prototype. But the visual builds I've done in my head works out fine. Plenty of room on the side over and under the PSU to put cables. It's mostly a case of making cable anchors around the case. These are made for pro builders, so I'm not too concerned about it.
But again, it would help tremendously if DonDan would include a lot of pictures from his protobuild to show us. He could watermark the entire pics with prototype wearing, if he is afraid it would reflect poorly on the project.
 
Sometimes you have to move a step back to recognise that the stuff you already made is better than thought. So I took the Saturday to take a deeper look at the prototypes, enjoying the design disassembly it and assembly it. After that day I got the feeling that you are right and I should focus on the current design and try to solve the problems. Maybe my reaction in the last week was a panic reaction because sometimes you come to a point where you think your product isn’t that good as thought and you have the feeling to start from zero. But I fall in love again with the current design.


So I am now back in CAD and try to optimized them and keep the design that you all love.


While working on this I have some question that maybe needs to be discussed:


  • Side I/O: I found out if you plug a USB drive or a headset in the side I/O it is very dangerous. The most customers will put the system next to them on the desk if you now making a wrong movement you can break the components inside the I/O because they stick out. The top I/O version will not have that problem.

  • Case stiffness: If you remove all panels the case stiffness is not very good. The reason for this is the attachment of the middle plate. I have to attach them in that way that you will not see screws in the side I/O area. If I will switch to the top I/O version the stiffness could be much better.

  • Case feet’s: The angled case feet’s that you can see in the rendering in the first post of this thread has two problems: The case balance point is top-heavy. With the angled one it is very easy to bowl down the case. The problem is also bigger if you don’t install a GPU. The second problem is, that both angled bottom plates (silver and black) has the problem that they are not mate 100% proper this will result in a wobbling case because one corner has no contact to the ground. I think I will use the top plate also for the bottom part with four A4 feet’s attached.
 
  • Side I/O: I found out if you plug a USB drive or a headset in the side I/O it is very dangerous. The most customers will put the system next to them on the desk if you now making a wrong movement you can break the components inside the I/O because they stick out. The top I/O version will not have that problem.

  • Case stiffness: If you remove all panels the case stiffness is not very good. The reason for this is the attachment of the middle plate. I have to attach them in that way that you will not see screws in the side I/O area. If I will switch to the top I/O version the stiffness could be much better.

Personally front I/O, side I/O or top is just for temporary use. Like for example to copy something to a USB stick and then unplug again. No matter the I/O orientation, there is always a risk to come against something sticking out of the I/O and break components. With side I/O facing towards me on the table, I don't see this issue being magnified compared to other positions. BUT that really depends on where you place your PC. I disagree that Top I/O would be safe in more situations. And especially audio ports on top and having cables coming from the top of the case back down is not a preferable solution imo. So I would rather like to see side I/O than top, as it complemented the design better.

As for the stiffness, can you comment on how big of a problem that is? A case like the NCase isn't really stable without the panels either and I never considered it a real problem.

But I am glad to hear that you are back to the previous designs :)
 
Side I/O: unfortunately, I experienced it before. The risk of breaking stuff plugged in is quite a common thing on bigger form factor cases, especially when the case is sitting under the desk (and has front I/O). With that said I'm not too sure that it's that dangerous on an SFF case since the case is at the bottom of the desk (close to a wall) and not that close to you (also, it depends if the I/O is on your side of the desk).

Case stiffness: is stiffness a problem even after installing all components?

Case feet: correct me if I'm wrong, the case is top-heavy only when the radiator is sitting at the bottom, right? - As for the second issue, an idea to minimize the wobbling would be to unify the rubber pads at the bottom (see pic below), it may not fix the problem entirely but it should alleviate it (or well, it depends on the softness of the pads).

c4-bottom01sxo.gif
 
Last edited:
AlMiDaX

Side I/O: I will show pictures later

Case stiffness: After installing all outer panels the stiffness is fine, but maybe I found a solution for this problem

Case feet: The bowl down problem exists in both orientation. At the first time with I/O at the right and power on the system it starts bowl down and I was able to rescue it in the last second. Same situation white putting in a USB stick with I/O on the left side.

Case feet: On my sample the backside left rubber pad was 1-1.5mm away from the desk. At first I thought my desk didn't have a straight surface but it was the case and not the desk.
 
dondan hmm at this point, I think Case Feet are no.1 priority. I like the A4 ones, but I will still think about an alternative solution. Hopefully, I end up with a suggestion or something.
If I'm correct, the A4 feet are only 5mm, with the C4 there's the chance to go up to 10mm.
 
AlMiDaX

Side I/O: I will show pictures later

Case stiffness: After installing all outer panels the stiffness is fine, but maybe I found a solution for this problem

Case feet: The bowl down problem exists in both orientation. At the first time with I/O at the right and power on the system it starts bowl down and I was able to rescue it in the last second. Same situation white putting in a USB stick with I/O on the left side.

Case feet: On my sample the backside left rubber pad was 1-1.5mm away from the desk. At first I thought my desk didn't have a straight surface but it was the case and not the desk.

Resolving case stiffness should help squaring up the case, and potentially alleviating the feet lifting off the tabletop. On the other hand, if there are concerns with center of gravity with the angled feet, can you test a fully populated case with different types of feet? e.g., less acute angle, vertical, and obtuse angle. These can be 3D printed, or just wood blocks. Or stay digital, and find CG with a top heavy assembly in CAD.

I'm glad you've renewed your passion for your prototypes!
 
Case feet:

First the unevenness. This is a problem with everything with more than 3 feet. If it's uneven I would say that's a reason to complain to LianLi. I mean other cases have 4+ corner contact patches too and seem to work fine. This is a quality assurance problem and should lie with Lian Li. This is a prototype. I'm sure once the tooling and processes are worked out it'll be fine.

Secondly the balance. Well the problem is that the support area is smaller than the case because it sweeps back. While this design looks cooler since it makes the case seem to float, there is really nothing we can do other than increasing the support area (or design it to be bottom heavy, but then we lose the ability to turn the case). This is limited by simple physics. So experiment with a design that instead of sweeping the feet in, sweeps them outward. Who knows it might look cool too.
Except there might be a little cheat: Instead of normal rubber use the kind of sticky rubber that is used on some phone holders. That would basically almost glue the case to the surface. We would have to experiment with the size of the rubber to get a good balance of stickiness vs ease of lift off.



IO:

I still prefer the side IO and I am willing to take the risk. Also for most people, because this is a small case, it will sit further back on the desk and thus not be totally close to the hands.


Question:
If you're making the case bigger for cable management. Would USB Type C port be an option now?
 
If the top I/O provides more structural rigidity, I'd much prefer that over something that feels flimsy. Besides, it looks great! I hope we can get USB-C included as well.

BTW, it looks very similar to the earlier Lian Li prototype:
Lian Li build the prototypes again and now the quality is very good. Here are some pictures of the top i/o version:

_dsc2629apr6x.jpg

_dsc2625zipko.jpg

_dsc26353bqol.jpg

_dsc2643otrrv.jpg

_dsc263898re2.jpg
 
IF dondan is going to keep the side I/O I don't think it is going to be that risky. As I said (on most scenarios) the case is not that close to the user.

desk-mincssqq.jpg



Daily suggestion - Feet variants:
feet-minruska.jpg

My humble opinion is that the original angled feet look the best, but if that doesn't work I'll still be happy with the A4 style feet. (I'm not too convinced on either of the variants above). Thoughts?
 
@AlMiDaX
I like the "angled type 2" variant from your example. It uses the same design language as the rest of the case.
 
  • Like
Reactions: Woxys
like this
@AlMiDaX
I like the "angled type 2" variant from your example. It uses the same design language as the rest of the case.
I think it looks just ok, problem is that it looks expensive to manufacture (due to the bending process).

Current Angled style and A4 feet look simple and clean. I think I prefer what has already been done by Dan.
 
The bends are similar—just inversed. I don't foresee how it'd be anymore expensive to manufacture. It should be stronger as well.
 
Last edited:
The bends are similar—just inversed. I don't foresee how it'd be anymore expensive to manufacture. It should be stronger as well.

They are similar but there are some structural differences, on the type 2 variant, the feet have a sort of "U" shape.
It is possible to implement this style but it may require special tooling that will surely increase production costs.
inversed9qsc7.jpg
 
A4: simple and minimal, nothing too fancy and that's a good thing.

InWin: not too sure if this will look good in real life but the nice thing about this style is that you may slide in a dust-filter. (See pic below)
301-6b0jsez.jpg


NZXT: squared more stable version of A4 maybe? I honestly don't love it lol.



AlMiDaX you can make the case...
Nah, leave that to dondan - I have neither the skills nor the knowledge for such a task. As you can see I can only do simple 2D illustrations :)
 
Last edited:
AlMiDaX
The problem is ONLY the area that the feet enclose. As long as it's the same or smaller than the original (assuming same height of the center of gravity), it won't change anything for stability.
It's the marked area (and by extension the angle) that determines how much angular (tipping) momentum the case can take without toppling over. (More rubber might slightly dampen impacts for our benefit)
So if you place the A4 or NZXT feet where you did, basically nothing would change.

(I am assuming a center of gravity in the middle of the case since that's the only configuration we can optimize to because the case is reversible
25so611.jpg


{NG}Fidel
I'd like to see some photos of that to prove me wrong.
 
Last edited:
I have a new question for you:

Whould it be a problem if the motherboard standoff will be a part of the side panel? This means you can't remove the panel without removing the motherboard. The advantage would be, that I don't need the middleplate. Then I will redesign the inner body that will have a bottom where you can screw the radiator. This will make mounting radiators much easier.

Disadvantages:
- you have to unscrew the motherboard to get access to its backside

Advantages:
- increasing stiffness of the case
- easier installation process for radiators
 
Back
Top