-
Save 15% on ALL SFF Network merch, until Dec 31st! Use code SFF2024 at checkout. Click here!
!nverse: A Highly Versatile Steam-Box Design
- Thread starter esplin2966
- Start date
Now that I think about it, the audio ports should be easily paintable with plastidip or something like that.
As I said on [H], you could make the CPU cutout smaller, especially to the back I/O of the mainboard. There are about 30mm reserved for the I/O in that location anyway.
As I said on [H], you could make the CPU cutout smaller, especially to the back I/O of the mainboard. There are about 30mm reserved for the I/O in that location anyway.
Thanks! The case feet are removable and replaceable, so you're free to put them wherever you want. An SD/microSD card slot usually comes in a huge package like this, so unfortunately I don't think I'll have space for it. If it comes in a smaller package, then possibly.
Thanks
Responded on [H]
The cutout could be made smaller, but the current size doesn't impact structural strength since it's not really taking much weight. I'll be happy to do it once I survey the motherboards out there to make sure that making it smaller doesn't block any M.2 slots or something.
There is a regidity tradeoff, as well as a quality control/costs tradeoff. Punching too many holes can distort a metal sheet, which can cause misalignment issues. You can fix this by flattening the sheet after punching the holes, but that increases the manufacturing costs.
Furthermore, I have found that having too many holes actually serves to worsen cooling within a case. While the volumetric flow rate of hot air out of the case remains the same, the speed of each heated particle is reduced, increasing the internal temperature.
I got the same impression as EdZ, and although I can't speak for him, my impression is not about the number of holes or the footprint of the entire venting regions, but about the balance between punched hole diameter and the spacing between adjacent holes ("the hole pattern"), i.e. the percentage of openness within a given rectangular "venting region".
And I would like to see your cooling statement confirmed in an actual test, as more restrictive vent holes worsen cooling from my experience on my tiny ITX case (BQ656). I don't know what exact conditions under which the fluid dynamics dictates the constancy of volumetric flow rate of hot air out of an enclosure and whether air how the air molecule speed distribution changes over various vent hole sizes/patterns with a fixed fan speed, so maybe "6mm vs 5mm hole diameter comparison" and "1.0mm vs 0.8mm dust filter fineness comparison" for instance are two different matters that can't be discussed with a blanket statement, I don't know. But I'd like to be enlightened with something convincing. (I'm aware of your M350 experiment, but that's cooling efficiency variance over different air flow pattern, and I'm more talking about how cooling efficiency changes over different openness ratio within a fixed, say, 120 x 120mm vent region where a 120mm fan is mounted over.)
Aside from this, I also had questions regarding the measurement and the internal but I asked them on [H].
Last edited:
The hole-to-nohole ratio of the side is about 15%.
Responded to your questions on [H]
Thanks esplin2966. 15% seems quite low, but again I know it's a trade off with rigidity (and noise.) Btw I added what I forgot to say about cooling efficiency in my previous post and your reply preceded my edit, in case it goes unnoticed.
- Jul 19, 2015
- 818
- 1,359
I do know they exist (here) using a USB pin header, as for where to source in qty or alternative suppliers, etc...no clue
The issue is more with the hole size itself: By having a lot of tiny holes instead of a few large holes, you end up with a greater airflow restriction - even when the 'hole area' is equal - due to the increased turbulence at the hole edges. 'Lots of small holes' has more edge length, and more edge-length-per-unit-open-area, than 'a few big holes'.
Clustering fewer larger holes with 'braces' of non-punched material between them (for load bearing) might work, but it all comes down to manufacturability: imperfect holes you can punch beat perfect ones you can't.
Clustering fewer larger holes with 'braces' of non-punched material between them (for load bearing) might work, but it all comes down to manufacturability: imperfect holes you can punch beat perfect ones you can't.
I agree with your thinking. We actually increased the hole size from 3mm to 4.5mm in this design update, but maybe it's still too small. I'll have to think of a test for this though.
Hmm, testing different openness ratio within the context of this case is going to be difficult since I'll need to make many prototypes with various hole sizes (very expensive). It may be possible to just conduct a general test, but I'll have to think about how to do it. Perhaps I can just make a small tunnel and then have different hole sizes to swap out at the end of the tunnel.
There are a few companies (e.g. Ponoko) that do on-demand laser cutting for relatively cheap. A few set size squares of material with different hole patterns and a tunnel test rig might work for comparing designs in an isolated manner. The tricky part would be choosing the flow rate and static pressure to resemble that of the inside of a case with non-ducted fans.
Thanks! I'll check them out. I think I may also use the laser cutter at my university to make the tunnel setup out of wood. That way, I can cut down costs while maintaining full control over hole size/patterns.
I don't think getting the right flow rate and static pressure is difficult. Just put your barebones setup (Motherboard, CPU, CPU cooler, RAM, Storage, PSU) inside the tunnel test rig and you'll be on your way.
Update 08/09/2015:
Added internal renders for the newest prototype design.
The second prototype is going through final surface finishing right now. Hopefully we can get it shipped out soon
Added internal renders for the newest prototype design.
The second prototype is going through final surface finishing right now. Hopefully we can get it shipped out soon
Building in this case is going to be interesting due to the inverted nature of it. Can't wait to see the final finish on the prototype v2
Edit: 10.31 liters. Snuck that in their I see. Lol. As long as it's smaller than the rvz02 it's ok with me. Those cases are fat and ugly tbh.
Edit: 10.31 liters. Snuck that in their I see. Lol. As long as it's smaller than the rvz02 it's ok with me. Those cases are fat and ugly tbh.
The building shouldn't be too weird. The interior frame is designed so that you can take it out completely and build it upright like a regular case. You just install the interior frame into the outer casing upside down.
It went from 9.98L to 10.31L after I made the outer casing 0.5mm thicker... Talk about slim margins.
I was disappointed as well after they boosted the rvz02 to 12.4L, though it was probably only a small difference in one or two dimension.
This would be a 3.3% difference, which is not only tiny, but would never be noticed by an end user practically speaking. With Nova, we've ended up adding a little more than 5% of volume, which is also unnoticeable, but yields huge usability and other advantages.
Being very selective about anything that adds volume is important when designing SFF cases, obviously, but so are the practical implications of the case itself. Most people will prefer an unnoticeably-larger case that's much improved in some regard, rather than the alternative. It's all about balancing compromises in just the right way.
If it adds rigidity and strength to the case I'm all for it. Like you said, I won't notice the extra space it takes up since its negligible, but I will notice the extra strength and rigidity it adds to the case. My guess is that espelin and his team added the extra thickness to avoid the top panel drooping and deforming when some weight is put on top. Which from what I've seen, is a problem with thin metals.
Similar threads
