The next NCASE project: a Steam Machine-style case (indeterminate)

@theGryphon - I wasn't talking about doing business at different country at all - I was comparing doing business locally as US or PL citizen. And my direct use of exchange value was an approximation which is quite close and because as PlayfulPhoenix noticed on last page this exchange rate is not stable but rather fluctuating. You keep trying to twist the meaning of what I said to support your thesis.

I hope that's enough to stop this flamewar over simple approximation and maybe talk about something more constructive :)
 
vipz said:
Wow. Good luck on your next ventures, whatever they might be. You've come a loooong way since the original "aluminum SG05" post.
Click to expand...

I wish it was the next venture, but I haven't made any progress b/c I don't have the time to make the commitment. I have a day job, that direction requires time that I don't have.

The good news on the hand, it's fitting for the SFF Collective that people have been talking about.
 
Last edited:
I wholeheartedly agree with w360. It would be nothing but awesome if we could pull this off. But I think it doesn't matter that our cases compete with each other, they are all a little different and everyone has different ideas of what their case should look like. Even cases with the exact same layout might appeal to different people depending on their design. Some people love brushed aluminium, others love flashy fronts and windows in their cases, everyone has different tastes. Not that I'd want anyone to just copy what I did, that was just an exaggerated example :)
 
Suggestion.
Rotate the GFX Card so MB and Card face the same way, that way you can have a clear side panel for people into that. :)
 
I think you've hit it wahaha, the problem with the collective is (as proven over the last page or so) that everyone has their own way and ideas, which when it comes to engineering is a messy endeavor.
The collective as it's self is not a bad idea, but maybe the application of it needs to be twisted. Essentially it would be nice to have a system a bit like I.materialise but not for profit and instead of 3d printers you have laser cutters and cnc folders etc...

initially kitting it out with 2nd hand (but not sh!t) machinery And building up from there.

Possible 2 production lines, one for volume orders of already proven designs and the other for prototypes.

Obviously we don't have to limit the above to sff, it could be a research bay for anything pc related.

This is just a pipe dream but you never know, Mr Musk might be reading!

I'm not commenting on which country it should be in!
 
lypse said:
Suggestion.
Rotate the GFX Card so MB and Card face the same way, that way you can have a clear side panel for people into that. :)
Click to expand...

Sounds easy when you say it, but that requires all sorts of changes to the design, as well as the inclusion of a shielded PCIe riser which would highly increase the cost. I personally like the idea a lot, there's a reason why I tried to design two cases with this concept, but this was probably already considered and ditched by Necere.
 
So is there a tangible future for this case if you can find a manufacturer that's within the limits ? Many people would be very happy to see Ncase realise more awesome designs !
 
I can manufacture this for you! I'm based out of the US and design/manufacture computer cases and systems for a living.

http://www.sliger.com/
http://www.quanmax.us.com/

We've been drawing up our own design for a layout that'd be similar to this. If you want to shoot me an email [email protected] id be happy to see what I can do to help you out!
I've been envious of the success you've had with the M1 and I'd hate to see this project stall out!
 
Apathy_Enthusiast said:
I can manufacture this for you! I'm based out of the US and design/manufacture computer cases and systems for a living.

http://www.sliger.com/
http://www.quanmax.us.com/

We've been drawing up our own design for a layout that'd be similar to this. If you want to shoot me an email [email protected] id be happy to see what I can do to help you out!
I've been envious of the success you've had with the M1 and I'd hate to see this project stall out!
Click to expand...

Nice!

Thanks for the offer.

For your own design that is similar, do you guys plan to use NCT or have you guys invested in tooling already?
 
Last edited:
Apathy_Enthusiast said:
I can manufacture this for you! I'm based out of the US and design/manufacture computer cases and systems for a living.
Click to expand...
I don't suppose you have a CNC stud welder? Finding a manufacturer that has the capability is our number one stumbling block, since it's a fairly specialized piece of equipment and not many have them. It's pretty integral to the clean look and toolless panels of the M1 and LRPC, and changing the designs would require some significant reengineering and not work nearly as well for us.
 
wahaha360 said:
Nice!

Thanks for the offer.

For your own design that is similar, do you guys plan to use NCT or have you guys invested in tooling already?
Click to expand...

We're using stocked punch tooling for all parts on ours, goal is to be as inexpensive as possible. It also has a 7" flex riser so that the GPU is flipped so it can be a tower or slim profile configuration with good airflow in either position.

Necere said:
I don't suppose you have a CNC stud welder? Finding a manufacturer that has the capability is our number one stumbling block, since it's a fairly specialized piece of equipment and not many have them. It's pretty integral to the clean look and toolless panels of the M1 and LRPC, and changing the designs would require some significant reengineering and not work nearly as well for us.
Click to expand...

We have one but you don't need to use a stud welder or change your design, going by this picture, http://www.pcper.com/files/review/2014-04-27/DSC_0035.JPG Lian Li is counter boring and using a hardware pressed blind ball stud.
 
Apathy_Enthusiast said:
We have one but you don't need to use a stud welder or change your design, going by this picture, http://www.pcper.com/files/review/2014-04-27/DSC_0035.JPG Lian Li is counter boring and using a hardware pressed blind ball stud.
Click to expand...
No - that would require punching through the panel, which isn't what they're doing. You can see their process in this video. It's capacitor-discharge stud welding.

They do use self-clinching studs in other areas - the motherboard standoffs, for example - but the panel studs are welded.
 
Necere said:
No - that would require punching through the panel, which isn't what they're doing. You can see their process in this video. It's capacitor-discharge stud welding.

They do use self-clinching studs in other areas - the motherboard standoffs, for example - but the panel studs are welded.
Click to expand...

It's an achievable look either with the stud welding or using a concealed head self clinching ball stud. We don't punch through the sheet, instead we CNC mill a blind hole ~3/4 the sheet thickness deep then press the self clinch hardware into that open side. This keeps the hardwareless appearance on the outside and will save you the trouble hunting down a stud welder.

PEM has a PDF of the CH fasteners we typically use for this: http://www.pemnet.com/fastening_products/pdf/chdata.pdf - they don't have the ball stud, but I'm sure some hardware company out there does.
 
Apathy_Enthusiast said:
It's an achievable look either with the stud welding or using a concealed head self clinching ball stud. We don't punch through the sheet, instead we CNC mill a blind hole ~3/4 the sheet thickness deep then press the self clinch hardware into that open side. This keeps the hardwareless appearance on the outside and will save you the trouble hunting down a stud welder.

PEM has a PDF of the CH fasteners we typically use for this: http://www.pemnet.com/fastening_products/pdf/chdata.pdf - they don't have the ball stud, but I'm sure some hardware company out there does.
Click to expand...
I see. I guess my main concern then is durability of self-clinching studs compared to welded studs. I actually managed to break one of the standoffs on the first M1 prototype (though admittedly I was putting a lot of pressure on it).

Given that you have access to both, is there any reason to prefer self-clinching over welded?
 
Necere said:
I see. I guess my main concern then is durability of self-clinching studs compared to welded studs. I actually managed to break one of the standoffs on the first M1 prototype (though admittedly I was putting a lot of pressure on it).

Given that you have access to both, is there any reason to prefer self-clinching over welded?
Click to expand...

Long reply but kinda three reasons:

1. Cost and minimum quantities

Self clinching I have negligible tooling costs and only ~30-240 seconds of setup time between hardware types. Stud welding needs to amortize a project specific jig, plus setups are easily ~30 minutes or more, scrap parts for test setup, and welder labor cost a lot more than general sheet metal labor. Even with that semi-automated CNC table it'd have to be 100+ pieces it starts to make some sense to program, set up and weld it.

2. Self clinching is the only option for dissimilar metals

Steel hardware into aluminum sheet. Stud welding is only possible when the application has same hardware/sheet material and needs the extra attachment strength. Spring steel is about the only thing we use ours for.

Aluminum hardware is also typically spendier and weaker, so it wouldn't make sense outside of your aluminum sheet -> aluminum ball stud, anything aluminum hardware is guaranteed RMAs.

3. Machine utility

That CNC stud welding table is easily ~10k or more, plus an annual service contract for something that does one thing. One thing that's almost the same as what a hardware press + mill can do. Sure hardware presses and mill are more expensive, but they have much more utility - I can stamp small parts, flatten dents, install hardware, remove hardware, not even mentioning the CNC mill.

You'll probably find few sheet metal fabricators outside like car part manufacturers and maybe aerospace (and Lian Li) have stud welders with any volume capabilities - it's simply a very specific tool that produces results not much different from what can be accomplished with another set of tools for with more diversity in use.

I use PEM SKC-F in our hardwareless aluminum covers - they achieve the same appearance but actuate by sliding into keyholes rather than clip straight in like ball pins do. Then it's secured with a screw or two on the rear flange.

PS somethings wrong with the weld if you could break one off without absolutely destroying your aluminum panel - or its a steel ball stud... any idea if it's ferrous?
 
Haha yeah, a little insight into the details of case fabrication is fascinating :)
 
Apathy_Enthusiast said:
Long reply but kinda three reasons:


2. Self clinching is the only option for dissimilar metals

Steel hardware into aluminum sheet. Stud welding is only possible when the application has same hardware/sheet material and needs the extra attachment strength. Spring steel is about the only thing we use ours for.
Click to expand...

I have talked about CH inserts with my own manufacturer and he advises against it because steel and aluminum react with each other, it will cause corrosion.
 
Apathy_Enthusiast said:
I use PEM SKC-F in our hardwareless aluminum covers - they achieve the same appearance but actuate by sliding into keyholes rather than clip straight in like ball pins do. Then it's secured with a screw or two on the rear flange.

PS somethings wrong with the weld if you could break one off without absolutely destroying your aluminum panel - or its a steel ball stud... any idea if it's ferrous?
Click to expand...

I didn't even know about the welding thing, I thought self clenching blind inserts were the only way to do this.

Personally, I've been able to break off the motherboard standoffs off my PC-Q12 where they are mounted in a 2mm thick aluminium sheet exactly the same way the ball studs are.
If it's true that they are welded and that you can only weld one type of metal together, the standoffs were probably made of aluminium, which would explain why it was that easy. The base of the insert is still sitting inside the panel, though.
Also, I did this on purpose, not sure if that matters so much :D
 
wahaha360 said:
Some things are just hard to give up on :)
Click to expand...

Couldn't be happier to read this! This idea is too well-concieved to not see daylight, or at least have a good shot at it.

Phuncz said:
I like this conversation, very interesting !
Click to expand...

WiSK said:
Haha yeah, a little insight into the details of case fabrication is fascinating :)
Click to expand...

No kidding - the amount of knowledge and straight-up lingo I've learned over the past few months, reading about metal fabrication and manufacturing here and elsewhere, is impressive. Incredibly engrossing stuff, too :) I have newfound respect for why many of my fellow students went into mechanical and electromechanical engineering.
 
QinX said:
I have talked about CH inserts with my own manufacturer and he advises against it because steel and aluminum react with each other, it will cause corrosion.
Click to expand...

It's true they react with each other, galvanic corrosion - but only if your case is exposed to consistent >80% humidity for ~50,000 hours (like 5-7 years) or are exposed to salt spray. There's zero risk of galvanic action so long as the computer is in a climate controlled environment <80% humidity. Ex right now it's raining outside and our outdoor humidity is only 40%, in the office it's 30% (only get 5% increments on my gauge).

Fastenal has a good white paper on galvanic action between materials:
http://www.fastenal.com/content/feds/pdf/Article - Corrosion.pdf

There's also a good explanation here regarding general properties of corrosion on aluminum:
http://www.aluminiumdesign.net/design-support/aluminium-corrosion-resistance/

If a case does have a 10+ year life cycle in a corrosive environment a manufacturer should be pointing the design to stainless steel hardware in gold alodine chromate conversion aluminum (this is what all our military cases are), all steel with zinc plating, or in extreme cases passivated stainless steel based on budget.
 
Apathy_Enthusiast said:
It's true they react with each other, galvanic corrosion - but only if your case is exposed to consistent >80% humidity for ~50,000 hours (like 5-7 years) or are exposed to salt spray. There's zero risk of galvanic action so long as the computer is in a climate controlled environment <80% humidity.
Click to expand...

Would anodization or powder coating offer additional protection, since you'd have a layer (oxide layer or polymer) insulating the metal from environmental exposure?
 
PlayfulPhoenix said:
Would anodization or powder coating offer additional protection, since you'd have a layer (oxide layer or polymer) insulating the metal from environmental exposure?
Click to expand...

Yes, however dip tanks cannot anodize aluminum with steel hardware installed - and installing it in after anodization strips small amounts of the oxide layer so you're still going to get oxidation around the hardware with extended high humidity conditions.

Powder coat is good because it can be sprayed right over installed hardware - still not ideal as the polymer layer has to be removed/masked at contact edges to form a contiguous Faraday cage for FCC and CE compliance.

My last post for this discussion, apologies to NCase for cluttering up your thread, I can move case fabrication discussion to another thread if anyone has interest.
 
Apathy_Enthusiast said:
My last post for this discussion, apologies to NCase for cluttering up your thread, I can move case fabrication discussion to another thread if anyone has interest.
Click to expand...
No worries - it's good info. We've been looking for alternatives to the stud+clip and the self-clinching keyhole fasteners might be a good option. Can you comment at all on how firm the fit is and whether there might be the potential for more vibration with these fasteners? One thing the plastic clips offer is they hold the panels pretty firmly to the chassis, which should help to prevent vibration.

Apathy_Enthusiast said:
PS somethings wrong with the weld if you could break one off without absolutely destroying your aluminum panel - or its a steel ball stud... any idea if it's ferrous?
Click to expand...
Sorry, I probably wasn't clear: the stud I broke was the self-clinching through-hole type. This pic shows how it was pushed through:

nS1u1Bkh.jpg


Obviously, breaking it this way isn't a concern with a blind hole.
 
Necere said:
Sorry, I probably wasn't clear: the stud I broke was the self-clinching through-hole type. This pic shows how it was pushed through:

Obviously, breaking it this way isn't a concern with a blind hole.
Click to expand...

Definitely should not be able to pop those out like that - that BSO was installed improperly is the only explanation. Is the prototype the only time you've seen this happen?

Here's a picture of a rackmount case aluminum front panel we make with a CH fastener installed in it

GNNpyBn.png


Thicker aluminum, but the appearance is consistent down to 062.
 
Apathy_Enthusiast said:
Definitely should not be able to pop those out like that - that BSO was installed improperly is the only explanation. Is the prototype the only time you've seen this happen?
Click to expand...
You know, now that I think about it it could actually be a blind hole and it simply broke through. The material would only be ~0.5mm there, so it's not hard to imagine that happening.

It's the only time I've seen it happen on the M1, but that's not surprising since a small brace was added to the back of the standoff for the production version. The other three motherboard standoffs are braced by the chassis.
 
Necere said:
You know, now that I think about it it could actually be a blind hole and it simply broke through. The material would only be ~0.5mm there, so it's not hard to imagine that happening.

It's the only time I've seen it happen on the M1, but that's not surprising since a small brace was added to the back of the standoff for the production version. The other three motherboard standoffs are braced by the chassis.
Click to expand...

Wouldn't leave that hexagon pattern if it was blind hole. BSO has that standard hexagonal head shape, the hex pattern keeps it from rotating when it's pressed into the metal. They also typically have a groove at the base that the displaced material fills in to keep it from popping out like that.

BSO.jpg
 
Apathy_Enthusiast said:
There's zero risk of galvanic action so long as the computer is in a climate controlled environment <80% humidity. Ex right now it's raining outside and our outdoor humidity is only 40%, in the office it's 30% (only get 5% increments on my gauge).
Click to expand...

Just want to point out that this is something that differs greatly depending on where you live. I live by the coast in sweden, and 80% is probably pretty close to our average constant humidity. Today looks to be a rainy day and we will be at >90% for the next 24 hours at least.
 
Apathy_Enthusiast said:
You'll probably find few sheet metal fabricators outside like car part manufacturers and maybe aerospace (and Lian Li) have stud welders with any volume capabilities - it's simply a very specific tool that produces results not much different from what can be accomplished with another set of tools for with more diversity in use...
Click to expand...

We tried to get a car parts mfg for this reason - anodizing is another factor of consideration.
 
xinux said:
Just want to point out that this is something that differs greatly depending on where you live. I live by the coast in sweden, and 80% is probably pretty close to our average constant humidity. Today looks to be a rainy day and we will be at >90% for the next 24 hours at least.
Click to expand...

Apathy_Enthusiast said "There's zero risk of galvanic action so long as the computer is in a climate controlled environment <80% humidity". If you have a dehumidifier (e.g. air conditioner), you shouldn't reach 80% indoors. But yes, your point is valid in a non-controlled climate environment.

Very interesting discussion, by the way. Learning a lot. :)
 
Just a small note, air conditioners are less common in Scandinavia, than the hotter parts of Europe.
More people probably have AC as a side effect of owning a heat pump than stand-alone air conditioners.
Hence the notion of climate control is not applicable to most of us during the short summers we experience. :)

Also, I would think leaking currents also could accelerate oxidation under the same humid conditions.
(Since Norway use IT ground, we are more prone to have this problem than other countries.)
 
lypse said:
Suggestion.
Rotate the GFX Card so MB and Card face the same way, that way you can have a clear side panel for people into that. :)
Click to expand...

iFreilicht said:
Sounds easy when you say it, but that requires all sorts of changes to the design, as well as the inclusion of a shielded PCIe riser which would highly increase the cost. I personally like the idea a lot, there's a reason why I tried to design two cases with this concept, but this was probably already considered and ditched by Necere.
Click to expand...

Also wish such a design was more common without making the case larger. But as Silverstone hasn't managed it then I'm not excepting a smaller company/individual to lead the way.
 
lypse said:
Suggestion.
Rotate the GFX Card so MB and Card face the same way, that way you can have a clear side panel for people into that. :)
Click to expand...

Tinkerer said:
Also wish such a design was more common without making the case larger. But as Silverstone hasn't managed it then I'm not excepting a smaller company/individual to lead the way.
Click to expand...

Lian Li's O-series cases are laid out that way, though they are much larger than the LRPC. Aside from wanting a window, there is no reason to do it, and plenty of reasons not to. For example, if we just flip the GPU, there's a couple of things that become apparent:

JjS0ib2.png


Right away we can see the PCIe slot will potentially be blocked by the GPU, and we'll need a much longer shielded flexible riser (which adds cost and potentially affects performance/reliability). We can move it up vertically or over horizontally to give the connector some room, but there's another problem:

8j51JCr.png


Having the GPU exhaust directly into the CPU area is the last thing you want if your goals include the words "cool" and "quiet." The current unflipped layout makes a lot more sense for both the riser connection and airflow:

DQDuFHX.png


Now, it's true that open-cooler GPUs exhaust from both sides, but the way video cards are designed (and the offset of the rear bracket in particular) means if you want to isolate the GPU like I've done with the LRPC, there's actually some room for air to flow:

x096n7Yh.png


Some GPUs are also designed for this side to be the primary or secondary (in blower cards) exhaust, so blocking it off is a poor idea.

Aside from those problems, there's the issue of how would the GPU, CPU, and PSU get air with a window? Do we punch the window full of holes? Dust filters are obviously out, as well.

So it would be bigger, perform worse, and accumulate dust. Ultimately, flipping the GPU just to add a window involves too many sacrifices, and makes little sense for the design goals of the LRPC.
 
You must log in or register to reply here.
Back
Top