NCASE M1: a crowdfunded Mini-ITX case (updates in first post)

WiSK said:
Interesting, the net cost of refunds is 0 at Stripe. That means, if not enough people order an M1 for a new production run, then the only loss would be the prototype and the $20 for a Squarespace pro account. But, would they really allow someone to do several hundred refunds at once?
Click to expand...

Although I have lots of experience using Squarespace, I haven't used Stripe, so I have no idea how practical they would be. Anecdotally, I know a few people that use the service through SS and through Stripe's own APIs, and they've been happy.

Zizu32 said:
Ok, now I can say it's REALLY quiet !! ;)

Few pages back thanks to Necere you could see my build with non-modular ATX PSU. It was nice setup, but I could still hear PSU vent from some distance.
So I look around what I can do about that... and went with this full modular quiet Queen from Corsair.
Click to expand...

I initially thought that any config with an ATX PSU would be clunky in the M1, but then I saw yours - Well, well done! You should keep the side off so you can always look at the internals. ;) Or cut out a window, perhaps, if you're feeling adventurous!
 
Stevo_ said:
I have tested mine with PWM fans an they work.
Click to expand...

That's because voltage control always works.

The finer PWM control only works on fans with PWM circuitry, but the fan will still largely work with voltage control as well.

Stevo_ said:
On my P8Z77-I I tried some non-PWM fans on the chassis headers and the fan control control had no effect, they just ran at a constant speed.
Click to expand...

Yes, because they just get the 12V and ignore the PWM signal.

Both headers on the P8Z77-I are PWM only. Many users were disappointed that their fans just ran at full speed. I think this is exactly the reason they changed it for the Z87I and the Impact.

What would be ideal is that all four headers could be both PWM and voltage controlled, depending on the fan, but this isn't the case. According to the ROG support forum, only the CPU header works this way. I don't know if that's a cost issue, or some kind of technical limitation.

All this confusion about it comes because the 4-pin PWM system is backwards compatible with the 3-pin voltage system. On the one hand that's a good thing, because both types of fan still work with both systems of control. On the other hand it means that fans aren't controllable as they could be.
 
WiSK and Aibohphobia are correct, chassis fan headers on the M6I are only voltage controlled - i even checked that with a multimeter a while ago, cpu is PWM or voltage depending on BIOS setting, but the rest is voltage only. PWM fans will work of course, but will not have as large range of operation.
 
Herstal said:
WiSK and Aibohphobia are correct, chassis fan headers on the M6I are only voltage controlled - i even checked that with a multimeter a while ago, cpu is PWM or voltage depending on BIOS setting, but the rest is voltage only. PWM fans will work of course, but will not have as large range of operation.
Click to expand...

I'm pretty sure the non-CPU headers are indeed voltage only but detecting a PWM signal would require an oscilloscope. There's some discussion on the topic here: http://www.overclock.net/t/867504/diy-bings-pwm-fan-controller/20_20
 
Sure, but you can see voltage regulation (5-12V iirc) or lack thereof (constant 12V) on pins 1-2 with a multimeter.
 
The advantage of running PWM fans on a PWM header is guaranteed RPM maintenance. This means, it will start and hold the minimum specified RPM. Like my Noctua U9B fan, 300 RPM idle is easy.

A Voltage controlled header will perform similarly across both PWM and non-PWM fans. Varying the voltage can change the fan speed. For full RPM usage, its no different, but for lower rpm usage, voltage control is more primitive, and less controlled. Although voltage control can sometimes allow you to get lower speeds than PWM headers, the innate method of voltage control does not guarantee that fan will start reliably if the voltage is set too low.

The Nexus silent 120 fan maxes out at 1000 rpm, but can be voltage dropped down to 4 volts to hold a stable 300ish rpm. It however will not start until the voltage is raised to 7V?, which corresponds to about 7-800 rpm. Thus, there is a range from 300-800 rpm where the fan cannot be guaranteed to start.

My build targets idle fan speeds of 300-500 across most fans, and thus, PWM fans are my only safe way to ensure low rpm compatibility.
 
So which ITX motherboards actually use proper PWM on all their fan sockets? Or do you need a fan controller?
 
Peter_364 said:
So which ITX motherboards actually use proper PWM on all their fan sockets? Or do you need a fan controller?
Click to expand...

Most boards have PWM on the CPU header. I only know of the Asus P8Z77-I that has PWM on the channel fan header as well. Graphics cards have PWM, but the connector is different, so you need an adapter.

As for fan controllers, usually they are voltage controlled. Only the Aquacomputer Aquaero 6 has more than one PWM header that can also provide voltage. It's quite an expensive solution. For very large builds with multiple radiators and 20+ fans then PWM is ideal and the Aquaero 6 is really more suited to those.
 
WiSK said:
That's because voltage control always works.

The finer PWM control only works on fans with PWM circuitry, but the fan will still largely work with voltage control as well.



Yes, because they just get the 12V and ignore the PWM signal.

Both headers on the P8Z77-I are PWM only. Many users were disappointed that their fans just ran at full speed. I think this is exactly the reason they changed it for the Z87I and the Impact.

What would be ideal is that all four headers could be both PWM and voltage controlled, depending on the fan, but this isn't the case. According to the ROG support forum, only the CPU header works this way. I don't know if that's a cost issue, or some kind of technical limitation.

All this confusion about it comes because the 4-pin PWM system is backwards compatible with the 3-pin voltage system. On the one hand that's a good thing, because both types of fan still work with both systems of control. On the other hand it means that fans aren't controllable as they could be.
Click to expand...

I realize that voltage control always works, and the reason why I added that the fan control had no effect on non-PWM fans, therefore no voltage control! My main point for the chassis was control was possible despite the fact that people were swearing up and down they were fixed headers on the P8Z77. On the M6I UEFI though it didn't mention voltage or PWM, the menu looked the same. The fact that the chassis fan lower limit was raised to 60% from 40% which would be 7.2V about the minimum starting voltage for many fans might be an indication though. Although the PZ88 has a 20% minimum and that was raised to 40% apparently
 
Last edited:
WiSK said:
That you fans/pump are running at particular RPMs, is not evidence that they are being PWM controlled. According to ROG forum support the board only has PWM control on the CPU header, the rest are voltage controlled.

You can use a needle to remove the blue wire+pin from the fan connector and see if you can still control the fan speeds without it.
Click to expand...

Phuncz said:
Most people won't even know the difference between PWM- or voltage-controlled, they just want to regulate their fans based on a temperature. That most motherboards aren't true PWM-controlled on all their fan headers is more common than many think, but so irrelevant to the result that most don't even know it is that way.
Click to expand...

Aibohphobia said:
They are 4 pin connectors but if you read the manual you'll realize that the fourth pin is just a steady 5V signal which just causes PWM fans to run at full speed. The fan control on the non-CPU headers are strictly voltage adjustments not PWM control. It's a bit deceptive on Asus's part.
Click to expand...

Very deceptive indeed. Four 4-pin connectors but only one that is truly PWM capable. Like Phuncz said, despite it all, temperature based voltage control and AISuite work for my setup. I guess there may be cases where multiple PWM is required - apologies. I obviously didn't read that part of the manual :eek:
 
Zizu32 said:
Ok, now I can say it's REALLY quiet !! ;)

Few pages back thanks to Necere you could see my build with non-modular ATX PSU. It was nice setup, but I could still hear PSU vent from some distance.
So I look around what I can do about that... and went with this full modular quiet Queen from Corsair.
img_0064g1lqd.jpg

img_0066wsxln.jpg
Click to expand...

How have you liked you short GPU? Good?
 
Herstal said:
WiSK and Aibohphobia are correct, chassis fan headers on the M6I are only voltage controlled - i even checked that with a multimeter a while ago, cpu is PWM or voltage depending on BIOS setting, but the rest is voltage only.
Click to expand...

Confirmed here as well now. 22.5% duty cycle on PWM line and 12v on positive for CPU FAN at idle, and 5v on PWM line and variable voltage on positive for CHASSIS FAN 2 depending on temperature.

Aibohphobia said:
I'm pretty sure the non-CPU headers are indeed voltage only but detecting a PWM signal would require an oscilloscope.
Click to expand...

Many relatively inexpensive multimeters now provide a duty cycle for PWM signals. Mine reports the same duty cycle as Fan Xpert 2 on the PWM header.

okwchin said:
The advantage of running PWM fans on a PWM header is guaranteed RPM maintenance. This means, it will start and hold the minimum specified RPM. Like my Noctua U9B fan, 300 RPM idle is easy.

A Voltage controlled header will perform similarly across both PWM and non-PWM fans. Varying the voltage can change the fan speed. For full RPM usage, its no different, but for lower rpm usage, voltage control is more primitive, and less controlled. Although voltage control can sometimes allow you to get lower speeds than PWM headers, the innate method of voltage control does not guarantee that fan will start reliably if the voltage is set too low.

The Nexus silent 120 fan maxes out at 1000 rpm, but can be voltage dropped down to 4 volts to hold a stable 300ish rpm. It however will not start until the voltage is raised to 7V?, which corresponds to about 7-800 rpm. Thus, there is a range from 300-800 rpm where the fan cannot be guaranteed to start.

My build targets idle fan speeds of 300-500 across most fans, and thus, PWM fans are my only safe way to ensure low rpm compatibility.
Click to expand...

I decided to try Cougar Vortex PWM fans for this build. They seem to work well under voltage control. Depending on the fan, they have minimum rpm speeds of 276 - 339, and minimum operating rpm speeds of 502 - 518. Fan Xpert 2 however won't allow operation below minimum operating rpm even though it starts fans above minimum operating voltage. Still that is a gap of less than 250 rpm, and 502 rpm is pretty good for voltage control.
 
I have my fans set at "silent" in the Asrock UEFI setup, which is about 50% I believe. When I start Windows, the Asrock A-Tuning app auto-launches and adjusts the fans to my setup that is more dynamic and performs with less noise. But this way, the fans always start.
 
Any recommendations on how to to simultaneously load CPU and GPU? It's my first fully wc'd rig, so i never had to do it before. Valley + Prime doesn't work that well. Maybe Prime + Cudaminer?

I played a bit with the finished rig over the last few days - did as some of you advised, moving the fans to separate headers and setting it up in AI Suite (gaaah, the bloat is killing me ;)). Did some tests with various pump speeds, it has a considerable impact on temperature. Tested in Valley, I got GPU and CPU temps of ca. 65*C @ 1200 rpm (0%), ~50-55*C @ 2000 rpm (40%) and ~45*C at 3000 rpm (100%) - all this with the fans maxed, CPU at 4.0 @ 1.3V (yeah...) and GPU at ~1150/3400 @ 1.212V.

With a slightly tweaked fan profile it is now reasonably silent at idle, so i'm satisfied. Load is more or less as it was, but that's mostly due to the PSU fan. Temps in Valley with the "real" fan profile do not exceed 60, so it's alright. Did a quick Firestrike, and i must say i'm very happy with the result - http://www.3dmark.com/fs/2098090. It is absolutely fantastic what you can fit in such a small and good looking case, while keeping it cool and relatively quiet. Thanks again, Necere and Wahaha :D.
 
Have you ever considered building a separate bigger case? Maybe compatible with 280mm radiator or 360mm? And maybe mATX? :)
 
Hey guys. I'm still not building anything until silverstone releases the 600 watt SFX, but new stuff keeps coming out any I keep morphing my future build around it.

This new C style cooler came out from CRYORIG.

http://www.cryorig.com/c1.php#video

This one has a really strange heat pipe design. But it seems to still make sense. It comes with their slim 140mm but since the M1 can handle the height, I think I'll buy their full size 140mm fan for it. With the thicker fan, the entire thing will be 86mm tall.

I want to get a 4770K for my build. The only thing I'm worried about is that there will still be a substantial gap between the fan and side panel.

Was wondering what temps you guys are getting with coolers of similar height, shape and fan size that are relatively far away from the side panel.

Also something to keep in mind, I plan to put a Sapphire R9 290X Tri-OC in there as well, so there will be some residual heat in the case that I have to deal with. I'll be putting a 92mm beQuiet fan in the rear fan mount to exhaust.
 
Thanks again Necere/Wahaha360. Just finished up my build. Turned out great besides the stock cooling. I plan on getting the parts for my custom loop sometime this month. You guys did an amazing job, probably the easiest build I have ever done.


 
is there any non-modular atx psu that fit in with a full sized GPU and have a decent amount of power output?
 
xxroxx said:
is there any non-modular atx psu that fit in with a full sized GPU and have a decent amount of power output?
Click to expand...
There are a few 700W+, see this list. Keep in mind that it's really not advisable unless you're willing to mod the cables, since the average ATX PSU has such an excess that it's next to impossible to stuff them all in the case.
 
I can advice Silverstone 600W. It's 140 mm tall, really quiet unit.
You should squize long GPU with some workaround with cables.
For me just wires sleeving wasn't acceptable.

img_9681qzuai.jpg
 
Aibohphobia said:
Not sure if it applies to all the Asus Z97 boards but at least on the Z97-Deluxe all the chassis headers can use either voltage or PWM control according to the manual (page 59): http://dlcdnet.asus.com/pub/ASUS/mb/LGA1150/Z97-DELUXE/E9061_Z97-Deluxe.pdf
Click to expand...

I do recall reading on some mobo manuals back when I was looking for my mobo that even higher end 87 boards have started offering this too. Good to see it being implemented in newer boards. Essentially, providing both means you are combining two fan controllers on the same header. So having 4 headers that are PWM and DC controllable, means you are getting 8 independent fan controllers, but can only use 4 at a time.

You can certainly understand why its not a feature thats provided on budget boards yet, as thats all extra costs here and there for a benefit that many people don't yet appreciate. But I know I certainly would LOVE to have that.

The other feature that is not commonly available, and a major selection point for my mobo, is the range of fan control, and the ability to control it via the bios. My Asrock board has a fan control range (for both PWM and DC) from 100% DOWN to 0%!! It has a 5 point adjustable temperature and output control, all from the bios. (good for running os's other than windows, so you are not limited to windows only fan control, and have less bloatware)

My Z68 Asus Maximus 4 board could only go down to 40% on the PWM header, and 60% on the voltage header. Fans slowed down, but were still in the audible category.

The Z87 Asrock board could stop fans, and you could find the exact point at which fans turn on, and control them from there. The only shortfall is having 2 headers, one PWM, one DC.
 
The Deluxe is a $400 board :eek: but hopefully all the ROG boards will also get PWM chassis headers.

You can override the minimum fan speed for the Asus boards (at least with Fan Xpert 2)

AI Suite, also, don't allow these fans to be set quiet. After Thermal Tuning AI Suite Sets them to min. 60% (obviously, AI Sute "reads" this from BIOS)
Currently I've made small "hack" in FanCalibrationData.xml in C:\ProgramData\ASUS\FanXpert
At this line I put this for both Assist Fans <maxdutytokeepstill>35</maxdutytokeepstill> which means Min. Duty Cuycle of 35%
Click to expand...

http://rog.asus.com/forum/showthread.php?36621-ASUS-Bios-fan-control-request&p=337135&viewfull=1#post337135
 
tekm said:
Does anyone know if it would be at all possible to physically cram a 2nd 240mm (30mm thick) radiator at the bottom of the case along with a couple of thin 120 mm fans (i.e. Scythe Slip Stream) with a watercooled GPU present. Obviously air flow would be an issue, though I'm mainly concerned if it would fit or it there is some other obstruction that I am not aware of (I don't personally have the case...yet). Since air flow would be greatly impeded in such a setup, maybe a blower fan/pci slot case fan would work as a better solution to pull air out through the pci slot.
Click to expand...

It is possible to build an M1 with dual radiators. The components are:
  • 120mm x 240mm x 25mm radiator (correction: 30mm)
  • 120mm x 120mm x 25mm radiator (correction: 30mm)
  • 3 120mm x 15mm fans
  • GPU water block
  • CPU water block with integrated pump
  • external reservoir
Oh, in addition, a 140mm ATX power supply.

Not yet tidy proof of concept:

s731b.jpg



Front of the case and 240mm radiator:

f6vt.jpg



Back of the case and reservoir:

6pty.jpg



Top of the case:

y19c.jpg



Interior of the case:

v4aem.jpg



Not much space between the fans and the water block. In places, however, the fans are not covered by the water block.

t80n.jpg



Wider interior view of the case:

odyd.jpg
 
Last edited:
Gandergray said:
It is possible to build an M1 with dual radiators. The components are:
  • 120mm x 240mm x 25mm radiator
  • 120mm x 120mm x 25mm radiator
  • 3 120mm x 15mm fans
  • GPU water block
  • CPU water block with integrated pump
  • external reservoir
Oh, in addition, a 140mm ATX power supply.
Click to expand...

:D :D :D
temps?
 
Gandergray said:
It is possible to build an M1 with dual radiators. The components are:
  • 120mm x 240mm x 25mm radiator
  • 120mm x 120mm x 25mm radiator
  • 3 120mm x 15mm fans
  • GPU water block
  • CPU water block with integrated pump
  • external reservoir
Oh, in addition, a 140mm ATX power supply.
Click to expand...

That's AWESOME! :D So many questions!:

1. Have you had this up and running yet? What temperatures are you seeing?
2. What specific radiator did you buy for the bottom? And what fans? Any links to these?
3. Does the radiator prevent you from using the front I/O (by blocking the cables or requiring that you remove them)? Can't quite tell by the images.
4. How difficult was it to route tubing? What components in particular tended to interfere with that? Any particular thinking behind the order of the components in the loop?
5. How did you go about mounting the radiator and fans to the bottom? Did you use any filters?

Many thanks for sharing your build, images, and insights! :)
 
majic12 said:
I come from time to time to see how is the revision of ncase going on.Recently i did saw this case acubic a90 http://www.abee.co.jp/Product/CASE/acubic/A90/ (only japan) but appart from the limited vga restriction it does have some nise ideas for the inside layout.Just for isnpiration for the ncase creators :)
Click to expand...

Interesting design, but wow.. that's kinda big for an ITX rig, right?

If I calculated correctly, it is exactly twice the size of the M1 (A90=25L, M1=12.5L). Looks pretty nice, though :)
 
AFD said:
Interesting design, but wow.. that's kinda big for an ITX rig, right?

If I calculated correctly, it is exactly twice the size of the M1 (A90=25L, M1=12.5L). Looks pretty nice, though :)
Click to expand...

Well, Majic12's last post linked to a case three times the size of the M1, so he's heading in the right direction.
 
AFD said:
Interesting design, but wow.. that's kinda big for an ITX rig, right?

If I calculated correctly, it is exactly twice the size of the M1 (A90=25L, M1=12.5L). Looks pretty nice, though :)
Click to expand...

What's the VGA restriction you are referring to? I'd love to get one of those, but the import cost is just sooo high......

Any larger, and we got ourselves another set of BitFenix Prodigy / Caselabs S3
 
Yup!Im still in search for the best case.I like so much the ncase but if im not mistaking,if the case is smow than heat will gather quicly and in order to have it stable the heat will need to be extracted quicly.Im just puzzled with the current ncase air cooling.I think it realy needs to be better and for that in my opinion it shoud grow in size a litle more.

To bloods question.The size limit of the gpu is 260mm.I shaw wait and see how the revision of the ncase goes because i realy want to build a small powerful workstation to get with me on a plane and go to another country in the near future.I did read a story with someone having a bitfenix prodigy on a plane next to him xd but mini itx shoud be the way too go.
 
majic12 said:
Yup!Im still in search for the best case.I like so much the ncase but if im not mistaking,if the case is smow than heat will gather quicly and in order to have it stable the heat will need to be extracted quicly.Im just puzzled with the current ncase air cooling.I think it realy needs to be better and for that in my opinion it shoud grow in size a litle more.
Click to expand...
If you think the Ncase is too small, you are in the wrong topic.

An alternative to the Accelero GPU cooler: http://www.bit-tech.net/hardware/2014/05/07/raijintek-morpheus-gpu-cooler-review/1
It looks like it might fit in the Ncase.

morpheus-9-300x292.jpg
morpheus-10-300x292.jpg


But this won't, probably:

morpheus-13-300x292.jpg
 
Last edited:
You must log in or register to reply here.
Back
Top