What's new

Custom Fan - Get More From Your SP3!

OP
Cothek

Cothek

Active Member
Actually you can build this without a 3D printer:
Mine is made with a desktop PC fan and thermoplast.
The fan can be easily snapped onto the surface and sticks to it very good due to the furrow of the surface.
Hence: easily removeable while adherent as hell while all sockets are still usable (dock support tested)

IMG_20140925_224112.jpg

Check out more pics at my google+ account
https://plus.google.com/102383660862021579153/posts/5Nv5EykKLUa



EDIT: I was able to use a big 92mm fan for more air flow with low noise
Total cost ~10 bucks
WIN_20140925_234456.JPG
Nice! This is what I like to see, the community coming together to solve a problem/feature (depends on how you look at it). You should post a XTU graph of the GPU for 10 min or so for comparison.

I ran the XTU graphics test with my Arctic fan.

View attachment 3785

As you can see, the TDP and frequencies are similar from what you get in your tests, but the graphics TDP take a bit of a drop after 10 minutes to 8W instead of 9-10W which is sat at for the first bit of the test. Also both throttling monitors go nuts, but the frequencies look similar to your tests so I'm not sure why.

I'm looking forward to a Rev2 that doesn't block the mini displayport. Finish it up already so I can buy one and try it out! xD
I've noticed that under the GPU test the CPU never goes over 1.7GHz but will drop if throttled.

To make it a safe fit for my adapters it would have to accommodate a half inch wide by 3/8 inch tall mini displayport plug.
Actually making the cutouts a little bigger and printing it over-night, I'll report in the morning.
 
OP
Cothek

Cothek

Active Member
I just realized that it may not be clear from my pictures or OP but the SP3 Duct is not blowing air on the back of the SP3 but into the vent holes on the side. This is a key difference between the Arctic Breeze fan and the one Flow28 posted and the SP3 Duct.

I'll update the OP to clarify.
 

Justing6

New Member
I just realized that it may not be clear from my pictures or OP but the SP3 Duct is not blowing air on the back of the SP3 but into the vent holes on the side. This is a key difference between the Arctic Breeze fan and the one Flow28 posted and the SP3 Duct.

I'll update the OP to clarify.

You should probably bold that or put it higher in your description as well. I didn't notice until you mentioned it and I took a closer look.
 

GreyFox7

Super Moderator
Staff member
What if you put the duct on the top and reversed the fan to pull air through the SP3, then you wouldn't have to work around the ports.
 

Justing6

New Member
What if you put the duct on the top and reversed the fan to pull air through the SP3, then you wouldn't have to work around the ports.

Well, whats nice about the current design is that the clips are passive and its more a combination of gravity and the usb connector being plugged in after the fan is placed that holds it in. If he moved it to the top he would have to redesign the clip system so the fan doesn't slide around or droop down the kickstand angle is moved.

Also, I'm not sure if that would even work thermal wise. The vent goes all around the surface, but its the right side where air is actually blown out of. I don't know if the top right vent actually leads to components or is just there for aesthetics.
 

Justing6

New Member
Nice! This is what I like to see, the community coming together to solve a problem/feature (depends on how you look at it). You should post a XTU graph of the GPU for 10 min or so for comparison.


I've noticed that under the GPU test the CPU never goes over 1.7GHz but will drop if throttled.


Actually making the cutouts a little bigger and printing it over-night, I'll report in the morning.

Good luck, after looking at the design closer I don't see how you could make a cutout for the displayport without making the vents for directing air into the surface considerably smaller. Please do another XTU with the new part when you get a chance. The last XTU you posted was just a broken image link.

Do you have an i5 or an i7 surface? With my i7 surface the CPU went to 1.9GHz several times and even 2-2.1GHz a couple times even after 10 minutes on a graphics test with just my arctic fan.
 

Flow28

New Member
Good luck, after looking at the design closer I don't see how you could make a cutout for the displayport without making the vents for directing air into the surface considerably smaller. Please do another XTU with the new part when you get a chance. The last XTU you posted was just a broken image link.

Do you have an i5 or an i7 surface? With my i7 surface the CPU went to 1.9GHz several times and even 2-2.1GHz a couple times even after 10 minutes on a graphics test with just my arctic fan.

Is a cutout for the display port this crucial? As you can connect an external monitor using the dockingstation, which utilizes the mag. plug if necessary
 

Justing6

New Member
Is a cutout for the display port this crucial? As you can connect an external monitor using the dockingstation, which utilizes the mag. plug if necessary

I could just use the docking station, but they cost $200 and are overkill for my port needs so I didn't intend on buying one. Still, if there is a performance decrease by adding in a spot for a display port I'll just have to pony up and deal with it.

However, if Cothek is more of an engineer at heart than we are (he already managed to think up this thing once) and manages to add it without sacrificing performance it would be a huge boon. I think it's at least worth some printed prototypes to see if it's possible.

Edit: To clarify, the main reason I would like the displayport open is for versatility sake. I plan to carry the duct in a pocket in my bag like I do with my arctic fan. More often than not I would use it without the displayport, but I occasionally use an external monitor to run games as I do not own a desktop computer. Yes I could buy a dock for use at home, but I plug in external monitors often at home, work, and in classes (labs). The duct might not be necessary in all these cases, but it sure wouldn't hurt.
 
Last edited:
OP
Cothek

Cothek

Active Member
Just pulled the newest fan off the printer. I added more clearance for the DP and some flow ribs:
IMG_20140926_102201.jpg


Here is an XTU GPU test run:
09-09 Intel GPU Balanced DPFan2.png


This is really tough to read but I super-imposed the first DP duct and the newest one:
09-09 Intel GPU Balanced DPFan Comparison.png


What i take from that jumbled mess is that the newest duct is working better (pay attention to the TDP lines), slightly, but better. I ran the duct with the DP slot last night playing Skyrim and it worked similarly to the one without the DP slot. This may sound weird but i think restricting the flow aperture (adding DP port) is helping because it causes a venturi effect and effectively speeds up the airflow.

What I'll do is make 2 versions available, one with DP and one without.

Also, to answer GreyFox's question, the SP3 internal fan is pulling air in at the top right corner so sucking air out by reversing the fan and placing it on the top right corner would starve the fan and would be much less effective. FWIW I've tried putting the fun up there and forcing more air though the internal fan but it does not work nearly as effectively as where it is now. In fact it will still throttle (extreme throttle where it drops CPU temp to 70C) with the fan on the top right.
 

GreyFox7

Super Moderator
Staff member
Cothek, have you conducted an air flow test to see where air is entering and exiting the SP3 around the edges? looking at the teardown I can't tell. its also not clear if the fan is just moving air over the heat pipes or if they are open to, the fan and its blowing air inside. Either way it would seem one is positive and the other is negative. air may be both entering and exiting in close proximity as well so if you are lined up precisely you may get optimal results.
 
OP
Cothek

Cothek

Active Member
Cothek, have you conducted an air flow test to see where air is entering and exiting the SP3 around the edges? looking at the teardown I can't tell. its also not clear if the fan is just moving air over the heat pipes or if they are open to, the fan and its blowing air inside. Either way it would seem one is positive and the other is negative. air may be both entering and exiting in close proximity as well so if you are lined up precisely you may get optimal results.
It depends on what you mean by "air flow test" but yes I have. I have tried moving the fan to multiple different locations along the top and I have watch this video which explicitly states how the internal fan works. Also, placing your hand around the edge while the internal fan is spun up will verify that the bulk of the exhausted air is coming out the right side around the ports. Along the top near the camera there is one small area where I have also felt hot air exiting the vents. Trust me when I say that I have tested this duct fairly extensively. In fact I have 8 old prototypes sitting on my desk.

What the SP3 Duct does is help exhaust the hot air much more quickly and most importantly keeps the external thermal sensor from throttling the unit.
 

GreyFox7

Super Moderator
Staff member
It depends on what you mean by "air flow test" but yes I have. I have tried moving the fan to multiple different locations along the top and I have watch this video which explicitly states how the internal fan works. Also, placing your hand around the edge while the internal fan is spun up will verify that the bulk of the exhausted air is coming out the right side around the ports. Along the top near the camera there is one small area where I have also felt hot air exiting the vents. Trust me when I say that I have tested this duct fairly extensively. In fact I have 8 old prototypes sitting on my desk.

What the SP3 Duct does is help exhaust the hot air much more quickly and most importantly keeps the external thermal sensor from throttling the unit.
If air only enters through the black duct from the right side above the display port and not from the top edge at upper right it would seem there's no advantage to air directed at points on the side below the display port. From what I can see in the video this is the case.

If air also enters from the top edge on the right then there's more intake space to work with for greater air flow. I don't think air comes in from the top but I can't be absolutely sure.

A wave of a punk stick (Or something else that could be seen depending on lighting available) with the fan running past the top and side should confirm that.
 
Top