msystems' gaming FC5 (Streacom FC5) 6.6L

Phuncz

Lord of the Boards
SFFn Staff
May 9, 2015
5,839
4,906
Awesome work ! They do sell those "pipe benders" in I would guess most hardware stores. You just need to check that they're for 6mm "pipe" :) I bought mine a while ago but have yet to try it on the heatpipes I have.
 
  • Like
Reactions: Lee.III.Will

Lee.III.Will

Caliper Novice
Jun 7, 2017
24
18
They do sell those "pipe benders" in I would guess most hardware stores.
They aren't as precise, or the one I bought wasn't. Mine was rated for 6mm 'pipe' and the bend came out okay, but I can promise that I got more than a .5mm distortion. Mine still work well but the bends in the last post look like what I'd expect from a factory bent heatpipe.
 
  • Like
Reactions: Phuncz

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
Awesome work ! They do sell those "pipe benders" in I would guess most hardware stores. You just need to check that they're for 6mm "pipe" :) I bought mine a while ago but have yet to try it on the heatpipes I have.

True,,, this tool is functionally no different than a pipe bender for regular copper pipe. It does have the ability to have a customized bend radius though, which is pretty nice.


I did get around to assembling the heatpipe cooling onto the GTX 1080 last weekend, using four heatpipes in an HDplex GPU block, just to see what would happen. That part went pretty well.

Incorporating the Accelero into the remaining four heatpipe slots is proving quite challenging. It took a lot of work to figure out how to disassemble the damn thing without causing damage to the heatpipes or fins. So I've pretty much destroyed the first Accelero, but at least I did figure out how to disassemble it :)

Should have a real update next weekend if I can get the whole thing to come together.
 
  • Like
Reactions: robbee

robbee

King of Cable Management
n3rdware
Bronze Supporter
Sep 24, 2016
861
1,349
So I've pretty much destroyed the first Accelero, but at least I did figure out how to disassemble it

I think we've all been there. What matters is that it got you somewhere. And if it didn't... For science!
 

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
New Loot

After months of waiting, I've just received two of the last items needed to complete the build, the HDPlex 400w dc-atx and 300w Ac-DC!





Here's a look at how they fit into the unused corner of the case:



This is by no means final, but it's a good place, for now.

Most importantly, finally a proper power plug!!!! And no more ghetto rigging the power.

Erm, don't worry, I wasn't trying to plug that in there :D



You might remember the Enermax SFX PSU from earlier which received a nice.. haircut.
Goodbye sweet prince! :D Your sacrifice will not be forgotten




Now that everything is here finally, its time to show all the things I've been accumulating to start work on the GPU side:

(Rest in peace to my credit card)


-Two HDPlex GPU assemblies, and a HDPlex CPU assembly. Thanks to SFF Forum member Lee Will for offering to provide me with these!

-Two Arctic Accelero S3 GPU coolers, because I am pretty sure I will accidentally destroy the first one.

-All the M2.5 Nuts and bolts I could ever use...

-Heatpipe brackets for the case wall. Special thanks to Streacom for providing me with extra brackets that are specifically designed for the FC5 and screw into the existing holes on the chassis (so I don't have to use the random ones I got from China).

-Liquid Ultra for chip surfaces, MX-4 for heatpipes, Arctic Alumina permanent thermal adhesive for... permanent things?

-A full array of Enzotech forged copper heatsink goodies in the 6.5mm, 10mm, and 13x78mm varieties to assist with VRM cooling in various areas.

-Kapton tape and two-pin thermistors for assisting with monitoring and measuring the effectiveness of the modded cooling heat transfer.

-Thermal pads and shims

-The 3d printed pipe bender, and a new tool I modeled and 3d printed needed to flatten my heatpipes to fit into the profile of Streacom's brackets.
 
Last edited:

AleksandarK

/dev/null
May 14, 2017
703
774
I think this will be very very cool build.

Are those two at bottom left corner coolpipes? For how much they are rated for?
 

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
GPU Part 1: GTX 1080 meets the FC5 chassis

I set out to test the FC5 chassis to determine how many watts it could handle and if my heatpipes would work.

Since the CPU side struggles with over 50 watts, expectations weren't exactly high...
and the GTX 1080 has a TDP of 180 watts. :eek:

First thing I did was strip off the stock cooling, which I found a bit interesting.



Most of the stock cooler is fancy plastic and brushed metal parts to just add to the looks. The heatsink itself is small. If they only took that money and put into a higher quality cooler, it might not even be necessary to replace it. EVGA also added a second heatsink to the rear of the unit which doesn't even contact the VRM. It seems ineffective, but it makes it look like the entire unit is a solid heatsink inside. Marketing?


A Cleaned up GP104. The memory will need heatsinks.


The VRM will need cooling, somehow.



This is a HDPlex block going on there. It weighs a lot.

The card needed to be supported somehow. Foam cubes won't cut it anymore.

I came up with this:

It's a 3d printed leg with a m2.5 screw hole. 43.5mm Height



"You must construct additional Pylons!"




Now the HDplex GPU block can be mounted.



I bent more pipes.






There's just one problem here:

Round peg in a square hole.

So I made this tool to flatten the pipe to the size of Streacom's brackets:




And after smushing:





And assembled:



I also added memory heatsinks. These are Enzotech 10 x 14mm forged copper.


A few of them had weaker copper I guess, and just fell over :D

Lol.. no, actually I had to bend them. You can see the reason:


They still work, I hope? :D

Here is the VRM cooling:

The heatsink is Enzotech 13 x 78mm forged copper. It is held on by a 3d printed bracket I made for the purpose. Unfortunately the bracket fell off later, perhaps from the heat, so it needs to be improved to be a little thicker.


Card is ready!


And fully assembled. I also added a second identical heatsink to the topside of VRM area. A temperature sensor was placed on the furthest heatpipe bracket.

Testing in Heaven



Huh? It actually handled stock 1.2V (180w) for a good 5 minutes+ before I dropped the voltage to .8V (115w) to keep temps below 80C. The chart above shows about 10 minutes of testing.

Even though the temps eventually reached 80, it appeared that the GPU side of the case was able to absorb almost double the wattage of the CPU side, exceeding 100 watts.


But that should be impossible... since they are identical.

It really didn't make any sense until I went to touch the side of the case above the temp sensor.
It was HOT, and I mean lava hot.

The reason I hadn't even considered was thermal resistance.


Above: The temperature sensor on the case wall registers only 9C lower than the GPU core.

Differences in Thermal Resistance:

1. The GPU Chip makes direct contact with a copper block, whereas the CPU side does not due to the IHS and it then makes contact with Streacom's aluminum block.

2. The GPU block has round heatpipes, which yield superior heat transfer compared to the CPU side which has already flattened heatpipes inside the evaporator block.

3. The GPU side has longer heatpipes with a larger bend radius which span the entire wall of the case, terminating in three evenly spaced areas, where as the CPU side has shorter heatpipes with a smaller bend radius which only cover 2/3rds of the case wall length.


So, mystery solved! The thermal capacity of one of the FC5 walls can approach 100 watts, *if* the thermal resistance is low enough. I'm inclined to leave the CPU side the way it is for now because the GPU heat even penetrates through the entire front faceplate of the FC5 and reaches the CPU side.

Next I will test the Accelero S3 and see how it performs on its own. If it is able to dissipate around 100 watts, then the combination of these cooling solutions should be able to handle the GTX 1080's TDP.
 
Last edited:

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
GPU Part II: The Accelero S3

I wanted to test the Accelero independently to get an idea of the capabilities. It's not as long as a full length card, but it is a bit wider than the standard 98mm card. This is no problem in the FC5's case layout which has plenty of horizontal space.


It comes with a nice backplate heat spreader as well, but it is too tall to be used here as the FC5 only has 8mm clearance above the top of the GPU pcb.



Official Specifications from Arctic:
Cooling Capacity:
Up to 135 Watts passively
Up to 200 Watts with Turbo Module

recommended with properly ventilated case only

Without the backplate and much ventilation, 135 watts is a stretch. 100 watts is a realistic goal.

Here is the card after install. The memory heatsinks were too tall to fit, so I had to take them off. I taped a heatsink on the VRM:



Installed in case and ready to test:



I loaded up Heaven for testing. Power limit to 50% (90 Watts)



Temps got out of control right away and the test had to be stopped. Fully passive, the performance was more like 40-50 watts in this scenario. The Accelero does not fare well sitting inside of an already warm case.

Not good, but that is not the whole story yet. I took the 3d printed fan duct made earlier and flipped it like this:



This can simulate modest airflow from adding an intake fan.



These 50mm fans only push 11.5 m3/hour (6.7 CFM) each. Datasheet
So it's realistic to expect that adding an intake fan here could match or exceed that.

Temps began to drop...


And stabilized at 67C...

Stable, at 50% Power limit (90 watts).

So the Accelero is capable of dissipating 90+ watts, given adequate airflow.
Now knowing this, it should be possible in theory to cool the GTX 1080 at full TDP in the FC5.

If only the Accelero can be linked up with the HDPlex copper block...
 
Last edited:

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
GPU Part III: Unforseen Consequences

In the last posts I tested each cooling system independently.
The next step is to connect the Accelero's four heatpipes to the four open slots on the HDPlex block:


Not necessarily those four slots though - the pipes can be ordered in different ways as long as the bends will permit it.



The Accelero's original heatpipes are too short - they only reach into the middle of the block, so they need to be removed and replaced with the spare heatpipes. This greater heatpipe length also allows the Accelero's position in the case to be offset to the right, allowing the largest possible intake fan.

I did some searching for information about how to go about disassembling a heatpipe cooler. The best resource I found was "How to make a heatpipe-based cooling system on your own" in which the author disassembles a Thermaltake cooler and successfully removes the heatpipes from the fins.

So I made a Plan A, and a Plan B.

Plan A was to just pull the heatpipes out of the Accelero. I examined a lot of review photos and it didn't look too bad. There was no glue or solder visible.



Unfortunately what I couldn't see until inspecting them in person, is that what was actually done by Arctic was much more permanent than glue or solder. Each fin is mechanically clamped onto the heatpipe at the factory. It left slight marks in the aluminum rings.


Pulling the heatpipes out would be impossible without damaging the fins.

Fortunately there was Plan B: to painstakingly remove each fin as done in the Silentmods guide.

I must warn you that this job is not for nervous people. The risk of damaging the pipies in the process is high as the pipe wall thickness is not much more than that of a sheet of paper and aluminium fins are tightly settled and fixed with epoxy adhesive… Think twice before ruining a good cooler.

With that in mind, I attempted to remove the fins, however this was extremely difficult (again, due to the mechanical clamping) and led to warping the aluminum fins. I didn't even make it past the first two fins before calling it quits on that.



I was pretty bummed at this point. I considered just cutting up the heatpipes to get the fins out. Another option was to see if maybe I could use the existing heatpipes. But even if that were possible, they were solidly connected to an existing bracket.



The pipes seemed to be glued or soldered on. It took a lot of work but I was able to pop off the top of the bracket by putting it in a vise.


Soldered and glued to the bracket. Bad because I don't have any experience or tools for desoldering.

I referred to Silentmods again:

Now you can unsolder the pipes on a hotplate. The cooler is soldered with a low-melt alloy the melting temperature of which is about 100° C. This procedure must be carried out wearing gloves. Put the cooler's base on the hotplate and hold on to the pipes with your hands. After the solder has melted rock the heatpipes slightly and then quickly take them out. Wipe the solder remnants at once with a rag. Don’t overheat the pipes and don’t cool them with cold water right after desoldering!

So I did try putting it on a heated buildplate to 100C in the hopes that it was a lower temperature solder. It wasn't. When that didn't work, I tried to pull the bracket apart with mechanical force by threading steel wire to the screw holes and reversing the vice. That took a lot of work and ended up doing a lot of damage.

I did get the bracket off, but by then, the heatpipes looked like this:



Not sure if this voids the warranty.



Who wants some free scrap metal!?!

Time to make a Plan C...
 
Last edited:

jeshikat

Jessica. Wayward SFF.n Founder
Silver Supporter
Feb 22, 2015
4,969
4,781
Poor Accelero, at least your sacrifice has advanced the scientific pursuit of high-wattage SFF passive cooling!
 
  • Like
Reactions: AleksandarK

Phuncz

Lord of the Boards
SFFn Staff
May 9, 2015
5,839
4,906
Sometimes it works out and sometimes it's manufactured so painstakingly well you wished it was done worse. Regretably you bumped into a huge obstacle at every turn with this part.
 
  • Like
Reactions: AleksandarK

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
Sometimes it works out and sometimes it's manufactured so painstakingly well you wished it was done worse. Regretably you bumped into a huge obstacle at every turn with this part.

Yeah thats kind of funny to think about. If only Arctic had made their product with lower quality, maybe it would be easier to disassemble.
 
  • Like
Reactions: AleksandarK

msystems

King of Cable Management
Original poster
Apr 28, 2017
786
1,373
GPU Part IV: Dr. Frankenstein's Laboratory




Previously when researching how to bend heatpipes, I found this advice pretty funny:

What an idiot. That's the last thing I would ever do.

Oh the irony...


This is Plan C.

once you heat them, they will rupture (if not explode) and be as useless as hollow copper tubing. heatpipes are specificly engineered for their specific interior volume and filled with a very specific amount of gas, when you bend them you will most likely change this volume and they will not work nearly as well, if it still works at all.

Right. So anticipating a face full of copper shrapnel, I brought out the second undamaged Accelero I had and heated the bracket until I could see the solder melting off. I was able to gently remove the bracket with pliers.





The heatpipes did not rupture. Actually they suffered no visible damage at all this time.



I did some light sanding to smooth out the melted solder.

Here is the first mockup. The position of the Accelero is ideal here, because it to the right of the case:




The heatpipe order here is not ideal though. There is no way for the last two pipes to be routed without more bending.

Either the long pipes have to be bent, or two of the Accelero's pipes will have to be bent, like this:



Here is how that went:



The heatpipe bender tool couldn't really be used here. Too cramped and the bend radius was too large to be useful. I used a round metal bar to help shape the curves, but really it was done by hand.



The last one was really tough to get in. I wanted it on the end, but there was no other way to do it without bending the pipe too tightly or having it stick out to the side or top where it will collide with other things. I did not want it in the center slots above the core... too risky to put only half of a heatpipe there.

Some of the bends look a bit sharp, but the diameter is above 4mm for the most part, so they are still functional.



Preparing to put it together, another problem surfaced.

In the CAD model, I only cared if things fit but not how they fit together, and I hadn't given any consideration to assembly.

First thing, the blocks are not able to create any clamping pressure without drilling out one side. I drilled out the top block.



Then, there wasn't an obvious way to attach the blocks to the card - The Accelero's fins block the screw holes on one side, and also the GTX 1080 blocks them from the other side. So I switched to machine screws and then countersunk the holes so it would not cause a clearance issue against the GTX 1080.



And, because the card is being mounted upside down in the case, unlike the HDplex H5, it is not possible to adjust heatpipe positioning after the assembly is mounted to the card because the screws are inaccessible.

I used the existing heatpipes as a guide for which to get the assembly positioned without needing to mount the card first.




The other block could be mounted now.



This fun problem was discovered next while checking alignment:





The PCIe riser was slightly blocked by one of the heatpipes.

Then this fun problem:



I had shims under the Accelero to support the weight and they were too high. The cooling assembly is about 40mm thick, but the card height is 43.5mm from the floor of the case and the GPU silicon package is about 2mm thick, so I made some 1.5mm shims to place under the Accelero to keep the entire card supported and not stress the PCB.

That was the last of the mounting problems. Moving on



I used Liquid Ultra this time:



And the card was assembled.. Here are a few last shots before it went in:








All rigged up:

 
Last edited:

AleksandarK

/dev/null
May 14, 2017
703
774
GPU Part IV: Dr. Frankenstein's Laboratory




Previously when researching how to bend heatpipes, I found this advice pretty funny:

What an idiot. That's the last thing I would ever do.

Oh the irony...


This is Plan C.


Right. So anticipating a face full of copper shrapnel, I brought out the second undamaged Accelero I had and heated the bracket until I could see the solder melting off. I was able to gently remove the bracket with pliers.





The heatpipes did not rupture. Actually they suffered no visible damage at all this time.



I did some light sanding to smooth out the melted solder.

Here is the first mockup. The position of the Accelero is ideal here, because it to the right of the case:




The heatpipe order here is not ideal though. There is no way for the last two pipes to be routed without more bending.

Either the long pipes have to be bent, or two of the Accelero's pipes will have to be bent, like this:



Here is how that went:



The heatpipe bender tool couldn't really be used here. Too cramped and the bend radius was too large to be useful. I used a round metal bar to help shape the curves, but really it was done by hand.



The last one was really tough to get in. I wanted it on the end, but there was no other way to do it without bending the pipe too tightly or having it stick out to the side or top where it will collide with other things. I did not want it in the center slots above the core... too risky to put only half of a heatpipe there.

Some of the bends look a bit sharp, but the diameter is above 4mm for the most part, so they are still functional.



Preparing to put it together, another problem surfaced.

In the CAD model, I only cared if things fit but not how they fit together, and I hadn't given any consideration to assembly.

First thing, the blocks are not able to create any clamping pressure without drilling out one side. I drilled out the top block.



Then, there wasn't an obvious way to attach the blocks to the card - The Accelero's fins block the screw holes on one side, and also the GTX 1080 blocks them from the other side. So I switched to machine screws and then countersunk the holes so it would not cause a clearance issue against the GTX 1080.



And, because the card is being mounted upside down in the case, unlike the HDplex H5, it is not possible to adjust heatpipe positioning after the assembly is mounted to the card because the screws are inaccessible.

I used the existing heatpipes as a guide for which to get the assembly positioned without needing to mount the card first.




The other block could be mounted now.



This fun problem was discovered next while checking alignment:





The PCIe riser was slightly blocked by one of the heatpipes.

Then this fun problem:



I had shims under the Accelero to support the weight and they were too high. The cooling assembly is about 40mm thick, but the card height is 43.5mm from the floor of the case and the GPU silicon package is about 2mm thick, so I made some 1.5mm shims to place under the Accelero to keep the entire card supported and not stress the PCB.

That was the last of the mounting problems. Moving on



I used Liquid Ultra this time:



And the card was assembled.. Here are a few last shots before it went in:








All rigged up:

Temps? Excellent heatpipe work.
 

Phuncz

Lord of the Boards
SFFn Staff
May 9, 2015
5,839
4,906
That's some special looking heatpipe noodle ramen you have there. I'm wondering if there is any benefit, because the GPU cooling looked almost done before adding the Arctic Accelero.