Cooling Thermalright AXP-100 Pure Copper

[tinyitx]

arrived today. definitely full copper with graphene coating.

I wonder how much the graphene coating contributes to the thermal performance. Unfortunately, you cannot compare with a non-graphene full copper one.
I think, one of the benefit is that copper colouring of the entire heatsink. It is visually attractive. Not that the black graphene coating looks bad, it does look good too but not as good as that reddish brown copper colour. So, it is a trade-off between having (slightly?) better thermal with the graphene coating or the (slightly) better look of the copper colour.

 

[gunpalcyril]

Its too bad you can't mount a fan underneath, would have been killer to have that be under 50mm height.

 

[Damascus]

The ty-100 is fantastic, I'm curious to see the results here

 

[Damascus]

Its too bad you can't mount a fan underneath, would have been killer to have that be under 50mm height.

You actually can mount a bottom fan. 12.5mm require no nodding, 15mm are a bit more involved

 

[jpmomo]

should be possible:

 

[gunpalcyril]

Wait really? This is assuming you use zipties right? The only possible orientation of the fan underneath the heatsink would have to be pushing air through the heatsink away from the motherboard, right? I'm really curious to see if this is possible, so I can use it for my am4 system.

 

[jpmomo]

If you look at 1 of my earlier posts, that is the default orientation of the noctua fan (bottom mount blowing up). I haven't actually tried this myself with the thermalright. that video did not give any details on how the fan was attached to the heatsink. there is also a plate that needs to fit between the bottom cold plate and the actual cpu to lock it down. In the picture above, this plate would have to fit between the fan and the cold plate. the plate is not that thick and could probably fit. You would have to screw the plate to the mounting brackets on the socket while the fan is removed. You could then slide the fan into the space between the plate and the rest of the cooler. That should make for a low profile cpu cooler that might do a good job of cooling a high tdp cpu. The noctua does a very good job as is and I am curious to see how the thermalright compares.

 

[Damascus]

Wait really? This is assuming you use zipties right? The only possible orientation of the fan underneath the heatsink would have to be pushing air through the heatsink away from the motherboard, right? I'm really curious to see if this is possible, so I can use it for my am4 system.

I'll have a post on my nodding of an axp-100 muscle eventually. I used velcro and a super modded fan to make it work

 

[Kmpkt]

I'm thinking we need @Thehack to design a bending tool and underside mounting bracket for this heatsink so we can put a 14mm Noctua underneath. Just saying ; )

 

[dcbn]

I know we don't have any tests yet, but, are the Thermalright AXP 100 supposed (copper or graphene) to be better than NH-L12s?

Thank you for your help and your attention.
I'm sorry for any inconvenience.

 

[Thehack]

I'm thinking we need @Thehack to design a bending tool and underside mounting bracket for this heatsink so we can put a 14mm Noctua underneath. Just saying ; )

Seems possible. Can anyone confirm that the AXP-100 Copper eddition is dimensionally the same as regular version? $80 is a bit too rich for my blood.

 

[Damascus]

I know we don't have any tests yet, but, are the Thermalright AXP 100 supposed (copper or graphene) to be better than NH-L12s?

Thank you for your help and your attention.
I'm sorry for any inconvenience.

Probably better, given the thermal properties of copper

 

[dcbn]

Probably better, given the thermal properties of copper

As someone who is not specialized on heatsinks (me!), would be possible to have an estimate? Would be something like 5 ~ 10ºC?

Thank you for your help and your attention.
I'm sorry for any inconvenience.

 

[Kmpkt]

I have one of these on order. As soon as the broker service I've chosen has it and ships I will be able to compare it to the other top dogs (LP53, C7 Cu, Etc) and let you guys know where it stands.

 

[jpmomo]

some initial results. I normally use server mb that have ipmi info but I couldn't get this cooler to fit on any of my server mb. That required me to use the asrock x99, install windows 10, install hwinfo64 and finally prime95. all of this was tested on an open bench platform and I tried to reduce as many variables as possible. I would still consider this testing just an approximation until someone with a better testing methodology can take a closer look. The first test was with the axp and its stock fan. after a few iterations of prime95 using the Small FFTs with a duration of about 10mins (the temps seemed to stabilize at around this time), it had a max temp of 71.5C. I then swapped just the fan with the new noctua nf-a12x25. This fan was noticeably quieter (no accurate measurements just by ear) but also ran a bit higher temps. This combo peaked at 75C for the same set of tests. Both of these setups were with the fan blowing down. I will post some pics later but the increased height of the noctua fan and its increase in cpu temps probably does not make up for the reduced noise for most folks. The next step is to swap the cooler with the noctua L12S and rerun the same tests. I will update this thread as soon as I am done with that test.
jp

 

[Damascus]

As someone who is not specialized on heatsinks (me!), would be possible to have an estimate? Would be something like 5 ~ 10ºC?

Thank you for your help and your attention.
I'm sorry for any inconvenience.

Hard to say, it's 100mm vs 120mm so the efficiency of copper could be nullified by the greater surface area. @IntoxicatedPuma should be able to say if the axp-100 cu winds up on the test bench

 

[jpmomo]

The noctua L12S beats both combinations of the axp with a max temp of 68C. One thing to keep in mind is that the fan blows up from the bottom. This seems to be very effective at keeping the cpu cool but the air coming out of the heat sink on top is very hot. The case should have plenty of ventilation just on top of the cooler. The height from the top side of the mb to the top of the fan or cooler is as follows:
68.20 for axp with stock fan
78.57 for noctua L12S
81.74 for axp with noctua nf-a12x25 fan

Also for those that were wondering if the axp with graphene was based on the pure copper cooler, the answer is yes! When I was cleaning off the thermal paste, I accidentally got some of the alcohol spray on the heat pipes and rubbed some of the graphene coating off. I will post some pics of the those heat pipes where you can see the copper coming through.

One interesting mod that might be worth giving a try is to take the axp and mount the stock fan underneath. This might be the best combination and also might give the most clearance as well.

some pics of today's testing:

 

[dcbn]

Hard to say, it's 100mm vs 120mm so the efficiency of copper could be nullified by the greater surface area. @IntoxicatedPuma should be able to say if the axp-100 cu winds up on the test bench

Thank you for the clarification and your attention!

I'm sorry for any inconvenience.

Have a wonderful rest of weekend.

The noctua L12S beats both combinations of the axp with a max temp of 68C. One thing to keep in mind is that the fan blows up from the bottom. This seems to be very effective at keeping the cpu cool but the air coming out of the heat sink on top is very hot. The case should have plenty of ventilation just on top of the cooler. The height from the top side of the mb to the top of the fan or cooler is as follows:
68.20 for axp with stock fan
78.57 for noctua L12S
81.74 for axp with noctua nf-a12x25 fan

Also for those that were wondering if the axp with graphene was based on the pure copper cooler, the answer is yes! When I was cleaning off the thermal paste, I accidentally got some of the alcohol spray on the heat pipes and rubbed some of the graphene coating off. I will post some pics of the those heat pipes where you can see the copper coming through.

One interesting mod that might be worth giving a try is to take the axp and mount the stock fan underneath. This might be the best combination and also might give the most clearance as well.

some pics of today's testing:

That's good to hear, thank you for all the explanation and the pictures!

Have a nice rest of weekend.

 

[jpmomo]

I finally was able to mount the stock fan underneath the axp. Bending the pipes into shape wasn't that difficult. trying to mount the entire cooler with the fan on the bottom was difficult. I also needed to modify the mounting brackets slightly. Two of the large thumb screws holding the rectangular mounting bracket needed to be replaced with smaller nuts so that the fan would clear. The net result was that it reduced the height by about 8mm from 68 to 60. It could probably be reduced further with some more bending/optimization. keep in mind that these measurements are probably not the same way that the manufacture takes them. But they can be used for a relative comparison. The other issue is that the fan when mounted on the bottom is very noisy. The thermals are also much worse. The prime95 test quickly passed 80c then I stopped the test. When I put the fan in "silent" mode, it was pretty quiet and idled about 50C (vs 34c with the fan on the top and in standard mode). Here is a picture with the fan mounted on the bottom and running:

 

[dumplinknet]

I finally was able to mount the stock fan underneath the axp. Bending the pipes into shape wasn't that difficult. trying to mount the entire cooler with the fan on the bottom was difficult. I also needed to modify the mounting brackets slightly. Two of the large thumb screws holding the rectangular mounting bracket needed to be replaced with smaller nuts so that the fan would clear. The net result was that it reduced the height by about 8mm from 68 to 60. It could probably be reduced further with some more bending/optimization. keep in mind that these measurements are probably not the same way that the manufacture takes them. But they can be used for a relative comparison. The other issue is that the fan when mounted on the bottom is very noisy. The thermals are also much worse. The prime95 test quickly passed 80c then I stopped the test. When I put the fan in "silent" mode, it was pretty quiet and idled about 50C (vs 34c with the fan on the top and in standard mode). Here is a picture with the fan mounted on the bottom and running:

Are you sure the fan was mounted in the correct direction? Blowing air upwards towards the heatsink.
As opposed to pulling air from the hestsknk and blowing it to the cpu?