How to Mount a Direct Contact Heatpipe cooler [Titan Fenrir]

rjk

rjk

Caporegime
Joined
8 Aug 2007
Posts
25,381
there have been many issues from customers experiencing high temperatures with their bundles

the problems are occuring due to people using a small amount of thermal paste.

when installing a normal cooler such as a Thermalright Ultra Extreme/Prolimatech Megahalems etc
this is fine:
IMG_3641.jpg

the flat contact plate spreads the thermal paste evenly

with the Titan Fenrir/Akasa Nero, you need to get rid of the air pockets.
the only way to do this is to use more thermal paste than you would with a flat contact plate

an S shape on the IHS may seem like overkill but it does work better with direct contact coolers
IMG_3643.jpg


you also need to bear in mind that the arctic silver TIM needs time to cure before it starts working well

also make sure the mounting hardware is as tight as possible to ensure good heat transfer
 
Man of Honour
Joined
18 Oct 2002
Posts
12,298
Location
Vvardenfell
I've actually found that TRUEs need a fair amount of paste as well. It's not the fabled "non-flat" baseplate either - it just seems to work better that way. I had one on and off three times with my new D0, and got the early ridiculous temps (85-90oc full load) down to a far more sensible 60-65oC at full load by nearly tripling the amount of paste (AS5). Those temps are in a room currently at 30oC (d*mn heatwaves).


M
 
Soldato
Joined
22 Mar 2009
Posts
5,712
**you also need to bear in mind that the arctic silver TIM needs time to cure before it starts working well**

How long do I have to wait for the TIM to cure and working well ? Do I need to start off full load or use normal idle for 24 hours before try out full load ?

Interesting of the S shape before installing Titan Fenrir on it. I just wish Titan Fenrir had a proper 4 x screw nuts to tight it proper as it very hard to tight it using a thumb and a finger that nearly hurt me !!! :(

Can you show a picture if the Titan Fenrir can face upward to the big fan of the top of the tower case on Antec 1200 case, so, I don't have to put a 120mm fan behind the tight memory ram.
 
Soldato
Joined
16 Jun 2009
Posts
7,664
Location
Cambridge
Interesting, thanks Rjkoneill.

@bulldog147 AS5 can take up to 200hours to cure according to the manufacturer. (That's not to say it doesn't work well in the first 200hours, it just works better after that time).
 
Last edited:
Soldato
Joined
9 Jul 2004
Posts
4,522
Location
Nottingham
My first impressions of the second image were :eek:

So you place the Titan/Sunbeam etc directly over the paste in exactly that state? Do you wobble it around a little when tightening up?

Can you show a picture if the Titan Fenrir can face upward to the big fan of the top of the tower case on Antec 1200 case, so, I don't have to put a 120mm fan behind the tight memory ram.

I've got my Titan facing upwards in my Antec 300, so the heat rises and is removed by the larger 140mm fan, as opposed to having it to the side, facing the rear 120mm fan. I must say I've never understood why most people seem to face the back of the heatsink towards the back of the case when they have a top fan. A) the top fan is usually larger, providing more CFM. B) Hot air rises.
 
Soldato
Joined
22 Mar 2009
Posts
5,712
I got my new delta fan 4000RPM arrive and try it out (my godness me, it very very loud like an aircraft - but temperature on the cpu drop amazing by 7C degrees at full load. Overclocked 4.2Ghz with intel turbo boost enabled and Extreme performance enabled and full load is now reduce from 67C to 60C !!!!!!

CPU temp using gigabyte easytune6: 39C
Real Temp cpu reading at full load: 60C at 4.20Ghz overclocked (turbo boost and extreme performance enabled) with HT enabled too.
North Bridge temp: 40C
GPU temp: 40C
HDD temp: 44C
Motherboard temp: 33C
Room Ambient temp: 25C
 
Last edited:
Soldato
Joined
1 Jan 2008
Posts
11,024
Blimey, that's a lot more paste than i'm used to seeing. I'm not using a direct contact heatpipe cooler, but i'm using a lot less than even the first pic. I applied my AS5 as per Arctic Silver's instructions on their website, would I possibly benefit from using a bit more then? My temps are acceptable but not as good as I feel they ought to be, but I blamed that on a chip that's like a mountain range.
 
Soldato
Joined
21 Aug 2006
Posts
7,507
This direct touch thing mean that the first set of fins sit directly onto the CPU rather than the kind of stilts i have on my Tuniq Tower?
 
Soldato
Joined
18 Oct 2002
Posts
7,813
Location
Scun'orp
After looking at pics of the base, the first thing I would do if I had one of these coolers would be to apply a film of tim to the cooler base itself, then draw a credit card across it just to see what you are dealing with level wise. If the cooler is any good there should be no tim left on the flattened copper parts and preferably none on the flat aluminium parts. Ideally there should only be tim in the obvious voids where the curved part of the tubes meets the aluminium in the slots. If this test shows any one of the copper tubes either standing proud or recessed to the others I would be wanting a replacement cooler, rather than throwing tim at it. Or at least you could gauge how much you think you could get away with lapping it. It is clearly already machined flat so there must tbe some tolerance to work with when they made the thing, the tube walls are hardly likely to be paper thin are they.
 
Soldato
Joined
3 Apr 2008
Posts
2,939
After looking at pics of the base, the first thing I would do if I had one of these coolers would be to apply a film of tim to the cooler base itself, then draw a credit card across it just to see what you are dealing with level wise. If the cooler is any good there should be no tim left on the flattened copper parts and preferably none on the flat aluminium parts. Ideally there should only be tim in the obvious voids where the curved part of the tubes meets the aluminium in the slots. If this test shows any one of the copper tubes either standing proud or recessed to the others I would be wanting a replacement cooler, rather than throwing tim at it. Or at least you could gauge how much you think you could get away with lapping it. It is clearly already machined flat so there must tbe some tolerance to work with when they made the thing, the tube walls are hardly likely to be paper thin are they.

If you have a read of Frosty Tech they have an accurate surface level guage they use when reviewing heatsinks. The Noctua's seem to have a problem with beingly slightly concave apparently.
 
Associate
Joined
31 May 2008
Posts
1,049
Location
York
With my Sunbeam HDT i put the paste directly onto the cooler, inbetween the heat pipes and everything seems ok. A good thick line as most people are saying that more than usual is needed.
 
Soldato
Joined
31 May 2006
Posts
7,564
Location
West London
Not paper thin no - somewhere around 0.7mm stock and they have already been shaped and machined so are thinner at the base. (in theory this should mean its relatively flat already)

imho I'd not lap it and risk spliting a pipe - but then I wouldn't buy one either.
 
Last edited:
Soldato
Joined
31 May 2006
Posts
7,564
Location
West London
And here is the reason why. I lightly lapped my OCZ Vendetta 2 just to flatten a raised middle heatpipe which was about .5mm higher than the outer ones. So I went with 400, 600 and finally 800 grit sandpaper and I wanted to polish it. as I applied the polish and lightly pressed on the base, it caved in and cracked. here is the proof:

lappingmistake9694687hb9.jpg


Can't link to source (rules and shops)

Speaks for it's self really...
 
Last edited:
Associate
Joined
28 Sep 2008
Posts
1,054
Location
nottinghamshire
nope still hitting 90c+ even doing it the way rich has in photo so theres gotta be something a miss somewhere,idle temps are 43,41,45,41 room temp is 22c pretty cool in here really

is it worth going watercooled if so is it hard to do?
 
Soldato
Joined
18 Oct 2002
Posts
7,813
Location
Scun'orp
I lightly lapped my OCZ Vendetta 2 just to flatten a raised middle heatpipe which was about .5mm higher than the outer ones.

0.5mm higher than the other ones?? Instant return if that had been me, that is ridiculously out of tolerance for something which relies so heavily on heat pipe contact. Yes it has become clear trying to lap out defects such as this is a non starter. No purchase for me, even if I wasn't water cooling.
 
Back
Top Bottom