Tuniq Tower woe's

[lavaheadache]

I've this cooler for a year and a half now. I've never been able to hit 3.6 with stability. I always blamed this on my 680i chipset.

This past week I switched out my 680i for the UD3p(awesome board btw). I initially "bench" tested the sytem and was blown away by the ease in overclocking ability. 3.6 ghz was very easy to attain.

Next, I mounted everything in the case and stability was immediately gone just as if I was using the 680i. I found out that by putting the system on it's side yields a much lower temp q6600 @3.6 (1.32v) on it's side is about 60 versus 80 standing upright and I couldn't get even get it stable.

I truly am baffled as to why it happens though. I can run prime for hours with the rig on it's side and stay around 60 degrees. Within 30 seconds of lifting it upright the temps start to rise slowly. After 3-5 minutes temps are in the 80's with cores 0-1 being hotter by 5-7 degrees and soon there after I crash out due to temps.

My rig has very good cooling. Antec nine hundred with cables tucked relatively neatly. Thumbscrews are bottomed out and the cooler seems snug. I have a near perfect application of as5. All motherboard screws are in place and firmly fastened. There are no wiggles......


Any suggeestions?





***update*** picked up a coolermaster v8 at the local pc shop for 69 bucks. 58 load on all 4 cores, 3.6 400x9 1.312 vcore. Problem solved. seems as though the tuniq's limit with my setup is good for 3.2 @1.28 vcore and like 55c, lifting the pc doesn't do much. 3.6 works @ 1.312 but only if the pc is laying flat, lifting the pc up will cause the cooler to loose efficiency and cause instability after a few moments.

I ended up cutting an access hole through the motherboard tray on my antec 900 so i could remount or remove the cooler without pulling the motherboard off(not really my favorite thing to do). I also had to trim off the corner of the side mounted 120mm fan to allow proper clearance of the v8 as it sits a little taller than the tuniq tower. I already had to cut the corner off the fan before because one of the fins from tuniq would hit it, just had to trim even more.

 

[Tweakin]

It sounds like the 800grams of that beast is flexing the board, causing the contact to weaken...just my thought.

 

[lavaheadache]

i wonder if it is the backplate flexing

 

[nyker96]

i used the tuniq on my ip35e before selling it for a dht cooler. when I had it it seems to be ok, but you relly have to tighten the screws to make sure it has good contact, it's very heavy. with that said, I be honest my vendetta 2 that I used to replace it didn't perform as well. about 1-2c higher in fact and I even replaced the push pin with bolt thru set. so I still say keep your tuniq not a bad heatsink although it is very heavy.

one tip for the tuniq if you can tigh a few strings/ziptie to secure the tuniq to you case frame, it will probably help to lighten it.

 

[lavaheadache]

Originally posted by: nyker96
i used the tuniq on my ip35e before selling it for a dht cooler. when I had it it seems to be ok, but you relly have to tighten the screws to make sure it has good contact, it's very heavy. with that said, I be honest my vendetta 2 that I used to replace it didn't perform as well. about 1-2c higher in fact and I even replaced the push pin with bolt thru set. so I still say keep your tuniq not a bad heatsink although it is very heavy.

one tip for the tuniq if you can tigh a few strings/ziptie to secure the tuniq to you case frame, it will probably help to lighten it.[/q]


I was actually thinking about doing that. I've actually gone and measured the amount of flex(less than 2mm) and the held it up with my hand but I still get the drastic rise in temps. Are these heat pipes filled with anything that may not be "moving" right being upright? I've tried just about everything aside from rotating the orientation of the tuniq on the cpu. I've placed a 120 mm fan in various places to see if it was an airflow problem... just can't seem to figure it out

 

[chizow]

Various sites have commented on this phenomena in the past and I believe they attributed the difference to the way the the heat transfer materials in the heat pipes react. Most heat pipes claim to have some liquid that quickly turns into a gas in the heat pipes, which then rise to the fins and again become liquid as heat is dissipated. If you think about it and look at the difference between horizontal and upright:

U or W (flat case, heatsink vertical) vs. E (upright case, heatsink horizontal)

It'd make sense that the flat cases configuration would seem as if it faciliated the natural phase change and use of gravity to return the cooling liquid to the base area would be more efficient. I personally didn't notice any such difference on my Tuniq Tower as I never ran it on its side. I did notice a pretty huge temp difference after upgrading to a TRUE though. But ya P45 is a huge improvement over Nvidia 6-series for Quad overclocking.

 

[LOUISSSSS]

or you could have shorted something out when trying to install it in the motherboard.

try removing it and testing again.. if its stable outside of the case ONLY and unstable inside the case ONLY, then you might have a grounding issue

 

[LOUISSSSS]

the cooler is NOT too heavy for a motherboard to support. i've used it for over 2 years without any flex or instability problems due to the cooler

 

[lavaheadache]

Originally posted by: LOUISSSSS
or you could have shorted something out when trying to install it in the motherboard.

try removing it and testing again.. if its stable outside of the case ONLY and unstable inside the case ONLY, then you might have a grounding issue

I didn't short something out... Did you read the post? If i lay the system flat temps are @60 at load and if i pick it up in any direction the temp climb up drastically.

I beginning t to think that it is a heatpipe issue cause I've tried lifting the pc up in every direction possible and once I cross about a 75 degree angle remps start to rise

 

[lavaheadache]

Originally posted by: chizow
Various sites have commented on this phenomena in the past and I believe they attributed the difference to the way the the heat transfer materials in the heat pipes react. Most heat pipes claim to have some liquid that quickly turns into a gas in the heat pipes, which then rise to the fins and again become liquid as heat is dissipated. If you think about it and look at the difference between horizontal and upright:

U or W (flat case, heatsink vertical) vs. E (upright case, heatsink horizontal)

It'd make sense that the flat cases configuration would seem as if it faciliated the natural phase change and use of gravity to return the cooling liquid to the base area would be more efficient. I personally didn't notice any such difference on my Tuniq Tower as I never ran it on its side. I did notice a pretty huge temp difference after upgrading to a TRUE though. But ya P45 is a huge improvement over Nvidia 6-series for Quad overclocking.

do you have any linkage? I've tried searching google but can't come up with anything

 

[chizow]

Benchmark Reviews TRUE vs. HDT-S1283

During our testing for the Best CPU Cooler Performance - Q1 2008 article, nearly every cooler that used a "U" shaped heat pipe rod in their design did better in the vertical (standard upright tower) position. There seemed to be one dissident though, because the Thermalright Ultra-120 eXtreme acted exactly opposite and performed better in every horizontal test we conducted.

and following the link to the Q1 2008 article they refer to:

Best CPU Cooler Performance - Q1 2008

In every case where the CPU cooler used a "U" shaped heat pipe rod in their design, having the unit positioned with the rods running horizontally proved to offer a cooling benefit. This was especially true for the Xigmatek Red Scorpion S1283, which had the most significant drop in temperature out of the bunch. This goes to show you that heat-pipes are in fact prone to suffer the effects of gravity in their design. Once the vapor cools and becomes liquid, it seems to have an easier time completing the thermal circuit from side to side than it does from top to bottom.

They have some Everest temp graphs that show the difference convincingly as the temp differential is evident immediately. It does look as if their results are different than yours however, finding most heatsinks perform better with the motherboard in an upright position (heat pipes horizontal), except for the TRUE which performed better with the heat pipes vertical. Their temp delta also isn't as extreme as yours.

Now I'm also wondering if the orientation of the heatsink makes a difference, like top to bottom or left to right, resulting in the fans blowing front to back vs. down to up. Seems as if a top to bottom pipe orientation would favor the top pipes, which might also help explain some of the temp differences seen on the cores of a Quad. If bolt through mechanisms weren't such a pain in the ass to disassemble I might be tempted to test this myself.

 

[nestlewater]

OMFG,, you kids are the least crafty generation ever,, Fix a string/wire/duktape/string of duktape and ur set.

 

[LOUISSSSS]

Originally posted by: lavaheadache

Originally posted by: LOUISSSSS
or you could have shorted something out when trying to install it in the motherboard.

try removing it and testing again.. if its stable outside of the case ONLY and unstable inside the case ONLY, then you might have a grounding issue

I didn't short something out... Did you read the post? If i lay the system flat temps are @60 at load and if i pick it up in any direction the temp climb up drastically.

I beginning t to think that it is a heatpipe issue cause I've tried lifting the pc up in every direction possible and once I cross about a 75 degree angle remps start to rise

so if u lift it up vertically (the mb) the temps go up crazy?

are you mounting it correctly? maybe one of your screws are loose. theres only so many things that can be wrong with a CPU heatsink, theres no moving parts besides the fan.

 

[LOUISSSSS]

Originally posted by: nestlewater
OMFG,, you kids are the least crafty generation ever,, Fix a string/wire/duktape/string of duktape and ur set.

why did you create this account?

 

[lavaheadache]

Originally posted by: LOUISSSSS

Originally posted by: lavaheadache

Originally posted by: LOUISSSSS
or you could have shorted something out when trying to install it in the motherboard.

try removing it and testing again.. if its stable outside of the case ONLY and unstable inside the case ONLY, then you might have a grounding issue

I didn't short something out... Did you read the post? If i lay the system flat temps are @60 at load and if i pick it up in any direction the temp climb up drastically.

I beginning t to think that it is a heatpipe issue cause I've tried lifting the pc up in every direction possible and once I cross about a 75 degree angle remps start to rise

so if u lift it up vertically (the mb) the temps go up crazy?

are you mounting it correctly? maybe one of your screws are loose. theres only so many things that can be wrong with a CPU heatsink, theres no moving parts besides the fan.

Trust me I'm baffled, I've remounted this thing a half a dozen times over the last day or 2 and no matter what same thing happens. lift up the case and the temps go up. It's only at 3.6 ghz that i loose stability with case vertical. 4x800 3.2 ghz every thing is ok, but I'm pretty sure that the temp rise occurs too when I left the case up.

 

[LOUISSSSS]

WELL, 3.6ghz on a Q6600 is very high. what vCore do you have?
i wouldn't be surprised if what you're doing is normal, having the temps rise when u move MB vertically.

also, what is your room temp? if your room temps are near 70F, i think its normal that 3.6ghz is too hot for the tuniq tower.

 

[lavaheadache]

Originally posted by: LOUISSSSS
WELL, 3.6ghz on a Q6600 is very high. what vCore do you have?
i wouldn't be surprised if what you're doing is normal, having the temps rise when u move MB vertically.

also, what is your room temp? if your room temps are near 70F, i think its normal that 3.6ghz is too hot for the tuniq tower.

vcore is 1.312 in windows and 1.337 in the bios. I very well may be expecting too much from the tuniq. It seems like the best option for me to run a good overlock is on water.

Chizow.... how good is that TRUE?

 

[LOUISSSSS]

the TRUE is not worth an upgrade over the Tuniq. but i guess since you're is dead, the up is the only way to go.

 

[Rubycon]

Originally posted by: LOUISSSSS
the TRUE is not worth an upgrade over the Tuniq. but i guess since you're is dead, the up is the only way to go.

WRONG!

The problem with the OP using a quad at 3.6 is you're reaching the limits of the three heat pipes of the Tuniq. When turned sideways this augments the problem as the thermal transfer capacity is hindered. Once the capacity is reached and ultimately succeeded temperatures will sharply rise. As a former Tuniq owner and discussing this issue at length with engineers this was their conclusion. The test rig had a strain relief device machined that braced the TT taking ALL strain off the mount AND the temperature still rose when the motherboard was lifted off the horizontal plane.

The TRUE with its SIX heatpipes has much more capacity to accommodate a heavy load such as an overvolted, overclocked quad. Even a B3. Temperature rise is not noticed with heatpipes UNLESS you're near capacity and reduction in said capacity due to gravity causes thermal run away.

Direct contact is more of a marketing gimmick as it CANNOT compensate for absolute (thermal) capacity. It may reduce IDLE temps (useless as the BURST rating of a HDD!) AND it may also reduce the load temperatures - albeit slightly - of dual cores with smaller cache. Again smaller thermal loads.

 

[JEDIYoda]

Originally posted by: Tweakin
It sounds like the 800grams of that beast is flexing the board, causing the contact to weaken...just my thought.

That`s not the problem at all!
Sounds to me like you are guessing!

 

[JEDIYoda]

Originally posted by: Rubycon

Originally posted by: LOUISSSSS
the TRUE is not worth an upgrade over the Tuniq. but i guess since you're is dead, the up is the only way to go.

WRONG!

The problem with the OP using a quad at 3.6 is you're reaching the limits of the three heat pipes of the Tuniq. When turned sideways this augments the problem as the thermal transfer capacity is hindered. Once the capacity is reached and ultimately succeeded temperatures will sharply rise. As a former Tuniq owner and discussing this issue at length with engineers this was their conclusion. The test rig had a strain relief device machined that braced the TT taking ALL strain off the mount AND the temperature still rose when the motherboard was lifted off the horizontal plane.

The TRUE with its SIX heatpipes has much more capacity to accommodate a heavy load such as an overvolted, overclocked quad. Even a B3. Temperature rise is not noticed with heatpipes UNLESS you're near capacity and reduction in said capacity due to gravity causes thermal run away.

Direct contact is more of a marketing gimmick as it CANNOT compensate for absolute (thermal) capacity. It may reduce IDLE temps (useless as the BURST rating of a HDD!) AND it may also reduce the load temperatures - albeit slightly - of dual cores with smaller cache. Again smaller thermal loads.

I am impressed!! What she said......heheheee

 

[LOUISSSSS]

so, what (on average) is the temp benefits of a TRUE over a Tuniq Tower?
lets say its 5c on average, is that worth the $60-70 for it? + needing it to be lapped.

 

[Rubycon]

Originally posted by: LOUISSSSS
so, what (on average) is the temp benefits of a TRUE over a Tuniq Tower?
lets say its 5c on average, is that worth the $60-70 for it? + needing it to be lapped.

Temperature differences are not the determining factor - it's a matter of capacity. Sites that test these high end coolers on dual core chips should be dismissed completely. Using a TRUE on a lower end dual core is akin to buying a GTX295 when you own a 19" monitor that does 1440x900.

 

[zagood]

Originally posted by: Rubycon
Using a TRUE on a lower end dual core is akin to buying a GTX295 when you own a 19" monitor that does 1440x900.

Not quite...I was able to get a e6400 up to 72c on stock heatsink maxing out at 3.8ghz, and 4.2ghz @ 60c on a TRUE. Stable.

I would say it's more like buying a GTX295 when you own a 15" CRT that you can force to do 1920x1440 - basically trying to get more out of your budget hardware then it was specced for. Is it worth it? Sometimes, but seems silly to those who can afford a LCD that does 19x12 natively.

EDIT: My mind was 3 years ago, the last time I looked at 24" monitors and they were $500+ - my comment doesn't make as much sense considering the LCD prices today.

 

[Tweakin]

Originally posted by: JEDIYoda

Originally posted by: Tweakin
It sounds like the 800grams of that beast is flexing the board, causing the contact to weaken...just my thought.

That`s not the problem at all!
Sounds to me like you are guessing!

Sorry op...statements like that are just useless...I edited out my reply comments.