I recently won a Thermalright XP-90 from Tek for a sig I made him. I was previously using a Thermaltake Venus 12. This is on an A64 3000+ at stock clock and voltage.

The problem is, with the Venus I was getting aroung 34-36 deg. cel. at idle at lowest fan speed and using Nanotherm TIM, and with the XP-90 it will not drop below 55 deg. even at high fan speed using 92mm Enermax UC-9FAB-B and 56 deg at lowest speed with AS5.

I tried removing and reseating the sink twice in both orientations (toward front and toward back) and get the same results. And yes I know how to apply TIM correctly.

I reinstalled the Venus and voila, the temps are now at the same 34-36 deg range at idle with AS5.

Do you think that the heat pipes on the XP-90 may have gassed out? I dont see any other reason why they would be so high with this sink. A stock cooler does better than this.

Any clues would be helpful.

Thanks.

I'm not a great fan of heatpipes on heatsinks. The XP-120 I got underperforms the Zalman by a huge margin. The orientation of the heatsink also affects the cooling. You want the cooling side of the pipe at the highest level possible above the hot side of the pipe. Otherwise you won't get enough cool fluid back down to the bottom. I tested this on a review using a graphics board heatpipe and the orientation did change the temperature.

It sounds like you are not making complete contact with your AMD 64 bit chip Luna. Make sure when you put on artic silver you only use a little bit, spread it out with your finger, and then really smooth it out with a credit card or such.

Also, make sure you AMD bracket in your board is locked in and secured to your Thermalright heatsink well. Sometimes it can tend to pull away before you lock it down, of even after.

This one is to BA...I have been testing the Zalman vs the XP-120 and I want to know how the Zalman got better numbers? The Zalman holds it owns under full speed fan specs, but with a max overclock and low speed under load it hits over 60C for me and has crashed my system many times.

With the XP-120, I can run the Panaflo fan(120mmX38MM) at low and still idle at 38C and load around 54C. My Zalman 7000B never did that. It would Idle at 39C and load around 56C to 62 C depending on the game I was playing. ALso to note this the fact that the Zalman at high speeds it louder than the Panaflo fan at lower speeds yet the XP-120 put out the same numbers at that speed. The Zalman 7000B was my favorite heatsink for over a year, but now I think the XP-120 has replaced it for me.

AND LUNA< I am glad you are enjoying those parts I sent you (with the trade for the etch mod which you still have), can I enjoy my etch and side panel now? Still no PM, giving you plenty of time. What is troubling you so much man, I can't understand?

Quote: Originally Posted by SpikeShot It sounds like you are not making complete contact with your AMD 64 bit chip Luna. Make sure when you put on artic silver you only use a little bit, spread it out with your finger, and then really smooth it out with a credit card or such.

i hope you're putting a plastic bag on your finger when you're spreading it out. if you're using your bare finger then you're adding in skin particles to the tim which will jack up the heat transfer ability of it.

Luna: Did you replace the stock retention bracket with the one included?
That is very important. Replace it and use the stock backplate if you have not already.
Also, try reverse bending the clips about 3-5MM to add a little extra downward tension.

For your reference, here is the XP-90 Installation Guide.

If we can't get this issue worked out I want you to send it back in for a refund, and I'll pay
the 15% restock and return S&H plus difference so you can get something else, OK?

I have done everything you guys recommend, except bending the brackets. And yes I put the retention bracket on. What do you guys think Ive never done this **** before, geez.

ENKO: I dont use my finger to spread TIM, I use a card to spread it thinly and evenly.


I have tried EVERYTHING. Its getting good contact with CPU, just the right amount of AS5 etc. etc. etc. I have changed it out twice with the the Venus and I get the same results. I have reseated the XP-90 6 times now double checking everything each time and its performance still reaks of ass.

Quote: Originally Posted by Lunamods I have done everything you guys recommend, except bending the brackets. And yes I put the retention bracket on. What do you guys think Ive never done this **** before, geez. ENKO: I dont use my finger to spread TIM, I use a card to spread it thinly and evenly. I have tried EVERYTHING. Its getting good contact with CPU, just the right amount of AS5 etc. etc. etc. I have changed it out twice with the the Venus and I get the same results. I have reseated the XP-90 6 times now double checking everything each time and its performance still reaks of ass.

Whoa, chill, I'm not doubting your intelligence, I go through steps to determine the real issue, you know that.

Check the temperatures with a probe...my NF7-S had readings that doubled when I swapped a CPU & HSF.

Onboard sensors are not to be trusted...get a probe in there.

Quote: Originally Posted by TekXoID Whoa, chill, I'm not doubting your intelligence, I go through steps to determine the real issue, you know that. Check the temperatures with a probe...my NF7-S had readings that doubled when I swapped a CPU & HSF. Onboard sensors are not to be trusted...get a probe in there.

Well in this case 34 deg with the Venus and 55 deg. with the XP, even using the onboard sensor still says there is a problem.

I didnt mean to be an ass. Some of the stuff said would be pretty obvious even to a noob.

Quote: Originally Posted by Lunamods Well in this case 34 deg with the Venus and 55 deg. with the XP, even using the onboard sensor still says there is a problem.

If a probe has the same story you are sending it back for something else, courtesy of yours truly.

Quote: Originally Posted by TekXoID If a probe has the same story you are sending it back for something else, courtesy of yours truly.

Ill check it with a CompuNurse.

Quote: Originally Posted by Lunamods I have done everything you guys recommend, except bending the brackets. And yes I put the retention bracket on. What do you guys think Ive never done this **** before, geez. ENKO: I dont use my finger to spread TIM, I use a card to spread it thinly and evenly. I have tried EVERYTHING. Its getting good contact with CPU, just the right amount of AS5 etc. etc. etc. I have changed it out twice with the the Venus and I get the same results. I have reseated the XP-90 6 times now double checking everything each time and its performance still reaks of ass.

luna, i know you've done this before. my comment wasn't to you, that's why i was quoting spikeshot who said he uses his finger to spread it. sorry if you thought i was directing it at you.

Quote: Originally Posted by SpikeShot This one is to BA...I have been testing the Zalman vs the XP-120 and I want to know how the Zalman got better numbers? The Zalman holds it owns under full speed fan specs, but with a max overclock and low speed under load it hits over 60C for me and has crashed my system many times. With the XP-120, I can run the Panaflo fan(120mmX38MM) at low and still idle at 38C and load around 54C. My Zalman 7000B never did that. It would Idle at 39C and load around 56C to 62 C depending on the game I was playing. ALso to note this the fact that the Zalman at high speeds it louder than the Panaflo fan at lower speeds yet the XP-120 put out the same numbers at that speed. The Zalman 7000B was my favorite heatsink for over a year, but now I think the XP-120 has replaced it for me.

How the Zalman got better numbers? By mounting it and recording the numbers, that's how. This is with a huge Delta 120mm that is much louder than the Zalman. You may have not had the Zalman mounted correctly, it can be tricky. I've owned or tested quite a few sinks in my day, and believe I can say with confidence that I know what I'm doing.

Quote: Originally Posted by BigAkita How the Zalman got better numbers? By mounting it and recording the numbers, that's how. This is with a huge Delta 120mm that is much louder than the Zalman. You may have not had the Zalman mounted correctly, it can be tricky. I've owned or tested quite a few sinks in my day, and believe I can say with confidence that I know what I'm doing.

I don't know BA, I don't feel like making a big deal about this. Maybe I mounted it wrong, maybe the Max3 OTES made for higher numbers because of the location right next to the heatsink, or maybe it had 11 months of dust in it. (I did clean it as good as I could)

I can say that I replaced it 2 times, let the AS5 settle in before testing again(loops of 3Dmark2001, cooling down, and then repeated again, last a day or so). Numbers were close on high speeds and higher on lower speeds when compared to the XP-120. I guess the only way to really know is to have exact same heatsinks, and same room temps.

I would say that both heatsink work great, and anyone would be find going with either one.

We're in agreement there, both would work fine in any rig they fit in.

Luna, If you are still getting temps that bad, I would suspect that one of the tubes may have lost vacuum pressure.

Quote: Originally Posted by Lunamods Well in this case 34 deg with the Venus and 55 deg. with the XP, even using the onboard sensor still says there is a problem. I didnt mean to be an ass. Some of the stuff said would be pretty obvious even to a noob.

Im not sure about the XP-90 but isnt it a Heatpipe?? Maybe something is wrong with the pipe? like the gas in it or something. Maybe its not dissipating the heat right? Something.. you never know. Was the heatsink doing this since you got it?


Matt

Quote: Originally Posted by MaTtsPc Im not sure about the XP-90 but isnt it a Heatpipe?? Maybe something is wrong with the pipe? like the gas in it or something. Maybe its not dissipating the heat right? Something.. you never know. Was the heatsink doing this since you got it? Matt

Quote: Originally Posted by BigAkita We're in agreement there, both would work fine in any rig they fit in. Luna, If you are still getting temps that bad, I would suspect that one of the tubes may have lost vacuum pressure.

Sorry didnt see that.. i didnt really read up on all the posts.. sorry pimps




Matt

Well I tested temps with the CompuNurse and get the same results,well a few degrees lower, but it still says that the XP-90 is obviously not working properly. My only guess is that one or more of the heat pipes are gassed out. Its going back to Jab. Thanks guys.