Heatkiller 3.0: Annoying rattle

[003]

I just installed my Heatkiller 3.0 and I am extremely pleased with the results coming from the Fuzion.

However, it seems to have a very annoying attribute. When I took it out of the box, I noticed I could hear something rattle a little bit if I shook the block. I just assumed that was how it was designed.

Now with it installed and in my loop, it seems to make this audible rattling sound, somewhat similar to heavy hard disk activity.

It is extremely annoying. Is this normal?

[century child]

No it's not. Something must have gotten into the block that doesn't belong there.

[defect9]

sounds like it has a free toy inside

[Bladesinger7x]

This might from the shafts holding the mounting plate or the injector plate.
I have the LC version and it also rattles a bit, but once mounted, it should
not rattle any more (mine is still in the box).

If it rattles when in use it might be something else...perhaps something in the
block or in the loop?

[dreamaxx]

sounds like it has a free toy inside

Rofl.

I wonder if maybe one of the little metal sticks which hold stuff in place has come loose.

[003]

Hmm. Maybe I was being over sensitive. The noise now seems not quite as annoying... but I'm nervous now... should I pull apart the loop and open the block? I hope I don't have to do that Bladesinger you say your LC rattles a bit as well, is it normal then? I have the LT.

[century child]

If the noise is lessening, it could be air in the system. If this is the case, the noise will eventually go away completely.

[Garrett]

I just installed my Heatkiller 3.0 and I am extremely pleased with the results coming from the Fuzion.

However, it seems to have a very annoying attribute. When I took it out of the box, I noticed I could hear something rattle a little bit if I shook the block. I just assumed that was how it was designed.

Now with it installed and in my loop, it seems to make this audible rattling sound, somewhat similar to heavy hard disk activity.

It is extremely annoying. Is this normal?

Explain pleased with the results coming from the FuZion... I mean what's the temp difference compared between the two?

Oh and I'm afraid the middle plate (stainless steel) got loose somehow inside the block... I guess you can only solve it by opening up the block, and unfortunately in your case that means draining the loop
The reason that it might not be the air is because you also heard it when you took it out of the box, and then there was no water in it yet

[003]

If the steel plate is loose, what would I do to make it snug again??

[dingdong555]

If the steel plate is loose, what would I do to make it snug again??

Only way to find out is to drain your loop and open up the CPU block.

[mindspat]

I am going to guess it's the restriction plate.

[jkresh]

003, I would also like to know what temp differences you are getting going from the fuzion to the heatkiller (and do you have the backplate or not). I am getting ok temps with the fuzion but my 920 gets pretty hot (around 73C for hottest core using coretemp with 3.9ghz at 1.4v, running 8 threads with prime95 for a few hours). And if the heatkiller would do a lot better (ie maybe keep the temps around that at 1.45 volts) I would consider the upgrade. Hottest core is about 62C at 1.33 volts and 3.6ghz with same load (and around 56 at 3.6 with 1.33 running folding at home multicore on 4, and 4 individual folding at home threads).

[003]

At 1.4125v with LLC (actually increases vcore a bit under load) and 191*21 MHz and HT enabled, the hottest core I have seen is 69C (using RealTemp 3.2 RC4) with 64-bit LinX 0.5.8 set to use 4096MB of memory, and LinX gets it quite a bit hotter than Prime95.

@mindspat,
If that is the case, then there really wouldn't be much I can do about it, would there?

EDIT:
With the Fuzion V2, I saw temps of up to 84C on LinX. Granted, on the Fuzion, the CPU was not lapped and I was using MX-2 TIM. This time I have lapped the CPU and am using Coollaboratory Liquid Pro. Also with the Fuzion, I had an MCW60-R2 in the loop (but it was not attached to anything), and I took it out with the HK 3.0.

[jkresh]

interesting, considering the fuzions flow I doubt the mcw60 made much of a difference temp wise (not sure about the mx-2 (which I am using) vs the coollaborary liquid pro (though I doubt its more then 1-2C)) so it seems that with your settings you got a 10C difference from the fuzion 2 to the heatkiller (since I am on a fuzion 1 I should do at least that much), seems like a worthwhile upgrade (though i wonder if any new blocks specifically designed for i7 are due out soon).

[hewi]

waking up a dead thread
So did ppl ever figure out what was causing the rattle, since my HK 3:0 LT sure as hell rattles. I opened the block and found these very tiny steel studs embedded into little holes on the edges of the block 4 in total. Im pretty sure they are the main reason for the annoying rattle....question is what are they for, and can I leave them out ? They seem to be there for a purpose, or just to piss ppl off, lol.

[scamps]

They should fix the mounting plate to the corpus of the cooler. And there is no rattling at my HK at all. Where should it come from, if your case is not rattling itselves?

[hewi]

The block rattles by itself if I shake it, nothing to do with case fan vibration etc. You can see the small studs im talking about on the edges of the block in the pics in this thread. And it's not the injector plate since the LT doesnt have one. The small studs arent part of the mounting hardware to my knoweledge...so what are they for ? they prolly should stay put in their little holes, but mine move back & fort against the delrin top making a rattle.

-Edit-

I must be going blind, since you just answered what the studs are for...lol
I'll have to see if I can glue them to be still or something..very annoying.

[Oberon]

Those studs are there to keep everything aligned properly when you put the block together. I wouldn't recommend taking them out.

[SAE]

The mounting plate will come off if you don't put them back in. They're there to hold the exchangable mounting plate in place and to make the block usable for several cpus/sockets.

[p2501]

I'm going out on a limb here but I don't think it's the injector plate by itself, there must be something wrong with the assembly. It's lying directly on the fins of the baseplate and is held in place by a rubber gasket, how in the world should that be able to rattle around? Either there is something in the block or they forgot the gasket altogether which would enable the plate to rattle around. Either way you're going to have to disassemble it to check. With a loose midplate you propably don't have it performing like it should too.

[WeeMaan]

I got my Heatkiller copper 1366 a week ago and when i shaked it it did rattle to!
Like it was something in it, but I did'nt bother to investigate.

[SAE]

Mates... don't be so uncreative. The rattling pieces are the little bolts to hold injector plate and mounting plate in place. It's been said and it does not change from wondering...

[WeeMaan]

Mates... don't be so uncreative. The rattling pieces are the little bolts to hold injector plate and mounting plate in place. It's been said and it does not change from wondering...

So it's normal that it's rattling?
Or is the plate loose wich results in worse temps?

[hewi]

If mine rattles while it's being used in a loop...it's a valid reason for not using any Watercool products in the future. It's these small things that annoy me, they could have used longer studs so that they wouldn't move back & forth
causing this damn rattle. *end of rant*

[SAE]

The plate is being forced down by the o-ring sealing it against the inlet area. The 2 little bolts are there to correctly align it so it cannot slip dis- and reassembling the block. The remaining 4 bolts are to hold the mounting plate and to avoid it being pulled out of the assembled block.
So the rattling is normal, if you are shaking the waterblock. But running it inside a loop as intended, it should not rattle at all.