Thoughts on the Jiushark JF13K Top-Down cooler?

[iolaus]

I actually typed this up as a response to an old post, so forgive the ancient links. There has been almost no discussion of this particular cooler on the site, so what little has occurred seemed relevant. Now that the Jiushark has been out in the wild for a while, I thought revisiting it might be interesting.

Here we go, the first review

That's the review that first put it on my radar. I wondered how well it would work in my GD09 case, but I've got an Assassin Mini on the 9700K in there that's working just great. When the AIO for the 9700K in my Corsair Air 540 case died, I stuck a salvaged single-tower 120 in it 'til I figured out something better. Seemed like a good time to try the Jiushark.

It's a PITA to install, and it blows the exhaust into the RAM. I saw someone somewhere mention that depending on your case and where you have the GPU, you can turn the orientation so that the RAM is clear.

It does appear to have dropped my temps a bit from what I had in there. I was idling at 35-40 before the swap; now I idle at 30-35, and dip down to 27-28 occasionally. I reset my min/max while typing up this post, so this reflects my temps during casual use...

...and while I don't have a pre-swap test to compare, this was after running Prime95 for about a half hour. It was peaking at around 150 watts while I was watching. Looking at it a half hour later, it's still exactly the same max temps.

I agree. I personally don't have a problem with top down fan directions. As you note, they tend to offer good cooling to sensitive components surrounding the socket. They also may interfere less with desired front to back flow of cool air through cases.

I'm wondering... with sufficient air flow, I imagine similar benefits would be achieved with a bottom up flow - i.e., fans pulling the air through, refreshing the ambient air below with fresh air from the sides.

I read somewhere that "push" fans focus more air-flow to specific regions (the ring below the fins) on the heatsink/radiator, while "pull" fans, given a little space to work with, can create a more even vacuum that affects a wider area - center axis area and the corners. With a high static pressure fan, that would seem to be a more efficient configuration for the heatsink itself, and might keep the underlying components cooler than a top-down flow.

Any why anyone would buy a low profile cooler just to spend extra to swap out the 15mm thin fans with standard 25mm fans, as suggested above, is just baffling to me. And they would still be 120mm fans which, as a general rule, have to spin faster (therefor louder) than a 140mm of equal quality to move the same amount of air.

LOL... My first thought before purchase was "Will better fans improve anything" Next thought was, "Can I use a couple of adapters to install 140mm fans?" For the moment, I'm sticking with the stock fans, but I'm willing to play with it.

I'm a big fan of overkill even if not needed, and I'm the type that tinkers with things just for the fun of tinkering (Heh...the mix of different salvaged fans in my Corsair case would horrify the aesthetically focused individuals). My case is relatively open, so the low-profile thing just gives me more room to play.

Looks like a MF'er to install. But uses the push pin design like stock coolers, those things suck.

Except with stock coolers, you push then make a 1/4 turn to lock in place. I don't see how these could be locked in place.

I gotta admit it's a PITA to install on an existing build. Screws are loosely held in place on the heatsink, go through the MB, and are held down with nuts on the other side. It's a PITA to get them all lined up proper and then hold in place while you flip it over and hold the backplate in place while you put the nuts on - worse if you are dealing with the whole case to do so. With all the rocking, I'd be pretty certain the paste is uneven now. It probably would have been easier to mount to the MB before installing - or rather, mount the MB to it.

 

[Bill_Bright]

I'm wondering... with sufficient air flow, I imagine similar benefits would be achieved with a bottom up flow - i.e.,

No. First, I have never heard of a bottom up flow CPU cooler - but my guess for that is because they just would not work. They would cool the CPU poorly because the actual force of the moving air (being pulled in from all directions) would be much less than top down. And they would offer zero cooling to the components surrounding the socket.

When a cooler is marketed as top-down, that typically mean opposed to side-firing.

Here's another review: Jiushark JF13K Diamond Review: Innovative Air Cooling, Ideal for SFF Systems | Tom's Hardware

 

[iolaus]

No. First, I have never heard of a bottom up flow CPU cooler - but my guess for that is because they just would not work. They would cool the CPU poorly because the actual force of the moving air (being pulled in from all directions) would be much less than top down. And they would offer zero cooling to the components surrounding the socket.

It's really just the old question of push vs pull - just with the heatsink in a different orientation - and everything I have found on that says that the difference is usually minimal, but not detrimental. In fact, some tests showed a slight improvement with pull, but only about a degree.

...and why would there be "zero" cooling for the surrounding components? It's not like there would be dead air in that space. Instead of hot air blowing down, cool air would be flowing in from the side on the way to the heatsink, and any heat the was pulled from them would be diluted by the rest of the air as it entered the heatsink. If anything, I would assume the CPU temps that might suffer from being downstream, but not the board components.

Here's another review: Jiushark JF13K Diamond Review: Innovative Air Cooling, Ideal for SFF Systems | Tom's Hardware

Yeah, I remember reading that a year or two ago, but didn't remember the content. I see he addresses upgrading the fans and concludes that the little bit of improvement isn't worth the effort.

 

[nomdeplume]

...and why would there be "zero" cooling for the surrounding components? It's not like there would be dead air in that space. Instead of hot air blowing down, cool air would be flowing in from the side on the way to the heatsink, and any heat the was pulled from them would be diluted by the rest of the air as it entered the heatsink. If anything, I would assume the CPU temps that might suffer from being downstream, but not the board components.

This is assumptive and likely as inefficient as has been suggested. With good reason cooling occurs with a fresh air feed cycling heat out an exhaust fan. Sucking heat off the board/processor/VRM through a heatsink. Or alternately even just from the still board level restricted airflow channel between RAM does not work.

I think this cooler is interesting. FWIW I ran a low profile Scythe Big Shuriken with 150mm Noctua fan in a case with max cooler height of nearly 200mm. Worked very well for Coffee Lake on a business oriented H370 CSM mobo. Very doubtful VRM and board components would've had nearly the level of performance during gaming with a tower cooler.

Anyhow, good luck and have fun experimenting.

 

[Harlow]

I use low profile coolers, e.g. Thermalright SI-100 in low profile SFF cases where there is very little headroom. For taller cases, I use full height coolers, e.g. Thermalright Peerless Assassin/Phantom Spirit, or Noctual NH-D14/D15. Well designed coolers have the advantage of being easier to mount (usually).

I might try JF13K if I was mad enough to fit a 200W CPU in a low profile home cinema system, but I was somewhat alarmed at the need to temporarily remove the AM5 metal socket surround to replace the back plate, with the increased risk of bending the delicate contacts on the exposed socket.

 

[Bill_Bright]

It's really just the old question of push vs pull

No its not. It seems you have your mind set and are not hearing what is said.

Push vs pull is definitely up for debate when it comes to case fans. It is easy to see how efficient cooling can be achieved if case air flow is front to back "OR" back to front - as long as not trying to do both at once.

But this is a CPU cooler, not case fans. It sits on top of the CPU where it is like a fan blowing into a brick wall.

Case fans either blow out into an open room, or into a cavernous case - neither offer much, if any resistance to that flow.

This cooler is even marketed as a top "down" cooler.

..and why would there be "zero" cooling for the surrounding components?

My bad. I was referring to typical downward firing CPU coolers - not this one where the "downward" firing fans are elevated a couple inches above the CPU. I should have been more clear there.

In fact, some tests showed a slight improvement with pull, but only about a degree.

I don't believe that. Show us these tests. Please post links that show "upward" firing CPU fans showed improvement.

Personally, I would never consider that Jiushark JF13K fan for my computers. It is too "gimmicky" for me - even though it works. I note the Tom's review said,

While it can be installed by a single person, it’s less of a hassle on Intel platforms if you have someone to help.

If it is not simple for me to install without someone else's help, I don't want it.

 

[iolaus]

No its not. It seems you have your mind set and are not hearing what is said.

No, I heard you. You essentially said it won't work because you've never heard of it; and then you presented an unsupported opinion that it would have less air moving through the heatsink.

My opinion is that a ring of pressure being directed against a similar surface of resistance would focus air through a smaller area of the heatsink than a more diffuse vacuum pulling through all available fins. I can see evidence of this in the dust patterns of uncleaned case air filters vs the dust patterns on radiators with push fans - the upstream filters that are 1/4" from the case fan get a diffuse covering while a push fan on a radiator creates a ring of dust. Logic would suggest that, all else being equal, a wider feed area would produce less effective resistance to the fan at the other end, and having the pull fan spaced away from the surface would produce slightly more airflow through the wider accessible surface of the heatsink.

Either way, I've already said that the reported effects were minimal, so it's hardly an important point.

But this is a CPU cooler, not case fans. It sits on top of the CPU where it is like a fan blowing into a brick wall.

Then what's the point of a fan? The air is moving through the fins, one way or the other.

This cooler is even marketed as a top "down" cooler.

As you say, "marketed," and as you said earlier, that just means as opposed to side-firing. I've never been one to follow instructions well. Using off-the-shelf modifications planned by others is boring; getting an idea and figuring out how to make it work... that's interesting.

I don't believe that. Show us these tests. Please post links that show "upward" firing CPU fans showed improvement.

Not "upward" because that's a fairly unique configuration, but addressing the push vs pull on a radiator... Here's one - https://tech4gamers.com/push-vs-pull-vs-push-pull-radiator . Beyond that, I'm not doing your homework for you.

If it is not simple for me to install without someone else's help, I don't want it.

Yeah, like I said, it was a PITA to install; but figuring out how to do things is part of the fun. LOL... I had paper baffles taped into this case to play with airflow once; and Hell, I've taken a dremel and a jigsaw to my GD09 HTPC case to make things work a few times!

 

[Bill_Bright]

You essentially said it won't work because you've never heard of it

No. You misinterpreted what I said.

I did NOT say it won't work "because" I never heard of it.

I said, in effect, I have never heard of it because they don't exist. Now again - I was talking about traditional downward firing CPU air coolers.

Then what's the point of a fan? The air is moving through the fins, one way or the other.

Ummm it should be obvious. The fan "pushes" more air (and thus moves much more heat) away from the heatsink than heat could just "float" away on its own with no fan.

If the fan was firing up (away from the heatsink), the force moving the air through the fins is dispersed (thus weaker) because it is being drawn in from all directions instead of all the air being pushed through one "nozzle" (the fan's housing).

Now they do have "passive" (no fan) heatsinks for "silent" operation, but they are HUGE as seen here.

My opinion is that a ring of pressure being directed against a similar surface of resistance would focus air through a smaller area of the heatsink than a more diffuse vacuum pulling through all available fins

You are describing exactly what I just said. That a downward firing fan - one that blows down on to the heat sink - a traditional fan - works best.

Do note my initial reply was in response to you suggesting a "bottom up" (upward) flow would be just as beneficial.

Not "upward" because that's a fairly unique configuration

I agree. But again, I was responding to you saying "bottom up" (which means "upward").

Your link clearly supports my claim. Note fan is blowing cool (blue) air into the radiator and hot (red) is exhausting the heat away. That said, push vs pull in your link, with a "water cooler" setup using a radiator and water block is totally different from an "air cooler" using a heatsink and fan(s) like the Jiushark cooler. You are talking apples and oranges there - both are fruit, but that's about it.

 

[iolaus]

...

Wow, talk about not hearing what's being said! LOL... A radiator is a heatsink, and a heat sink is a radiator; they both move heat out to fins to be dispersed by air movement. It's more like, Honeycrisp vs Golden Delicious.

Okay, whatever. I've said my piece, and I think I was pretty clear in how I presented it, even if you want to misinterpret and deny what's been said. I'll let you have the last say and let others judge.

*EDIT...
My apologies. I know I said I was dropping it, but I realized the misunderstanding on this point was my fault.

But this is a CPU cooler, not case fans. It sits on top of the CPU where it is like a fan blowing into a brick wall.

Then what's the point of a fan? The air is moving through the fins, one way or the other.

Ummm it should be obvious. The fan "pushes" more air (and thus moves much more heat) away from the heatsink than heat could just "float" away on its own with no fan.

I was not questioning the point of a fan - I know convection is the weakest of forces. I was laughingly responding to your portrayal of there being a "brick wall" - hence my observation that air is flowing whether it's a push or a pull force creating the flow. Subtleties get lost in text, and I phrased it poorly.

 

[Bill_Bright]

*EDIT...
My apologies. I know I said I was dropping it, but I realized the misunderstanding on this point was my fault.

Thanks for this. We are good!

A radiator is a heatsink, and a heat sink is a radiator; they both move heat out to fins to be dispersed by air movement.

Just to clear up one point for those reading. The above is not exactly true. While heatsinks and radiators are very similar, they are very different too and work in very different ways.

A heat sink is a "passive" device. Passive implies NO moving parts. No coolant flowing through it. NO fan. And NO "air movement" required. It works by absorbing heat directly from the hot CPU surface and transferring that heat to the surrounding environment with its increased surface area provided by the fins. The transfer is by convection and radiation and NOT air movement.

Now if there just happens to be air movement caused by case cooling or an accompanying fan mounted to the heatsink (AKA, HSF or heatsink fan assembly), that's fine. But that then is a "heatsink fan assembly" and not a heatsink. And a key difference there too is, if a fan is added, it blows air through the heatsink's fins - not the entire heatsink.

"IF" the total surface area of the heatsink fins are large enough AND there is sufficient space above the heatsink to allow the heat to move away through the process of convection, no fan is required at all to properly cool the CPU.

A radiator, on the other hand, is an "active" device. It does not sit directly on the CPU's hot surface. The coolant or water block sits on top of the CPU, and coolant is pumped (part 1 of the "active" component) to the radiator through hoses or pipes. Then a fan(s) (part 2 of the "active" component) blows air through the radiator to move the heat away.

Heatsinks don't require a fan to remove heat. Radiators require a fan, coolant, and a pump (or pumping action via thermosiphon) to circulate the coolant through the radiator. No coolant circulation = hot CPU. No fan blowing air "through" radiator = hot CPU.

The Jiushark JF13K cooler is a radiator cooler, as clearly indicated on the box - twice! One might argue the coolant or water block (the solid hunk of aluminum that mates with the CPU's IHS surface) is a heatsink, but it has no fins and is totally useless without the pipes and coolant flowing through them.

I will point out too that the Jiushark JF13K cooler, at least by the Tom's review is an excellent cooler - once mounted. And that's the main thing. For someone with a slim case, it's low profile might be the perfect solution.

***

Oh, and OT but just for future reference since you are very new here, I didn't watch your video above and not many of the regulars here will either. Why? because, just like that one, they take 25+ minutes to say what could have been said in 5 minutes or less - clearly filling it up with extraneous, non-essential footage to create more exposure for their sponsors. I'm just saying... ...that's why I quickly searched for other reviews and found Tom's - one I could quickly read in less in 5 minutes.

Anyway, like I said, we are good and thanks for posting about this very unique, and effective cooler.