Thermalright SI-97A for both K7 & K8!

Mats · Post by **Mats** » Mon Sep 27, 2004 5:35 pm

Here.
Looks a lot like the XP-90, but the fins doesn't go all the way down to the base. It even got the same dimensions, but the weight is only 280 g. Maybe it would have been better (at least for ducting) if the upper part were vertical, just like a Shuttle ICE heatsink.
Good to see new socket A and S370 heatsinks entering the market!

Edit #1: Changed the title to get the attention without making a new thread and spreading the discussion....

Edit #2: New model SI-97A! Fits socket 370, 462, 754, 939, 940.

msm_zgok · Post by **msm_zgok** » Mon Sep 27, 2004 6:04 pm

Looks like another winner from Thermalright. Hard to tell how this thing will be mounted on the mobo. My only concern is whether you need extra space between the motherboard and PSU to fit this thing as most of the socket A motherboards are near the top.

acaurora · Post by **acaurora** » Mon Sep 27, 2004 6:10 pm

It is interesting how there is that gap between the bottom and the top. How will that effect cooling performance?

I smell a review coming on!

Edward Ng · Post by **Edward Ng** » Mon Sep 27, 2004 6:20 pm

I will have to procure a sample for review...

-Ed

I procure it, but I don't do the review; another SPCR editor will.

mathias · Post by **mathias** » Mon Sep 27, 2004 6:23 pm

What on earth did they do that for?

I don't get this design at all. It's completely heatpipe dependant, yet it has the fan blowing onto or away from the base.

The weight is good though, and besides that one peculiarity, it is a proven design. If it's cheaper than it's cousins, it should be a good deal, and I think I might get one.

I think it could be modded by welding a passive cooler to the base. Like maybe an old orange zalman northbridge heatsink(NB32?), but with the pins bent out in all directions.

Edward Ng · Post by **Edward Ng** » Mon Sep 27, 2004 6:26 pm

Maybe to prove a point?

The fact also remains that this heatsink will clear anything the motherboard manufacturers can throw at it!

-Ed

sthayashi · Post by **sthayashi** » Mon Sep 27, 2004 8:19 pm

I wonder if it requires mounting holes around the socket....

/probably

acaurora · Post by **acaurora** » Mon Sep 27, 2004 8:35 pm

Edward Ng wrote:I will have to procure a sample for review...

-Ed

I procure it, but I don't do the review; another SPCR editor will.

/me eyes the "SPCR Editor" position - while acquiring a SilverStone Lascala SST-LC01 case

pony-tail · Post by **pony-tail** » Tue Sep 28, 2004 9:50 am

Ed , is there any way you can find out if they are going to use the Stock retention clip as it seems to be light enough .
I have been trying to find a cooler for an XP3200+ on an Asus A7N8X - VM400 as it has IDLE temps of 55 deg.C with the stock cooler . There have not been a lot of higher end coolers for AMD boards without the holes.

Post by **Rusty075** » Tue Sep 28, 2004 10:16 am

My guess:

If you look at the base of the Si-97, there's a flat spot in the short little fins that protrude from it. That flat area looks to be in the right spot, and be the right size, for a clip retention spring to mount over.

Don't know any reason why they wouldn't use the clips...its lighter than some of their other clip-mounted heatsinks.

I have some curiousity about whether mounting orientation will have an effect on performance....one way or another, the "hot" end of half the heatpipes is going to end up being mounted above the "cold" end. (unless you have a desktop case)

Tibors · Post by **Tibors** » Tue Sep 28, 2004 11:27 am

Rusty075 wrote:[...]I have some curiousity about whether mounting orientation will have an effect on performance....one way or another, the "hot" end of half the heatpipes is going to end up being mounted above the "cold" end. (unless you have a desktop case)

They probably use the same type of heatpipes as in the XP-120. So it won't matter. See the revisions MikeC made to the XP-120 review after remarks in the discussion thread about it.

pony-tail · Post by **pony-tail** » Wed Sep 29, 2004 11:16 am

I wonder if it requires mounting holes around the socket....

I just got an email back from tech support and YES they are going to use the 6 lug AMD clips not the mobo holes.
So count me in for 1 of these

sthayashi · Post by **sthayashi** » Wed Sep 29, 2004 9:03 pm

pony-tail wrote:I just got an email back from tech support and YES they are going to use the 6 lug AMD clips not the mobo holes.
So count me in for 1 of these

Woohoo!!!! I might have to give them a spin if they'll fit on my dual setup...

Post by **MikeC** » Wed Sep 29, 2004 11:04 pm

Looks to me they didn't really think it through. The primary usefulness of a heatpipe is to move the heat farther away from the source. They do that, but then they have the fan blowing the heat right back down to the CPU.

If I get one, I'd probably try bending the heatpipes so that the side profile is an L (like a HS from a Shuttle SFF) instead of a sideways U. Then it could be used with a fan blowing parallel to the motherboard surface rather than down aginst it or up from it. If the CPU socket is oriented right, an exhaust case fan might be enough.

mathias · Post by **mathias** » Thu Sep 30, 2004 8:54 am

MikeC wrote:Looks to me they didn't really think it through. The primary usefulness of a heatpipe is to move the heat farther away from the source. They do that, but then they have the fan blowing the heat right back down to the CPU.

That's great to hear, if it sucks at blowing, that should drive the price down. I guess this is why the XP's work rather well with the fan sucking.

Maybe the final version will be different, like at least a bit tilted, they could do that since it's no longer connected to the base. Hopefully it's only like this because it's made from XP-90 parts.

Edward Ng · Post by **Edward Ng** » Thu Sep 30, 2004 9:27 am

mathias wrote:...if it sucks at blowing...

Sort of like a solar-powered flashlight?

Mats · Post by **Mats** » Thu Sep 30, 2004 9:29 am

Edward Ng wrote:
mathias wrote:...if it sucks at blowing...
Sort of like a solar-powered flashlight?

.....................HAHAHA!!!

acaurora · Post by **acaurora** » Thu Sep 30, 2004 9:29 am

or dehydrated water - just add water!

or the wind powered fan?

mathias · Post by **mathias** » Thu Sep 30, 2004 9:54 am

acaurora wrote:or the wind powered fan?

I think that could work, kind of like a bicycle.

Post by **Rusty075** » Thu Sep 30, 2004 10:37 am

Edward Ng wrote:Sort of like a solar-powered flashlight?

Laugh all you want....but I own one of those. (and it works great, by the way.

)

Edward Ng · Post by **Edward Ng** » Thu Sep 30, 2004 11:56 am

I do mean without a cap or battery.

DG · Post by DG » Thu Sep 30, 2004 3:50 pm

solar-powered flashlight = mirror

Anyway, back ontopic now, do you think the SI-97 would cool better than a SP-97, with a low cfm fan (for ex. nexus 92mm)? Copper vs Aluminium, you know the story..

Edward Ng · Post by **Edward Ng** » Thu Sep 30, 2004 3:58 pm

I know the XP-90 easily outperforms SP-94; how this translates to SI-97 vs. SP-97 is difficult to say, however. We will have to wait until a formal review is completed. We expect a sample within a couple weeks.

-Ed

pangit · Post by **pangit** » Thu Sep 30, 2004 7:30 pm

MikeC wrote:If I get one, I'd probably try bending the heatpipes so that the side profile is an L (like a HS from a Shuttle SFF) instead of a sideways U. Then it could be used with a fan blowing parallel to the motherboard surface rather than down aginst it or up from it. If the CPU socket is oriented right, an exhaust case fan might be enough.

I like that idea, especially as you could position the fins right up against the exhaust fan with the right mobo/case (or use a short duct)

But would it be possible to bend heatpipes without cracking them or compromising their efficiency? I don't know how bendable they normally are, has anyone tried it with any other heatpipes?

Edward Ng · Post by **Edward Ng** » Thu Sep 30, 2004 7:34 pm

The heatpipes on the Zalman ZM-2HC1 are incredibly flexible. I've picked up hard drives mounted into them by mistake just squeezing the two opposite pipes together almost like a grip (none to tightly, mind you, although I do have quite the kung fu grip), and found the pipes bent in several mm.

Gholam · Post by **Gholam** » Thu Sep 30, 2004 8:43 pm

I'm thinking about mounting a fan under the heatsink and having it blow up and out of case side duct... in theory, this way the fan will have less interference on the intake, and the heat will be evacuated from the case directly. Would this be workable?

pony-tail · Post by **pony-tail** » Fri Oct 01, 2004 2:04 pm

I want to try and reverse the fan so it blows upwards as my intended use is in a rather small micro ATX mini tower (InWin V450a ) on an Asus A7N8X -400VM board and it looks like it will almost reach the Case door just add a foam plastic gasket and a side vent and all the heat is outside the case - The CPU is an XP3200+ and runs at 45 to 50 deg C, just surfing the web with the stock cooler - Currently it is running a TT Golf 325 which gets good temps with a 2000 rpm fan(an educated guess)but is utter cr*p with an undrvolted LI fan -
so the Thermalright HAS to be an improvement even in the worst case possible .

Please Note -: for the TT supporters out there that I am not knocking the Golf 325 it is a very good heatsink with a Panaflo M with a 27 watt resistor (voltage unknown & exact speed unknown) and is not terribly noisy but definately NOT a candidate for silent computing ( I bought it because A. it was the cheapest heatsink I could get B. it was available locally on the day I needed a heatsink .)

DanceMan · Post by **DanceMan** » Fri Oct 01, 2004 2:13 pm

Ed and Mike: after the regular testing is completed, might the tester try bending the radiator section 90 degrees? If much resistance is felt, a little assistance could be supplied by a heat gun.

Or not: from efcoins' post in another thread: "Heatpipes rupture at about 150C, less if they have bends in them, DO NOT HEAT THEM"

pony-tail · Post by **pony-tail** » Fri Oct 01, 2004 5:16 pm

I'm thinking about mounting a fan under the heatsink and having it blow up and out of case side duct... in theory, this way the fan will have less interference on the intake, and the heat will be evacuated from the case directly. Would this be workable?

I believe with a little engineering and creativity it would be possible - but how much and what kind of advantage would there be against just having the fan on top and blowing upwards .

Post by **MikeC** » Fri Oct 01, 2004 5:37 pm

pony-tail wrote:
I'm thinking about mounting a fan under the heatsink and having it blow up and out of case side duct... in theory, this way the fan will have less interference on the intake, and the heat will be evacuated from the case directly. Would this be workable?
I believe with a little engineering and creativity it would be possible - but how much and what kind of advantage would there be against just having the fan on top and blowing upwards .

IMO, none. The disadvantage is higher impedance - on both sides of fan instead of just one, so higher noise.