Battle of the titans: Aerocool HT-101 Vs SP-97 + Nexus 92

jones_r · Post by **jones_r** » Thu Sep 23, 2004 6:50 pm

Well, I'm building a quiet PC for my friend. He has an Athlon XP Thoroughbred 2400 (2000mhz, 68.3W).

He doesn't want to OC, but also doesn't want to DC, and want the quietest possible operation. He also bought the mCubed fan controller.

After much searches, I've narrowed it down to these two. They seem to be the quietest solutions possible today for Athlon XP.

Which do you think will better fit the job ?. The SP-97/Nexus combo cost $20 more, but my friend doesn't care about it, as long as it will prove to be more quiet.

Somehow the HT-101, with its much separated fins, seem to be more fit to very low RPM tasks.

What do you think ?, is there a clear winner ?.

Trip · Post by **Trip** » Thu Sep 23, 2004 6:53 pm

What board are you going to use? (the sp-97 may not fit)

jones_r · Post by **jones_r** » Thu Sep 23, 2004 6:55 pm

MSI K7N420-PRO.

http://www.gamepc.com/images/labs/rev-m ... fileLG.jpg

It's not on the SP-97 list, but I think it is because this mobo is simply ancient and forgotten.

Edit:
Weird, I've just checked and it is on the SLK-900A list. What does this mean ?.

Trip · Post by **Trip** » Thu Sep 23, 2004 7:13 pm

I see holes too, I think it'd fit. Silvervarg or Bluefront may have gotten ahold of an HT-101.

jones_r · Post by **jones_r** » Thu Sep 23, 2004 7:17 pm

I think that the only way that the SP-97 will fit is with its pipes on the outer edge of the mobo. The SLK-900A didn't have pipes, so maybe the SP-97 can not mount with the pipes at that certain position, and this is why the K7N420 PRO is out...

Post by **MikeC** » Thu Sep 23, 2004 7:20 pm

Why don't you go with your instinct and try the HT-101? According to the specs, its weight is fairly low at just 470g w/o the fan, which is good considering how tall it is -- so the cantilever force on a vertical motherboard won't be too bad even with a fan. I doubt the supplied fan is quiet; clear plastic fans rarely are.

Then you can come back and tell us how successful your project is.

PS -- I don't know if the fins are really any farther apart than on the SP97... Will have to get a sample and see.

jones_r · Post by **jones_r** » Thu Sep 23, 2004 7:30 pm

PS -- I don't know if the fins are really any farther apart than on the SP97... Will have to get a sample and see.

Maybe this will help:

so the cantilever force on a vertical motherboard won't be too bad even with a fan.

I'm not going to attach the fan to the HT-101. It is going to hang in the air, near the HT-101 (at the right direction), and something is going to hold it to a base which will be put on the carpet. No connection of the fan to the computer, no vibes.

Maybe I will get the Nexus 80mm with the HT-101 ?, or should I try the 92mm ?.

Edit:
This is interesting:

Dobby · Post by **Dobby** » Thu Sep 23, 2004 8:14 pm

Maybe this test will help you in decision http://www.skenegroup.net/fi/artikkelit ... ol_ht101.4.

Sorry, text is in Finnish, but I think you'll understand graphs and different fans used in the test. Apparently, it seems that HT-101 is slightly better when low cfm fan is used.

Bluefront · Post by **Bluefront** » Fri Sep 24, 2004 2:36 am

I do have one of these coolers.....still working on the project. I did come up with a method to mount the Ht-101 in any direction on a P4 board.

I think the fins are relatively close together....but very thin, with lots of surface area for the weight. The main thing for me is the ducting potential of the unit.

I'm considering ducting the airflow up, directly into a PSU with a bottom 120mm fan. I can then construct an intake duct to the lower side of the cooler, completely eliminating the CPU fan. I already have a version of this setup working, using an NCU 1000......this HT 101 should be an improvement.

jones_r · Post by **jones_r** » Fri Sep 24, 2004 3:57 am

Is there any point of using a 120mm (or 90mm) fan with the HT-101 ?.

The bigger the fan, the slower the RPM you can use, and still achieve the same CFM. But on the other hand, the HT-101 is constructed in a very directional way (not like the SP-97), so there might be two problems with a bigger than 80mm fan:

1) The areas of the fan outside of the plastic case of the HT-101, will do very little to cool the HT-101.

2) A bigger than 80mm fan, also has a larger "dead spot" in its center, so this mean less airflow for the middle of the HT-101.

Merging these two deficiencies together, one might think that the 120/90mm fan will have to rotate at the same speed, or even at a higher speed than the 80mm fan, in order to achieve the same cooling performance.

BUT (there is always a "but"), what happens if you decide to use the HT-101 in an unconventional way:

By removing the plastic case, you drastically enlarge the heat transfer space, drastically!. By doing so, you can now use a larger than 80mm fan, and let it rotate real slow.

Could it be that the only reason that Aerocool put this plastic case in the first place, was that there would be a possibility to attach a fan to it ?. Otherwise its just another NCU-2000, only this one can't run without a fan...

Btw, about the issue of which has more space between its fins, the HT-101 or the SP-97, I think that this pic will put a rest to the assumptions:

SP-97 is on the left. HT-101 on the right.

Dobby · Post by **Dobby** » Fri Sep 24, 2004 9:13 am

If you are planning to cool only stock xp2400+, I think its waste of time trying to fix larger fan HS. I don't have HT101, but Sharkoon HPS1 which according to tests should have slightly lower capacity. However, its perfectly capable to handle cooling of similar processor with nexus 80 mm fan undervolted to ~9V in suck mode. Being inside case thats already practically dead silent solution.

Trip · Post by **Trip** » Fri Sep 24, 2004 9:23 am

an 80 to 70mm fan adapter should allow you to keep the duct.

jones_r · Post by **jones_r** » Sat Sep 25, 2004 8:47 am

Dobby,
So, are you telling me that your Sharkoon HPS1 together with the Nexus 80mm undervolted to 9V, can not be heard at all inside the case ?. Is that the only fan in your computer ?.

I don't have HT101, but Sharkoon HPS1 which according to tests should have slightly lower capacity.

Actually, at least according to this comparison review, the Sharkoon HPS1 took #1 place at low RPM.

If you read that review, you see that both the Sharkoon and SP-97, scored a little bit better than the HT-101, but, in contrast to the HT-101, they do not have a cage which decrease the heat transfer space of the fins with their surrounding. This might lead one to think, that by removing the plastic cage from the HT-101, it is going to be #1.

Dobby · Post by **Dobby** » Sun Sep 26, 2004 7:38 am

jones_r wrote:Dobby,
So, are you telling me that your Sharkoon HPS1 together with the Nexus 80mm undervolted to 9V, can not be heard at all inside the case ?. Is that the only fan in your computer ?.

At default it's not of course only fan, but I made some tests where other fans were stopped for a short while and case was closed. Voltage is fanmate approximate.

I
this comparison review, the Sharkoon HPS1 took #1 place at low RPM.

Yes, I have read it and for some reason the test placed HT-101 surprisingly low in ranking. That is contradictory to other other test made with HT-101. Personally, I made choice between these two purely because of the orientation in NF7-s motherboard is towards back of the case in HPS1.

silvervarg · Post by **silvervarg** » Sun Sep 26, 2004 10:58 pm

If you read that review, you see that both the Sharkoon and SP-97, scored a little bit better than the HT-101, but, in contrast to the HT-101, they do not have a cage which decrease the heat transfer space of the fins with their surrounding. This might lead one to think, that by removing the plastic cage from the HT-101, it is going to be #1.

One of the major benefits with a tower heatpipe cooler is that it is perfect for ducting to a case exhaust or at least pushing the air towards an exhaust fan.
If you want to use a cooler this way you will benefit from having a shroud around the heatsink.
Since most reviews are done without case a cooler without a shroud will perform better. Mounted in a case you will benefit a lot from having a shroud and/or duct, so you should have that in mind when choosing a cooler.

jones_r · Post by **jones_r** » Mon Sep 27, 2004 3:08 am

Ok, I asked Aeorcool this question:

I would like to know why did you enclose so much of the heatsink of the HT-101, inside a plastic cage ?, doesn't it reduce the heat trasnfer between the metal fins and the surrounding area ?. Why haven't you done a similar approach as you did with the DP-102, where there is no cage ?.

Their reply:

Hi !! Thank you for your interest in our HT-101 cooler.

The reason why the HT-101 fan cap enclose so much of the heatsink is because this will allow the air to travel from one side of the fins to the other side. This actually improves the heat dissipation of the HT-101 cooler. The air will cover more surface areas of the cooler.

Thank you and have a great day !!

Best Regards,

Tony Lin
Aerocool Advanced Technology

I'm not so happy with this answer.

Dobby · Post by **Dobby** » Mon Sep 27, 2004 7:35 am

jones_r wrote:
I'm not so happy with this answer.

I think there's no straight black&white answer to this. If you for example have PSU with 92/120 mm bottom fan, tower HS without a shroud will get cooling aid from PSU fan. You can most probably cool cpu with less cfm, but warmer air adds risk that PSU fan speeds up. With a shroud or duct you will likely need slightly more airflow with CPU fan, but on the other hand PSU gets colder air.

jones_r · Post by **jones_r** » Mon Sep 27, 2004 11:06 am

My PSU is phase-change cooled...

jones_r · Post by **jones_r** » Thu Oct 07, 2004 9:57 am

Finally, I got all the parts.

I did some testing, and I really don't understand the results posted in this table:

They used a XP2200 Thoroughbred, and I use a XP2400 Thoroughbred.

They undervolted the original HT-101 fan to 7 volts, which gives 1650 rounds per minute, yet I use the T-balancer fan controller in order to produce only 16%, which gives me 398 rounds per minute.

My ambient temps are 25 degree celsius, and the HT-101 fan is the only fan in my computer (I don't know the equivalent details for their setup, but I think its safe to assume it is more forgiving than mine).

And now for the weird thing, they got 54 celsius for idle and 60 celsius for load, yet I gt 56 for idle and 61 for load.

To top everything, the HT-101 that I got has one of the fins missing! (near the top).

How can you explain this ?.

No matter what is the explanation, I'm very impressed with the results. Cooling a 2000mhz CPU using a quiet 80mm fan spinning at 398 (!!!) rpm, and getting only 61 degrees at full load, is something to behold.

I also received the quiet Panaflo 80mm L1A Fan, but for some reason it stops working when I try to make it spin lower than ~760 rpm. This is WAY too fast for my quiet needs.

Post by **MikeC** » Thu Oct 07, 2004 10:13 am

jones_r wrote:How can you explain this ?.

Easily: CPU temperature reporting inaccuracy. On the reviewers' part and/or yours. They are using a different CPU and a different motherboard. There is no assurance that the temps reported are accurate from either their motherboard or yours, nor can we know what heat your respective CPUs are putting out.

At least try calibrating your CPU temp reporting system... see Russ's article Calibrate Your CPU Temp Reporting

jones_r · Post by **jones_r** » Thu Oct 07, 2004 10:22 am

MikeC,
Good point. I'll try to calibrate.

For what its worth, the former CPU in this motherboard (1.13ghz Athlon thunderbird), gave me 66 degrees @ load, yet it worked for 3 years with no problem (so if it was actually 76 degrees, it wouldn't have survived that long), also, I touch the pipes now, and they are only warm to the touch at full load, and it is possible to keep the finger on them all the time.

I guess what I'm trying to say is that if 398 rpm cool this CPU under load to a lower temperature than the older CPU, which worked flawlessly for years, I think I'm safe. But that's a whole different story I guess.

Interitus · Post by **Interitus** » Thu Oct 07, 2004 11:04 am

What's really funny is that they say an SP-97 with a Vantec Tornado at full speed is leaving them at 47C Idle.

If that's true I've lost all hope for CPU cooling with air...my Zalman at 5v keeps my Mobile XP @ around 40 idle running at 200x9 ..either their SP-97 really sucks or this guy can't type lol