Search
Click here to Search by Part Number
Store Index
Memory
MacGurus RAM
SATA (Serial ATA)
Host Cards
Burly Enclosures
Cables
Hard Drives
RAID
Ciprico MediaVault
SuperDrive
Burly DVD-R Burner
SCSI
Enclosures
External Cables
Internal Cables
Terminators
SCSI Hard Drives
Adapters
SCSI-Firewire Adapters
FireWire
Case Kits and Drives
Multi-Drive Enclosures
Host Cards
FW 400 Cables
FW 800 Cables
Enclosure Parts
Burly Parts
ATA
Cables
ATA Hard Drives
Hard Drives
ATA Hard Drives
SCSI Hard Drives
SATA Hard Drives
Notebook Hard Drives
PowerBooks
Express34 and Cardbus Expansion Cards
Momentus Drives
Special Deals
Specials&Closeouts
Tech Guides
Guides to Acceleration
Building a Photo Database
Storage Acceleration
Photoshop Acceleration
Storage Guides
DeskTop Drive Setup
Preparing New Drives
Move Users 10.5
Move Users 10.4
SATA Host Card Guide
Port Multiplication Guide
Table - Front Side Bus
Giga CPU Upgrade
Roll Your Own SATA
Hot Swap SATA Guide
Roll Your Own RAID
Motherboard Layouts
Drive & RAID Database
|
Temperatures
Since I planned on installing a Seagate Cheetah 15K.3 on an OEM ATTO
UL3D, I wanted to
monitor the temps. I placed a temp. probe in the gap between the PSU
and
optical/zip drives cage to get an idea of the air/case temps before the
air gets
sucked into the PSU. I used Hardware Monitor
for S.M.A.R.T. hard drive temps; there's also the freeware Temperature
Monitor for this. The stock Maxtor, located in Bay 2
(middle), didn't yield S.M.A.R.T. temps, so that drive is ignored.
I monitored the DA for
hours during normal activity and under load, e.g. Cinebench, FCP3
rendering, cloning, etc., both before and after the upgrade. DA was
under a table with nothing behind it;
so there was ample ventilation. These are max. readings unless range is
given.
Temperatures
(°C)
|
733MHz
|
2 x 1.33GHz |
Ambient
|
22-23
|
22
|
| Temp. probe |
31
|
40.6 - 41.4 |
120GB
7200.7 - bottom U
bracket
|
41-44
|
46
|
80GB Cuda IV - top U bracket
|
38
|
41
|
80GB Cuda IV - bay 1
|
41
|
41
|
The jump
in temp. probe readings were expected. Not just because of the faster
and second Giga 1.33GHz CPUs, but also because the stock heatsink cover
with an exhaust fan was removed and not put back, as per GigaDesigns'
installation
instructions. The heatsink cover in stock configuration probably acts
as a heat trap, with the small built-in fan serving to exhaust some of
that heat generated by the 733MHz CPU.
Without that cover, some of the (additional) heat generated by the Giga
upgrade with two fans can swirl around the case; and the temp probe,
close to Giga, was bound to catch it. Some
hot air does get pushed out of the vents near the CPU however.
Even with four hard drives in the DA, hard drive temperatures, while
higher with the Giga, are well below 60° C max.
operating
temperature for the Seagate Barracudas.
Additional
Cooling
|
 |
|
Digital
Audio G4 with
GigaDesigns M5D-1213Q and Quicksilver CPU fan housing & fan.
|
With the Cheetah 15.3K on the horizon, I was curious whether I could
lower the temps without using a slot cooler. When I installed the Giga,
I was sorry to see the
CPU exhaust fan go.
Since air was getting forced out the vents near the CPU, I decided to
help matters.
While there was the option of cutting out the fan section of
the stock plastic heatsink cover, I didn't consider it seriously. I
wanted to keep the original parts intact. More
importantly, I noticed that the Quicksilvers' rubber CPU fan housing
had the
same
form and size as that section of the DA's plastic heatsink cover. It
looked
like it would fit. So for US$40 I bought a new QS fan housing and fan
(Sunon KDE 1206 PTV1 - 33.5 dbA,
23.5 CFM) and tried it my DA. The 60mm Sunon was part of the deal, and
probably not a bad thing since the stock CPU fan was a bit noisy
after years of use.
The QS housing fits
perfectly where the
DA's CPU fan used to be without any case modifications, although
different
mounting holes and machine screws are used. I reversed the QS fan so it
exhausts -- in a QS
it's an intake fan. I stuck a piece of tape over the hole on top of the
housing to help direct the air out of the case, much
like
the DA's original heatsink cover which didn't have a hole on the
beveled section. The DA's stock CPU fan plugged into the 733MHz CPU
daughtercard. So to power the fan I made my own leads that run around
the motherboard,
just like the Giga M-series 12V leads. One end was tailed with a 2-pin
fan connector (male) so the fan's connector plugs right in, which can
be seen below. The other end was tailed with a standard 4-pin Molex
PATA HD
connector. Using the 2-pin fan connectors, rather than splicing, makes
it easy to remove the fan for cleaning or replacment.
 |
| Left: CPU vents and
mounting |
Right: Fan modifications
and power
|
I wired the QS CPU fan for 12V, and it dropped both the case
(temp
probe) and HD
temps. But at 12V the fan was little
too much like a leaf-blower for my tastes. So I lowered the voltage to
7V (using the standard wiring trick explained here),
and it still kept things cooler than no CPU exhaust fan. Here is the
complete set of results for easy comparison:
Temperatures (°C)
|
733MHz
DA CPU fan
|
2 x
1.33GHz
No QS CPU fan
|
2 x
1.33GHz
QS CPU fan @ 12V
|
2 x
1.33GHz
QS CPU fan @ 7V |
Ambient
|
22-23
|
22
|
22
|
23-24
|
| Temp. probe |
31
|
40.6 - 41.4
|
35.6 - 36.4 |
37 - 37.5 |
120GB
7200.7 - bottom U
bracket
|
41 - 44
|
46
|
42 |
44 - 45 |
80GB Cuda IV - top U bracket
|
38
|
41
|
38 |
39 |
80GB Cuda IV - bay 1
|
41
|
41
|
38 |
39 |
At 7V I can't hear the CPU fan over the case and PSU fans; temp.
probe/case temps are still lower; and HD
temps are
still close to stock temperatures, even with an ambient temperature a
tick higher. So I've left the fan at 7V since it hits the right balance
between noise and cooling -- and no slot lost to a slot cooler.
Next: Part
III: System Benchmarks
The Gurus Certify This Review Rules
|