It's a well known fact that Indigo2's are noisy. Sadly, because of the cooling requirements, most people just put up with the noise, or try and locate the system unit out of hearing range. Sometimes that's not possible or convenient though. My Indigo2 has a CD-RW drive and a floppy drive located internally, so I need close access to the machine. I decided to take a look at ways to reduce the noise level of the machine without making in inaccessible or difficult to work with.
A quick search round the Internet revealed that a new product called Akasa pax.mate. This sticky-backed sound absorbing foam can be applied to the inside of your computer case to absorb sound and damp vibration. There are several reviews out there on the Internet, but PC type cases are a little less elaborate that the Indigo2. Fitting pax.mate to an SGI looked like a bit of a challenge and nobody else seemed to have attempted it, so I grabbed a box from Maplin and gave it a shot.
Please note: I cannot be held responsible for any damage you do to your machine, yourself or others if you attempt this yourself. I am simply documenting my experience. If you choose to follow these instructions, you are responsible for any consequences which may arise!
Opening the packing.
The pax.mate comes in four pieces; two large square pieces intended for the side panels of a midi-tower computer case, and two smaller rectangular pieces for the top and bottom of the case. The beer is optional and not included in the package.
The target machine
As you can see, the Indigo2's case is a pretty tight design. Just placing a large piece on the inside of the lid would cause it not to close, as it contacts the chassis in several places. This is to separate the different sections of the case to control the airflow. Some careful measuring is required to cover just the parts of the lid which are next to an open space in the machine.
The sides of the lid.
Fortunately, the sides of the lid have a gap between them and the chassis of the machine. There are ribs at intervals along the panels, which provided an ideal place to put the pax.mate without it getting snagged when the lid is closed. I don't believe these channels are part of the airflow of the case because they're outside the actual chassis. I cut the pax.mate to fit and fixed it inside.
Vents in the top cover.
The other side of the cover has slots in it where air is drawn in over the graphics cardcage. I cut thin strips (1/8" wide) and stuck them along the plastic runs so as not to impede the airflow. This was the most time-consuming part of the process and you could just not put any pax.mate in these areas if you were in a hurry.
Removing the mainboard.
I removed the mainboard and placed pax.mate beneath it. I couldn't see as airflow beneath the board was important, since there are no rear-mounted devices on it. I placed the sheet of pax.mate on the board and pressed down with the backing paper still on to mark where I needed to cut for the mounting holes.
The bottom sheet of pax.mate in place.
I messed up a bit here, I have to admit. Having got the pax.mate perfectly cut and stuck down, I found that the pins and solder joints on the bottom of the motherboard would not allow it to be slid back into position, since it's held mainly by keyhole type slots. It took quite some effort to get it back where it should be. With hindsight, I would have left the backing paper on the pax.mate to allow it to slide with the motherboard. Still, it's back in place now.
Other places needed covering too.
With the motherboard back in place, I set about lining the other spaces in the machine with pax.mate. I made sure that absolutely no gaps got covered up with pax.mate so as not to impede the air flow. I only stuck the pax.mate where there was a clear, flat surface.
Drive bays.
The drive bays already have some sound insulation between the chassis and the outer skin, but there was enough gap to place another layer in the right hand side without reducing the air flow. If anything, it should make the air move faster past the hard disk.
The front cover.
Finally, I applied the foam to the inside of the front cover. Although you can't really see it from this picture, the slots in the lower portion have not been covered. It was just applied in thin strips along the plastic runners.
It took me at total of about 5 hours to get everything covered, and I had quite a bit of the foam left over. Maybe I'll put some in an external SCSI hard drive I have on my O2.
So, I'm sure if you've made it this far, you'll be wanting to know the result. The machine sounds distinctly more muffled, and is a lot less irritating because the tonal quality has changed. The biggest difference is that I can't hear the hard disk spinning, although you can still hear it seeking. Also some of the vibration has disappeared, leaving purely an airflow noise. Overall I'm pretty impressed, but the next stage would be to start changing fans. The first candidate would be the CPU fan, it's a little 60mm one and they always scream. I'll report back once I've tried that.
In conclusion, Akasa pax.mate gets the thumbs up. I wouldn't say it reduced the actual noise level that much and I don't have any equipment to measure that, but it's much more easy on the ear.
