[soviet]

Looking for some upgrade fans for an o2 the ones installed are quite noisy and not in the best shape (cpu and psu fans).
Sourced some random fans for the same size locally but after some tests, they don't move as much air as the original ones.

Don't want to cook the guts of the o2, any suggestions about good models for this?

[johnnym]

Don't want to cook the guts of the o2, any suggestions about good models for this?

Can't really help you with that, as I don't use my O2s on my work table but they rather run in a sort of machine room, so their noise is not so troubling me.

But I recently stumbled upon this article here, where a guy explains what he considers important for a replacement fan for an O2:

https://bukosek.si/hardware/collection/s...ods-psufan

...i.e. static pressure.

I find the remainder of the site also pretty cool, especially the highres pictures of SGI equipment.

In addition I believe Ian (Mapleson) also did some fan mods for the O2 PSU, so maybe ask him, too.

[jan-jaap]

...i.e. static pressure.

That's the magic word. I like the Noctuas, but not in my SGIs.

I modded my CISCO FDDI concentrator with a pair of temperature controlled Noctuas but I reverted back to the original fans because even with the Noctuas at full speed it would still overheat the optical transceivers, causing them to drop the connection. You don't want something like that happening with your SGI internals.

[weblacky]

Sorry to bring this back from the dead, but I'd rather continue a thread versus start a new one:

I'm likely going to try O2 PSU rebuilds next, and I was again asked by someone about Fan replacement. Looking through this: https://bukosek.si/hardware/collection/s...ods-psufan

I THINK I found a BETTER (faster @ 4K RPM) fan than the original (Panaflo FBA08A12U1A: https://bukosek.si/hardware/collection/s...asheet.pdf) with SLIGHTLY LESS noise output: https://www.mechatronics.com/pdf/G8020.pdf

Model: G8020E

Anyone see a reason we can see this wouldn't be a modern replacement for the O2 fans wearing out? One DB less noise and HIGHER static pressure and CFM...I cannot locate anything better that meets or beats OEM specs/requirements.

Thoughts?

[johnnym]

[...]
Model: G8020E

Anyone see a reason we can see this wouldn't be a modern replacement for the O2 fans wearing out?  One DB less noise and HIGHER static pressure and CFM...I cannot locate anything better that meets or beats OEM specs/requirements.

Thoughts?

Maybe you perform a preview test - not sure if this is the correct wording, I mean like "test-hear" the fan installed. The RPMs are considerably higher (4,000 to 3,450 for the Panaflo FBA08A12U1A) which could lead to different sound effects from resonance on the plastic casing. Well, it might be different from machine to machine in the end, but if the O2 starts humming like wild with the replacement fan installed, it might not be such a pleasure to use it in the end. But maybe any disk drive will easily outperform the fan in that regard, so not so much of a problem.

[weblacky]

Yeah, on paper it’s faster and quieter than the original OEM drive! It’s not that much quieter, just 1DB. But of course DB is logarithmic so it’s still quieter by a noticeable amount.

The plan was more replacing worn out old fans. Not so much finding me a whisper quiet replacement. More being able to do so and I’m not having to worry about underperformance.

I have to tear apart one or more supplies and gather a build list of materials in order to put an order so it will take me some time. I just wanted to put this out there in case someone’s eyes were sharper than mine on specifications or some thing that would make this a no-go candidate.

Also, the truth is theat NMB still makes the original fan within specifications. Of course, it’s gone from being Japanese to Chinese, and the quality may have slipped or not be the same so either way there’s going to be a change.

[johnnym]

Yeah, on paper it’s faster and quieter than the original OEM drive!  It’s not that much quieter, just 1DB. But of course DB is logarithmic so it’s still quieter by a noticeable amount.

I didn't mean the noise produced by the fan itself, but the noise possibly produced from resonance in the plastic case of the O2 due to the higher RPMs of the fan.

E.g the Octane I use to make the compilatons for OpenBSD/sgi tends to produce an annoying humming sound due to the operation of the two 15K disks I use(d) for source code storage, object files and distribution files.