Computer server room noise problems

Posted: Thu Mar 08, 2007 9:49 am

Here is an acoustic problem that is not directly related to the focus of this board, but I hope some of you can either give advice or point me to where I can get some.

The problem is my department's computer server room. We (Virginia Tech math department) are currently planning to add a lot of server capacity and knock out some walls to increase the size of our server room. The servers that the IT people are planning to buy are rather loud. I'll get SPL readings on the current volume level soon, but there is already pretty significant transmission to adjacent offices and the internal SPL level is annoying.

The IT guy consulted me and said his goals were to
(a) reduce sound transmission levels, and
(b) reduce internal SPL levels.

I told him that (b) was a big problem - the sound has to go somewhere, so we focused more on problem (a). (Am I correct in this? Any ideas on reducing noise levels in the room other than buying quieter servers?)

Concerning isolation I showed him some of the chapters in Rod's book, and of course he kind of balked at the time, space, and expense (in about that order) of double wall construction. So I'm trying to investigate some "quick fixes" that could give some STC improvements without too much trouble. I'm considering
(1) Buying better doors. Rod mentions Overly. Any other suggestions? I figure we'd need to do this even if we broke down and did the wall construction.
(2) Adding mass to the existing wall structure. What would be the relative benefits of adding another layer of drywall and/or lining the existing walls with 703 or 705 (covering with fireproof fabric of course)? And how effective would that be relative to double wall construction?

If there are any sites that address this specific problem that I have missed, I'd appreciate a pointer. Thanks, Bob

Posted: Thu Mar 08, 2007 12:00 pm

Bob,

Sorry, but there are no quick fixes, nor are there simple products you can buy. Sound isolation requires construction. If you already have Rod's book, you already have everything you need.

[editorial mode]
If they're dropping a hundred grand or whatever on a bunch of new computers, and tearing down walls anyway, but are unwilling to spend the extra few percent to make double walls, they seriously need to reassess their priorities. I suggest you ask them if they're willing to move their own offices into the adjacent rooms. Smile

[/editorial mode]

--Ethan

Posted: Thu Mar 08, 2007 7:06 pm

Sorry Ethan. "Fixes" was a bad word choice on my part. I've shown him what he needs to do it for a true fix. But it may not be feasible. The IT guys are the ones with the adjacent offices, so it's mostly their problem. However, the current problem isn't all that bad, and the doors are obviously a big source of leakage. I'm in an office across the hall and in my office the computer noise is much lower than the HVAC system. I'll take some measurements tomorrow and see what kind of attenuation we really need.

BTW, It is a state run institution after all, so having priorities out of order is not exactly unusual (always easy to get money for computers, very hard to get it for construction.)

Posted: Thu Mar 08, 2007 8:52 pm

call me a tool, but what is "Rod's Book"? I have a feeling I NEED it
Rolling Eyes

Posted: Thu Mar 08, 2007 9:58 pm

Don't answer that, I found it and am paying my MasterCard bill and buying it (in that order).

Posted: Fri Mar 09, 2007 4:03 am

Bob,

You're in a professional environment, which can have very specific problems.
Sound can pass through ventilation ducts. Sound can pass over suspended acoustic ceilings, computer floors, ...... Frequency is important, etc.....

Invite a pro from an acoustic company. In 15 minutes he notices what you need days and days to grasp.

It's possibly free, or you must have small budgets for maintenance, undefineds, whatever, which doesn't be to be planned years in advance or approved by 10 other departments.

Just use the yellow pages under acoustics, insulation, sound insulation, whatever...

Warm regards
Eric

Posted: Fri Mar 09, 2007 6:14 am

eric_desart wrote:

....You're in a professional environment...

You have a lot to learn about American universities Rolling Eyes

Posted: Fri Mar 09, 2007 6:41 am

Bob,

I didn't meant anything bad.
I indeed don't know American Universities, but know some here.

And I know that companies like to help Universities (within reason of course), since being consulted by them is as a quality label. And I think that's universal.
I assume being a University, you have a maintenance department, used to consult outside people?

My simple advice is based on experience of doing 30 years acoustics mainly in noise control.

Suppose you find the answers yourself. This assumes you know the questions and understand all contributing sound paths. Does it then become easier to solve that problem? I assume you're not going to DIY this solution?
Anyhow, Ethan is correct in saying that this can be a complicated and extensive thing.

Best regards
Eric

Posted: Fri Mar 09, 2007 7:27 am

Eric- I didn't take offense. I was just joking around. You have to have a sense of humor to work at this place.

I'm going to take my spl meter in today, measure the current levels, lend them Rod's book, and then leave the problem to them. I definitely will recommend that they get a pro in, but I'll doubt they'll take my advice. I think they asked me to help because they thought it was a matter of "lining the wall with acoustical tiles."

Posted: Tue Apr 03, 2007 10:42 am

Just came across this post. You might want to contact people within your own university to look at the problem. Some eager students may welcome the challenge...

Posted: Tue Apr 10, 2007 10:57 am

Jeff,

Congratulations with your Sherlock Holmes qualities.

But you're 100% right of course. A lot of knowledge is readily available inside this Univ.

Warm regards
Eric

This topic sponsored by: