6

Our server room is a converted closet, and we have a pair of wall-mounted air conditioning units to cool the room. These are both pointed towards the front of the server racks, and there are a couple small vents for hot air. We definitely don't have an ideal environment, but it's what I'm working with.

My question is related to an ionization feature of the A/C units. I know that this feature will basically spit out negative ions that attach to dust and other particles, and make them heavy enough to drop to the ground. With this basic understanding, I am trying to figure out if it is better for a server room to have this turned on or off.

I would imagine turning it on would mean purer air, which would mean the intake of air to the servers would have less dust (ie, good to turn it on).

However, I could also imagine that the ionized air would contain bigger, heavier dust particles that would normally drop to the ground, but could get sucked up into the servers because of the air flow. This would lead to bigger dust particles in the servers, which would be more likely to cause issues (ie, good to turn it off).

I haven't found much looking around online, so I'm hoping that someone out there might have done some deeper research into this already. Or, hopefully someone can factually show me that I'm considering something that is so minor it won't make a difference.

voretaq7
  • 79,345
  • 17
  • 128
  • 213
Paul Kroon
  • 2,220
  • 16
  • 20
  • 1
    As an aside have you considered orienting your chillers to cool the hotter exited air rather than feed cooler air into the system? modern systems actually prefer 'ambient' temperatures to cooler ones and super-modern data centers do away with chillers altogether, simply force-extracting the hotter air. Just a thought. – Chopper3 Aug 25 '11 at 17:35
  • The way the room is set up, we wouldn't be able to move these wall mounted units to the back, although I would agree this would be more efficient for us. On a similar note, though, we know our biggest cooling problem is with venting the hot air, but we have not been allowed to do this more efficiently in any way just yet. – Paul Kroon Aug 25 '11 at 17:46
  • @Paul - Re: venting, sometimes it helps to make the 5-year business case (venting your hot air and running (filtered) ambient office air through the servers can be a big savings in cooling power - especially if your office is somewhere like mine where electricity costs per watt what gold costs per ounce. – voretaq7 Aug 25 '11 at 17:55
  • @voretaq - I definitely agree. They gave us a second A/C unit rather than vent air, no matter how much we tried to make the case of the problem being the hot air. By the way, I know this is more of a meta question, but should this be tagged "cooling" when my question is about the ionization/dust, and not the cooling? – Paul Kroon Aug 25 '11 at 17:59
  • I added cooling because we don't really have an ionization tag (and if we did this would probably be the only question in it), but we do have a bunch of guys with DC cooling knowledge that exceeds mine - hopefully they watch [tag:cooling] and will come in & contribute :) – voretaq7 Aug 25 '11 at 18:07

1 Answers1

11

The ionization option is a mixed bag: It will lead to "cleaner air" but there are definitely drawbacks:

First one that comes to mind is that your server room is a largely-contained environment, so any ozone generated by the ionizer will stay in the room. Being a largely-contained environment (with filters in the AC units) also means the air will be somewhat cleaner by nature.

The second is that ionizing the particles in the air will get them to stick to anything with an opposite charge. This includes the ionizer's collector plate, and hopefully the AC filters, but may also include your servers, heat sinks, etc -- Anything that has a slight positive potential will attract the particles. I've no practical experience with ionizers in datacenter AC units (not even sure they have 'em?) , but I would be concerned about the cooling ports of servers getting obstructed over (a long) time.

voretaq7
  • 79,345
  • 17
  • 128
  • 213
  • So you would tend to agree with the idea that the ionization would do more harm than good? – Paul Kroon Aug 25 '11 at 17:47
  • that's what my gut tells me -- I've never seen it on datacenter HVAC equipment so I have no real-world experience with it, but in a sealed room filters should handle most of the dust. You will still need to (damp)mop occasionally but an ionizer wouldn't really prevent that... – voretaq7 Aug 25 '11 at 17:53
  • The fact that you don't see ionizers on datacenter HVACs is telling. I think the general consensus is that ionizers and datacenters aren't a good fit. If you're concerned about dust, add filters (at one point, our server room had furnace filters mounted over the AC to reduce dust). – Klox Aug 25 '11 at 21:19
  • 1
    Also, ionizers in a confined area lead to more rapid chemical breakdown of things made out of rubber and similar artificial polymers. (Ozone is the natural enemy of many polymers.) – David Schwartz Aug 25 '11 at 22:54
  • @Klox - I've never seen a room *without* filters on the AC (usually on the intake to protect the evaporator coils from getting dusty and nasty) – voretaq7 Aug 26 '11 at 14:55
  • @DavidSchwartz - Good point Re: ozone damage to polymers/rubbers - There aren't many rubber items I can think of in your typical datacenter, but I'm sure there are at least a few. – voretaq7 Aug 26 '11 at 14:56
  • @voretaq7 - These were in addition to the built-in AC filters; furnace filters from the local hardware store screwed right into the front face of the AC unit :P – Klox Aug 26 '11 at 15:34
  • @Klox - *that* might be getting a bit obsessive :-) Gotta be cheaper to change out than the framed AC filters though – voretaq7 Aug 26 '11 at 16:05
  • I'll mark this one as the answer and go with it, but I'd still love to get more info from anyone else who might know something about this. This is the direction I was leaning towards, so I'll be leaving this option turned off unless I learn otherwise. – Paul Kroon Sep 01 '11 at 16:22