CPI Passive Cooling™ Solutions Dispel Common Heat Management Theories | Chatsworth Products

CPI Passive Cooling™ Solutions Dispel Common Heat Management Theories

(Thermal Management) Permanent link

TeraFrame Cabinet

It is no longer news to anyone that the effects of Moore’s Law has created computer equipment heat loads that data centers are ill-equipped to effectively manage. Emerging conventional wisdom is beginning to settle on a response to this crisis that either involves dispersing the offending servers to reduce the heat densities or utilizing supplemental liquid cooling to remove the heat.

Chatsworth Products, Inc. (CPI) Passive Cooling™ Solutions is the best choice because it: 

  • reduces energy consumption 
  • lowers construction and operational costs 
  • meets Tier IV operating requirements

CPI Passive Cooling Solutions focus on controlling airflow by isolating and removing heat from the data center allowing you to maximize the cooling capacity in your space.  Unfortunately passive air cooling is often excluded because of several misconceptions:

Myth 1: There is a limit to passive air cooling capacities that are well below today’s potential heat load densities.

Truth:  The relationship between volume of air, heat load and temperature rise is often misapplied by considering the total temperature rise through the rack. The perceived limit to air cooling capability is based on how much air can be delivered out of a perforated access floor tile. CFD models show that chilled air is consumed by the bottom half of the cabinet, creating hot spots at the top of the cabinet. CPI Passive Cooling Solutions eliminates the dependency on chilled air from a perforated floor tile, as well as eliminates hot spots created by re-circulated air.

Myth 2: High density passive air cooling systems create unmanageable high return air temperatures.

Truth: Chilled water computer room air conditioners (CRAC) increase cooling capacity with higher return air temperatures. In instances when the return air temperature exceeds the cooling properties of economizer air, then a small amount of bypass air will suffice. Since the ideal embodiment of the passive cooling solution includes isolated return air in a ceiling plenum, bypass air may be delivered through an open ceiling grate, helping to regulate pressure in the room.

Myth 3: Lower acquisition costs for passive air solutions are overshadowed by higher operating costs than liquid cooling.

Truth: With the improved efficiency of a complete isolation between supply air and return air, close-coupled systems lose their operating cost advantage and the total cost of ownership clearly favors the well-engineered passive cooling solution.

Please refer to the complete white paper for detailed explanations behind the products and solutions of CPI Passive Cooling. For additional documentation visit our CPI Passive Cooling page.Kim Ream, eMarketing Designer

Posted by Kim Ream, eMarketing Designer at 11/07/2016 02:04:53 PM


Any concerns with Blade Chassis stacked 4 high in a cabinet with their fans causing air flow concerns with the passive cooling concept
Posted by: Steve Otis at 9/19/2008 2:20 PM

We have had third party CFD testing conducted by Flomerics with four HP c7000 blade centers in one of our TeraFrame cabinets configured with the complete Passive Cooling Solution. The blade center’s fan systems were profiled in zero static pressure – i.e., on the workbench and then those readings were compared to the server performance in our cabinet. The testing substantiated our claims regarding the creation of low pressure in the rear of the cabinet at extremely high heat loads. Due to this measured ∆P between the front of the cabinet and the rear of the cabinet, the server fans actually performed better inside the cabinet than they did on the workbench. - Ian Seaton, CPI Technology Marketing Manager
Posted by: Ian at 10/2/2008 6:19 AM

With all this talk about data center cool, obviously where most of the power consumption is, what is being considered for smaller telecom rooms? IE, 1-4 Cisco 3750's in one TR. When you have 100 rooms of this size on a campus, power is still a concern. What mitigating factor, aside from turning down the A/C, are there available? Is any one looking into it?
Posted by: jtodd5dot1 at 1/12/2009 4:53 PM

Unfortunately, these small TR’s are not getting anywhere near the attention that the big multi-megawatt data centers are, as you probably suspected, by your question. The basic principles for efficient equipment cooling are still the same, however, whether we are talking about hundreds of cabinets and megawatts of power or if we are only talking about a single rack. You do what you can to keep your supply air and return air as isolated from each other as possible. For example, my wife has a “data center” in her small business that fits in one cabinet, which sits in a small multi-purpose office. The rear of the cabinet sits just in front of a ceiling return air grill, so most of her exhaust heat is captured in that return air path before it does much mixing with the rest of the chilled air in the room. So any location strategy like that, in combination with the ubiquitous blanking panels in all unused rack mount spaces will help with that separation and therefore reduce waste and keep a nice temperature rise going back to the AC. Also, the better you can do with that separation, the more room you have to work with to raise your AC set points, because you are not feeding that exhaust air into your switches. Remember also to check your switches’ environmental specifications – most switches will have higher inlet temperature specifications than most servers. Finally, in spaces where no conditioned air is available, you might want to look at the non-condensing self-contained AC units from Kooltronic or Stulz that can be mounted on the side or top of a cabinet. Again, to maximize the efficiency of these units, you’ll want to be sure you’ve got structures inside the cabinet that help to separate the supply path from the return path.
Posted by: Ian Seaton at 1/16/2009 6:06 AM

what about using a passive cabinet with no defined return air plenum? i.e. just using the duct to get the heat up. Any value in that?
Posted by: Bob at 3/18/2009 1:52 PM

Leave a comment
Name (required)
Email (optional, not published)
Your URL (optional)


Note: Conversation is encouraged and expected. However, moderation of comments is necessary to prevent spam, per- sonal attacks, profanity, or off-topic commentary. Comments related to specific product support or customer service issues will be addressed separately rather than posted here. Please email techsupport@chatsworth.com for assistance with these matters.

4/27/2017 3:50:34 AM