Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations pierreick on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

19 inch Rack Boxes

Status
Not open for further replies.

drawoh

Mechanical
Oct 1, 2002
8,947
I am working with a customer's 19" rack mounted box, and I am thinking about writing a design procedure. There is just so much stuff to do wrong.

Question: In what direction should cooling air flow through a 19" rack? I prefer in the front and out the rear. What do you all think out there?

--
JHG
 
Replies continue below

Recommended for you

In a rack I had up to 10kW dissipation and the approach was to keep the hot side hot and the cold side cold so blocker pieces were added to the installation to have all the intakes from a rear plenum as most of the instruments were rear intake, side exhaust. It worked pretty well not having the lower instruments pre-heating air for the upper ones.

 
3DDave,

Side exhaust only works if the 19[ ]rack is open on the sides. There also must be a way for the hot air exiting the sides to completely exit the rack. Even on commercial server racks, this is not a good assumption.

Let's assume we are designing boxes to go into a customer's rack, and that you no control over this.

--
JHG
 
The rack was fed from a plenum in the back and exhausted out the top. Almost all the rackable instruments were rear intake, side exhaust. We weren't able to tell Agilent or NI to redesign their equipment. Our rack, like most every equipment rack I've see, was buttoned up on both sides. Most computer centers seem to have caught on to the same idea and are careful to route air via ducting to and through the racks. No longer the days of a 60F room to keep the mainframes cold.

I suppose in your case, follow what competitors and adjacent suppliers are doing and include an air intake temp sensor to shut down if the screw up the installation and try to run 100+F air into your module.
 
Just to clarify - the 10kW was the heat expended inside the cabinet as part of testing full load outputs of smaller transmitters and motor drivers. We never got a full list from the customer of what testing they did because the tests and test adapters were contracted literally all over and had built up over decades and procurement cycles. The customer did not have that list.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor