© 2018 by Design Management Group

570-299-7520

info@dmgeng.com 

2007 PA-315 / Suite 202 / Pittston, Pa 18640

1 East Broad St. / Suite 320 / Bethlehem, Pa 18018

mo

Please reload

Recent Posts

Taking the fear out of Halloween for Facility Managers

October 20, 2017

1/1
Please reload

Featured Posts

HVAC that Predicts the Future

March 3, 2017

We do quite a lot of data center work, and one topic that comes up now again is using instantaneous IT power usage to control cooling capacities.  In data centers this is pretty straightforward as the load is fairly consistent and primarily based on one variable (IT infrastructure power). Further simplifying this, the equipment is normally monitored already to track infrastructure utilization.

 

At first this might seem over the top, but think about how the energy is converted from electrical to thermal in a data center process.  It goes something like this:

  1. Processing demand draws power

  2. Power drawn to servers is converted almost 100% to heat (no mechanical work being done and minimal IT side energy being transferred to data users via technology infrastructure)

  3. Server components heat up

  4. Server cooling air heats up (or room air if whole room cooling without containment is used)

  5. Thermal management increases mechanical cooling capacity to reduce temperature

  6. Equipment cooling water or refrigerant absorbs heat

  7. Load is transferred to the mechanical components

  8. Mechanical components respond

OK, so admittedly some of these steps take milliseconds...

 

...but depending on the type of thermal management system you have, it could take seconds or minutes to fully respond and drop temperatures.  Essentially, if you have a large amount of thermal inertia, such as a centralized chilled water system, the time between the instantaneous IT load and refrigeration system response grows.

 

So what's the point?  The point is that HVAC systems in general are reactive in nature, but they don't have to be.

 

By preemptively controlling cooling capacity based on IT load monitoring, small optimizations can be made over periods of time. This means reducing the amount of time you're playing catch-up and allowing the equipment to run steadier and more efficiently.  Anytime a thermal control system needs to increase or decrease temperature in a short period of time, the power draw of that system increases.  So, the longer you have to modulate your capacity, the more efficiently you can do so! You're predicting the future temperature of the medium being controlled by monitoring the source of the heat. Pretty cool..... literally.

 

Now here's where you may think I'm reaching....

 

I believe that this can be leveraged in some basic comfort cooling applications. The basic principles are the same, and technology to monitor the number of people in the building, solar loading, ventilation loading, internal loading are all coming down in cost and size.  The fact of the matter is that buildings are becoming smarter everyday, so why not use that to your advantage and save cash while minimizing occupant complaints?

 

If this were to become the norm, enormous amounts of data could be collected and binned by building type, location, etc. and empirical data could be used to further optimize equipment sizing and reduce initial cost of new installations.  Furthermore, if engineers had access to that empirical data, human error and excessive safety factors could be avoided in the design phase of projects. This should (if used properly) create more consistency in facility performance.

 

To apply rightsizing concepts to your projects, please contact us to find out how we can help you achieve...

 

Thermal Management, Smarter...

Share on Facebook
Share on Twitter
Please reload

Follow Us
Please reload

Search By Tags