[BLDG-SIM] atrium natural ventilation

[Jeremy Cockroft]

Nick,
I assume you are modelling a nodal airflow network in TAS, i.e. each
horizontal slice of your atrium is represented by a single node.
That means that buoyancy effects will only be effective on the bulk air
flow through the atrium (upwards if atrium nodes are at a higher air
temperature than external).  That means that at each level (slice), if
you have a net heat loss through the atrium walls, the air will cool as
it rises, which is what you are reporting. To see more complex air flow
effects you would need to include at least two nodes per level, which
would allow air exchange due to localised horizonal air exchanges (not
sure if TAS has this capability; ESP-r does, maybe others), and also the
possibility of an air circulation within the atrium (cooler air
travelling down the external wall, and warmed air up the other side, for
example).  You would have to arrange these nodes intelligently, as you
are making an assumption about how air circulates within the atrium.
ESP-r has a CFD domain for simple (cuboid) spaces, integrated within the
thermal (nodal) nodelling domain, so you could study timewise (dynamic)
evolution of airflows and temperatures.  I'm not aware of other tools
with this capability.
Hope that helps.
Jeremy Cockroft
ESRU
University of Strathclyde
75 Montrose Street
Glasgow G1 1XJ
phone    +44(0)141 548 5765
email       jeremy at esru.strath.ac.uk
website   www.esru.strath.ac.uk

-----Original Message-----
From: BLDG-SIM at gard.com [mailto:BLDG-SIM at gard.com] On Behalf Of Nick
Doylend
Sent: 18 October 2007 10:28
To: BLDG-SIM at gard.com
Cc: mguin at foremanroberts.com
Subject: [BLDG-SIM] atrium natural ventilation

Hi, I wonder if anyone can give some advice on this one.  My colleagues
and I are considering a naturally ventilated atrium in the UK.  We're
trying to establish appropriate opening areas at top and bottom to limit
summertime overheating conditions.  At the moment we are using Tas for
this analysis but now suspect its bulk air movement calculations are not
sophisticated enough to model the stratification in the space.

The atrium is occupied at the ground floor and extends up four stories,
it has been zoned into four horizontal slices.  Rather than showing an
increase with height of air temperature we see the opposite - the lower
zone is warmer than the upper zones.  Maybe the ground floor zone has a
higher sensible gain due to the occupants but in a high vertical space,
shouldn't the heat rise to the top of the space?  I wonder if Tas is not
considering the convection across the horizontal zone boundaries or is
overestimating the downward air circulation?

Does anyone have experience of using Tas or alternative techniques for
this kind of analysis?  We would prefer to run a dynamic simulation that
can predict annual overheating, rather than a detailed CFD snapshot in
time.

Any suggestions?

Thanks,

Nick Doylend

PS It would also be interesting to get a show of hands from the Tas
users on this forum - eQUEST seems to be the most popular simulation
tool here judging by the volume of posts.

======================================================
You received this e-mail because you are subscribed to the
BLDG-SIM at GARD.COM mailing list.  To unsubscribe from this mailing list
send a blank message to BLDG-SIM-UNSUBSCRIBE at GARD.COM


===========================
You received this e-mail because you are subscribed 
to the BLDG-SIM at GARD.COM mailing list.  To unsubscribe 
from this mailing list send a blank message to 
BLDG-SIM-UNSUBSCRIBE at GARD.COM