[bldg-sim] Make-up air systems and heat recovery

Jeff Hirsch Jeff.Hirsch at DOE2.com
Fri Aug 26 13:10:26 PDT 2005

This discussion is relative to current versions of DOE-2.1E and
DOE-2.2 that we know of and/or currently distribute.  Older
out-of-date versions may not have all features discussed here.

No version of DOE-2 that we distribute, or know about, has
heat recovery equipment "built it" as an option for fan coil systems;
the 2.1E RECOV-MODE and 2.2 ERV keywords are "unused" for
those system types.

In 2.2 it is possible, using expressions, to override this and force the
FC system to use the ERV model; this is not recommended to do since
the ERV keywords and variables are "system" level variables and
the FC system is a "zonal" system; for a single zone per FC system this
may work (I'm not sure as I have not tried) but you would need to
carefully examine the source (in the DOE-2.2 release .zip archive) and
hourly output to ensure it is working correctly.

In 2.1E you can also override this unused set of keywords with a set
of "FUNCTION" commands, however, you would need to implement the
heat recovery algorithms into the fancoil subroutine (FCOIL) at FUNCTION
point FCOIL-1Z as fan coil systems do not allow economizers (unlike 2.2,
which allows economizers, DCV and cycling FC fans with zone load)
thus that routine does no call the ECONO routine (where the recovery
keywords are used) when establishing the supply air temperature.

But the question was really on recovery for systems that are set up as
OA supply systems so that defined FC systems can reference them using
the OA-FROM keyword.

In general, no version of DOE-2 that we have or know of can correctly
model OA supply systems with energy recovery systems.  This is due to
the lack of iteration between the OA supply system and the systems that
are supplied the treated OA.  The return air conditions for the OA supply
system do not reflect the return air conditions from the systems that are
supplied the OA; thus the recovery system acts on the return air conditions
of the OA supply system without including the effects of the other AHU's
and their zones.  If the OA supply system only has a small fraction of its
air going to the other AHU's then the results may be close (but not correct);
but usually (and what the DOE-2 OA-FROM keyword is commonly used
to model) the OA system supplies 100% of its air to the other AHU's and
the calculated return conditions for the OA supply system (on which the
recovery algorithms act) is the OA system supply conditions not the common
return from all the supplied AHU's ... the resultant recovery calculation
and thus OA system load are incorrect (thus the effect of the recovery system
will not be correctly reflected into the results.)

The OA-FROM DOE-2 keyword can do a good job of modeling an OA treatment
system when that system is 100% OA, the treated OA air is supplied at constant
volume flow directly to other AHU's, and the OA system is controlled to supply
OA at a constant, scheduled (OA temp reset or not) or floating (based on the OA
system capacity being low or the OA into the unit being within the controller
setpoint range) temperature to the AHU's.

If the OA system is VAV,  OA system or AHU's have movable OA dampers, the OA
system has zone controlled supply temperature, or either the OA system or AHU's
cycle on independent controllers, then the DOE-2 simulation is not reliable ... it can be
done, but very careful input and examination of output is required to get a result
and determine if it is correct.

In 2.1E there is no correct sizing of OA supply system or error checking on any
incorrect input for these systems; 2.2 has sizing of these systems and a lot of
error checking ... please see the latest Vol 6 Manual (new features) for more

2.1E assumes the supplied AHU systems OA fraction is to equal the OA-FROM
systems OA amount; thus making is not usually possible to supply more than one
AHU with an OA supply system.  2.2 has not such limit; it can correctly model
an OA supply system that supplies any number of other AHUs with each AHU
OA flow being able to be set independently (via input or default OA requirements
specified for that AHU)

Both 2.1E and 2.2 require the OA system to be defined in the BDL input file prior to
any AHU it supplies; 2.2 issues an error if this is not found to be true while
2.1E issues no errors on any incorrect OA-FROM input, including many
types of errors that produce completely incorrect results ... it just performs an
incorrect simulation.  2.2 has many error messages about many aspects of OA
supply systems, and does not allow a completely invalid simulation to be performed.
Jeff Hirsch
James J. Hirsch & Associates
Building Performance Analysis Software & Consulting
12185 Presilla Road
Camarillo, CA 93012-9243 USA
phone: (805) 553-9000
fax: (805) 532-2401
email: Jeff.Hirsch at DOE2.com
web: http://DOE2.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 

More information about the Bldg-sim mailing list