--- Begin Message ---
It sounds like you have two different issues here: the "small timestep"
warning from the pollutant transport solver, and the range error. They
may be completely unrelated, or the error may proceed from the warning.
To understand the "small timestep" warning, you need to keep in mind
that the pollutant solver actually uses two types of time steps. The
first, "outer" time step is the one that most users think about.
That's the time step between scheduled events, and that shows up in the
output file. For example, suppose your simulation has its first
scheduled event ten minutes into the simulation. Then you would get
pollutant output at t=10 minutes.
But the pollutant transport solver can't necessarily jump right from
t=0 to t=10. Instead, it has to take smaller, "inner" timesteps, to
work its way up to that first scheduled event.
That "small timestep" error comes when the pollutant solver tries to
determine an appropriate length for the inner timestep.
There is little you can do to affect the inner timestep calculation,
unless you are willing to change the airflow system. Here's why. To
estimate an appropriate inner timestep, the pollutant solver runs
through every zone, calculating its time constant as the zone volume
divided by the sum of the mass flows into the zone (actually, I think
close inspection of the code would reveal it divides the zone volume by
the sum of the absolute values of all flows both in and out of the
zone, but no matter). Increasing the flows through a zone (or
decreasing its volume) decreases its time constant, making the zone
concentration change more rapidly with time. Therefore the zone with
the smallest time constant determines how long COMIS can make that
inner timestep. Thus, the pollutant solver calculates its inner
timestep as a fraction of the worst (the smallest) of these zone time
constants.
However, the pollutant solver won't reduce its inner timestep below 1
second. In other words, even if you have an extremely small zone with
very high flows (say because you're trying to represent a duct junction
as a zone), that ten-minute outer timestep will never get made up of
any more than 600 inner steps of 1 second each.
If you really want to locate the cause of that 1-second warning, look
for small zones with large flows. Since a one-second time step should
be somewhat reasonable for most applications, the program may be trying
to warn you of a modeling error (your model doesn't represent the
physical situation you think it does).
I suspect the second, "range" error relates to a pollutant
concentration that exceeds some level that COMIS considers an upper
limit (say, 0.5 or something like that). However, that error may come
from some other part of the program, including TRNSYS. The rest of my
comments assume that the range error comes from the COMIS pollutant
solver.
That range error might relate to the small timestep warning in the
following way. Suppose the pollutant solver really needs an extremely
small inner timestep, say, of 1e-3 seconds (I'm just making that number
up for illustration), in order to control the errors in the solution.
Then taking relatively long one-second steps could allow the pollutant
solution to "blow up" and exceed the range of concentrations that COMIS
considers meaningful for its calculations.
However, that range error might not relate to the small timestep
warning. It may be that your problem defines such a large release,
relative to the zone volumes, that an exact solution would still find
the pollutant going out of what COMIS considers its safe range.
-Dave
David Lorenzetti
Lawrence Berkeley National Laboratory
> I am trying to model CO2-controlled ventilation with TRNFLOW
> (TRNSYS15 with comis3.1). But I have a problem modelling pollutant
> flow. I have a simple one zone-building with inhabitants producing
> CO2. I use this concentration to control a fan. The problem is that
> when I run the simulation, I get the following warning at every
> timestep.
>
> ***** WARNING ***** UNIT 37 TYPE 56, TIME = 4.188E+00
> COMIS MESSAGE ***** WARNING *****
> Pollutant transport solver : Small timestep
> needed (Tau < 10 sec), but is limited to 1 sec
>
> But, when I decrease the time step below 56.25 sec (0.5^6*1hr), I get
> a range check error during the simulation (the first couple of hours
> are plotted before the error). Does anyone have experience with this?
> What causes the error? And is such a small time step always needed
> when simulating pollutants or does it depend on e.g. the zone volume?
------------------------ Yahoo! Groups Sponsor --------------------~-->
1.2 million kids a year are victims of human trafficking. Stop slavery.
http://us.click.yahoo.com/.QUssC/izNLAA/TtwFAA/NhFolB/TM
--------------------------------------------------------------------~->
To Post a message, send it to: comis@yahoogroups.com
To Unsubscribe, send a blank message to: comis-unsubscribe@yahoogroups.com
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/comis/
<*> To unsubscribe from this group, send an email to:
comis-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
__________ NOD32 1.1330 (20051220) Informatie __________
Dit bericht is gecontroleerd door het NOD32 Antivirus Systeem.
http://www.nod32.nl
--- End Message ---
--- Begin Message ---
Dear Stefan
This message concerns the internal time step of the pollutant transport
model and has nothing to do with the time step of the thermal model. The
used time step of the pollutant modell is 1/10 of the smallest room time
constant, which is defined by the zone volume devided by the total volume
flow of this zone. It is limited in the code to a minumum of 1 sec. If you
have small zone volumes with big volume flows (e.g. in a HVAC network.) then
the smallest time constant may be below 10 s and this message warns you
that transient concentration values in those zones might be not correct. If
the time step of your thermal model (TRNSYS time step) is much bigger than
the discribed time constant then the transient concentrations will not play
any role in these zones as the steady state concentrations will be reached
in these zones at every TRNSYS time step.
with kind regards,
Andreas
Empa
Swiss Federal Laboratories for Materials Testing and Research
Energy Systems and Building Equipment Laboratory
Überlandstrasse 129
CH-8600 Dübendorf
Tel. ++41 ( 0)44 823 47 22
FAX ++41 (0)44 823 4009
www.empa.ch
>>> s.dekool@chri.nl 20.12.2005 10:29 >>>
Hello,
I am trying to model CO2-controlled ventilation with TRNFLOW
(TRNSYS15 with comis3.1). But I have a problem modelling pollutant
flow. I have a simple one zone-building with inhabitants producing
CO2. I use this concentration to control a fan. The problem is that
when I run the simulation, I get the following warning at every
timestep.
***** WARNING ***** UNIT 37 TYPE 56, TIME = 4.188E+00
COMIS MESSAGE ***** WARNING *****
Pollutant transport solver : Small timestep
needed (Tau < 10 sec), but is limited to 1 sec
But, when I decrease the time step below 56.25 sec (0.5^6*1hr), I get
a range check error during the simulation (the first couple of hours
are plotted before the error). Does anyone have experience with this?
What causes the error? And is such a small time step always needed
when simulating pollutants or does it depend on e.g. the zone volume?
Thanks a lot,
Stefan de Kool
To Post a message, send it to: comis@yahoogroups.com
To Unsubscribe, send a blank message to: comis-unsubscribe@yahoogroups.com
Yahoo! Groups Links
------------------------ Yahoo! Groups Sponsor --------------------~-->
Get Bzzzy! (real tools to help you find a job). Welcome to the Sweet Life.
http://us.click.yahoo.com/KIlPFB/vlQLAA/TtwFAA/NhFolB/TM
--------------------------------------------------------------------~->
To Post a message, send it to: comis@yahoogroups.com
To Unsubscribe, send a blank message to: comis-unsubscribe@yahoogroups.com
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/comis/
<*> To unsubscribe from this group, send an email to:
comis-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
__________ NOD32 1.1330 (20051220) Informatie __________
Dit bericht is gecontroleerd door het NOD32 Antivirus Systeem.
http://www.nod32.nl
--- End Message ---