Confusion over T_BEGIN in the users manual.

143 views
Skip to first unread message

louiea

unread,
Feb 20, 2008, 9:58:36 AM2/20/08
to FDS and Smokeview Discussions
The FDS 5_User_Guide (version 5, SVN:1161) Section 6.2 Page 23. the
last paragraph reads:

"If you want the timeline to start at a number other than zero, you
can ues the parameter T_BEGIN (Time
Begin) to specify the time written to file for the first timestep.
This would be useful for matching timelines
of experimental data or or video recordings."
(yes, "use" is spelled incorrectly in the manual)

Then in the box right under it says:

"This does not run any of the simulation prior to the T_BEGIN value.
It is only used to offset the reported
starting time from zero."

As I understand it, If you had a video recording of a fire that
started at 300 seconds, and you made a simulation of the same fire
that started at t=0, you can set T_BEGIN=300, to make the timelines
match up for easy comparaison.

but the box says that none of the simulation before T_BEGIN is run, so
does that mean if your simulation was set to be 600 seconds long
(T_END=600.) that you will only get 300 seconds on simulation? (From
T_BEGIN=300 to T_END=600) or does it do the full simulation but dosn't
output any results until T=T_BEGIN. Do Time dependent devices activate
based on the actual simulation time, or the T_BEGIN offset?

I guess I am just really confused on what is meant by "does not run
any of the simulation prior to T_BEGIN." and whether or not the entire
time line offsets, so the timeline_END would end up really end up
being T_END + T_BEGIN.

thanks

dr_jfloyd

unread,
Feb 20, 2008, 10:24:42 AM2/20/08
to FDS and Smokeview Discussions
Yikes. We need to clarify that portion of the manual and I think
modify the RAMP inputs.

Internally FDS has time start at 0. So if you input T_BEGIN and
T_END, FDS gets the runtime from T_END - T_BEGIN and when writing
outputs merely adds T_BEGIN to any time quantity. Therefore, the DEVC
output for TIME is corrected for T_BEGIN, meaning that a SETPOINT
would be based on time starting at T_BEGIN rather than 0.

The not running anytime before was intended to address a case where
you have a heat release RAMP from 0 to 1000 s. If you set T_BEGIN to
400 s, FDS will not know anything about the heat release up to 400 s,
i.e. at 400 s everything will start at its default conditions.

RAMP inputs are also offset by T_BEGIN. Though it has now occurred to
me that we don't always want that to happen. Some RAMPs we want to be
relative (like specifying an HRR for a surface with an ignition
temperature followed by a predefined HRR). Look for a keyword in a
future maintenance release to address this.

Tom

unread,
Feb 20, 2008, 2:29:52 PM2/20/08
to FDS and Smokeview Discussions
That was very interesting -- I had a question adhering to relative
times when using RAMP that I was just going to post!

Am I right in assuming that the addition of this new keyword will make
it possible to ramp vents after activating them (by a device or other
critera) some initially unknown time T into the model, thus making the
specified times in RAMP relative, just like you say?

Just to clarify, I would like to have a smoke detector (or whatever
device) activate a vent, but then RAMP up the vent. Seeing as the RAMP
starts at a user specified time, and the vent will only be activated
after whatever (unknown) time the smoke detector activates, I couldn't
quite get it to work...

Cheers

dr_jfloyd

unread,
Feb 20, 2008, 2:39:38 PM2/20/08
to FDS and Smokeview Discussions
Currently in FDS all time RAMPs are adjust based on T_BEGIN (which
defaults to zero). When a VENT is opened, a wall cell ignites, etc.
that RAMP is interpolated using the relative time from when cell
became active. So if T_BEGIN = 0, there is no problem. When a cell
with a RAMP becomes active, the first time the ramp is interpolated it
is done at a value of T - T_activate = 0.

The issue ,which was realized with the previous post, was for T_BEGIN
<> 0. For example, if T_BEGIN were 100, then all time ramps are
adjusted down by 100. So now when a wall cell activates at T-
T_activate = 0, it would actually use the input file value
corresponding to 100.

So in summrary if T_BEGIN is 0, don't worry about it. If T_BEGIN <> 0
and you need a time ramp for a wall cell, either make T_BEGIN 0, or
adjust the time RAMP.

In your case, this should work fine as I have run similar simulations
without a problem (but with T_BEGIN = 0). I recommend you create a
very simple test case with just the vent you want to open and control
it with a TIME DEVC. Once you get that to work, then add the
complexity of the smoke detector.
> > > thanks- Hide quoted text -
>
> - Show quoted text -

dannyj...@googlemail.com

unread,
Sep 12, 2019, 11:59:08 AM9/12/19
to FDS and Smokeview Discussions
But if we want the vent parameter, e.g. volume flow to interpret the ramp relative to T_BEGIN = 0, not the activation time of the vent, how is this done?

dannyj...@googlemail.com

unread,
Sep 12, 2019, 12:10:18 PM9/12/19
to FDS and Smokeview Discussions
Perhaps some context:

If we have a solid with a surface temp that varies with time and we want to mimic the moving of that solid by creating obstructions on time triggers - is there means to do this without having to manually offset the RAMPs for each solid created?

Randy McDermott

unread,
Sep 12, 2019, 12:34:19 PM9/12/19
to FDS and Smokeview Discussions
Probably can be done.  You can do nearly anything with RAMP, DEVC, and CTRL.  I'd start by looking at the "Controlling a RAMP" of the user guide.

On Thu, Sep 12, 2019 at 12:11 PM dannyjhopkin via FDS and Smokeview Discussions <fds...@googlegroups.com> wrote:
Perhaps some context:

If we have a solid with a surface temp that varies with time and we want to mimic the moving of that solid by creating obstructions on time triggers - is there means to do this without having to manually offset the RAMPs for each solid created?

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/da09194d-6293-4846-a559-bdcca7e46181%40googlegroups.com.

dannyj...@googlemail.com

unread,
Sep 12, 2019, 12:45:30 PM9/12/19
to FDS and Smokeview Discussions
Thanks Randy - I’ll take a look.

dannyj...@googlemail.com

unread,
Sep 12, 2019, 1:01:11 PM9/12/19
to FDS and Smokeview Discussions
It seems quite straightforward - create a timer DEVC and direct the RAMP to that quantity.

Cheers.

Reply all
Reply to author
Forward
0 new messages