Topic Options
#11868 - 06/25/07 09:37 AM Time History Analysis for Blast Overpressure
Alok Offline
Member

Registered: 04/12/07
Posts: 23
Loc: UK
I want to carry out a time history analysis for an blast on an offshore platform.The blastpeak over pressure is 0.5 bar, assume it acting in X and Z direction. The profile for blast overpressure over time is triangular i.e. it reaches to peak pressure of 0.5 Bar at 250 ms and it is linear.

Now to generate the force against time in Spectra. I have model from static case, whch gives geometry as well diameter, length nodes, etc.

I want to generate force profile in spectra. Please suggest how to proceed from here. The force neds to be calculated for a specfic node. for e.g node 60 for the whole range of pressure/time. Once we claculate peak foce due to peak pressure, it can be linear interpolated for all other time intervals

I have done a calculation for the same, however I am not sure whether it is correct approach.

So your valuable suggestion from previous project are welcomed and appreciated with an short example with pipe running in all the three direction namely X, Y and Z.

regards

Top
#11870 - 06/25/07 11:30 AM Re: Time History Analysis for Blast Overpressure [Re: Alok]
Captain Kenny Offline
Member

Registered: 09/09/05
Posts: 72
Loc: Scotland
You're taking an interesting approach to the scenario, and I'm interested in how you get on too.
I would think you wll have to arrive at a whole bunch of inputs to the time history, one for each node in the area with the applied blast force distributed loads being converted into equivalent nodal forces. Don't forget to include the negative rebound phase of the blast wave as well into the time history. Are you going to establish a nominal blast center and have finite differences between the blast front hitting elements at different distances from the center.
Just remember the blast front moves very fast in relation to the typical dimension of piping so the active loading times are very short [i.e. for a blast front travelling at 300m/s it will take 0.7msec to pass a 8" line]
Best of luck - it's a massive undertaking
_________________________
Kenny Robertson

Top
#11879 - 06/25/07 10:19 PM Re: Time History Analysis for Blast Overpressure [Re: Captain Kenny]
MPB Offline
Member

Registered: 09/08/06
Posts: 51
Loc: Perth
Design for blast is full of uncertainties but there are now quite a few references from various organisations including UKOOA, API, FABIG.
My understanding is this - the blast overpressure generates a shock wave which as Kenny mentioned passes the pipe very quickly. Therefore the pressure equalises on a pipe almost instantly and this pressure is ignored for analysis (this is not the case for a large vessel or a blast wall).
However an air/combustion product wave is also generated which causes drag loads on the pipe. The timeform of the drag wave is I believe normally taken to be the same as the overpressure pulse. The equivalent pressure of this drag wave (0.5rho.v^2) is significantly less than the static overpressure - some guidelines say about 0.3 times the static overpressure. There is an ASCE standard on design of blast resistant buildings which gives an empirical formula.
The appropriate drag factor is also an area of uncertainty. So is dynamic amplification but your approach attempts to address this.
I ran such a time history analysis recently as a backup to a hand calc. You need to simulate the load as a series of point forces. If the outer fibre stress hits more than about 1.3*yield the pipe goes plastic across the entire section and forms a plastic hinge assuming ductile pipe. This invalidates the stress model. On the other hand you may wish to keep your stresses below yield anyway.
That opens up the question of your design criteria – is it practical to keep within code or yield stress when designing for say a 1 in 10,000 year event. Most of the references recognise or instruct in the use of ‘plastic design’.

Top



Moderator:  Denny_Thomas, uribejl 
Who's Online
0 registered (), 42 Guests and 2 Spiders online.
Key: Admin, Global Mod, Mod
April
Su M Tu W Th F Sa
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30
Forum Stats
12065 Members
14 Forums
16973 Topics
75151 Posts

Max Online: 303 @ 01/28/20 11:58 PM
Top Posters (30 Days)