I am pleased to welcome you to the Online Workshop Simulation in NDT 2010. The workshop will take place during the whole month of September. This workshop wants to bring together research, providers and NDT practitioners.
I wish you to enjoy and to benefit from this workshop.
Introduction to SimNDT 2010In Reply to Rolf Diederichs (Editor) at 09:00 Sep-01-2010 (Opening).
First, I would like to thank the participating authors for their contributions. I extend a special thanks to Rolf Diederichs for organising this workshop.
The first Workshop on this site was the Transducer Workshop in September 1996[>1]. May 1997 was the last such dedicated workshop on NDT.net (Applications[>2] was the topic then). The Workshop format seemed extremely popular as an information resource so Rolf decided to not just hold the occasional workshop, but instead he decided to make the discussion sessions a full-time feature of NDT.net. In October 1997 the first NDT.net "Forum" was available to provide an ongoing "Workshop" for discussions in all NDT fields.
The SimNDT2010 Workshop uses a format similar to that introduced in 1996. Presented papers are provided on a variety of specific related subjects and the guests are invited to ask the authors questions. 13 years on and technology has made it easier to improve on the variety and quality of online presentations. In fact, not all presentations are just "technical papers". In SimNDT2010 Philippe Dubois' presentation is available as a 30 minute real-talk video. Looking back to 1997 we can see that although some questions related to the specific papers, the papers often provided just a starting point and the discussions branched to other areas.
I hope the presentations here can again provide initiation points for discussions.
For many years I have been a user of the computer-tools of modelling and simulating NDT testing. The ability to mock-up on a computer display the set of parameters I am dealing with in the field, has provided me with better understanding of the problems of the test. But the assembly of the details to complete the simulation has also forced me to look more closely at the basic theory behind the NDT methods I am using. Its quite one thing to be told the theory of beam spread and diffraction as part of your Level 2 training. But when you get to actually see the effects in realistic models indicating what is happening to the beam as it interacts with flaws and other boundaries, you can better understand the theory you took for granted (or failed to pay attention to) in your NDT training classes.
The presentations in this workshop are a mix of theoretical modelling and practical simulation. I invite you to review the presentations and to ask questions concerning modelling and simulation. SimNDT 2010 is listed as a Special Session and will eventually be replaced by a future Special Session. However, the discussions will not disappear. They will be transferred to the Technical Sessions section and may inspire further discussions on the forum in the future.
18:57 Sep-04-2010 José Vicente R & D, - Technological Services on NDT Technological Institute of Construction, Spain, Joined Aug 2010 4
Values for Damping Parameters Hello to Everybody:
I am user a numerical simulaton software called Wave 2000 (cyberlogic).
From the moment, I use only the time-of-flight information becuase my objectives is to see if elastic modulus in the simulations are fit to the experimental tests that we perfome using sonic-ultrasonic waves in cement-based materials (in extend construction materials-elements).
Nowadays, I want to include in my simulations the information of amplitud, to reproduce the scattering process that suffers the ultrasonic wave in the porous microstructure.
There is some information about attenuation coefficients of elementary materials: cement, aggregates, porous,...but I do not hear/see anything about damping paramters. The interested parameters are: eta (Pa*s) & phi (Pa*s), do you know how to proceed in this subject?? The problem is that I couldn't do experimental test (in aggregates or cement powder) because there are not solid materials.
01:31 Sep-06-2010 Ed Ginzel R & D, - Materials Research Institute, Canada, Joined Nov 1998 1196
POD in CIVA I found it interesting in Philippe Dubois' presentation to learn that the new release of CIVA incorporates a POD computation. This has been a "requirement" in pipeline applications for compliance with DNV OS F101 for over a decade now.
The presentation on SimNDT2010 indicates that several parameters may be analysed simultaneously using a Monte Carlo approach. But when we analyse a "detection" we compare it to a threshold for the binomial distribution. The threshold is, in UT terms, based on the response relative to a target (e.g. a 3mm diameter SDH). Does the POD function in CIVA 10 compute the various responses of the variations relative to a specified reference target? Does it then allow the thresold to be adjusted to see what the required threshold would be to achieve a 90|95 PoD?
As for me, modeling consists in building a model, mathematical or not, to establish predictions in a given field. In NDE, the modeling of a specific phenomenon can therefore predict the outcome of experience in the field of validity of the model.
The simulation is the practical implementation (computing) of the model. This implementation is therefore accompanied by a discretization of the model (whether finite elements or integral methods) and therefore with further approximations to modeling.
In other words, the simulation is the representation of a real phenomenon, based on an underlying theoretical model, itself valid in a bounded domain.
Thanks Ed for this question. You are right, POD is a new feature of CIVA 10 and may interest people working on POD as the cost of building these curves is very high.
As for calibration : it is possible in CIVA to make simulation adjusting levels to a simulated calibration (based on side drilled holes or flat bottoms holes) directly. If one may want to make calibration on other flaws, it is also possible, but then two calculations will be done, one for calibration and one for simulation. Then a simple formula can adjust the results for simulation according to calibration.
As for DNV document, the rule applied called 29/29 is based on the fact that 29 is the minimum number of flaws to obtain 90% of Hit and 0% of Miss with 95% of confidence (using binomial law). As we want to calculate the famous A90/95, we have to have 29 results. This binomial method is more simple to implement that the Berens Method that is used in CIVA but CIVA can be used to apply this law.
Anyway, the threshold given in the POD in CIVA can be easily linked to the threshold obtained on some calibration flaws and A90/95 can then be measured on the curve.
If you need further information, dont hesitate to ask.
23:38 Sep-24-2010 Linxiao Yu R & D, NDE in Oil & Gas; Energy; Manufacturing BP, USA, Joined Aug 2006 13
Re: where to get software for simulation? free or unfree?In Reply to Philippe Dubois at 19:20 Sep-24-2010 .
For starters, there are many small simulation tools (Java Applets) illustrating key NDE concepts at "NDE resource center" from Center for NDE at Iowa State university. The materials are free and the link to the website is: http://www.ndt-ed.org/EducationResources/CommunityCollege/communitycollege.htm
00:33 Sep-25-2010 Rolf Director, Editor, Publisher, Internet, PHP MySQL NDT.net, Germany, Joined Nov 1998 602
Re: where to get software for simulation? free or unfree?In Reply to Linxiao Yu at 23:38 Sep-24-2010 .
This site shows some principles for education purpose, but I do not see any practical modeling software for free use. However, in CNDE at Iowa they did research with a tool called UTSim but I do not know where this is free or commercial available.
12:46 Sep-25-2010 José Vicente R & D, - Technological Services on NDT Technological Institute of Construction, Spain, Joined Aug 2010 4
Re: where to get software for simulation? free or unfree?In Reply to qing sun at 12:54 Sep-24-2010 .
another SW is WAVE that comes from Cyberlogic and can combine ultrasonic propagation and electromagnetic techniques. The most useful is the version for cylindrical probes and the 3D. As the major of the SIM SW tools, the CAD part is not friendly. so the 3D is hard too work and expend much time to get good approach when models become complex. Requires import the models in PCX or GEO formats,...so you have to build the models out and after converts to these formats and import.
Anybody can expline more about SIM that directly get ultrasonic images...?
Thank you in advance.
15:39 Sep-25-2010 Ed Ginzel R & D, - Materials Research Institute, Canada, Joined Nov 1998 1196
Re: where to get software for simulation? free or unfree?In Reply to José Vicente at 12:46 Sep-25-2010 .
José...I think your question about explaining more about simulation software is a good idea! Your specific interest is for simulation software that directly gets ultrasonic images. Perhaps it would be informative if someone (or a group) of those expert in the field) could assemble a detailed description of how the various modelling/simulation methodologies are approached. In 2007 I assembled a list of software packages in a paper on NDT.net (http://www.ndt.net/article/modellingNDT2007/2.pdf). Some of these software programmes are no longer available but many are now updated and perhaps new ones exist. In the table of the softwares I had a column that described the "approach". There are programmes for Ultrasound, Guided Wave, Acoustic Emission, Eddy Current and X-ray (at that time I see that CIVA did not include X-ray).
Some of the papers presented in this Workshop (like the paper by Prof. Honarvar) provide a bit of background in the methodology used.
Perhaps having a single source of comparison of capabilities and underlying principles for the various simulation options in NDT would be useful for people looking for the correct software for their specific applications.
Do any of the readers have any suggestions where this is available or are there any volunteers to write such a descriptive paper?
20:59 Sep-28-2010 Rolf Director, Editor, Publisher, Internet, PHP MySQL NDT.net, Germany, Joined Nov 1998 602
Simulation of Weld inspection for small thickness materials One part of the following message is addressed to this workshop theme. Is it possible or more difficult to perform simulation in a small thickness range?
Board: Technical Discussions at 10:55 Sep-28-2010:
Subject: Weld inspection for small thickness materials.
I would like to enquire about the possibility of doing weld inspection on small thickness pipes. Thickness ranging from 2mm to 8mm by using advanced Ultrasonic Testing (phased array or ABI scan method). Are there special probes or equipments or softwares available in the current market that can be used to obtain these inspections as mentioned ?? Also are there any relevant softwares that can be used to come up with these kind of test simulations to see the results ?
18:15 Oct-04-2010 Daniel Richard Engineering, Product Management Zetec inc., Canada, Joined Oct 2010 1
Re: Simulation of Weld inspection for small thickness materialsIn Reply to Rolf at 20:59 Sep-28-2010 .
It is possible but challenging to address those small thickness pipe.
Commercial instrument and software like those provided by Zetec will allows data gathering capability and efficient acoustic field simulation.
On the other hand, I see most of the challenge being related to the probe design itself for those kinds of inspections.
free book Hi,I find one amazing free to download book about Modeling and simulation.This book collects original and innovative research studies concerning modeling and simulation of physical systems in a very wide range of applications, encompassing micro-electric-mechanical systems, measurement instrumentation, catalytic reactors, bio mechanical applications, biological and chemical sensors, magneto-sensitive materials, silicon photonic devices, electronic devices, optical fibers, electro-microfluidic systems, composite materials, fuel cells, indoor air-conditioning systems, active magnetic levitation systems and more. You can find it here: http://www.intechopen.com/books/show/title/modelling_and_simulation
Hello, I am a student. Now, i still do my final report about NDT. I want to ask something. I confused with my problem in my simulation. I simulate impact echo methode on the concrete. I try to find the good result of P-wave speed of concrete. If i change number of data from 4096 to be 8192, is that will affected the error results?. My result of simulation is different when i use 4096 or 8196 (same material, same dimension, same paramater like time increment, force of impact). Is that any explanation?.
Second, Is that any relation between force of impact and contact time of impact with P-wave speed of concrete?or is that any relation force of impact and contact time of impact with time arrival in the receiver?. Thank you very much for your explanation.
There cannot be a simulation without a model but there can be a model without a simulation.modeling is static process in contrast, simulation is a dynamic process .however, both try to imitate the real situation. modeling and simulations are used for analyzing parameters which could be expensive if it were to be studied on the real process or device.
for example, if one want to study how a paper ball moves in the wind. making of paper ball becomes a model by which one can measure parameters(static) such as diameter, weight, color etc.,but if the same ball moves in the wind it becomes a simulation where the parameters(dynamic) like velocity,acceleration etc., can be analyzed. thus both can be distinguished. A NDE simulation software can have always a mathematical model as a back end or as its base.
pls correct me if i am wrong. thanks