XGSLab FAQ
XGSLab FAQ
XGSLab™
SCIENCE FOR ENGINEERING
FAQ
SINT Srl
1995 – 2023
(Release 10.3.1 - 06/23)
FAQ
SINT Srl reserves the right to revise and improve its products as it sees fit.
This document describes the state of this product at the time of its publication, and may not reflect the product at all times
in the future. Moreover, due to the continuous evolution of the XGSLab software, some screens or numerical obtained by
last XGSLab version could be slightly different from those in this document.
The software described in this document is delivered under license agreement.
SINT Srl - Via Cristoforo Colombo, 106 - 36061, Bassano del Grappa (VI) – Italy (www.xgslab.com)
Possession or use of the XGSLab software is authorized only pursuant to a valid written license agreement from SINT Srl.
Before using XGSLab please read the document “End User License Agreement “.
While every precaution has been taken in the preparation of this User’s Guide, SINT Srl assumes no responsibility for
errors or omissions, or for damages resulting from the use of the information contained herein.
Trademark
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights.
SINT Srl shall not be held responsible for identifying any or all such patent rights.
Windows is a registered trademark of Microsoft.
AutoCAD is a registered trademark of Autodesk.
Google Earth is a registered trademark of Google.
Other Trademarks are used in this document in an editorial fashion with non intention of infringement.
All trademarks are the legal property of their registered owners.
June 2019 9.1.1 Zig-Zag Transformer + Pipe Type Cable + Fragmentation Rules
September 2019 9.2.1 XGSA_TD extended to 100 MHz + Corona Effect Tool
September 2021 10.0.1 New module SHIELD + New CAD based on OpenGL
The support period may be renewed on an annual basis for periods of 1, 2 or 3 years.
Support is included in Annual Subscriptions during the validity period.
Software requirements:
- Operating system: Windows® 8 or later, Windows® 10 or later are anyway suggested (Windows® 11 included)
- Microsoft® .NET Framework 4.8 installed
The entire calculation process runs on the CPU and does not use any GPU features, so the graphics driver does not affect
calculation performance. XGSLab uses parallel computing and some processes can be executed simultaneously, so multi-
core CPUs can offer visible advantages in calculation performance, at least with a core number up to about 40.
Essentially, the calculation speed grows with the product “cores number * clock frequency”.
Some multi-core CPUs have a different clock frequency for cores, in such cases previous rule is not valid.
For the *GSA* modules, the following RAM hardware requirements should be considered. As the maximum elements
number is limited depending on the license profile, the following RAM requirements are suggested:
- Models with up to 2000 elements: at least 4 GB
- Models with up to 5000 elements: at least 4 GB for GSA and 12 GB for *_FD and XGSA_TD
- Models with up to 16000 elements: at least 20 GB for GSA and 64 GB for *_FD and XGSA_TD
- Models with more than 16000 elements: depending on the available RAM as in the figure below (note the log-log scale)
The following figure shows the RAM requirements as a function of the module and maximum elements number.
The hardware key is shipped via courier. Delivery time is usually in the range 1 – 4 working days.
The software key is sent via email. Delivery time is usually in the range 1 – 2 working days.
The program can be downloaded from our helpdesk web site. Accounts are usually sent in 1 – 2 working days.
All deadlines refer to the purchase order date.
Figure 1-2: Error message in case of use of stand alone key in remote way
Conductor Sizing:
- European standard: HD 637 S1:1999
- European standard: EN 50522:2010
- European standard: EN 50522:2022
- USA standard: IEEE Std 80-2000
- USA standard: IEEE Std 80-2013
Split Factor:
- European standard: HD 637 S1:1999
- European standard: EN 50522:2010
- European standard: EN 50522:2022
- USA standard: IEEE Std 80-2000
- USA standard: IEEE Std 80-2013
Lightning Shielding:
- International standard: IEC 62305-3:2010 (equal to EN 62305-3:2010)
- USA standard: IEEE Std 998-2012
Reference Standards considered by XGSLab are accepted in many countries in the World.
In any case, a limit of a few tens of MHz is conservative and should not be understood in an absolute sense.
This limit means that starting from a few tens of MHz, calculation accuracy gradually decreases.
The frequency range from DC to a few tens of MHz contains all power system frequencies, and the most significant
frequency spectrum of the electromagnetic transient as represented in the following figure.
Corona effects are partially out of the application range.
In practical cases, a single soil model is not able to represent a large site, in all places and possible environmental
conditions.
Therefore it is useless to refine soil models that inevitably are correct only in some specific point of the site.
A double layer soil model represents a good approximation because touch and step voltages depend mainly on the
resistivity of the upper layer, where the electrode is usually buried, and the lower layers can be represented by one
equivalent layer which mainly affects the equivalent resistance of the electrode.
In few words, in general, a double layer soil model is not a compromise but the only reasonable alternative to a uniform
soil model.
In the following a true double layer soil model case.
The measurements at different sites indicate about the same model.
Figure 2-2: Soil resistivity measurements in case of true double layer soil model
Figure 2-3: Soil resistivity measurements in case of false double layer soil model
Figure 2-4: Soil resistivity measurements in case of false double layer soil model
When frequency grows, resistivity and permittivity change substantially and is fundamental to consider these effects in
order to avoid large calculation errors.
There is not a general consensus about the soil parameter frequency dependence model.
XGSLab consider the following models:
- Messier
- Visacro – Portela
- Visacro – Alipio
- CIGRE TB 781 Model
In time domain calculation involving high frequency spectrum signals, if the soil parameters frequency dependence is
neglected, calculation errors could be unacceptable.
2.6 WHY IT IS IMPORTANT A MODEL THAT CONSIDERS BOTH SELF AND MUTUAL
IMPEDANCES?
Self impedances cannot be neglected in all systems where physical size is not negligible if compared to the wavelength
length.
In practical cases and industrial frequency (50 or 60 Hz) self impedances cannot be neglected when the system size is
over a few hundreds of meters.
In same conditions, mutual impedances cannot be neglected in case of large systems when soil resistivity is in the common
range from 10 to 100 Ωm.
In general, neglecting self and mutual impedances can lead to very large errors.
Anyway, a parametric study showed that is not sufficient consider self impedances only.
The effect of mutual impedances can be significant and neglecting this parameter, the calculation error can be more than
25%.
In general, effects of self and mutual impedances grow with the grounding system size, with soil conductivity (inverse of
resistivity) and with frequency.
2.8 WHAT THE MESSAGES “TOO SHORT ELEMENT” AND TOO LONG ELEMENT”
MEAN?
XGSLab works with “thin” elements.
Thin mean that the length of the element must be much greater than its diameter.
n
U= U
2
i
i =1
where:
U = U1 + U 2 + 2 U1 U 2 cos
2 2
where:
- Ui (V) = effective value of the induced potential from system i
- γ (deg) = angle between induced potentials
In general the angle between induced potentials is unknown, and conservatively the formula to use is the following:
U = U1 + U 2 + 2 U1 U 2 = U1 + U 2
2 2
Starting from release 10.2 the maximum number of elements has been increased from 16000+ to 32000+.
Then, previous approach by parts can be avoided, but it is anyway useful in case of limited RAM resources.
The PEEC is a BEM (Boundary Element Method) developed in about 1990, and allows to simulate systems of conductors
integrated with circuit components like generators and impedances.
Many people do not know the difference between BEM and FEM (Finite Element Method) but the difference is crucial.
Essentially with FEM the discretization is applied to the propagation medium, with BEM the discretization is applied to the
sources.
FEM is a numerical method for solution of partial differential equations and is really useful in some applications, but hard
to use for engineering purposes because the modelling difficulties in case of unbounded scenarios and the hardware
requirements.
BEM is numerical method in integral form perfectly suitable for engineering purposes.
All main software in the market like CDDEGS®, WinIGS®, CYME®, ETAP®, ELEK® are based on BEM.
The module NETS is based on circuit theory (graphs theory and Kirchhoff laws), on power lines, cables and transformers
theory and the numerical method PCM (Phase Components Method).
Many people knows the method SCM (Sequence Component Method) but do not know PCM.
In particular many people do not know the application limits of the SCM.
We used the PCM because this method is for general purposes and can be applied with multiphase and multiconductor
systems, symmetrical or not, balanced or not and in particular can consider multiple grounded systems and problems that
involve currents to earth.
In general, the SCM cannot be used in case of multiple grounded systems.
WHY SOMETIME POTENTIAL CALCULATED CLOSE TO ELEMENTS IS NOT AS EXPECTED?
The modules GSA, GSA_FD, XGSA_FD and XGSA_TD are based on the PEEC numerical method.
The system of conductor is divided in short elements.
With a first calculation step, the program determines charges, currents and potentials on each element.
With a second calculation step, the program determines potentials and fields in the propagation medium as superposition
effects of all elements.
The calculation of potentials assumes a uniform charge distribution (or leakage current distribution) along each element.
As a consequence, the potential close to the elements will not be uniform and then will be in general different from the
actual element potential.
The potential distribution related to a uniform charge distribution in a single element can be represented using ellipsoids in
the 3D space.
Far away from the elements these ellipsoids becomes spheres and calculation is as expected. But close to the elements
potential distribution differ from ellipsoids and there are no analytical equations to express them.
= 3162
f
where:
- λ (m) = wavelength
- ρ (Ωm) = soil resistivity
- f (Hz) = frequency
For instance:
- If f = 100 kHz and ρ = 100 Ωm, λ = 100 m
- If f = 10 kHz and ρ = 100 Ωm, λ = 316.2 m
- If f = 1 kHz and ρ = 100 Ωm, λ = 1000 m
- If f = 100 Hz and ρ = 100 Ωm, λ = 3162 m
Considering a simple system with a single wire and injection point located at an end, the behaviour is the following:
- The impedance decrease when length increase and length is much shorter than wavelength
- The impedance increases when length increase and length is comparable to wavelength
- For even greater lengths the behaviour is oscillating but damped by the dissipative medium
You could avoid to represent the concrete. The buried concrete is hygroscopic and tends to attracts moisture and salts
from surrounding soil, and with time (years) the concrete resistivity becomes similar to the soil resistivity.
The other option is to consider the concrete as a cylindrical shell around buried conductors. The shell thickness is usually
in the range 3 - 5 cm and represents the minimum concrete thickness between conductors and soil.
After that, you can set the conductor you want to consider as concrete encased with the specific created new conductor.
The prospective permissible touch and step voltages with additional resistances (below indicated with a symbol +) without
and with soil covering layer (SCL) are calculated by using the following equations:
U SSP + = ( RB + 2 R f + 2 Rshoe ) I B
e e
Rf = = = 3.125e without SCL
4b 4 0.08
Cs s C
Rf = = s s = 3.125Cs s with SCL
4b 4 0.08
where:
- RB (Ω) = body resistance
- Rf (Ω) = ground resistance of one foot
- Rshoe (Ω) = additional resistance of a shoe
- Rglove (Ω) = additional resistance of an insulating glove
- IB (A) = tolerable body current
- ρe (Ωm) = ρe or ρ1 for uniform or multilayer soil model respectively
- ρS (Ωm) = soil covering layer resistivity
- CS = reduction factor related to the soil covering layer
- b (m) = equivalent radius (0.08 m)
For instance, if tf = 0.5 s and Body Weight = 50 kg, IB = 0.164 A, USTP = 189.7 V and USSP = 266.6 V.
If the resistance of a single shoe is 2000 Ω and no gloves are considered, the new permissible touch and step voltages
will be respectively:
If you click on “Commit” you make the new items present in the data base of NewItemsPrj available also for all the future
projects that will be created in 2nd Pc.
Moreover, once executed the “Commit”, if you want to have the new items also in a project (let us call it “OldPrj2ndPc”)
created before the commit, you have to open OldPrj2ndPc and click “Update” in the section of the main tool bar circled in
the picture here above.
These conditions have been introduced recently and they face databases built with old and more “relaxed” constraints.
This implies that it is very likely you will find a message like this.
In case this message pops up there are two alternative ways to overcome it:
- Rename the record(s) pointed out (in the picture here above “fence post” in the “conductor library”)
- Delete the record(s) pointed out
Finally consider also that, at the moment, when you update the software, the database of libraries is again the initial one.
To make it contain all the customizations you did in the past you have to open a project created before the updating of the
software and execute a commit from there.
2 R
m =
1 1 1 1
− + −
c a+c c + 4b2
2
(a + c)
2
+ 4b2
where:
- R (Ω) = Wenner or Schlumberger resistance
- a (m) = spacing between voltages probes
- b (m) = probes depth
- c (m) = spacing between voltages and current probes (in case of Wenner method c = a)
The apparent soil resistivity can be calculated also starting from the soil model parameters.
For instance, for a multilayer soil model it is possible to calculate:
where:
- a (m) = spacing between voltages probes
- c (m) = spacing between voltages and current probes
- ρi (Ωm) = resistivity of layer i
- hi (m) = thickness of layer i
The calculation of layer parameters (resistivity and thickness) can be done using a minimization algorithm which lead to
the minimum of the following squared error function:
( a , c ) − c ( ai , ci , 1 , 2 , n , h1 , h2 , hn−1 )
2
N
( 1 , 2 , n , h1 , h2 , hn−1 ) = m i i
i =1
m ( ai , ci )
where:
- a (m) = spacing between voltages probes
- c (m) = spacing between voltages and current probes
- ρi (Ωm) = resistivity of layer i
- hi (m) = thickness of layer i
- N = number of measured resistivity values
Previous equation can include weight functions, but this is not essential for the following considerations.
The number of unknowns in previous minimization problem is P, a resistivity and a thickness for each layer except the
bottom layer with infinite depth. It follows:
P = 2L − 1
where:
- L = number of layers
As known, in case of a linear system, the number of independent equation must be the same of the unknown.
In the specific case, the minimization algorithm can find also more unknowns than equations, but in XGSLab the following
conservative checks has been implemented:
- If the number of independent measurements N is lower than 2P, the following message is shown “Warning: poor
number of measurements with different electrode spacings”. In this case the calculation is done
- If the number of independent measurements N is lower than P, the following message is shown “Error: insufficient
number of measurements with different electrode spacings”. In this case the calculation is not done
Independent measurements means Wenner and Schlumberger measurements with different electrode spacings.
For instance, if L = 5, P = 9 and then, the minimum number of independent measurements is 9, the minimum suggested
number of independent measurements is 18.
In order to import external file in the application (eg. dxf) or export file project use these instructions:
You can click \\tsclient\Storage, which is pinned to Quick access, and will be taken directly to where uploaded files are.
Then you have to select “All files (“.”) as file type to be able to see