Για κατάρριψη των ισχυρισμών του Λακαφώση και των cfd του αρκεί αντίθετο cfd που να καταρρίπτει τα παρακάτω και καταθέσεις πυροσβεστών και φωτό ''χιονισμένου'' πεδίου :


(τα οποία βέβαια δεν υπάρχουν)
Σε 0.4 sec η ποσότητα ελαίων σιλικόνης δεν μπορούσε να αποδράσει απο τους μετασχηματιστές βάση των φωτό που υπάρχει στην έκθεση
490 This is further confirmed by examination of the outside casting of the three transformers, showing that the
quantity of silicone oil that could possibly escape during the first 0.3-0.4 seconds would not be enough for the
large fire ball that was observed (Figure 67).
Furthermore, the absence of white dust (Silicone dioxide, SiO₂) at the scene of the accident, in quantities that
would justify the combustion of a large quantity of silicone oil, is another clear indication that such an occurrence
did not take place.
Αναφορά σε Pisa Ghent
Both the
University of Pisa and the University of Ghent offered an opinion on the scientific validity of CFD analysis
to modelise deflagrations. Furthermore,
EODASAAM contracted the University of Ghent to
review the work on
CFD analysis of the Tempi accident, as was already done by EDAPO. This review resulted in a report that offered a
technical opinion on the validity of the variables and parameters for modelling that had so far been used in the
CFD simulations, in order to recreate the event as recorded by the cameras with the most realistic scenarios. The
adjustment of some of the variables and parameters led to an improved and more reliable set of CFD simulations
and analysis, that resulted in the following findings:
a. Among different types of hydrocarbon fuels that were tested, it was found that the results were very similar
for many different types of fuels with similar physical quantities, with the
most important parameter being
mass of fuel. LPG, gasoline fuel, naphtha mixture, etc. of the same total mass provide almost the same result
(similar fireball size and duration) with only very small differences in the shape of the fireball.
b. Calculations, with the mathematical CFD models, indicate that approximately
2500 kg of generic hydrocarbon
fuel is needed to recreate the 3 distinct stages (1000, 1200 and 300 kg respectively) of the explosion, the
fireball and the secondary pool fires that were recorded in the 3 videos that show the event from 3 different
angles.
c. The duration of the release of fuel during stage 2 (approximately 4 seconds) and the observed result (fireball
size growing from 40 to 80 meters in diameter) point out to a quantity of fuel which is estimated at around
1200 kg by the CFD simulations.
If true, this rules out the possibility of liquid fuel (silicone oil from the
transformer or any other liquid fuel from an unknown source) "splashing all over" the Restaurant Car, because
such an action could not hold and release 1000 kg of fuel for 4 seconds. The only such possibility would be for
the liquid fuel to enter the Restaurant Car through its broken windows, but the 6 Restaurant Car survivors do
not support such an occurrence, nor do they have any burn injuries consistent with 1000 kg of fuel burning
from inside their carriage.
d. It is further estimated that a quantity of 300 kg of Hydrocarbon fuel may have been involved in the pool fire
#2, consuming the Restaurant Car, that continued to burn for more than 2 hours feeding also on silicone oil
8
FDS and Smokeview are free and open-source software tools provided by the National Institute of Standards and Technology
(NIST) of the United States Department of Commerce.
Report RL01-2025 Published by EODASAAM Page 107 of 178
and the materials of the internal fittings of the carriage. It is not possible, however, to accurately calculate this
quantity due to the unknown time of secondary ignition of additional fuels involved, which is also the case for
the amount of fuel that was burned in the other pool fire.
e. The examination of video footage and corresponding CFD analysis for the 2nd stage of the deflagration point
towards a single source for the 2nd stage of the deflagration. Although it is not possible to confirm that the fuel
for the 1
st stage of deflagration has the same origin as the fuel for the 2nd stage, it is most probable that all 3
stages of the phenomenon are burning the same source of fuel.
489 Using the assumptions of RI.SE (484), an extra CFD simulation was run using the smallest droplet diameter
mentioned (0.5mm) and complete atomization of the whole mass of fuel, which resulted in the following findings:
-
Silicone oil of the type used in the power transformers of the locomotives could not be made to explode and
create a large fireball in any way at all. Even if large sources of heat and open flames were added as a possible
ignition source and even if the silicone oil was sprayed in various ways against a wall in order to create
dispersion and smaller droplets, the only possible result was a very short ignition of 0.3-0.4 seconds that did
not propagate to the rest of the quantity of the oil.
490 This is further confirmed by examination of the outside casting of the three transformers, showing that the
quantity of silicone oil that could possibly escape during the first 0.3-0.4 seconds would not be enough for the
large fire ball that was observed (Figure 67).
Figure 67. Large cracks of the transformer still attached under the 120-022 locomotive (Left). Bumps, folds, rips and large cracks of
remaining locomotive transformer 120-023 (right), see also Figure 61.
491 The above simulation also corresponds with the finding that there is no mechanism
to create a fine mist of droplets
at this stage, since the original point of impact is now 5-6 seconds in the past, 30 meters to the south and with the
wind blowing in the opposite direction.
This fact alone should rule out any fuel with similar characteristics to
silicone oils.
Report RL01-2025 Published by EODASAAM Page 108 of 178
492 Even if the hypothesis of ignition and creation of the initial fireball could be accepted, the 2nd stage of the fireball is more difficult to explain with silicone oil as fuel
the fire plume is observed as coming out of a fuel source already
burning fiercely, obviously without any additional heat source and under ambient temperature. Survivors from
the restaurant car did not report any significant temperature change during this stage and do not show any burns
due to radiation. These elements point to the fact that the unknown fuel is volatile and very flammable.
493 The detailed information on CFD initial conditions, technical parameters and test results can be found in Appendix
to this report.
4.4.3.3. Conclusion
494
The presence of a flammable substance with the characteristics as determined in the previous analyses is not
mentioned on the transport documents of freight train 63503. Also, the recently released videos of train 63503,
although not yet made available to EODASAAM by the judicial investigation, do not make any transport of a
representative amount of flammable material visible in a location consistent with the above findings. EODASAAM
recognises this fact, but this does not dismiss the above observations (4.4.3.1) that are clearly documented.
495 These observations, together with the performed simulations, confirm the conclusion that was also made by Prof.
Konstandopoulos: “The large fireball at the Tempi railroad accident that occurred within 0.5 sec from the collision
of the trains cannot be attributed to the transformers oil (PDMS) which although combustible at high enough
temperatures (> 400) it could not - due to reaction kinetics reasons - ignite in the prevailing environment.
Furthermore, the absence of white dust (Silicone dioxide, SiO₂) at the scene of the accident, in quantities that
would justify the combustion of a large quantity of silicone oil, is another clear indication that such an occurrence
did not take place. Hence the origin of the fireball consistent with a few tons of a flammable volatile fuel needs to
be searched elsewhere.”
496 The mass of this flammable volatile liquid has been estimated, via the CFD modelisation, in the order of magnitude
of 2,5 tons. So far, the investigation has not identified any substance, with the properties and total mass known
to have been in the rolling stock, which could have had this role. Similar to the previous steps (488), EODASAAM
will contract a renown institute to review the validity of the variables and parameters that were used in the revised
CFD analysis.
497 Despite this uncertainty about the source of the observed phenomenon, EODASAAM has explicitly chosen not to
postpone the publication of this report. First of all, the fireball and the subsequent fire "only" had an impact on
the consequences of the accident, without contributing to its causes. More importantly, the Greek railway system
needs to know and accept the results of this accident investigation as soon as possible, in order to be able to start
the necessary improvement process.