4
After the completion of the 3 heating cycles under-
ground it was evident, that trench depth affected the heat
transmission between the cylinder and host rock (Figure 6).
Even though evidence for the hypothesis was detected, the
results were not as consistent as expected. This was mainly
due to inaccurate sensor positioning possibilities on a in-
situ scale.
Since trench depth had to be increased for every phase,
the rock salt had to be dug out after each phase and then
filled back in. After filling it back in, the sensors had to be
placed again in the same depth. Even little displacement
was detected to have an impact on measurement accuracy.
In order to get more accurate measurements, the experi-
ment was decided to be repeated on a laboratory scale with
the sensors being fixed in the same position for all 3 phases.
Laboratory Testing
For laboratory testing, a new experimental environment
had to be prepared. In order to simulate a down-scaled
disposal drift, concrete curbstones were arranged in a way
that they resemble a trench. The bottom of the model was
created with a concrete cast plate. The first heating phase
was started at a depth that is equal to 30% of the heat-
ing dummies diameter and then repeated for 15% and 0%
respectively.
The order was reversed in the laboratory experiment
since it was easier to decrease the depth of the trench by
filling it with concrete for each phase than chiseling it out.
After placing the POLLUX in the trench of the con-
crete cast floor, approximately 11 dm3 of backfill material
were filled into the shape. Afterwards, heating was initiated
and continued for 3 hours in each phase. For safety reasons
the test was not run overnight, therefore a shorter heating
phase was conducted.
Between each phase there was a time frame of at least
24 hours for the freshly cast concrete left, so it can suffi-
ciently set before another heating cycle was initiated.
The results from the laboratory experiment met the
expectations regarding their accuracy. The logged data
clearly showed a connection direct correlation between
trench depth, contact surface and a shifting heat transmis-
sion ratio towards backfill and host rock (concrete cast).
Figure 5. Trench depths in different phases
Figure 6. Temperatures in Heating Cycles (Alsalamin, 2023)
After the completion of the 3 heating cycles under-
ground it was evident, that trench depth affected the heat
transmission between the cylinder and host rock (Figure 6).
Even though evidence for the hypothesis was detected, the
results were not as consistent as expected. This was mainly
due to inaccurate sensor positioning possibilities on a in-
situ scale.
Since trench depth had to be increased for every phase,
the rock salt had to be dug out after each phase and then
filled back in. After filling it back in, the sensors had to be
placed again in the same depth. Even little displacement
was detected to have an impact on measurement accuracy.
In order to get more accurate measurements, the experi-
ment was decided to be repeated on a laboratory scale with
the sensors being fixed in the same position for all 3 phases.
Laboratory Testing
For laboratory testing, a new experimental environment
had to be prepared. In order to simulate a down-scaled
disposal drift, concrete curbstones were arranged in a way
that they resemble a trench. The bottom of the model was
created with a concrete cast plate. The first heating phase
was started at a depth that is equal to 30% of the heat-
ing dummies diameter and then repeated for 15% and 0%
respectively.
The order was reversed in the laboratory experiment
since it was easier to decrease the depth of the trench by
filling it with concrete for each phase than chiseling it out.
After placing the POLLUX in the trench of the con-
crete cast floor, approximately 11 dm3 of backfill material
were filled into the shape. Afterwards, heating was initiated
and continued for 3 hours in each phase. For safety reasons
the test was not run overnight, therefore a shorter heating
phase was conducted.
Between each phase there was a time frame of at least
24 hours for the freshly cast concrete left, so it can suffi-
ciently set before another heating cycle was initiated.
The results from the laboratory experiment met the
expectations regarding their accuracy. The logged data
clearly showed a connection direct correlation between
trench depth, contact surface and a shifting heat transmis-
sion ratio towards backfill and host rock (concrete cast).
Figure 5. Trench depths in different phases
Figure 6. Temperatures in Heating Cycles (Alsalamin, 2023)