South Korea is said to be one of the world’s most mountainous countries with 70% of its landmass in rugged terrain. It also has one of the highest population densities in the world. To meet the needs of the growing population and industrial development, infrastructure projects are underway to improve transportation links between major population centres.

Given the predominant terrain and populous nature of the nation, preference is given to a ‘straight line’ method requiring significant use of tunnel and bridge for both road and rail construction. Granite and limestone rock are dominant, thereby making drill and blast the preferred method for tunnel advance.

One large-scale project underway is the Kyung-Chun Railway Line Construction that is providing a complete double track link between the South Korean city of Chunheon and the capital of Seoul.

Opportunity arose to fire three ‘full face’ tunnel rounds in the Kang Chon 1 Tunnel of the 7th Zone of the Kyung-Chun Line using the i-kon Electronic Blasting System (EBS) and perform some preliminary assessment versus conventional pyrotechnic signal tube initiation. It also allowed for a fresh look at initiation timing possibilities for large civil tunnels. This paper is a summary of the results obtained from a series of five firings in the Kang Chon 1 Tunnel – two preliminary bench marking blasts using conventional non-electric (NE) initiation followed by three firings by EBS and modified timings.

It should be noted that the results were obtained under ‘production conditions’ whereby data collection and trial quality were dictated by operational needs.

This paper will describe these limitations and attempts made to normalise the results to give performance indicators opposite key aspects of the tunnel cycle including: perimeter control, advance, fragmentation and vibration.

The Kang Chon 1 Tunnel

The Kang Chon 1 Tunnel is one of several on the 7th Section of the Kyang-Chun Line. The completed length will be 2500m with cross sectional dimensions of 10.6m wide by 8.4m high and a face area of 88m2. The major rocks present are granite gneiss or banded gneiss with sporadic quartz dykes and scattered cubic jointing observed. A rock mass rating (RMR) gives a result of 66 indicating a medium hardness rock.

A 12-hour cycle is worked which includes: mark-up, drill, charge, fire, excavate, scale, bolt and shotcrete. Typical blasting parameters give 116 x 45mm diameter shotholes drilled with a V Cut (wedge), with a centre stab hole, for 3.5m round depth. Shotholes are charged with 32mm diameter emulsion cartridge explosives (MegaMITE) and stemmed with wet sand plugs. Perimeter holes are charged with a 17mm pipe charge (FINEX). Initiation uses a range of millisecond (20ms) and long period (100ms) non-electric detonators (HiNEL). To offset the timings the face is fired in 4 bunches which are fired with 0 to 42ms trunkline delays. A conventional face timing is shown in figure 1.

Electronic Initiation Design

Given the accuracy and flexibility of the electronic blasting system it was decided to redesign the face timing based on a number of good blasting principles. The V Cut timing was left the same as conventional, thus to minimise risk related to cut failure, giving three holes firing at any one instant.

Once outside the V-Cut, timing was set to give single hole firing within 20ms, working in direct opposites from the initial cut. With the cut extension removed the face holes were set at 10ms, off-setting either side of centre. The perimeter was to be fired in 2 halves. One half was fired in instantaneous batches of 6 holes to gauge the effect of post-split perimeter control, while the other half was fired on 10ms delay interval.

The delay versus batch firing sides were swapped on alternate firings to negate the effect of geology. The lifter holes were designed to fire on 10ms delay offset either side of centre. Total blast duration was reduced from 3.2 seconds to 1.3 seconds (figure 2).

Tunnel conditions

During the evaluation period a porphyry intrusion entered the left-hand side of the tunnel face. This caused a number of poor rounds both with the two non-electric blasts and the first electronic blast. The tunnel contractor decided to increase the number of shotholes drilled from 116 to 132. This change may have masked some results, however attempts were made to normalise the outcomes by swapping perimeter technique left and right and to compare results only in similar ground.

Vibration analysis

All blasts were monitored for vibration and airblast. Up to 4 monitors were used for each blast spaced along the tunnel at distances of 131m to 604m from the tunnel face. To give statistical credibility a further two non-electric blasts were included that were monitored the week previous (Table 1 and 2 and figures 3, 4, 5, 6).

Furthermore a comparison of the near field vibration traces was conducted. The available vibration recording time was only two seconds therefore latter stages of the non-electric blast is missing.

The following conclusions can be drawn from the above data:

Comparison of the regression analysis shows that both systems gave similar results for the K (geological constant) and n (reduction level). This is expected in that all the results are taken in the same rock mass with similar explosive weight per delay and indicates that rock mass is not a significant factor for the difference in result.

It can however be seen that the regression line for the electronic blasts gives an r2 value of 0.953 compared to 0.653 for the non-electric system. The r2 value represents the Coefficient of Determination for the regression fit, being the degree of correlation between the estimated and actual y values. This indicates that the vibration prediction will be far more reliable with electronic blasting.

Comparing the vibration traces; the ppv generated in the cut was slightly higher for the electronic blasting system. This is not surprising as the design allowed for three shotholes firing same delay and the accuracy of the EBS would guarantee instantaneous firing of these holes. The non-electric system gave higher ppv in the cut extension, due to the fact of more shotholes firing within a tight time window, and the EBS now operating on true single hole firing at 20ms interval. This is similar with the face where the EBS has true single hole firing and the NE system now operating with multiple holes firing at the upper end of the delay range. The EBS gave higher readings in the perimeter, probably due to multiple holes firing in batches on the same delay compared to the NE system. The highest vibration came from the EBS lifter area firing on 10ms delay. It is not fully understood why and more work is needed to assess if this is related to reinforcement due to a tight time window.

Fragmentation size analysis

Fragmentation analysis was conducted for all five blasts using photographic image analysis. The non-electric detonator showed the average (P50) fragmentation size of 34.9cm versus 31.2cm for the EBS, indicating a size reduction of 11% with the overall results showing a uniform distribution. No direct conclusion can be draw here.

Perimeter control

A series of images were taken of the perimeter at the completion of each blast for digital analysis, by assessment of Half Cast Factor (HCF) (figure 7). Owing to the change in rock type between left-hand and right-hand side, the results are presented as a series of comparisons between non-electric versus electronic, simultaneous versus delay, and the difference between simultaneous and delay in each side of the tunnel face.

The comparisons in figure 8 were obtained using the above analysis technique.

Table 4 (in figure 8) shows that all non-electric perimeters gave a HCF of 49.7% versus all electronic perimeters with 57.6%. This result is taken in both the left and right side of the tunnel face and with delayed and instantaneous firing.

Table 5 compares the EBS instantaneous firing with the EBS delayed firing of perimeter in both sides of the tunnel. Overall instantaneous firing gave a HCF of 68.85% versus delay firing of 44.75%.

Table 6 gives a comparison of instantaneous versus delayed firing in the right hand side of the tunnel where the RMR was 71 (classified as ‘excellent’ rock). Instantaneous firing gave 82.6% HCF versus 63.1% HCF with delay firing.

Table 7 shows the difference between instant and delay firing in the left hand side of the tunnel where the RMR was 41 (classified as ‘good’). Instantaneous firing gave 55.1% HCF versus delay firing with 26.4% HCF.

From these results it can be concluded that electronic blasting system firing has a positive influence on perimeter control compared to NE firing. It is also concluded that instantaneous firing of perimeter yields better result than delayed firing. The HCF analysis technique confirmed those observations made in the field where an obvious difference between system was noticed.

Tunnel advance

Measurements were taken of tunnel advance for each firing. Unfortunately, after some disappointing results due to a change in geology, for both the bench mark blasts and the first electronic blast, more shotholes were added to the round. No direct conclusions can be drawn here, other than the addition of more shotholes is significant to face advance (figure 9).

Conclusions

There are a couple of good pointers to where Electronic Blasting Systems (EBS) may offer value to the Tunneller. The firing and measurements were taken in a production environment. This meant no special controls were taken opposite drill accuracy and charging. It could be said that the positive results are most powerful in that the sole change was initiation timing and accuracy.

Most obvious was the improvement to perimeter control. Also the fact that vibration prediction proved to be more reliable will have value to tunnellers. Often cautious limits are set for maximum charge weight to overcome variability in vibration results. Charge weight limitation usually results in shorter drill length and therefore reduced advance per cut. Being able to maximise advance for a given set of criteria is of enormous value.

The electronic timing designs were not optimised for vibration control. Opportunities exist to further refine these designs to give true single hole firing to minimise charge weight per delay.


ACKNOWLEDGMENTS

The authors would like to thank the following people for their contribution to this work
Duk-Young Kim, President of HongLim Co. Ltd for his vision for electronic blasting in Korea.
The Hanwha Blasting Engineering Team for their dedication and support in conducting ‘round the clock’ measurements at critical stages of the tunnelling cycle.
T&TI would also like to thank the International Society of Explosives Engineers (www.isee.org), Ohio, USA, for its permission to publish this version of the paper ‘Civil Tunnelling with Electronic Blasting Systems – Early Experiences from Korea’ originally presented at the 31st Annual Conference on Explosives and Blasting Technique, held in Orlando, 2005.

The Kang Chon 1 Tunnel Fig 1 – Conventional non-electronic timings with MS and LP series detonators Fig 2 – Generic electronic timing used in the blasts Change of rock quality in the face. The face is split almost vertically right down the middle, with the left hand side of the face in RMR = 47, and the right in RMR = 77 Table 1: Non-Electric vibration and airblast data Table 2: Electronic vibration and airblast data Fig 3 – Vibration trace electronic firing Fig 4 – Vibration trace non-electronic firing Fig 5 – Non-electric System Administrator Fig 7 – HCF (Half Cast Factor) Fig 6 – Electronic blasting system Table 3: Fragmentation analysis results Fig 8 – Tables 4, 5, 6 & 7 Fig 9 – Advance ratio comparison