Originally posted by eshmun
What a load of rubbish. The competent person, unlike you, has explained that the dd holes are more representative and has given reasons. The company was constrained by time, equipment and the pressure of the milestones imposed by the opportunistic takeover offer which was a massive distraction that rushed the processes and has influenced the results as explained.
Here is what has been said about the sampling. If you can read, then read it, if you need help give me your address and I can send my 10 year old over to help. Esh
“Samples collected by the drill hammer were delivered to a Metzke Splitter for sub-splitting.
The splitters were specifically purchased for the 2017 Tampia hard-rock resource definition program as they provided a superior split over the more industry-standard cone splitter. However, at Mace, these devices had little positive effect, as sticking clays caused sample splitting issues.
Recoveries were generally low, with an average of 65% in samples over 0.2 g/t Au. To evaluate the sample quality and determine its fitness for the purpose of resource estimation, results of 14 diamond drilling control samples were investigated.
The diamond core was of PQ size and collared as close to existing RC holes as possible. Holes were drilled along the long axis of the deposit, thereby providing a representative sub-set of data for comparison. Samples were submitted to the laboratory as whole-metre intervals, where they were crushed, pulverised and leached. The results of samples within a 5m buffer of RC samples, with diamond core recoveries of more than 80%, and within the low-grade mineralisation domain were compared using a quantile-quantile plot (Figure 9). This resulted in 55 sample pairs, sufficient for statistical comparison.
The results show that the RC results are biased low above approximately 0.6 g/t Au, with the diamond control samples being higher on average than the RC drilling. The means of the two data sets (with diamond data set capped at 63 g/t, the maximum grade in RC population) are markedly different (9.4 g/t for diamond and 4.1 g/t for RC), which is due to the several high-grade ‘outliers’ in the diamond control data set that are not present in the RC dataset.
It is clear that the RC drilling does not efficiently recover coarse gold, likely due to a combination of drilling type, splitting process and analytical process, whereas in the diamond core drilling and sampling, apart from perhaps minor plucking effects, all gold in the primary sample is reported in the final assay. Sonic drilling, followed by whole-sample leaching would be the best drilling and sampling method for this type of mineralisation and host rock.
Chutes on the splitters were adjusted to deliver a maximum-size sample split to the laboratory, in the order of 3 – 4 kg. This weight was assumed to be fit-for-purpose, rather than determined by a nomogram, as there was little to no pre-existing sample data for Mace available. The coarse nature of the gold does not make this a suitable split size, which is reflected by the diamond control sample results. The performance of splitting was monitored on a per-sample basis by collecting a duplicate split sample for each metre. The difference in sample weight acted as a proxy for sample split consistency, which was monitored in a spread sheet in real-time. Reasonable precision was demonstrated form the analysis of duplicate primary sample splits (Figure 10). The Competent Person audited this performance throughout the campaign and, apart from minor issues, deems the sample splits of reasonable quality and, following comparison with diamond drilling control samples, marginally fit for the purpose of resource estimation.
The sample splits were submitted for the same sample preparation process as the samples for the Tampia hard-rock deposit. Samples were weighed wet (on site) and wet (at the laboratory), and then split in a Rocklabs Boyd RSD Combo, which allowed a percentage linear split to be specified for each sample. The split weights were optimised for pulverising in Essa LM-2s and their percentage passing size monitored consistently. Samples were then milled in the LM-2s before a manual split of around 200 g was put in brown paper bags. The final 50 g charge weight was weighed from this. Duplicate samples were inserted at each of the splitting stages, to monitor precision. Duplicates were collected at >10% from mineralised zones only. Samples that were duplicated in the field were also flagged for duplication at subsequent splitting stages at the laboratory. The Competent Person notes that such a splitting process is not necessarily optimal for this style of mineralisation but considered acceptable for the purpose of mineral resource estimation and appropriate classification.
Duplicate results for primary splits, secondary (crush) splits and pulp splits were monitored, and no significant bias was noted in the splitting processes.
In the Competent Person's opinion, the sampling and sub-sampling were not always accurate (leading to lower-grade bias), and often imprecise. However, based primarily on the results of the diamond control drilling, the methods are considered marginally fit for the purpose of resource estimation. The likely under-reporting of Au content provides a potentially significant upside to the project but can only be proved with better-quality drilling or mining.”
hey esh on further reading i see there is one instance where the RC width to mineralisation is 3m longer than the twin diamond hole - so it amy not have been the sampling above ground but smearing/inability to clean out the hle properly of gold in each meter, hence the lower grade /meter and the wider zones - and in the instance they commented on - the grade/meters between the RC and twin diamond hole were nearly identical to each other!!! - so potentially no amount of duplicates reassaying etc as i suggested would solve the problem - it has happened at the hammer face and one cannot do much about it now??!!
