TSD is a dynamic and forward-thinking enterprise specializing in mining engineering, metal trading, and corporate commercial brokerage services.

Dubai, United Arab Emirates
Joined November 2025
Recovery shortfalls are rarely a flotation problem. Before touching reagents, we reconcile four things in sequence: t/h against design, P80 against design grind, mill head grade against block model, and weightometer calibration first. #Mining tsdconsultancy.ae
2
Two competent persons can classify one drillhole database differently — both defensible. The gap is rarely grade. It is variogram range versus drill spacing, and whether JORC 2012 Table 1 was documented or asserted. #JORC2012 tsdconsultancy.ae
Before inspection the file reads well: a factor of safety per sector. After: as-built geometry vs design, logged piezometer data vs model inputs, deposition vs permitted raise rate. Does the file hold? #GISTM tsdconsultancy.ae
1
A 72-hour performance test proves a plant can start. It does not prove it can run. Reconcile the test feed's Bond work index with the LOM composite, and read surge capacity in hours, not tonnes. #DueDiligence tsdconsultancy.ae
6
Feasibility is an accuracy claim, not a document class. We test the band before the NPV: 1) priced take-offs vs carried factors, 2) contingency ≠ scope growth, 3) escalation separated. Worth pressure-testing before close? #Mining tsdconsultancy.ae
3
Vendor qualification is usually decided in the first hour. Three checks: effective date versus tonnes mined since, the code applied verbatim (JORC 2012, NI 43-101), and the schedule re-run with Inferred set to zero. #MiningFinance tsdconsultancy.ae
1
In vendor DD, open the royalty returns before the production report. Both count the same tonnes; only one was prepared for a party with audit rights. Gaps usually sit in moisture basis, payable factor or NSR deductions. #MiningDD tsdconsultancy.ae
10
Permits are issued against a flowsheet, not an orebody. If locked-cycle testwork, the lodged reagent schedule and the EPC design freeze drift by more than one study cycle, the permit no longer describes the plant. #mining tsdconsultancy.ae
5
Recovery shortfalls are an accounting problem until proven otherwise. Close the balance first: weightometer vs shipped tonnes, then the two-product formula. Is the figure monthly accounting, or metal paid at the shipment? #Metallurgy tsdconsultancy.ae
2
Illustrative: every reference material in the QA/QC log sits in one grade range, none near the cut-off that drives the pit. The database looks clean; the Indicated tonnes are untested. Worth pressure-testing before close? #JORC tsdconsultancy.ae
4
Inspectors rarely argue with a stated 1.3 static factor of safety. They test the instruments behind it. A register listing 54 piezometers with 40 returning data isn't a design dispute — it's a monitoring gap. #GISTM tsdconsultancy.ae
3
Plant availability is usually the first assumption to fail on the crusher floor. A model at 92% means little if stoppages under 15 minutes are logged "minor" and excluded. Define the term before accepting the number. #DueDiligence tsdconsultancy.ae
2
Feasibility studies rarely fail inside a chapter. They fail at the handover. Control point one: does the Reserve declared under JORC 2012 run the same cut-off grade, schedule and price deck as the financial model? #Mining tsdconsultancy.ae
4
A vendor pack can be current by date and stale by basis. If the mine plan was rebuilt twice since the NI 43-101 report, the warranty schedule is drafted against the wrong one. Worth pressure-testing before close? #MiningFinance tsdconsultancy.ae
4
Hypothetical: a vendor pack shows 61,400 dry tonnes at loadport; royalty statements show 60,100 at final settlement weights. Neither figure is wrong. The revenue model reconciles with neither. Pull the royalty deed. #MiningDD tsdconsultancy.ae
A depressant swap that lifts recovery also moves effluent chemistry. The locked-cycle result is defensible; the discharge consent still names the original reagent suite. Is this risk on your permit timeline? #MiningFinance tsdconsultancy.ae
A mass balance that closes every month may be closing by definition: tails grade back-calculated via the two-product formula, never cut from a full stream. Design 89%, accounting 85.6% — which number is measured? #Metallurgy tsdconsultancy.ae
2
Same drillhole database, two Competent Persons: Indicated on 50 x 50 m spacing in one estimate, Inferred in the next. Geology did not change — the evidence accepted for grade continuity did. Which one holds at close? #JORC tsdconsultancy.ae
1
Before inspection: a stated inter-ramp angle, a monitoring plan annexed. After the walkover: which piezometers read that week, how many depressurisation wells are commissioned rather than drilled. #geotechnical Tested in yours? tsdconsultancy.ae
2
92% mechanical availability, carried flat from year one to fifteen, is a modelling choice, not a plant. Availability is not utilisation: process-limited idle hours rarely reach the spreadsheet. Is this tested in your feasibility? #MiningFinance tsdconsultancy.ae
1