Gee, turn of events saying you can't trust the DFS. Clearly hitting DFS targets is a problem in WA deposits and the simple answer is lack of MET testing and lack of scoping
One of the key variables in hitting DFS targets is resource grade and understanding your resource, which was PLS problem.
If you think about how a pegmatite forms, it is an igneous rock, so it is essentially an intrusion. It goes in between existing geological structures and there are differing zones in the pegmatite itself and the key to a good pegmatite system is differentiation been visible in the zones themselves (and width and depth and size helps that differentiation) - to understand what I am saying have a read of Greenbushes and its zonal structures in the pegamatite system, and especially the spodumene aspect of that pegmatite system
http://www.portergeo.com.au/database/mineinfo.asp?mineid=mn508.
The thicker the pegmatite, then the zones are likely to be more apparent in the pegmatite itself and become wider and the interplay - read impurities with many actually happening in other zones, meaning they are not attached to the spodumene - so are more easily removed (because they are not essentially in the lattice of the spodumene itself) a
nd/or do not appear significantly in the spodumene layering itself zone (i.e. look at the diagrams in figure 4 below embedded post).
But where the pegmatite isn't thick the zones are likely (not always) to interplay with the (smaller) spodumene zones and that becomes the issue with a lot of the hard rock plays, and why greater reliance on floatation is required in spodumene deposits by and large to hit stated 80% recovery rates, but the quality of deposits is really demonstrated around what element of that deposit is amenable to DMS (because only the floatation stage really removes impurities in a process flowsheet, hence if your crystals are too fine or there are impurities in the lattice of the spodumene then DMS cannot be used realistically).
The spodumene zone has your typical easily removed gangue. but if the zones are not spread out well then you get your (greater) issues around mica and Fe203 especially, even though these two can and do present in the spodumene zone as well. The two key impurities are essentially fluorine (hence why essentially the concerns with mica IMO) and Fe203. If not properly removed they end up having a clinkers effect in downstream processing (i.e. conversion to lithium chemicals) and in effect reduce recovery rates of spodumene in producing LCE, and that translates to a lower purchase price for you. Explained further in an embedded post below:
Figure 4 of the attached especially give the idea - especially read and understand what the term homogeneous means in a zonal structure.
https://www.researchgate.net/public...ent_granitic_pegmatites_Northwest_Territories
Going to AVZ, from this article you get a key to pegmatiate structures and what DMS actually means -
http://www.meadegunnell.com/assets/PDF/17-003LithiumHardrock.pdf:
"In summary, DMS can be an effective means to pre-concentrate spodumene from hard rock operations with the benefit of less feed to the grinding and flotation circuits and subsequent capital, energy and operating cost savings. However, the effectiveness of DMS is highly dependent on the degree of spodumene liberation at coarse particle sizes. Poor spodumene liberation will result in significant lithium losses to the float product and negate economic justification for its use."
In liberating spodumene note that mica and Fe203 really cannot be removed from the spodumene itself at the DMS stage, so liberation at that stage is also where they are unattached (and that is how DMS is viable at that stage been large crystals and low deleterious elements in the spodumene itself). Certainly at floatation you can remove impurities but IMO you don't want to be significantly reliant on floatation to achieve 80% recovery rates IMO (note: AVZ will retrofit floatation and based on previous DMS Ann, IMO over half the more than 80 percentage point recovery rate because of the high DMS rate likely with floatation will come from DMS btw).
You may have read some posts of mine harping on about AVZ been able to get more DMS product than other lithium plays - well the answer is the pegmatite itself been homogeneous and less need for floatation - Post #:
40459181. Your cost overruns are more likely when you are solely reliant on floatation is the point and your ore feed is not high width and length and depth, and that is why the other day I said on LTR I was surprised they were considering that option, albeit I suspect they will settle on DMS/floatation - Post #:
42977372.
The other element to potential cost overruns is compactness of resource - again the AVZ resource is very compact evidenced by the fact AVZ has a substantially higher indicated and measured resource compared to other lithium plays, despite undertaking far less drilling. Refer: Post #:
43048559
If your mine plan footprint is compact and strip ratios are low there is also less possibility IMO of the types of cost overruns at minesite you have seen in opex for PLS and AJM (not withstanding there greater need for floatation itself, especially the former where the capex wasn't scoped properly scoped thus impacting floatation recoveries). Obviously transport risk and cost overruns can occur and are a definate risks for AVZ, so the DFS will need to scope that well, but what reduces transport risk IMO is rail options (as you need less transport movements than a road option, so you want more rail than road in AVZ's transport solution IMO).
Ultimately you can't have a DFS that others are confident in plus or minus 25%, but as I said the other day apart from sovereign risk in the DRC, Australia has given evidence itself that risks of meeting opex and capex costs estimates are evident in operating in Australia btw, (probably as a result of the resource itself for PLS and need for more capex than typical processes especially and/or given the regulatory environment in Australia itself IMO - LNG cost overruns are the examples generally used to show that btw). Refer: Post #:
43031668
So if you want to talk about departing from DFS variables, likely less of a problem for AVZ at minesite - - note transport is a post minesite cost - because of the pegmatite structure and high width and depth making zones more definitive and isolated in a pegmatite structure (and government commitment to the project itself IMO).
Take PLS, I suspect by not reporting their METs properly they are having difficult in dealing with the the flourine in the mica (and probably Fe203 as well) especially, (given thiner layerings of its pegamtite system and how mica works and forms in the 'zones of spodumene pegamatites' and interplay. That is IMO without any evidence btw, but just a thought in the absence of the METs data.
Then because of the interplay - meaning the rock structures zones go into the feed - they are probably also having difficulty in configurating the SAG and ball mill (which are the two units before floatation) hence the need for more capex spend there (i.e. the capex spend is targetting the floatation end from recollection). Given floatation requires p80 of 106 microns (against DMS which is essentially in mm - been larger ore feed size arising from the HGPR unit in the process flowsheet) if your ore also has variability that can be an issue as well - mica schist, if it is in the ore especially, especially IMO given it is flakey and could clog the SAG/Ball Mill efficient operation if not adequately addresses before the ball milling stage for example IMO.
The point is the better your resource and availability of DMS, as compared to floatation, the less likely IMO you are going to have problems with your DFS targets, given your less reliance on floatation especially in achieveing an 80% recovery rate. For others I suggest revist this Ann by AVZ which explains impurity levels etc - a good read for new investors here (Post #:
41005358 - page
https://hotcopper.com.au/threads/ann-shareholder-letter.5025479/
With this post I will take a break from posting on AVZ until an Ann arises again. And let others try to provide insights and share information as I am getting sick of my own voice on these threads today LOL and the past week.
For avoidance of doubt am hypothesing as well above, severely hypothesing above, as a number of the spodumene players haven't given full detailed METs to make judgements from, and therefore all you have is the pegmatite layering widths in their Anns to hypothesise what that composition maybe in that pegmatite. Using information on what a pegmatite looks like (refer back to Greenbushes example and do a bit of google and read relevant articles on pegmatite formations and how they look etc etc from outer wall to the other outer wall etc etc.
All IMO and time to sign out and drnk some VB
All IMO