【0207】7萬(wàn)能材料試驗(yàn)機(jī)CAD,0207,萬(wàn)能,材料,試驗(yàn),實(shí)驗(yàn),cad Batu Hijau——七年的運(yùn)行與持續(xù)發(fā)展 Batu Hijau從1999年開(kāi)始投產(chǎn)運(yùn)營(yíng)，從2001年開(kāi)始不斷致力于改進(jìn)系統(tǒng)的形式,以使其品牌生產(chǎn)能力每天超過(guò)12萬(wàn)噸。通過(guò)由礦山選礦廠，使廠礦集中，并在礦石處理中不斷改善突破生產(chǎn)瓶頸。 本文所要討論的改進(jìn)措施主要包括以下幾個(gè)方面： ●擴(kuò)建了磨凹陷破碎卵石、混合電路 ●增強(qiáng)燃爆和混合程度 ●碾磨有效性 ●凹陷磨襯板、壁爐、除漿升降機(jī)的優(yōu)化 ●磨機(jī)鋼球大小和負(fù)載 ●氣旋和氣旋水泵改良升級(jí) ●升級(jí)過(guò)程控制 ●成本控制措施 緒論 Batu Hijau1999年9月委托制作的品牌容量每日120噸碾磨機(jī)其可靠性達(dá)到92％。家的粉碎巡回是由最初的回旋決定的，直徑36尺寸的SABC凹陷碾磨機(jī)由陸上運(yùn)輸和貯存到供給兩條平行的磨合線構(gòu)成，直徑20英寸的小鋼球和1000個(gè)小圓石排進(jìn)行撞擊。McClaren et. al.描述設(shè)計(jì)時(shí)說(shuō)到，建筑和第一個(gè)兩年運(yùn)作重在大的細(xì)節(jié), 不會(huì)再修改實(shí)施文件，除非為今后斷續(xù)改善提供相關(guān)資料。第一年運(yùn)作由于被很多設(shè)備的可靠性，實(shí)用性困擾，其生產(chǎn)能力相對(duì)于品牌產(chǎn)品低了一些。在2001年間通過(guò)改進(jìn)操作和保養(yǎng)程序以及許多小型設(shè)備的改進(jìn)使設(shè)備的可靠性和碾磨生產(chǎn)能力得到提高。 在這段時(shí)間里由于銅和黃金價(jià)格降低對(duì)Batu Hijau生產(chǎn)率和營(yíng)業(yè)成本的提高產(chǎn)生重大壓力。通過(guò)2001 年戰(zhàn)略經(jīng)營(yíng)的回顧得出，通過(guò)消除薄弱生產(chǎn)加工環(huán)節(jié)達(dá)到了提高生產(chǎn)率的目的。從總體上增加單位技術(shù)含量并降低相對(duì)成本支出，抓住這個(gè)時(shí)機(jī)將會(huì)在未來(lái)幾年內(nèi)獲得更多利潤(rùn)。這些論述在文件里有詳細(xì)描述。 生產(chǎn)歷史 由于操作程序改進(jìn)和礦石堅(jiān)硬程度具有易變性，Batu Hijau處理生產(chǎn)率也是在改變的。不過(guò)技術(shù)改進(jìn)對(duì)礦石堅(jiān)硬增量的彌補(bǔ)超過(guò)了過(guò)去幾年，允許操作持續(xù)超出的設(shè)計(jì)容量(參見(jiàn)圖1)。 圖1: 2000－2006 生產(chǎn)率和混合物 銅和金生產(chǎn)率如下圖2所示，并且隨季節(jié)變化逐步采用介入開(kāi)采的采礦計(jì)劃，旱季期間從坑的底部，雨季開(kāi)始再?gòu)目拥撞炕氐缴线?。磨料成分中銅可能從干季的少于0.5％到濕季的超過(guò)0.9％。這些季節(jié)性波動(dòng)使清潔器浮選精礦和浮選精礦處理操作時(shí)在高干旱季節(jié)里存在很大難度。 圖2: 銅等級(jí)的季節(jié)性變動(dòng) 漂流碾磨小卵石的擴(kuò)展擊碎流程 小卵石擊碎電路最初被安裝在Batu Hijau (下頁(yè), 圖3) 被許多操作程序干擾, 其維護(hù)困難和容量問(wèn)題如McClaren (2001)所描述。一個(gè)重大操作問(wèn)題是有限的超負(fù)荷能力導(dǎo)致流入混合的頻繁溢出，要求直接裝貨人員親自清潔。在2000年末對(duì)溢出滑道進(jìn)行了修改，傳動(dòng)機(jī)被安裝在回歸容器溢出處，直接傳送混合物到下陷磨房，消除了對(duì)裝貨人員自行清潔的要求。 小圓石傳送路線包括垂直傳送的鏟斗，將巖石投入設(shè)備過(guò)程中對(duì)其造成頻繁損傷，并且鏟斗產(chǎn)生頻繁顫動(dòng)，從而引起整體安全，維護(hù)和清潔問(wèn)題。2000年和2001年打擊襯墊過(guò)早破壞主要?dú)w因于這些桶傳送機(jī)和鋼珠殘缺不全，這些缺陷不能解決之前，小卵石擊碎過(guò)程有效性有待提高。2001年正式批準(zhǔn)推翻原方案，全部按傳統(tǒng)傳送小圓石裝置重新完成傳送裝置設(shè)計(jì)和壓碎器設(shè)施設(shè)計(jì)。傳送機(jī) 圖3: 原始的Batu Hijau 粉碎流程圖 系統(tǒng)包括四塊大的Eriez傳送帶磁體，其中兩塊在交叉帶子處，另兩塊被安裝在傳動(dòng)機(jī)調(diào)動(dòng)點(diǎn)以改進(jìn)鋼珠移除方式。該設(shè)施已經(jīng)完成并且現(xiàn)有的兩臺(tái)小卵石壓碎器開(kāi)2002年2月被調(diào)遷到現(xiàn)場(chǎng)固定安裝。該設(shè)施改善了小卵石線路的實(shí)用性以及名義上的碾碎率，其效率更高超過(guò)每天5000噸。然而鋼球從壓碎器移除時(shí)由于數(shù)量很多，碾碎筒篩相對(duì)磨房料庫(kù)過(guò)大而有反回的可能，所以這仍然是一個(gè)有待解決的問(wèn)題。典型大小20－30mm的多邊形廢棄物球很快通過(guò)漂流到達(dá)下陷碾磨機(jī)礦石篩產(chǎn)生巨大的爆炸聲。如果這些小鋼珠被理沒(méi)在傳動(dòng)機(jī)上，那么他們將不能被有效地吸附在磁鐵上。磁鐵運(yùn)轉(zhuǎn)更遠(yuǎn)處并且在附近存在磁鐵礦，那么磁鐵的磁性就將會(huì)被削弱了。 金屬探測(cè)器作為壓碎機(jī)進(jìn)料器的最后防線能夠有效的使鋼珠保留在壓碎器處面。設(shè)備運(yùn)行中要求操作人員能夠?qū)核槠鞯慕饘偬綔y(cè)器可能出現(xiàn)的差錯(cuò)作出反應(yīng)，并在小卵石中找出被埋沒(méi)的鋼珠。延長(zhǎng)初級(jí)壓碎器的維護(hù)周期和適當(dāng)?shù)男菝邔?duì)于貯存并迅速進(jìn)入球磨房進(jìn)料器很有必要。每隔幾分鐘金屬探測(cè)器運(yùn)行到壓碎機(jī)進(jìn)料器時(shí)小卵石巡回導(dǎo)致小卵石壓碎器變得不能動(dòng)作直到主壓碎器在聯(lián)機(jī)控制下使探測(cè)器返回。 2003年期間對(duì)球磨機(jī)作了更改使礦石篩和廢棄物隔離，將廢棄物帶到垃圾填埋場(chǎng)，并使噴涂劑用量減少，這樣消除了由于小卵石撞擊引起的損傷。 粉碎回路的容量傳統(tǒng)上受到磨碎機(jī)或小卵石的限制?；仡?002年消除薄弱環(huán)節(jié)的方案，使其吞吐容量增加到10000噸，10mm以下的小卵石直接進(jìn)入磨碎機(jī)，10mm以上的小卵石再次被送回進(jìn)行磨碎，這樣磨碎機(jī)基本上處于開(kāi)放的狀態(tài)。工廠條件下試用結(jié)果證明產(chǎn)品改進(jìn)將使小卵石不再循環(huán)時(shí)磨碎機(jī)效率提高，對(duì)于新進(jìn)原料每一噸被擊碎的小卵石中有0.5到0.6噸沒(méi)被回收到自動(dòng)磨碎機(jī)進(jìn)料器中。改進(jìn)后的流程圖如圖4（下頁(yè)），需要增加第四臺(tái)MP1000壓碎器（在2003年3月完成），并且在7.3m振動(dòng)篩基礎(chǔ)上制成3.6mm振動(dòng)篩（2003年7月完成）。 由于消除了從磨碎機(jī)回收擊碎的小卵石，使初始小卵石的生產(chǎn)率下降了20%。把磨碎機(jī)開(kāi)口從60mm增大到80mm以提高小卵石的生產(chǎn)率使其達(dá)到先前水平。由于礦石篩自重問(wèn)題，小卵石在機(jī)械和操作上遇到了一些困難。礦石篩機(jī)械件的高度磨損，要求另處的預(yù)防性維護(hù)，其篩選強(qiáng)度降低，礦石篩篩選效率同時(shí)降低。由于使小卵石進(jìn)入氣旋產(chǎn)生強(qiáng)烈擠壓，旋風(fēng)進(jìn)給泵和水力旋流器襯墊磨損率大幅上升。 小卵石壓碎器的運(yùn)行改進(jìn)使有效性從少于60%到超過(guò)80%，安裝的馬達(dá)由于更加均勻的進(jìn)料和礦石篩重新布置最大負(fù)荷提高其平均壓碎器牽引力從原來(lái)的60%提高到80%，在磨碎機(jī)礦石篩卸料斜槽安置靜態(tài)的篩子以從壓碎器進(jìn)料中除去鋼珠，改善從磨碎機(jī)小卵石脫水情況。另外由于早期鋼珠的損傷使壓碎器襯墊疲勞直至破裂。 圖4：提高開(kāi)礦爆破和混合力度 磨碎機(jī)進(jìn)料大小對(duì)生產(chǎn)率產(chǎn)生很大影響，2006年較有影響的論述是被(RQD)收錄的一篇論述，其磨碎機(jī)進(jìn)料大小和磨碎機(jī)生產(chǎn)率關(guān)系如圖5（下頁(yè)）。2002年分裂照相機(jī)設(shè)施被安置在磨碎機(jī)傳送器上，允許發(fā)展改進(jìn)的礦石特性描述手段，針對(duì)巖石類(lèi)型設(shè)計(jì)爆破方案，從而減少了流入磨碎機(jī)進(jìn)料器石塊大小的可變性。在2004年改進(jìn)過(guò)的標(biāo)準(zhǔn)爆破方案進(jìn)入實(shí)踐操作階段，在堅(jiān)硬的巖石中使用高濃度火藥粉，在軟的斷裂的巖石中用低濃度火藥粉。 在最堅(jiān)硬的礦石方面生產(chǎn)量提高10－15%的預(yù)期計(jì)劃2004年實(shí)踐操作證實(shí)了該數(shù)據(jù)。由于礦石混合注入進(jìn)料器，新爆破技術(shù)的應(yīng)用，生產(chǎn)率相比年計(jì)劃提高了2－7%。 礦石類(lèi)型特性描述技術(shù)的改進(jìn)使磨碎機(jī)礦石生產(chǎn)能夠預(yù)期交付，其混合優(yōu)化也使磨碎機(jī)的生產(chǎn)量和金屬生產(chǎn)率得到最大程度的提高。 圖5：RQD Effect on SAG feed size and SAG Throughput, afterBurger et al (2006) 磨碎機(jī)在有效性方面改進(jìn)之后，其壽命可靠性由設(shè)計(jì)的92%提高到94-95%超過(guò)以往的幾年。這些進(jìn)步是改善維護(hù)計(jì)劃施行維護(hù)計(jì)劃，改進(jìn)碾碎組分，輔助設(shè)備設(shè)計(jì)及某一設(shè)備升級(jí)的結(jié)果。主要改善是 l 2004年磨碎機(jī)重新布置同時(shí)在巡回路線上提供更加頻繁的維護(hù)機(jī)會(huì)，使整個(gè)線路更加整潔。每年跟蹤管道停機(jī)48小時(shí)維護(hù)一條磨碎生產(chǎn)線，另一條線也同時(shí)停機(jī)維護(hù)。 l 將氣旋進(jìn)給泵從GIW24/22升級(jí)到Krebs28/26，同時(shí)在2004年對(duì)暖泵進(jìn)行例行維護(hù)的時(shí)間由原來(lái)的五個(gè)星期延長(zhǎng)到十二個(gè)到十三個(gè)星期。 l Batu Hijau在每個(gè)球磨機(jī)只安置一臺(tái)氣旋進(jìn)給泵，減輕本身重量，減輕振動(dòng)引起的破壞，停工維護(hù)以級(jí)磨碎機(jī)的不斷改進(jìn)提高了工作有效性超過(guò)0.5%。 l 從2001年開(kāi)始橡膠式礦石篩被安裝在磨碎機(jī)和球磨機(jī)面板上，面板設(shè)計(jì)中擴(kuò)大了磨損范圍，使最大磨損區(qū)的使用壽命從原來(lái)的2-3個(gè)星期延長(zhǎng)到12個(gè)星期。其中聚氨酯面板的磨損貢獻(xiàn)數(shù)幾乎是1%不計(jì)算在內(nèi)。 l 2001年停工檢修期間發(fā)現(xiàn)大型卵石和碾磨球進(jìn)入旋風(fēng)進(jìn)給泵和氣旋，因此適當(dāng)?shù)念A(yù)防性維護(hù)成功的避免的重大事故的發(fā)生。 l 2002年改進(jìn)小卵石巡回路，從小卵石回路中除去了垂直的桶傳送機(jī)，提高小卵石可用性到90% 。 l 在磨碎機(jī)上安裝了100mm 厚的橡膠襯里塊和陶瓷磚，放電洗滌槽和廢水坑，減少了修理這些設(shè)備的停工頻率，同時(shí)也減少了排除泄漏的無(wú)計(jì)劃停工維護(hù)時(shí)間。 圖6：Annual Total Mill Availability 磨碎機(jī)襯墊，壁爐和漿液舉升機(jī)的設(shè)計(jì)優(yōu)化 磨碎機(jī)襯墊設(shè)計(jì)－試驗(yàn)和檢驗(yàn) 原始的磨碎機(jī)襯墊的設(shè)計(jì)在Batu Hijau上是72行頂部呈帽子形式的小凸起，并與齒面角成12度。選擇襯墊和舉升機(jī)的最優(yōu)組合并在磨碎機(jī)上進(jìn)行測(cè)試，最大限度的提高磨碎機(jī)生產(chǎn)率延長(zhǎng)襯墊使用壽命。對(duì)兩臺(tái)磨碎機(jī)送入同樣物料，保證在一定時(shí)間內(nèi)同樣條件下接收到相同進(jìn)料，性能對(duì)比表明，并不受貯存分離程度的影響。McClaren et. al. (2001)所記述的48行的帽形襯墊設(shè)計(jì)與72行的襯墊相比其生產(chǎn)率和磨損率相去甚遠(yuǎn)。與MEI制造廠商合作設(shè)計(jì)出36行的舉升機(jī)和襯墊并于2001年投入使用，使磨碎機(jī)在其生命周期內(nèi)的生產(chǎn)率提高了2－3%。一些根據(jù)技術(shù)細(xì)則手冊(cè)設(shè)計(jì)的36條墊板的并發(fā)的產(chǎn)生,在生產(chǎn)率上沒(méi)有和第一個(gè)測(cè)驗(yàn)中所顯示的那樣表現(xiàn)出同樣的進(jìn)步.磨機(jī)的性能和72條機(jī)構(gòu)一樣。36條設(shè)計(jì)繼續(xù)遭受墊板的過(guò)早破損,導(dǎo)致了在墊板壽命的最后一到兩個(gè)月里未意料到的停工期。 磨機(jī)墊板設(shè)計(jì)的DEM模型說(shuō)明代25度齒面角提升機(jī)構(gòu)的36條提升機(jī)構(gòu)和盤(pán)狀墊板能提高生產(chǎn)率.在2002年12月這套墊板的安裝沒(méi)有顯示出明顯生產(chǎn)率的提升。這套墊板過(guò)早破損而迫使在使用了僅僅三個(gè)月之后就要去除這些墊板。 36條墊板在運(yùn)行中引起了一些問(wèn)題.在使用一樣的卸料篩,裝球量和充料速度下與72條墊板設(shè)計(jì)相比36條墊板平均形成了多出50%的卵石.由于卵石的增加卵石非常臟,惡化了卵石循環(huán)輸送和低下的卵石碰撞性能所帶來(lái)的問(wèn)題.在精細(xì)的礦石填充的情況下36條設(shè)計(jì)磨機(jī)不能有效的提升,因?yàn)楸?2條設(shè)計(jì)磨機(jī)更低的生產(chǎn)率.而在粗糙礦石填充的情況下,36條設(shè)計(jì)比72條墊板設(shè)計(jì)更勝一籌。 36條墊板設(shè)計(jì)中的墊板盤(pán)的過(guò)早破損和涉及到在生產(chǎn)率沒(méi)有任何改善下卵石循環(huán)性能的運(yùn)行問(wèn)題的結(jié)合,導(dǎo)致了2005年兩種磨機(jī)72條設(shè)計(jì)被重新使用的決定。72條有15度的提升機(jī)構(gòu)齒面角,并且別加強(qiáng)來(lái)增加墊板總磨損壽命。將來(lái)對(duì)于墊板設(shè)計(jì)的修正把目標(biāo)定在在不影響生產(chǎn)率的情況下增加磨損壽命。 SAG磨機(jī)篩和礦漿提升機(jī)構(gòu) SAG磨機(jī)篩和礦漿提升機(jī)構(gòu)以及卸料機(jī)構(gòu)一直在經(jīng)歷著為提高可維護(hù)性,卵石制造率和磨機(jī)卸料率方面的改進(jìn)。篩孔從一開(kāi)始就在穩(wěn)定的增加。在2000年一個(gè)25毫米和60毫米結(jié)合篩發(fā)展到2001年40毫米和60毫米結(jié)合篩.到2002年所有60毫米篩的使用為維持卵石制造和磨機(jī)生產(chǎn)率成為了必要。卵石碰撞篩的安裝之后90毫米篩試驗(yàn)不久便開(kāi)始。當(dāng)前篩結(jié)構(gòu)使用的是比80毫米篩更均衡的80毫米和60毫米的結(jié)合篩.三分之一篩的替換維持了新和部分舊篩的混合和貫穿篩總壽命的平均篩孔和開(kāi)口區(qū)域。 以前所有篩的替換由于更低的開(kāi)口區(qū)域和篩孔尺寸最初減少了磨機(jī)生產(chǎn)率。 篩孔尺寸的每個(gè)改動(dòng)符合壓低卵石生成量的粗糙充料量和礦石硬度的的增加。隨著粗糙充料量降低更少卵石率,卵石生成率已經(jīng)典型的達(dá)到了15-22%。 隨著2002年磨機(jī)生長(zhǎng)率的提高, 因?yàn)槟C(jī)突然停止時(shí)觀測(cè)到的礦漿池的存在,SAG磨機(jī)卸料能力在非常高的生產(chǎn)率下變成了一個(gè)阻礙。研磨最后的完全卸料由Metso贊助進(jìn)行重新設(shè)計(jì)使篩與磨機(jī)前部分離,這樣是為了在靠近磨機(jī)出料口制造一個(gè)更深的礦漿提升機(jī)構(gòu)髓室和增加礦漿排出能力。篩磨損的觀測(cè)說(shuō)明部分內(nèi)篩很少在磨機(jī)上的礦石內(nèi)部充料機(jī)構(gòu)里所以對(duì)去除磨機(jī)內(nèi)礦漿或者卵石沒(méi)什么作用。然而從篩背面的磨損可以看出它帶來(lái)了礦漿明顯回流到滾筒的現(xiàn)象。外篩被延伸內(nèi)篩的剩下部分被封鎖。這種改變減少了在重新?lián)Q板停工期間需要替換的量,通過(guò)減少回流到研磨室的量提高了礦漿排出量。 研磨鋼球尺寸和填充 Batu Hijau半自動(dòng)磨機(jī)被設(shè)計(jì)來(lái)使用伴隨著12-14%125毫米鋼球的使用率的改變。2002年裝球量逐漸增長(zhǎng)到18%并且產(chǎn)生了研磨生產(chǎn)量上的明顯提高.2003年的一項(xiàng)試驗(yàn)中減少磨機(jī)中鋼球填充量到14%,證實(shí)了生產(chǎn)量5-10%的降低.鋼球填充量現(xiàn)在維持在16-18%。 133和140毫米大鋼球的試用最初旨在增加磨機(jī)產(chǎn)量但是可預(yù)測(cè)的影響沒(méi)被重視.但是隨鋼球尺寸的增加總鋼球消耗量減少了,隨著80毫米篩孔的采用這變得更加重要.現(xiàn)在的磨機(jī)普遍采用133毫米鋼球,鋼球消費(fèi)率為0.4kg/t。 主動(dòng)控制成本 Batu Hijau管理者從車(chē)間工廠實(shí)際出發(fā)，不斷促進(jìn)文化進(jìn)步并尋找公司新的理念。這些想法也許對(duì)于普通雇員實(shí)施起來(lái)很快很容易，但在管理者那里具體操作過(guò)程中就是要節(jié)省維修及其它費(fèi)用。實(shí)行對(duì)多種操作能力的員工的小褒獎(jiǎng)制度，引入持續(xù)競(jìng)爭(zhēng)的思想。 另外一些更加鮮明的主動(dòng)節(jié)省成本的描述包含在下面這個(gè)報(bào)告中： l 從磨碎機(jī)碾磨房廢棄物中回收鋼珠加入球磨房 l 在運(yùn)轉(zhuǎn)過(guò)程中下部機(jī)構(gòu)支撐區(qū)域保持動(dòng)力供給 l 選擇更加有效的試劑作反應(yīng)物 l 地方與全球供應(yīng)商同步預(yù)購(gòu) 結(jié)論 Batu Hijau磨碎機(jī)的連續(xù)改進(jìn)表明未來(lái)將設(shè)計(jì)生產(chǎn)量比過(guò)去幾年更加優(yōu)越的磨碎機(jī)。經(jīng)過(guò)對(duì)正常的商業(yè)案例的瓶頸和發(fā)展的系統(tǒng)鑒定，證明獲取重要過(guò)程收益的關(guān)鍵在于降低基礎(chǔ)設(shè)施成本。這些工程已經(jīng)改變了規(guī)格不等的設(shè)計(jì)和材料設(shè)備供應(yīng)商,為流程變革投入重要的工程和資本開(kāi)支。 參考書(shū)目 B. Burger, K. McCaffery, I. McGaffin, A. Jankovic, W.Valery, D. La Rosa, 2006, “Batu Hijau Model for Throughput Forecast, Mining and Milling Optimisation and Expansion Studies”, pp.461-479, Advances in Comminution, Edited by Komar S. Kawatra, SME, 2006. McClaren, D., Mitchell, J., Seidel, J., Lansdown, G., 2001, “The Design Start-up and Operation of the Batu Hijau Concentrator”, International Autogenous and Semautogenous Grinding Technology 2001, Volume IV of IV, pp, IV-316-335, Edited by Derek J. Barratt, Michael J. Allan, Andrew L. Mular, Pacific Advertising Printing & Graphics, Delta, BC, Canada, 2001 BATU HIJAU – SEVEN YEARS OF OPERATION AND CONTINUOUS IMPROVEMENT ABSTRACT Since the 1999 commissioning of the Batu Hijau concentrator, a systematic approach of continuous improvement has allowed the concentrator to exceed its name plate production capacity of 120,000 tonnes per day since 2001. Production improvements were achieved by de-bottlenecking the ore treatment process from the mine through the concentrator. This paper will discuss improvements in the following areas: ? Expansions of the SAG Mill pebble crushing circuit ? Enhanced mine blasting and blending efforts ? Mill availability ? SAG mill liner, grate and pulp lifter optimisation ? Mill ball size and load ? Cyclone feed pump and cyclone upgrades ? Process control upgrades ? Cost control initiatives INTRODUCTION The Batu Hijau Concentrator was commissioned in September 1999 with a nameplate capacity of 120,000 tonnes per day at 92% mill availability.The comminution circuit comprised primary gyratory crushing, overland conveying and stockpiling to feed two parallel grinding lines in SABC configuration with one 36 foot diameter SAG mill, two 20 foot diameter ball mills and one MP1000 pebble crusher per line. McClaren et. al.(2001) described the design, construction and first two years of operation in significant detail and will not be revisited in this paper except to provide context for subsequent improvements. The first year of operation was plagued with numerous equipment availability issues and less than name plate throughput rates. By mid 2001 equipment availability and mill throughput rates had improved to design levels through improved operating and maintenance practices and numerous minor equipment modifications. Around this time copper and gold prices were depressed which added significant pressure to production rates and operating costs at Batu Hijau. A strategic operating review was conducted in 2001 to identify de-bottlenecking opportunities from the mine through the process plant with the aim of increasing production rates. The study concluded that opportunity existed for significant throughput increases from a range of incremental unit improvements at relatively low capital expenditure.These opportunities were captured on a master list and tackled in turn over subsequent years. The major initiatives are described in the body of this paper. PRODUCTION HISTORY Process production rates have been variable over the life of Batu Hijau due to process improvements and variable ore hardness. Progressive increases in ore hardness have offset throughput gains over the past several years but have allowed the operation to continue to exceed design capacity (see Figure 1). Copper and gold production rates, as shown in Figure 2, have also varied due to seasonal phasing of mining plans which involve mining as much high grade from the bottom of the pit during the dry season but then pulling out of the pit bottom at the start of the wet season. Mill feed grades can vary from less than 0.5% Cu to more than 0.9% Cu between wet and dry seasons. These seasonal fluctuations have placed pressure on cleaner flotation and concentrate handling capacity during the high grade dry season periods. EXPANSIONS OF THE SAG MILL PEBBLE CRUSHING CIRCUIT The pebble crushing circuit originally installed at Batu Hijau (next page,Figure 3) was plagued by a number of operating, maintenance and capacity issues as described by McClaren (2001). A significant operational problem related to limited surge capacity in the crusher feed bin producing frequent overflows of the surge bin, requiring immediate loader cleanup. In late 2000 the overflow chute was modified and conveyors installed to return bin overflow directly to the SAG mill feed belts to eliminate the loader rehandle requirement. The pebble conveying circuit included vertical bucket conveyors which generated a range of safety, maintenance and cleanup problems from the bucket conveyors flinging rocks into the plant and frequent damage to the bucker conveyor belts. Poor pebble crushing circuit availability in 2000 and 2001 can be largely attributed to these bucket conveyors and incomplete removal of steel balls from the pebble crusher feed resulting in premature failure of crusher liners. In 2001 approval was given for addition of one crusher and complete redesign and installation of conventional conveyors to transport pebbles. to and from the new crusher installation. The conveyor system included four large Eriez belt magnets two in cross belt and two installed on conveyor transfer points to improve steel ball removal. The installation was completed and the existing two pebble crushers relocated to the new crusher building in February 2002. This installation improved pebble circuit availability significantly and increased nominal milling rates by more than 5,000 tonnes per day. Removal of steel from crusher feed was still a problem with the new circuit however, due to the practice of returning large quantities of ball mill trommel oversize to the SAG mill feed stockpile. Reject balls from the ball mill are typically 20-30mm polygonal steel shapes which pass directly through the SAG mill and report to the SAG mill trammel oversize. If these small steel balls are buried on the conveyor they can not be effectively recovered by the magnets. Magnet operation was further complicated by the presence of some magnetite in the ore so that magnet sensitivity was reduced. Metal detectors on crusher feeders have been effective as the last line of defence to keep steel balls out of the crushers. Operating practices require operators to respond to a metal detector trip of the crusher feeder and locate the steel ball buried in the pebbles. Periods of extended primary crusher maintenance have necessitated dozing of stockpiled ball mill scats into the mill feeders. The resulting flood of ball chips into the pebble circuit resulted in the pebble crushers becoming. inoperable until the primary crusher came back online as metal detectors were tripping crusher feeders every few minutes. By mid 2003 modifications to both SAG and ball mill trommels screens and sprays reduced the volume of ball mill scats so they could be effectively segregated and taken to a landfill site, this has eliminated steel ball chips from pebble crusher feed. Capacity of the comminution circuit had been traditionally SAG mill or pebble crusher limited. A review of de-bottlenecking options in 2002identified a change to the pebble crushing circuit would produce a nominal 10,000 tpd increase in total circuit throughput by directing the minus 10mm portion of pebble crusher product directly to the ball mills and the plus 10mm portion returned to the pebble crushers, essentially open circuiting the SAG mills. The production improvement was justified on plant trials demonstrating increased SAG mill feed rates when pebbles were not recycled, with increases of 0.5-0.6 tonnes of fresh feed for every 1 tonne of crushed pebbles not recycled to the SAG mill feed. The modified flowsheet shown in Figure 4 (next page), required the addition of a fourth MP1000 crusher (completed in March 2003) and two 3.6m by 7.3m single deck Schenck vibrating screens (July 2003). On start up of the pebble screens a 20% drop in pebble production rate from the SAG mill was noted due to the removal of recycled crushed pebbles from the mill feed. This decrease allowed the mill discharge grate apertures to be opened from 60mm to 80mm to increase pebble production rates to previous levels. The pebble screens have suffered from mechanical and operating problems due to a combination of light weight screen design, high wear on screen structural members requiring additional preventative maintenance and poor screening efficiency due to thick operating bed depths. Wear rates of cyclone feed pumps and hydrocyclone liners also increased due to the introduction of sharp crushed pebbles into the cyclone feed. Pebble crusher operation improved from less than 60% effective utilisation to over 80% and average crusher power draw increased from 60% to 80% of installed motor power as a result of more consistent feed rates due to the recirculating load of screen oversize, improved dewatering of SAG mill pebbles through installation of static grizzly screens in SAG trommel discharge chutes and elimination of steel balls from crusher feed. Crusher liners are now worn out rather than cracking early due to ball damage. ENHANCED MINE BLASTING AND BLENDING EFFORTS The effect of SAG mill feed size on throughput rates is significant and is discussed by Burger et al (2006) who illustrated the influence of in-situ fracture spacing in the rock as measured by Rock Quality Index (RQD)on SAG mill feed size and mill throughput rate is shown previously in Figure 5 (next page). Installation of Split cameras on SAG mill feed conveyors in 2002 allowed development of improved ore characterisation and tailoring blast designs to rock types to reduce SAG mill feed size variability. Modified blast designs were standardised into a drill and blast cookbook in 2004 which use high powder factors in competent ores and conserve powder in softer/fractured ores. Throughput improvements of up to 10-15% on the hardest ore types were modelled and subsequently verified with operating data in 2004.Due to the mix of ore types in mill feed the application of the ‘cookbook’blast designs have improved throughput rates by 2-7% on an annual basis. Improved characterisation of ore types has improved mill throughput predictions and optimisation of mine ore delivery blends to maximize mill throughput and metal production rates. MILL AVAILABILITY Mill availability and utilisation (Figure 6, next page) has improved over the life of the project from less than design of 92% to 94-95% over the past few years. These improvements have been the result of improved maintenance planning, execution of maintenance plans, modifications to mill component and ancillary equipment designs and complete upgrades of certain equipment. Major improvements have been. ? Simultaneous SAG mill relines commenced in 2004 to provide more frequent maintenance opportunities on the tailings pipeline and cleaner flotation circuit. Annual tailings pipeline maintenance with only one SAG mill line being worked on resulted in 48 hours of lost availability on the other grinding line. ? Upgrade of cyclone feed pumps from GIW 24/22 to Krebs 28/26 and Warman Ash 650 pumps in 2004 extending time between maintenance shutdowns from 5 weeks to 12-13 weeks. As Batu Hijau only has one cyclone feed pump installed per ball mill reducing frequency and duration of pump driven downtime improved mill availability by >0.5%. ? Installation of rubber trommel panels in SAG and ball mills from 2001 and improved panel designs extended wear life from 2-3 weeks to 12 weeks in the highest wear zones. Failure of polyurethane panels contributed to almost 1% unplanned ? downtime in 2001 and allowed large pebbles and grinding balls into the cyclone feed pumps and cyclones. These types of failures are successfully avoided with proper preventative maintenance. ? Pebble circuit upgrade in 2002 removed the vertical bucket conveyors from the pebble circuit, increasing pebble circuit availability to 90%. ? Installation of 100mm thick rubber liner blocks and ceramic tiles in mill discharge launders and sumps has reduced thefrequency of downtime required for this repair work and unplanned downtime to repair leaks. SAG MILL LINER, GRATE AND PULP LIFTER OPTIMISATION SAG Mill Liner Design – Trials and Tribulations The original SAG mill liner design at Batu Hijau was a 72 row Hi-Lo top hat style liner with 12 degrees of lifter face angle relief. Several alternative lifter/liner designs have been tested in one mill in an attempt to increase mill throughput rates and extend liner life. The use of a tripper to feed the two SAG mill feed stockpiles ensures both mills receive the same feed so long term comparisons of mill performance are not influenced by stockpile segregation etc. The first liner trial of a 48 row Hi-Hi top hat liner design is described by McClaren et. al. (2001) as a failure with mill throughput rates and wear rates far poorer than the 72 row liners. Working with the liner manufacturer MEI, a 36 row lifter and plate design was installed in late 2001 and showed a 2-3 % increase in millthroughput rates over the life of the liner set. This liner set was removed prematurely due to excessive breakage of the liner plates.Several subsequent generations of the 36 row liner design from MEI did not show the same improvement in throughput rates observed in the first test. The mill performance was on par with the 72 row design. The 36 row designs continued to suffer from premature liner plate breakage resulting in significant unplanned downtime in the final month or two of the liner life. Ball mills are operated with a 33% ball charge which draws maximum motor power. The mills were operated with a 25:75 mix of 90mm and 65mm balls from 2001 to 2004 before the 90mm balls were changed for a harder 80mm ball to reduce ball consumption. Current operation use a 15:85 blend of new 65mm and 80mm, with 90mm diameter SAG mill reject balls added to reduce new ball consumption. SAG mill reject balls collected by the pebble conveyor magnets are hand sorted and added to the ball mills, a ball sorting machine is being designed and built this year. SAG reject ball addition comprises more than 30% of total ball addition to the ball mills and has reduced new ball consumption