當今科技所需求的手機電(dian)(dian)池除了要能夠長時間供應穩定(ding)電(dian)(dian)源外,體積(ji)小重量輕也(ye)是(shi)關鍵(jian)。縮小電(dian)(dian)路板面(mian)積(ji)、增(zeng)長供電(dian)(dian)時間與減少成本該(gai)如何畢其功(gong)于一(yi)役？將(jiang)眾多電(dian)(dian)源管(guan)理組件整合(he)在單一(yi)芯片上將(jiang)是(shi)解決問題的最好途徑。

    早(zao)期的(de)(de)行(xing)動(dong)電(dian)(dian)(dian)話(hua)不(bu)是體積笨重龐大,就(jiu)是必須受到汽(qi)車(che)電(dian)(dian)(dian)池的(de)(de)束縛,但(dan)經過長(chang)時(shi)間的(de)(de)發展,今天的(de)(de)行(xing)動(dong)電(dian)(dian)(dian)話(hua)已變得非常(chang)輕巧,除了(le)電(dian)(dian)(dian)話(hua)功(gong)(gong)能(neng)(neng),它們還會做許多(duo)(duo)事。新(xin)型3.xG智(zhi)能(neng)(neng)型手機把傳統(tong)的(de)(de)2G行(xing)動(dong)電(dian)(dian)(dian)話(hua)和多(duo)(duo)種其它功(gong)(gong)能(neng)(neng)結合在一(yi)起,包括PDA、數(shu)字相(xiang)(xiang)機、音(yin)樂播(bo)放機（MP3）以及全球(qiu)定位系統(tong)（GPS）。如此多(duo)(duo)元的(de)(de)功(gong)(gong)能(neng)(neng)需(xu)(xu)要許多(duo)(duo)零(ling)件(jian),其中絕大多(duo)(duo)數(shu)的(de)(de)電(dian)(dian)(dian)源電(dian)(dian)(dian)壓并不(bu)相(xiang)(xiang)同(tong),電(dian)(dian)(dian)流需(xu)(xu)求則不(bu)斷增加,使(shi)得它們需(xu)(xu)要更多(duo)(duo)電(dian)(dian)(dian)力。(圖一(yi))是從2G語音(yin)電(dian)(dian)(dian)話(hua)升級到3G視訊電(dian)(dian)(dian)話(hua)后,功(gong)(gong)率需(xu)(xu)求增加的(de)(de)估(gu)計值。

 

 

    圖一功(gong)耗值

    在(zai)此同時,消費者卻想(xiang)要更(geng)(geng)精(jing)巧的手機(ji)。本文介紹兩種電(dian)(dian)源(yuan)管理系(xi)統(tong),它們(men)可(ke)以協助智能(neng)型手機(ji)設計人員在(zai)彼此沖突(tu)的目(mu)標(biao)間(jian)(jian)取(qu)得平(ping)衡,例如將封(feng)裝減至(zhi)最(zui)小,同時支持更(geng)(geng)大的功率(lv)需求;實現最(zui)佳效率(lv),讓電(dian)(dian)池提供最(zui)長的使用時間(jian)(jian);以及將電(dian)(dian)源(yuan)噪聲和漣(lian)波降(jiang)至(zhi)可(ke)接受(shou)水平(ping),以支持新世代的行動電(dian)(dian)話(hua)。

    選擇電池

    選擇充電電池是電源管理系統設計的首要工作之一,鎳氫電池和鋰離子電池則(ze)是目前僅有的(de)(de)兩種(zhong)實際選擇(ze)。鋰離子(zi)電(dian)(dian)池(chi)的(de)(de)單(dan)位(wei)體(ti)積蓄(xu)電(dian)(dian)量(liang)為(wei)270～300Wh/l,單(dan)位(wei)重量(liang)蓄(xu)電(dian)(dian)量(liang)為(wei)110～130Wh/kg,都高(gao)于(yu)鎳氫電(dian)(dian)池(chi)的(de)(de)220～300Wh/l以及(ji)75～100Wh/kg,因(yin)此在(zai)同(tong)樣(yang)蓄(xu)電(dian)(dian)量(liang)下,鋰離子(zi)電(dian)(dian)池(chi)的(de)(de)體(ti)積和重量(liang)都小于(yu)鎳氫電(dian)(dian)池(chi);另外,鋰離子(zi)電(dian)(dian)池(chi)的(de)(de)3.6V工(gong)作電(dian)(dian)壓也(ye)高(gao)于(yu)鎳氫電(dian)(dian)池(chi)的(de)(de)1.2V。

    行(xing)動電(dian)話的(de)多(duo)數功耗都來自于1.2V和3.3V電(dian)源,要讓(rang)交換(huan)式(shi)電(dian)源轉(zhuan)換(huan)器發揮最大(da)工作(zuo)效率,較有(you)效的(de)方法通常是從高(gao)電(dian)壓(ya)轉(zhuan)換(huan)至(zhi)低電(dian)壓(ya),而不是從低電(dian)壓(ya)轉(zhuan)換(huan)至(zhi)高(gao)電(dian)壓(ya),因(yin)此鋰(li)離子電(dian)池是最佳選擇。

    要讓充電電池提供最長使用時間,適當的電池管理和控制就顯得格外重要。電池管理包含三個部份：充電控制、電池監視和電池保護。從使用外接導通組件的線性控制器開始,到內建開關組件且效率更高的交換式控制器,充電控制組件已有長足進步。電池充電器必須處(chu)理500mA到1500mA范圍內的電流,以(yi)便提供快(kuai)速的充(chong)電周期時間。

    電(dian)(dian)池(chi)(chi)監視(shi)和(he)保護組(zu)件(jian)通常都與(yu)電(dian)(dian)池(chi)(chi)封(feng)裝在(zai)一起(qi),電(dian)(dian)池(chi)(chi)監視(shi)組(zu)件(jian)可以是簡單(dan)的(de)(de)「電(dian)(dian)荷計量(liang)器(qi)」（coulombcounter）,由中央處(chu)理器(qi)負責計算電(dian)(dian)池(chi)(chi)剩余電(dian)(dian)力(li);也可以是內建(jian)微控制器(qi)的(de)(de)電(dian)(dian)池(chi)(chi)電(dian)(dian)力(li)量(liang)測(ce)組(zu)件(jian)（gasgauge）,由它透(tou)過(guo)DSP與(yu)處(chu)理器(qi)之間(jian)的(de)(de)簡單(dan)界面,直接提(ti)供(gong)剩余電(dian)(dian)力(li)、剩余供(gong)電(dian)(dian)時間(jian)、電(dian)(dian)池(chi)(chi)電(dian)(dian)壓、溫(wen)度和(he)平均(jun)電(dian)(dian)流量(liang)測(ce)值(zhi)等資料。

    電源拓樸

    接著(zhu),設計工程師必須(xu)決(jue)定電源(yuan)轉(zhuan)(zhuan)(zhuan)換組件(jian)的(de)種類,它或(huo)許是(shi)以(yi)(yi)(yi)電感(gan)為(wei)基礎、并且內建(jian)FET開關(guan)的(de)交(jiao)換式(shi)電源(yuan)轉(zhuan)(zhuan)(zhuan)換器(qi)(qi)(qi)(qi)、無電感(gan)的(de)交(jiao)換式(shi)電源(yuan)轉(zhuan)(zhuan)(zhuan)換器(qi)(qi)(qi)(qi)（電荷泵浦(pu)）或(huo)是(shi)線(xian)性穩壓器(qi)(qi)(qi)(qi)。這(zhe)些轉(zhuan)(zhuan)(zhuan)換器(qi)(qi)(qi)(qi)各有其(qi)優點。就(jiu)效(xiao)率(lv)(lv)而(er)言,以(yi)(yi)(yi)電感(gan)為(wei)基礎的(de)轉(zhuan)(zhuan)(zhuan)換器(qi)(qi)(qi)(qi)擁有最高的(de)整體效(xiao)率(lv)(lv),其(qi)次是(shi)電荷泵浦(pu),最后才是(shi)線(xian)性穩壓器(qi)(qi)(qi)(qi)。成本(ben)通常反比于效(xiao)率(lv)(lv),因此線(xian)性穩壓器(qi)(qi)(qi)(qi)成本(ben)最低,然后是(shi)電荷泵浦(pu),最后則是(shi)以(yi)(yi)(yi)電感(gan)為(wei)基礎的(de)轉(zhuan)(zhuan)(zhuan)換器(qi)(qi)(qi)(qi)。

  線(xian)性穩(wen)壓(ya)器沒有輸出(chu)漣波,電(dian)(dian)荷(he)泵浦有一(yi)些輸出(chu)漣波,交換式穩(wen)壓(ya)器的輸出(chu)漣波則在三者之間(jian)最高(gao)。就(jiu)整個解決方案(an)的體(ti)積(ji)(ji)來(lai)看,線(xian)性穩(wen)壓(ya)器的體(ti)積(ji)(ji)最小,通常(chang)只需輸入和輸出(chu)電(dian)(dian)容,電(dian)(dian)荷(he)泵浦除了輸入和輸出(chu)電(dian)(dian)容外(wai),還需一(yi)顆或兩(liang)顆「飛馳」（flying）電(dian)(dian)容,交換式穩(wen)壓(ya)器則需要電(dian)(dian)感器,因此其封裝體(ti)積(ji)(ji)會有很(hen)大(da)差異。

    無論DSP或模(mo)擬(ni)數(shu)字(zi)轉換器(qi)等(deng)數(shu)字(zi)零(ling)(ling)件,或是電(dian)(dian)源(yuan)管理(li)系(xi)統等(deng)模(mo)擬(ni)零(ling)(ling)件,2G電(dian)(dian)話幾(ji)乎不(bu)提(ti)供任何的(de)功能(neng)整(zheng)合(he),系(xi)統設計(ji)人員在發(fa)展電(dian)(dian)源(yuan)管理(li)系(xi)統時,通常會(hui)以(yi)成(cheng)本(ben)和(he)體(ti)積為優先考(kao)慮,而不(bu)是轉換效(xiao)率。線(xian)性穩(wen)壓(ya)器(qi)只能(neng)將輸入電(dian)(dian)壓(ya)轉換成(cheng)更低(di)的(de)輸出(chu)電(dian)(dian)壓(ya),因此電(dian)(dian)池電(dian)(dian)壓(ya)必須高于3.3V,此時可利用低(di)電(dian)(dian)流(liu)(liu)或中電(dian)(dian)流(liu)(liu)的(de)線(xian)性穩(wen)壓(ya)器(qi)進行電(dian)(dian)壓(ya)轉換,以(yi)便提(ti)供電(dian)(dian)力(li)給(gei)至2.8V范圍內的(de)其它電(dian)(dian)源(yuan)需求。

    在3G芯片組(zu)中,基頻處(chu)(chu)理(li)器(qi)現已(yi)包含DSP、微處(chu)(chu)理(li)器(qi)／微控(kong)制器(qi)、模(mo)(mo)擬數字轉換(huan)(huan)器(qi)和數字模(mo)(mo)擬轉換(huan)(huan)器(qi),用(yong)來(lai)控(kong)制射頻訊(xun)號和音頻訊(xun)號處(chu)(chu)理(li)。這顆處(chu)(chu)理(li)器(qi)的(de)(de)核心(xin)電(dian)(dian)(dian)壓(ya)已(yi)降至(zhi)1.2V或是更低(di),I/O和外圍(wei)電(dian)(dian)(dian)壓(ya)也開始減(jian)少至(zhi)2.5V至(zhi)3.0V范圍(wei);由于3.xG電(dian)(dian)(dian)話的(de)(de)電(dian)(dian)(dian)流(liu)需求通常(chang)都超過2.G電(dian)(dian)(dian)話,3.xG設計(ji)人員需要(yao)效(xiao)率高于線性穩壓(ya)器(qi)的(de)(de)直流(liu)電(dian)(dian)(dian)源轉換(huan)(huan)器(qi),以(yi)便提供更長的(de)(de)電(dian)(dian)(dian)池使(shi)用(yong)時間。為(wei)進一步延(yan)長電(dian)(dian)(dian)池壽命(ming),許多設計(ji)人員必須盡量利用(yong)鋰離子(zi)電(dian)(dian)(dian)池電(dian)(dian)(dian)力,直到其電(dian)(dian)(dian)壓(ya)降至(zhi)最小值為(wei)止;在此過程中,如(ru)何(he)產生3.3V電(dian)(dian)(dian)壓(ya)就變(bian)成一項挑戰。

    從(cong)表面(mian)上來看(kan),設計人(ren)員若能繼續使(shi)用(yong)(yong)(yong)電(dian)(dian)(dian)(dian)池直到(dao)2.7V,并(bing)(bing)利用(yong)(yong)(yong)正電(dian)(dian)(dian)(dian)源降壓―升壓轉換(huan)器(qi)(qi)或是(shi)(shi)SEPIC轉換(huan)器(qi)(qi)提供(gong)3.3V電(dian)(dian)(dian)(dian)源,可攜式(shi)裝(zhuang)置的電(dian)(dian)(dian)(dian)池壽命就會大幅延長(chang),但是(shi)(shi)根據(表一)針(zhen)對600mAh電(dian)(dian)(dian)(dian)池所(suo)做的簡單分(fen)析可發現情形并(bing)(bing)非如此,因為無(wu)論是(shi)(shi)采(cai)(cai)用(yong)(yong)(yong)效率更高(gao)的降壓轉換(huan)器(qi)(qi),并(bing)(bing)將電(dian)(dian)(dian)(dian)池使(shi)用(yong)(yong)(yong)到(dao)3.3V,或是(shi)(shi)采(cai)(cai)用(yong)(yong)(yong)SEPIC之類的轉換(huan)器(qi)(qi),并(bing)(bing)將電(dian)(dian)(dian)(dian)池電(dian)(dian)(dian)(dian)力完全用(yong)(yong)(yong)盡,這兩種方式(shi)的供(gong)電(dian)(dian)(dian)(dian)時間幾(ji)乎沒有(you)任何區(qu)別。

    

 

    表一60mAh電池分析(xi)

    除此(ci)之外,無論(lun)是(shi)(shi)使(shi)用兩(liang)顆電感的(de)SEPIC轉(zhuan)換(huan)(huan)(huan)器(qi),或是(shi)(shi)某(mou)些(xie)效率更(geng)(geng)高(gao)(gao)的(de)新型正電源(yuan)(yuan)降壓(ya)―升壓(ya)轉(zhuan)換(huan)(huan)(huan)器(qi),它們的(de)成本都更(geng)(geng)高(gao)(gao),因此(ci)在做整體評估(gu)時,只使(shi)用3.3V以上的(de)電池電力,然后利(li)用高(gao)(gao)效率交換(huan)(huan)(huan)式(shi)電源(yuan)(yuan)轉(zhuan)換(huan)(huan)(huan)器(qi)提(ti)供3.3V電源(yuan)(yuan)的(de)方(fang)法不但(dan)更(geng)(geng)有效率,還(huan)可能是(shi)(shi)更(geng)(geng)具吸引力的(de)選擇。以下介紹的(de)離(li)散解(jie)決方(fang)案(an)(an)就是(shi)(shi)使(shi)用降壓(ya)轉(zhuan)換(huan)(huan)(huan)器(qi)提(ti)供3.3V電源(yuan)(yuan),整合式(shi)解(jie)決方(fang)案(an)(an)則采用SEPIC轉(zhuan)換(huan)(huan)(huan)器(qi)。

   系統概述

    不(bu)同(tong)的(de)(de)(de)(de)(de)智能型手機零(ling)件有(you)著(zhu)不(bu)同(tong)的(de)(de)(de)(de)(de)電(dian)源(yuan)(yuan)需求,(圖(tu)二(er))是行動電(dian)話中(zhong)需要(yao)(yao)電(dian)源(yuan)(yuan)的(de)(de)(de)(de)(de)主要(yao)(yao)零(ling)件簡單(dan)方塊圖(tu),例(li)如(ru)射(she)頻(pin)(pin)單(dan)元的(de)(de)(de)(de)(de)壓控振蕩器（VCO）以(yi)及鎖相回路（PLL）就需要(yao)(yao)極(ji)低噪聲和(he)很高電(dian)源(yuan)(yuan)拒斥比的(de)(de)(de)(de)(de)電(dian)源(yuan)(yuan),確保它(ta)們提供(gong)最(zui)高的(de)(de)(de)(de)(de)傳送和(he)接收(shou)效能,因(yin)此(ci)雖(sui)然線性穩壓器的(de)(de)(de)(de)(de)效率(lv)(lv)不(bu)高,但由于它(ta)沒有(you)輸出漣波,所以(yi)是這類電(dian)源(yuan)(yuan)供(gong)應的(de)(de)(de)(de)(de)最(zui)佳選擇;同(tong)樣(yang)重(zhong)要(yao)(yao)的(de)(de)(de)(de)(de)是將直流轉換器的(de)(de)(de)(de)(de)開關頻(pin)(pin)率(lv)(lv),還(huan)有(you)它(ta)們的(de)(de)(de)(de)(de)二(er)階(jie)和(he)三(san)階(jie)諧波,都(dou)保持在中(zhong)頻(pin)(pin)頻(pin)(pin)帶(dai)之外。

    由于(yu)DSP和中央(yang)處理器(qi)(qi)的核心電(dian)壓已(yi)降(jiang)至1V左右,以電(dian)感(gan)為(wei)基礎的高效率(lv)交(jiao)換(huan)式降(jiang)壓轉換(huan)器(qi)(qi)是理想(xiang)選擇(ze)。至于(yu)屏幕背光照(zhao)明(ming)所(suo)使用的白光二(er)極管,其電(dian)源可來自電(dian)荷泵浦或電(dian)感(gan)式升壓／降(jiang)壓轉換(huan)器(qi)(qi)。

    圖(tu)二(er)智(zhi)能型手(shou)機電源方塊(kuai)圖(tu)

    動態電壓調整（DynamicVoltageScaling）

    從圖(tu)一可看出,電(dian)(dian)源需(xu)求最(zui)高的兩(liang)顆零件(jian)是在射(she)頻單(dan)元,分別是發(fa)射(she)機的功率(lv)放(fang)大(da)器(qi)(qi)和基頻處(chu)理器(qi)(qi)。隨著電(dian)(dian)話與基地臺之間的距離不同,功率(lv)放(fang)大(da)器(qi)(qi)在通話過(guo)程(cheng)中最(zui)多(duo)消耗75％的總(zong)功耗,待命模式(shi)則只有30％。采(cai)用非線(xian)性功率(lv)放(fang)大(da)器(qi)(qi)的舊型GSM電(dian)(dian)話發(fa)射(she)機的典型工作效率(lv)約為50％,但是WCDMA等較新標準(zhun)卻同時需(xu)要振(zhen)幅及相位調變,這(zhe)只有工作效率(lv)在25％至35％之間的線(xian)性放(fang)大(da)器(qi)(qi)可以提供。

    除此之外,CDMA20001x手(shou)機的(de)正常基頻處理(li)器(qi)負載需求是在60至120mA范圍,因此提供最有(you)效率的(de)電源給功(gong)率放大器(qi)和處理(li)器(qi)就顯(xian)得極為重要。

    動(dong)(dong)態／可適(shi)性電(dian)(dian)壓(ya)調(diao)整(zheng)技術（DVS/AVS）與高(gao)整(zheng)合度組件所使用的(de)(de)(de)(de)方式很類似,它會(hui)把閉(bi)回路系統(tong)中(zhong)的(de)(de)(de)(de)處理器和(he)穩壓(ya)器連結在(zai)一起,并在(zai)確保系統(tong)正常工作(zuo)(zuo)的(de)(de)(de)(de)情(qing)形下,將(jiang)數(shu)字電(dian)(dian)源(yuan)供應的(de)(de)(de)(de)輸出(chu)電(dian)(dian)壓(ya)動(dong)(dong)態調(diao)整(zheng)至最小值。功(gong)率(lv)(lv)放(fang)大(da)(da)(da)器會(hui)被最佳化,使它在(zai)最大(da)(da)(da)傳送(song)功(gong)率(lv)(lv)下擁有(you)最高(gao)效(xiao)率(lv)(lv)。由于絕大(da)(da)(da)多數(shu)手機(ji)都在(zai)基地(di)臺附近工作(zuo)(zuo),手機(ji)的(de)(de)(de)(de)無線電(dian)(dian)功(gong)能會(hui)在(zai)維持通訊質量的(de)(de)(de)(de)前(qian)題下,將(jiang)傳送(song)功(gong)率(lv)(lv)降至最低水平(ping)。當功(gong)率(lv)(lv)放(fang)大(da)(da)(da)器在(zai)較低的(de)(de)(de)(de)功(gong)率(lv)(lv)水平(ping)下工作(zuo)(zuo)時(shi),它的(de)(de)(de)(de)效(xiao)率(lv)(lv)會(hui)受到影響,從(圖三)可以看(kan)出(chu),利(li)用動(dong)(dong)態電(dian)(dian)壓(ya)調(diao)整(zheng)技術來調(diao)整(zheng)功(gong)率(lv)(lv)放(fang)大(da)(da)(da)器的(de)(de)(de)(de)電(dian)(dian)壓(ya),它的(de)(de)(de)(de)工作(zuo)(zuo)效(xiao)率(lv)(lv)會(hui)增加10％至20％。

 

    圖三功(gong)率放大器效率

    數字處(chu)理(li)(li)(li)器(qi)的功耗(hao)正比于電(dian)壓平方,因此中央處(chu)理(li)(li)(li)器(qi)也能采(cai)用動態電(dian)壓調整(zheng)技術(shu);當(dang)中央處(chu)理(li)(li)(li)器(qi)進(jin)入(ru)待(dai)命(ming)模式或(huo)其(qi)它功能精簡模式,它就能在(zai)較低的頻(pin)率頻(pin)率下工(gong)作,此時可將處(chu)理(li)(li)(li)器(qi)電(dian)壓降低,以便減少功耗(hao),提升(sheng)工(gong)作效率,延長電(dian)池壽命(ming)。

   ; 就以OMAP1510為例,假設(she)它的(de)電源是(shi)由(you)TPS62200供應,并使(shi)用(yong)1安培小時的(de)3.6V鋰離子電池輸入,其它特性包(bao)括：

    ●睡眠模式（TPS62200采用(yong)PFM調變）未(wei)用(yong)動(dong)態電壓調整：Vout=1.5V@300μA;效率=93％

    ●正常工作(zuo)模式(shi)（TPS62200采(cai)用PWM調變）：Vout=1.5V@100mA;效率(lv)=96％

    假設此組件(jian)95％時(shi)(shi)間處于睡(shui)眠模式(shi)(shi),5％時(shi)(shi)間處于正(zheng)常工作模式(shi)(shi),則從(cong)輸出功率與時(shi)(shi)間的關系圖(tu)可看出,將動(dong)態電壓調整技術用于睡(shui)眠模式(shi)(shi),電池壽命(ming)會最多延(yan)長9個小(xiao)時(shi)(shi)。

離散解決方案

    (圖(tu)四)是利用離散組件實作的電源管理系統,電池電壓限制(zhi)為3.3V。

    圖四利(li)用(yong)離散組件實作(zuo)的(de)電源管理(li)系統

    在這(zhe)個解決方(fang)(fang)案中(zhong),就算鋰離子電(dian)池下降(jiang)至3.3V左(zuo)右(you),在100％負(fu)載周期模(mo)式下工作(zuo)的(de)高(gao)效(xiao)率TPS62200降(jiang)壓(ya)(ya)轉(zhuan)(zhuan)換(huan)(huan)器(qi)(qi)(qi)仍能(neng)提(ti)(ti)供3.3V的(de)I/O電(dian)壓(ya)(ya)。上述所有零件都采用SOT-23封裝(zhuang),除了bq24020電(dian)池充電(dian)組件、TPS61020升(sheng)壓(ya)(ya)轉(zhuan)(zhuan)換(huan)(huan)器(qi)(qi)(qi)以(yi)及TPS61042白光二極管驅動(dong)組件之外(wai),它們是采用3×3平方(fang)(fang)厘米的(de)QFN封裝(zhuang)。TPS61040和TPS61042還內建上端FET晶(jing)體管,每(mei)顆組件只需要一個外(wai)接二極管。bq24020、TPS622xx、TPS61020和線性穩壓(ya)(ya)器(qi)(qi)(qi)組件全都內建FET晶(jing)體管,功(gong)率放(fang)大器(qi)(qi)(qi)和中(zhong)央處理器(qi)(qi)(qi)電(dian)源采用的(de)動(dong)態電(dian)壓(ya)(ya)調(diao)整技術可以(yi)提(ti)(ti)高(gao)每(mei)顆零件的(de)效(xiao)率,進而協助降(jiang)低功(gong)耗。

    整合解(jie)決(jue)方案(an)

    最(zui)新制程技術使得工程師(shi)更(geng)容易結合(he)、迅速修改以及(ji)／或是利用現有的離散組(zu)件(jian)(jian)設計,以便提供不同(tong)整(zheng)合(he)程度的半導體芯片,例(li)如通(tong)用的雙(shuang)(shuang)通(tong)道(dao)交換(huan)式轉換(huan)器(qi)和(he)電(dian)(dian)源(yuan)(yuan)(yuan)拒斥比很高而噪聲很低的雙(shuang)(shuang)信道(dao)線性(xing)穩(wen)壓器(qi)、特殊應(ying)用白光二極管的電(dian)(dian)源(yuan)(yuan)(yuan)供應(ying)以及(ji)行(xing)動電(dian)(dian)話、PDA和(he)數字相機的多電(dian)(dian)源(yuan)(yuan)(yuan)管理解(jie)決(jue)方案,這些產品都已開始供應(ying)。專門(men)支持(chi)終端設備(bei)的電(dian)(dian)源(yuan)(yuan)(yuan)組(zu)件(jian)(jian)則會內建各(ge)種外圍,其范圍從行(xing)動電(dian)(dian)話的響鈴器(qi)和(he)蜂鳴器(qi)到(dao)PDA的通(tong)用I/O接(jie)腳,例(li)如圖(tu)四整(zheng)合(he)解(jie)決(jue)方案所(suo)使用的TPS65010就是這類組(zu)件(jian)(jian)。

 

    圖五整合(he)式解(jie)決方案

    在此解決方案中,3.3VI/O電源是由SEPIC轉換器提供,它讓應用系統能充份利用鋰離子電池電(dian)(dian)力(li),直到電(dian)(dian)池電(dian)(dian)壓(ya)降至最(zui)低水(shui)平（大約(yue)2.7V）。和離散解決方案一樣(yang),穩壓(ya)器(qi)(qi)輸出(chu)(chu)也(ye)來(lai)自3.3V輸入電(dian)(dian)源,以便提(ti)高(gao)工作效率。TPS65010采用(yong)(yong)48只接腳QFN封裝(zhuang),這些組件(jian)都內(nei)建FET晶體管。TPS61130SEPIC轉換(huan)器(qi)(qi)采用(yong)(yong)4×4平方厘(li)米QFN封裝(zhuang),并(bing)且內(nei)建FET晶體管,最(zui)高(gao)達到90％以上效率,TPS5100則是(shi)三(san)通道輸出(chu)(chu)控制器(qi)(qi),專門用(yong)(yong)來(lai)提(ti)供電(dian)(dian)源給(gei)顯示器(qi)(qi)。功(gong)(gong)率放大器(qi)(qi)和中央(yang)處(chu)理器(qi)(qi)電(dian)(dian)源使用(yong)(yong)的動態電(dian)(dian)壓(ya)調整技術可以改善(shan)每顆零件(jian)的效率,進而協(xie)助(zhu)降低功(gong)(gong)耗。

    離(li)散或整合(he)？

    如何在(zai)離(li)散(san)(san)或整合解(jie)決方案(an)之(zhi)間做出(chu)抉擇？一般說來,整合組件的成本會(hui)低于同樣等(deng)級的多顆離(li)散(san)(san)零(ling)件;除此(ci)之(zhi)外,如同(圖六)的電(dian)路(lu)板布(bu)局(ju)所示,相較(jiao)于執行同樣功能的多顆離(li)散(san)(san)零(ling)件,TPS65010以及與其搭(da)配(pei)的被動(dong)零(ling)件只需較(jiao)少的電(dian)路(lu)板空(kong)間,這主(zhu)要(yao)是因為離(li)散(san)(san)零(ling)件之(zhi)間需要(yao)額(e)外空(kong)間來容(rong)納(na)訊號(hao)線路(lu)。由于TPS65010還包(bao)含原來由離(li)散(san)(san)零(ling)件提供的其它功能,例如電(dian)源(yuan)供應順序、振(zhen)動(dong)器和二極管驅(qu)動(dong)組件,因此(ci)整合解(jie)決方案(an)可以節(jie)省更(geng)多電(dian)路(lu)板面積。

  

 

 

    圖六(liu)TPS65010與同等級離(li)散解決方案的電路板布比(bi)較

    整合(he)(he)組件(jian)(jian)過去主要支(zhi)(zhi)持(chi)特殊應用,彈(dan)性也(ye)不(bu)是很高,因此在設計(ji)流程(cheng)(cheng)后期,它們就無法再進(jin)行重大的(de)設計(ji)變(bian)更。然而新的(de)制程(cheng)(cheng)技術,包括支(zhi)(zhi)持(chi)可(ke)程(cheng)(cheng)序輸出電(dian)壓以(yi)及封(feng)裝后調整的(de)整合(he)(he)式EEPROM,卻(que)使得工程(cheng)(cheng)師能(neng)以(yi)更低成本,更簡(jian)單快速的(de)對(dui)現有(you)組件(jian)(jian)（也(ye)就是不(bu)同固定輸出電(dian)壓的(de)組件(jian)(jian)）重復進(jin)行簡(jian)單修改。另一方(fang)面,整合(he)(he)組件(jian)(jian)的(de)供(gong)貨商(shang)通常只有(you)一家,這可(ke)能(neng)迫使廠商(shang)必須(xu)采用離散解決方(fang)案。

    未來挑戰(zhan)

    消(xiao)費者(zhe)想(xiang)要操(cao)作時(shi)間更長的智能型(xing)手(shou)機,新發展(zhan)的半導體(ti)制程(cheng)技術已能減少泄漏(lou)電(dian)流和(he)阻抗(kang)(kang)（有時(shi)透過銅覆蓋層）,使得(de)FET晶體(ti)管的靜態電(dian)流更低,導通阻抗(kang)(kang)也變得(de)更小。然(ran)而不同(tong)于持續進步(bu)中的半導體(ti)技術,電(dian)池技術卻沒有任何重大進展(zhan),無法(fa)在不增加電(dian)池體(ti)積的情形下延長供電(dian)時(shi)間。

    電(dian)(dian)(dian)(dian)容器(qi)(qi)(qi)技(ji)術的(de)某(mou)些進(jin)展使(shi)得充(chong)電(dian)(dian)(dian)(dian)電(dian)(dian)(dian)(dian)池(chi)和電(dian)(dian)(dian)(dian)容器(qi)(qi)(qi)之間(jian)的(de)界限(xian)日(ri)益模糊,許(xu)多可攜式產品(pin)已(yi)開始使(shi)用高(gao)能量超(chao)級電(dian)(dian)(dian)(dian)容器(qi)(qi)(qi)（supercapacitor）,做為消費者更換(huan)電(dian)(dian)(dian)(dian)池(chi)時(shi)(shi)的(de)暫時(shi)(shi)電(dian)(dian)(dian)(dian)力(li)來源;另外(wai),高(gao)能量暨(ji)高(gao)功(gong)率的(de)超(chao)高(gao)電(dian)(dian)(dian)(dian)容器(qi)(qi)(qi)（ultracapacitor）還能在短時(shi)(shi)間(jian)內提供(gong)很大電(dian)(dian)(dian)(dian)流,讓電(dian)(dian)(dian)(dian)池(chi)不必瞬間(jian)供(gong)應(ying)龐大電(dian)(dian)(dian)(dian)力(li),可以延(yan)長(chang)電(dian)(dian)(dian)(dian)池(chi)的(de)使(shi)用時(shi)(shi)間(jian)。這些超(chao)高(gao)電(dian)(dian)(dian)(dian)容器(qi)(qi)(qi)會整合至電(dian)(dian)(dian)(dian)池(chi)封裝(zhuang)內,并在系統電(dian)(dian)(dian)(dian)力(li)需求不太高(gao)時(shi)(shi),利用微小電(dian)(dian)(dian)(dian)流充(chong)電(dian)(dian)(dian)(dian)。

    燃(ran)料電(dian)池(chi)近來是熱門(men)話題,但由于外(wai)形包(bao)裝(zhuang)尚未(wei)標準(zhun)化,使得燃(ran)料電(dian)池(chi)的廣泛(fan)應用(yong)受到影響(xiang)(xiang),商業化過程也(ye)(ye)不太順(shun)利。燃(ran)料電(dian)池(chi)的輸出瞬(shun)時(shi)響(xiang)(xiang)應也(ye)(ye)很糟糕,因此至少在最初階段(duan)燃(ran)料電(dian)池(chi)只會做(zuo)為普通(tong)電(dian)池(chi)的補強裝(zhuang)置,無法取代普通(tong)電(dian)池(chi)。

    消費者(zhe)還希望產(chan)品的體積更(geng)小,功能更(geng)加(jia)強大(da),創新的電源管(guan)理組(zu)件(jian)設(she)計以及封裝和制程技術(shu)(shu)的進(jin)步都(dou)能幫助實現此目標。日益精密的制程技術(shu)(shu)可以制造(zao)出越來越小的FET晶體管(guan),讓(rang)晶粒和封裝的體積更(geng)小,工作(zuo)電壓更(geng)低,閘極電容更(geng)少,使得晶體管(guan)的開關速度(du)更(geng)快

    ―對于以電(dian)感為基礎的交換式(shi)電(dian)源(yuan)供應(ying),更(geng)快的開關速(su)度意味(wei)著(zhu)更(geng)小的電(dian)感。新(xin)封(feng)(feng)裝(zhuang)技術則能(neng)在更(geng)小的封(feng)(feng)裝(zhuang)中容納更(geng)多功(gong)能(neng),并且(qie)承受更(geng)大的功(gong)耗,例如內(nei)建FET開關的鋰離子電(dian)池線性充電(dian)組件bq24010就采用3×3平(ping)方厘米(mi)的QFN封(feng)(feng)裝(zhuang),它在普(pu)通室(shi)溫環境下,最(zui)高(gao)能(neng)承受1.5W功(gong)耗。

    要在較低(di)的(de)(de)工作電(dian)壓(ya)下提供(gong)更(geng)(geng)強大功(gong)能(neng),電(dian)源(yuan)管理(li)單元和低(di)噪聲布局的(de)(de)容忍要求(qiu)通(tong)常也會(hui)(hui)變(bian)(bian)的(de)(de)更(geng)(geng)嚴格,例如系統若要求(qiu)1.2V電(dian)源(yuan)的(de)(de)誤差小于(yu)±3%,就(jiu)表(biao)示輸出電(dian)壓(ya)變(bian)(bian)動(dong)幅度不(bu)能(neng)超過±36mV;相形(xing)之下,使用3.3V電(dian)源(yuan)就(jiu)表(biao)示在同(tong)樣的(de)(de)±3%誤差限制下,它能(neng)容忍的(de)(de)電(dian)壓(ya)變(bian)(bian)動(dong)高達±99mV。由于(yu)電(dian)源(yuan)電(dian)壓(ya)不(bu)斷降低(di),未來幾年內對于(yu)誤差更(geng)(geng)小、電(dian)流(liu)(liu)更(geng)(geng)大、效率更(geng)(geng)高和電(dian)磁干(gan)擾(rao)極低(di)的(de)(de)直(zhi)流(liu)(liu)電(dian)源(yuan)轉換器的(de)(de)需求(qiu)將會(hui)(hui)增加(jia)。除此之外,隨(sui)著封(feng)裝縮小,可(ke)供(gong)散(san)熱(re)的(de)(de)面(mian)積也會(hui)(hui)減少,讓這些(xie)高功(gong)耗組件的(de)(de)熱(re)管理(li)繼續成為(wei)困難挑戰。

    整合的力(li)量(liang)

    本(ben)文介紹的(de)電(dian)(dian)源(yuan)(yuan)解決方案使用(yong)不(bu)同(tong)整(zheng)合程度(du)的(de)電(dian)(dian)源(yuan)(yuan)組(zu)件(jian)。把(ba)部份(fen)或全部的(de)模擬電(dian)(dian)源(yuan)(yuan)組(zu)件(jian)和基(ji)(ji)頻(pin)處理器(qi)等數字(zi)(zi)零件(jian)整(zheng)合在一起會帶(dai)來許多優(you)點,包括(kuo)節省更(geng)多的(de)電(dian)(dian)路(lu)板面積,并且降低總(zong)成本(ben)。復雜電(dian)(dian)子(zi)系統的(de)每個部份(fen)都有著不(bu)同(tong)的(de)需(xu)求,這是(shi)過去(qu)實現更(geng)高(gao)階(jie)數字(zi)(zi)和模擬零件(jian)整(zheng)合的(de)障礙之(zhi)一,例(li)如(ru)數字(zi)(zi)基(ji)(ji)頻(pin)單(dan)元(yuan)需(xu)要(yao)高(gao)密度(du)制(zhi)程以(yi)支持數字(zi)(zi)訊號(hao)處理,模擬基(ji)(ji)頻(pin)和電(dian)(dian)源(yuan)(yuan)功能(neng)需(xu)要(yao)電(dian)(dian)壓(ya)更(geng)高(gao)的(de)組(zu)件(jian);射頻(pin)單(dan)元(yuan),特別是(shi)鎖相回(hui)路(lu),則需(xu)要(yao)最適合支持高(gao)頻(pin)操作(zuo)的(de)BiCMOS組(zu)件(jian)。

    傳統上,制(zhi)(zhi)(zhi)程(cheng)發(fa)展是由(you)數(shu)字(zi)設(she)計(ji)人員(yuan)負責管理(li),他們(men)通常只(zhi)會推動高(gao)密度(du)制(zhi)(zhi)(zhi)程(cheng)發(fa)展,電(dian)(dian)路若需要高(gao)電(dian)(dian)壓組(zu)件,就必須(xu)采用(yong)不同制(zhi)(zhi)(zhi)程(cheng),這(zhe)表示他們(men)需要獨立的(de)(de)數(shu)字(zi)組(zu)件。半導體廠商(shang)不但(dan)開始發(fa)展「最小閘極(ji)長度(du)」更(geng)短的(de)(de)BiCMOS制(zhi)(zhi)(zhi)程(cheng),以便提(ti)供很高(gao)的(de)(de)組(zu)件密度(du)和工作(zuo)速度(du),還有更(geng)高(gao)電(dian)(dian)壓的(de)(de)汲極(ji)延伸型組(zu)件（drainextendeddevices）,它(ta)們(men)已(yi)用(yong)于更(geng)多的(de)(de)模(mo)擬和電(dian)(dian)源(yuan)應(ying)用(yong)。包括電(dian)(dian)源(yuan)管理(li)在內的(de)(de)許多模(mo)擬和數(shu)字(zi)功能(neng)最后都會整合成單顆(ke)芯片。

    不同(tong)程度的(de)(de)組(zu)(zu)件(jian)(jian)(jian)整合正在簡(jian)化可(ke)(ke)攜(xie)式電源設計,尤其(qi)是可(ke)(ke)攜(xie)式產品的(de)(de)系(xi)統設計人員,他們不必再擔心(xin)組(zu)(zu)件(jian)(jian)(jian)的(de)(de)電源需(xu)求管理(li),整合程度不同(tong)的(de)(de)電源管理(li)組(zu)(zu)件(jian)(jian)(jian)可(ke)(ke)以幫助他們讓電池提供最長供電時間,同(tong)時將電路板(ban)面積和成本減(jian)至(zhi)最少。