TW406270B - A charge pump circuit includes a plurality of pumping units connected - Google Patents

A charge pump circuit includes a plurality of pumping units connected Download PDF

Info

Publication number
TW406270B
TW406270B TW087103901A TW87103901A TW406270B TW 406270 B TW406270 B TW 406270B TW 087103901 A TW087103901 A TW 087103901A TW 87103901 A TW87103901 A TW 87103901A TW 406270 B TW406270 B TW 406270B
Authority
TW
Taiwan
Prior art keywords
transistor
substrate
voltage
voltage source
output terminal
Prior art date
Application number
TW087103901A
Other languages
Chinese (zh)
Inventor
Ki-Hwan Choi
Seung-Keun Lee
Kang-Deog Suh
Original Assignee
Samsung Electronics Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Samsung Electronics Co Ltd filed Critical Samsung Electronics Co Ltd
Application granted granted Critical
Publication of TW406270B publication Critical patent/TW406270B/en

Links

Landscapes

  • Read Only Memory (AREA)

Abstract

A charge pump circuit includes a plurality of pumping units connected between a first voltage source and an output terminal, the pumping unit comprising a transistor having a bulk of a floating condition in which active regions of the transistor are formed, and a plurality of discharge transistors connecting the bulks of the transistors to a second voltage source in response to a first control signal. The circuit also provides a precharge transistor connecting the output terminal to a third voltage source in response to a second control signal. The circuit is employed for generate positive and negative high voltages on a nonvolatile memory.

Description

2832pif.doc/002 406270 - 五、發明説明(I ) 本發明是有關於一種具有充電幫浦電路(charge pump circuit)之半導體記憶體(semiconductor memory),且特別是 有關於一種利用充電幫浦電路產生高壓之半導體記憶 體,例如唯讀記憶體(read only memory)中的快閃電性抹除 和電性編程動作(flash electrically erase and program)。 在目前攜帶型電腦的發展趨勢中,越來越要求供應電 源的降低。爲了減少電力的浪費,在大部分個人電腦中的 積體電路,都會重新設計,以因應低電壓的要求。一般攜 帶型電腦中的電路或元件,其操作電壓大都被設計爲約在 5V左右,或是3.3V甚至更低,用以降低此種電腦所需的 電力。很不幸的,在攜帶型電腦中有些元件是需要高壓 的,例如在最近的攜帶型電腦中,快閃唯讀記憶體(flash EEPROM)用以儲存基本輸出入系統(Basic Input/Output System ; BIOS)資料。當欲改變基本輸出入系統(BIOS)資料 時,此快閃唯讀記憶體只要藉由進行一個小的更新程式 (update program),就不需要從電腦中取出,即可進行抹 除與編程的動作。然而,快閃唯讀記憶體的抹除與編程動 作需要超過10V以上才能有效的完成,此種高壓是普通低 壓的攜帶型電腦所無法提供的。而在一些其他領域的電性 安排中,充電幫浦電路的功能通常都用以從低電壓源中提 供出一高電壓。 利用充電幫浦電路來產生筒壓的方式被揭示於1976年 的期刊“IEEE Journal of Solid State Circuit”中,在冊數爲 Vol.SC-11,頁數爲 pp.374-378,題目爲“〇n-chip High- (請先閱讀背面之注^h項再填寫本頁) 本紙張尺度適用中國國家標準(CNS > A4规格(210X297公釐) 2832pif.doc/002 406270 at __B7 五、發明説明(2 ) voltage Generation In NMOS Integrated Circuit”中的第 1 (請先閲讀背面之注意事項再填寫本頁) 圖。在此習知技藝中描述到在電路中進行的幫浦運作,其 中會串聯複數個幫浦單元,而幫浦單元是由多個電容C1-Cn組合而成。以及選擇多個NMOS加強電晶體MN1-MNn(enhancement transistors)。每個加強電晶體介於節點 N1與輸出端T1之間,節點N1係透過先行充電電晶體 MNO(precharging transistor)連接至源極電壓Vcc,而輸出 端T1可產生高幫浦電壓Vpp(pumped voltage)。上述幫浦 單元中電晶體的閘極與汲極係直接連接於電容,而源極連 接至下個電晶體的閘極與汲極,至於電晶體MNl-MNn的 基底則保持接地。幫浦單元可根據幫浦計時器P1和 P2(pumping clocks)輪流交替,計時器P1和P2具有互補的 相位差,如第2圖所示,計時器P1應用於偶數的電容 C1,C3...,計時器P2應用於奇數的電容C2,C4...。若是計 時器P1的狀態設在高,計時器P2的狀態則設在低。在初 始態時,一旦記憶體元件被供電,則節點N1在電壓Vcc-Vth下被充電,其中Vth爲電晶體ΜΝ0的起始電壓 (threshold voltage)。對應計時器P1和P2的昇和降,從N2 到T1的電壓層級會逐漸的增加,直到到達Vpp爲止,此 電壓Vpp爲完成抹除(erase)和編程所需的電壓(program)。 利用第1圖的電路結構來進行幫浦充電時,在源極與 基底之間的電壓VSB,會漸漸增加,而到達電晶體的起始 電壓,所以源極電壓會升高,而基底的電壓仍是接isr*,此 稱爲整體效應(body effect)。請參照第3圖,源極對基底的 5 本紙张尺度適用中國國家橾準(CNS ) A4規格(210X297公釐) 2832pif.doc/002 40 627〇 A? 五、發明説明(,) 電壓VSB由OV到3V逐漸增加,使痦閘極對源極的電壓 亦會移動,由0.8V到1.3V,因爲汲極對源極電流Ids的零 基準點。以及起始電壓Vt隨著正比於電壓VSB的增加曲 線而增加。此種起始電壓Vt的移動會降低幫浦充電的效 率,導致無法獲得高電壓(或幫浦電壓)來進行低壓記憶體 元件中的抹除和編程動作。 有鑑於此,本發明的主要目的,就是在提出一種具有 充電幫浦電路之半導體記憶體,可以更加強推動效率 (boosting efficiency) ° 本發明的另一目的,就是在提出一種具有充電幫浦電 路之半導體記憶體,可以避免整體效應。 本發明的另一目的,就是在提出一種具有充電幫浦電 路之半導體記憶體,在幫浦充電之後,基底電壓會很快的 下降至預定的電壓。 本發明的又一目的,就是在提出一種具有充電幫浦電 路之半導體記憶體,即使此記憶體在低壓下操作,亦可以 有很好的幫浦充電效率。 爲達上述之目的,本發明提供一種具有充電幫浦電路 之半導體記憶體,其中包括:一記億單元陣列,一行解碼 器與一列解碼器。還有一第一充電幫浦電路,此第一充電 幫浦電路包括多個幫浦單元,連接在一第一電壓源極和一 第一輸出端之間,第一輸出端連接至行解碼器。而幫浦單 元包括電晶體,其設於一浮置的基底上,此基底中有多個 主動區。以及多個放電電晶體,其設於一基底上,將此基 6 I I I— I . 訂 餘 . - (請先閱讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) 2832pif.doc/002 A7 B7 五、發明説明(f) .底連接至一第二電壓源極,對應一控制訊號。再來,還有 一第二充電幫浦電路,此第二充電幫浦電路包括多個幫浦 單元,其連接在一第三電壓源極和一第二輸出端之間,第 二輸出端連接至行解碼器。而幫浦單元包括一電晶體,其 設於一浮置的基底上,且基底中有多個主動區。以及多個 放電電晶體,放電電晶體設於一基底上,再將此基底連接 至一第四電壓源極,對應一控制訊號。此外,第一充電幫 浦電路更包括一電晶體,此電晶體連接上述第一輸出端至 一第五電壓源極,對應第一先行充電控制訊號。以及第二 充電幫浦電路更包括一電晶體,此電晶體連接第二輸出端 至一第六電壓源極,對應一第二先行充電控制訊號。 爲讓本發明之上述和其他目的、特徵、和優點能更明 顯易懂,下文特舉一較佳實施例,並配合所附圖式,作詳 細說明如下: 圖式之簡單說明: 第1圖,其所繪示的是習知一種非揮發性記憶體中的 充電幫浦電路示意圖; 第2圖,其所繪示的是第.1圖的時序示意圖(timing diagram); 第3圖,其所繪示的是習知起始電壓受整體效應影響 而改變之示意圖; 第4圖,其所繪示根據本發明,一種非揮發性記憶體 結構的區塊示意圖(block diagram); 第5圖,其所繪示的是根據本發明之第一較佳實施例, 7 裝 訂 梦 (請先閱讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) 2832pif.doc/002 406270 A7 B7 五、發明説明(Γ) 用於第4圖的一種充電幫浦的電路示意圖; 第6圖’其所繪不的是根據第5圖,充電幫浦的部分 結構示意圖; 第7圖’其所繪示的是根據第5圖,充電幫浦操作時 的時序示意圖; 第8圖’其所繪示的是根據本發明之第二較佳實施例, 用於第4圖的一種充電幫浦的電路示意圖;以及 第9圖,其所繪示的是根據第8圖,充電幫浦的部分 結構示意圖。 實施例 請參照第4圖,其所繪示根據本發明,一種具有充電 幫浦電路之NOR型快閃EEPROM結構的區塊示意圖。列 (row)與行(column)的部分電路120與160,每個選擇字元 線WL1到WLn中的一個字元線以及選擇位元線bli到 BLn中的一個位元線。這些字元線以及位元線以矩陣方式 分佈在記憶單元陣列100中,記憶單元例如爲快閃 EEPROM單元。行部分電路160連接選擇的位元線至對應 的感測放大器(未顯示)。 列部分電路120,驅動所有具負高電壓Vpp的字元線。 此負電壓Vpp·係在抹除動作時,由負的充電幫浦電路150 所提供。因此形成在基底上的記億單元會被抹除。伴隨著 負的充電幫浦電路150所產生用以抹除與修復抹除的電壓 VPP·’正的充電幫浦電路140提供正高電壓Vpp+給列部分 電路120,用以進行編程與確認編程的動作。藉由列部分 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) ----------- (請4'閲讀^面之注^1^項再填寫本頁} *1Τ 2832pif.doc/002 406270 A7 B7 五、發明説明(6) 電路120的選擇操作,此正電壓Vpp+提供給被選擇的字元 線。在正電壓Vpp+的輸出端,以及電壓節點V〇D之間(此 電壓大小約爲源極電壓Vcc),PMOS電晶體142的閘極會 耦合至訊號ΦΕΝΟϋΒ處。在正電壓Vpp·的輸出端,與基底 電壓Vss之間(或接地),NMOS電晶體152的閘極會耦合 至訊號(DENODB處。PMOS電晶體142與NMOS電晶體 152的基底,分別連接至電壓節點VOD與基底電壓Vss。 在幫浦充電之前或之後,電晶體142將Vpp+的輸出端拉下 至VOD的電壓大小,對應訊號ΦΕΝΟϋΒ的啓動。而在幫 浦充電之前或之後,電晶體152將Vpp-的輸出端拉上至 Vss的電壓大小,對應訊號ΦΕΝΟϋΒ的啓動。 請參照第5圖,其所繪示的是根據本發明之第一較佳 實施例,對應第4圖的一種產生Vpp+的正充電幫浦電路 M0示意圖。其中,在電源供應電壓Vcc與Vpp+的輸出端 之間,串連多個幫浦單元UP1到UPn。每個幫浦單元是由 一個耦合電容(coupling capacitor)、一個PMOS選擇電晶體 (switching transistor)與兩個寄生二極體(parasitic diode)所 組成。幫浦單元UP1到Upn都有相同的結構,以及都連接 到放電電壓VBDp(discharge voltage) ’此VBDp電壓在 NMOS電晶體BDT1到BDTn中,而其中每個閘極均耦合 到放電訊號ΦΕΝΒϋ。 舉例來說,在第一幫浦單元UP1中,電容C1的-個 電極連接到幫浦計時器ΦΡ1,而另一個電極連接到PMOS 選擇電晶體MP1的閘極,此閘極更連接到汲極節點(或是 9 本纸張尺度適用中國國家標準(CNS ) Α4規格(2丨0 X 297公釐) (請先閱讀背面之注意事項再填寫本頁) 裝· 訂 2832pif.doc/0022832pif.doc / 002 406270-V. Description of the invention (I) The present invention relates to a semiconductor memory having a charge pump circuit, and in particular to a method for utilizing a charge pump circuit Semiconductor memory that generates high voltage, such as flash electrically erase and program in read only memory. In the current development trend of portable computers, there is an increasing demand for reduced power supply. In order to reduce the waste of power, the integrated circuits in most personal computers are redesigned to meet the requirements of low voltage. Generally, the operating voltage of circuits or components in portable computers is designed to be about 5V, or 3.3V or lower, to reduce the power required by such computers. Unfortunately, some components in portable computers require high voltage. For example, in recent portable computers, flash EEPROM is used to store the basic input / output system (Basic Input / Output System; BIOS). )data. When you want to change the basic input / output system (BIOS) data, this flash read-only memory can be erased and programmed as long as you perform a small update program without removing it from the computer. action. However, erasing and programming of flash read-only memory requires more than 10V to complete effectively. Such high voltage cannot be provided by ordinary low-voltage portable computers. In some other areas of electrical arrangements, the function of the charging pump circuit is usually used to provide a high voltage from a low voltage source. The method of using the charging pump circuit to generate the cylinder pressure was disclosed in the 1976 journal "IEEE Journal of Solid State Circuit" in Vol. SC-11, pages pp.374-378, and titled " 〇n-chip High- (Please read the note ^ h on the back before filling this page) This paper size applies to Chinese national standards (CNS > A4 size (210X297 mm) 2832pif.doc / 002 406270 at __B7 V. Invention Explanation (2) Figure 1 in the “Voltage Generation In NMOS Integrated Circuit” (please read the precautions on the back before filling this page). The pump operation in the circuit is described in this conventional technique, which will be connected in series. A plurality of pump units, and the pump unit is composed of a plurality of capacitors C1-Cn. And a plurality of NMOS strengthening transistors MN1-MNn (enhancement transistors) are selected. Each strengthening transistor is located between the node N1 and the output terminal Between T1, node N1 is connected to the source voltage Vcc through a precharging transistor MNO (precharging transistor), and the output terminal T1 can generate a high pumping voltage Vpp (pumped voltage). The gate and sink of the transistor in the above-mentioned pump unit Extremely straight Connected to the capacitor, and the source is connected to the gate and the drain of the next transistor, and the base of the transistor MNl-MNn is kept grounded. The pump unit can alternate according to the pump timers P1 and P2 (pumping clocks) Timers P1 and P2 have complementary phase differences. As shown in Figure 2, timer P1 is applied to even-numbered capacitors C1, C3 ..., and timer P2 is applied to odd-numbered capacitors C2, C4 .... If yes The state of timer P1 is set to high, and the state of timer P2 is set to low. In the initial state, once the memory element is powered, node N1 is charged at the voltage Vcc-Vth, where Vth is the voltage of transistor MN0. Threshold voltage. Corresponding to the rise and fall of timers P1 and P2, the voltage level from N2 to T1 will gradually increase until it reaches Vpp. This voltage Vpp is required to complete erase and programming The voltage VSB between the source and the substrate will gradually increase when using the circuit structure in Figure 1 to charge the pump, and it will reach the starting voltage of the transistor, so the source voltage will rise. High, while the voltage of the substrate is still connected to isr *, this is called integral Body effect. Please refer to Figure 3. The 5 paper sizes of the source to the base are applicable to China National Standard (CNS) A4 (210X297 mm) 2832pif.doc / 002 40 627〇A? V. Invention Explanation (,) The voltage VSB gradually increases from OV to 3V, so that the voltage of the gate-to-source will also move, from 0.8V to 1.3V, because the zero reference point of the drain-to-source current Ids. And the starting voltage Vt increases as the curve increases in proportion to the voltage VSB. Such a shift of the starting voltage Vt will reduce the efficiency of the pump charging, resulting in the inability to obtain a high voltage (or pump voltage) for erasing and programming in a low voltage memory element. In view of this, the main object of the present invention is to propose a semiconductor memory with a charging pump circuit, which can further enhance the boosting efficiency. Another object of the present invention is to propose a charging pump circuit The semiconductor memory can avoid the overall effect. Another object of the present invention is to provide a semiconductor memory having a charging pump circuit. After the pump is charged, the substrate voltage will drop to a predetermined voltage quickly. Another object of the present invention is to provide a semiconductor memory having a charging pump circuit. Even if the memory is operated at a low voltage, it can have a good pump charging efficiency. In order to achieve the above object, the present invention provides a semiconductor memory with a charging pump circuit, which includes: an array of 100 million cells, a row decoder and a column decoder. There is also a first charging pump circuit. The first charging pump circuit includes a plurality of pump units connected between a first voltage source and a first output terminal, and the first output terminal is connected to the row decoder. The pump unit includes a transistor, which is disposed on a floating substrate, and the substrate has a plurality of active regions. And a number of discharge transistors, which are set on a substrate, this base 6 III — I. Reservation.-(Please read the precautions on the back before filling out this page) This paper size applies Chinese National Standard (CNS) A4 Specifications (210X297 mm) 2832pif.doc / 002 A7 B7 V. Description of the invention (f). The bottom is connected to a second voltage source, corresponding to a control signal. Furthermore, there is a second charging pump circuit. The second charging pump circuit includes a plurality of pump units connected between a third voltage source and a second output terminal, and the second output terminal is connected to Line decoder. The pump unit includes a transistor, which is disposed on a floating substrate, and the substrate has a plurality of active regions. And a plurality of discharge transistors. The discharge transistors are arranged on a substrate, and the substrate is connected to a fourth voltage source corresponding to a control signal. In addition, the first charging pump circuit further includes a transistor, which is connected to the first output terminal to a fifth voltage source and corresponds to the first advance charging control signal. And the second charging pump circuit further includes a transistor, which is connected to the second output terminal to a sixth voltage source and corresponds to a second advance charging control signal. In order to make the above and other objects, features, and advantages of the present invention more comprehensible, a preferred embodiment is given below in conjunction with the accompanying drawings for detailed description as follows: Brief description of the drawings: FIG. 1 , Which shows a schematic diagram of a charging pump circuit in a conventional nonvolatile memory; FIG. 2 shows a timing diagram of FIG. 1; FIG. 3 shows Shown is a schematic diagram of a change in the conventional starting voltage affected by the overall effect; FIG. 4 is a block diagram showing a non-volatile memory structure according to the present invention; FIG. 5 , Which shows the first preferred embodiment of the present invention, 7 Binding Dream (please read the notes on the back before filling this page) This paper size is applicable to China National Standard (CNS) A4 (210X297 mm) ) 2832pif.doc / 002 406270 A7 B7 V. Description of the invention (Γ) A schematic circuit diagram of a charging pump used in Figure 4; Figure 6 'cannot be drawn according to Figure 5, the part of the charging pump Schematic diagram; Figure 7 'its drawing FIG. 5 is a timing diagram of the charging pump operation according to FIG. 5; FIG. 8 ′ is a schematic diagram of a charging pump circuit according to the second preferred embodiment of the present invention for FIG. 4; And FIG. 9 is a schematic structural diagram of a charging pump according to FIG. 8. Embodiment Please refer to FIG. 4, which is a block diagram showing a NOR flash EEPROM structure with a charging pump circuit according to the present invention. The column and row partial circuits 120 and 160 each select one of the word lines WL1 to WLn and one of the bit lines bli to BLn. These word lines and bit lines are distributed in a matrix manner in the memory cell array 100. The memory cells are, for example, flash EEPROM cells. The row portion circuit 160 connects the selected bit line to a corresponding sense amplifier (not shown). The column part circuit 120 drives all the word lines having a negative high voltage Vpp. This negative voltage Vpp · is provided by the negative charging pump circuit 150 during the erasing operation. Therefore, the billion-units formed on the substrate will be erased. Along with the negative charging pump circuit 150, the voltage VPP for erasing and repairing is erased. The positive charging pump circuit 140 provides a positive high voltage Vpp + to the column circuit 120 for programming and confirming the programming action. . The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) by listing part of the paper. ----------- (Please read the 1 ^^^^^^ item on this page before filling in this page} * 1T 2832pif.doc / 002 406270 A7 B7 V. Description of the invention (6) Select operation of the circuit 120, this positive voltage Vpp + is provided to the selected word line. At the output terminal of the positive voltage Vpp +, and the voltage node V〇D (This voltage is about the source voltage Vcc), the gate of the PMOS transistor 142 will be coupled to the signal ΦΕΝΟϋΒ. Between the output of the positive voltage Vpp · and the base voltage Vss (or ground), the NMOS voltage The gate of the crystal 152 is coupled to the signal (DENODB. The substrates of the PMOS transistor 142 and the NMOS transistor 152 are connected to the voltage node VOD and the substrate voltage Vss, respectively. Before or after the pump charging, the transistor 142 changes the voltage Vpp + The magnitude of the voltage pulled down from the output to VOD corresponds to the activation of the signal ΦΕΝΟϋΒ. Before or after the pump is charged, the transistor 152 pulls the output of Vpp- to the voltage of Vss, corresponding to the activation of the signal ΦΕΝΟϋΒ. FIG. 5 is a drawing showing The first preferred embodiment is a schematic diagram of a positive charging pump circuit M0 for generating Vpp + corresponding to FIG. 4. Among them, a plurality of pump units UP1 to UPn are connected in series between the power supply voltage Vcc and the output terminal of Vpp +. Each pump unit is composed of a coupling capacitor, a PMOS switching transistor and two parasitic diodes. The pump units UP1 to Upn have the same structure , And both are connected to the discharge voltage VBDp (discharge voltage) 'This VBDp voltage is in the NMOS transistors BDT1 to BDTn, and each of them is coupled to the discharge signal ΦΕΝΒϋ. For example, in the first pump unit UP1 One electrode of the capacitor C1 is connected to the pump timer ΦP1, and the other electrode is connected to the gate of the PMOS selection transistor MP1, and this gate is further connected to the drain node (or 9 paper sizes are applicable to China) Standard (CNS) Α4 specification (2 丨 0 X 297 mm) (Please read the precautions on the back before filling in this page) Binding · Order 2832pif.doc / 002

40627Q ί\7 Β7 五、發明説明(q) 第一幫浦節點)N1。電晶體MP1連接至電源供應電壓Vcc, 其源極透過寄生二極體Dls連接至基底節點B1,其汲極透 過寄生二極體Did連接至基底節點B1。除此之外,電晶 體MP1的基底也連接至基底節點B1。寄生二極體Dls和 Did分別形成在源極區和汲極區,與基底區的交界面上, 如第6圖所示。放電節點B1透過電晶體BDT1連接至 VBDp。以幫浦單元UP2作爲UP1的下一個推動單元,電 容C2會對應到幫浦計時器ΦΡ2,而其具有與幫浦計時器 ΦΡ1互補的相位。電容C2的另一個電極連接至PMOS選 擇電晶體MP2的閘極,此閘極更連接到汲極節點(或是第 二幫浦節點)N2。汲極節點N2透過寄生二極體D2d連接至 基底節點B2。電晶體MP2的源極透過寄生二極體D2s連 接至基底節點B2。就如在電晶體MP1中一樣,電晶體MP2 的基底會連接到基底節點B2,而寄生二極體D2d與D2s 分別形成在源極區和汲極區,與基底區的交界面上。基底 節點B2透過電晶體BDT2連接到VBDp。相似結構的幫浦 單元’重複的安排於線路上,形成一連串的幫浦單元鏈。 幫浦計時器ΦΡ1與ΦΡ2則交替的應用於電容Cl-Cn,計時 器ΦΡ1應用於偶數的電容C1,C3...,計時器ΦΡ2應用於奇 數的電容C2,C4.·.。如第4圖所示,PMOS FET的先行充 電電晶體142連接至Vpp+的輸出端。 請參照第6圖,其所繪示的是根據第5圖,充電幫浦 的部分結構示意圖。根據此第6圖可以瞭解寄生二極體的 形成。選擇電晶體MP1-MP2,係分別形成於他們自己所屬 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) -----------裝------irI^-----.#' . - (請先閱讀背面之注意事項再填寫本頁) 2832pif.doc/002 406270 A7 B7 五、發明説明(y) 的Ν·(微摻雜Ν·型離子)井2中,此Ν·井2係用以做電晶體 的基底用。Ν·井2係形成於ρ·(微慘雜ρ·型離子)半導體基 底1中’且此半導體基底1透過Ρ+區(濃摻雜ρ+型離子)連 接至Vss(例如0V) °在Ν·井2中,Ρ+主動區域係形成用以 做源極和汲極用’而N+主動區域9係形成用以在必要時接 受井偏壓(well bias)用。至於N+主動區域7和8形成於半 導體基底1中,用以做放電電晶體,例如BDT1,的汲極 和源極用。此放電電晶體的N+主動區域8均共同連接至放 電電壓VBDp。 放電電晶體,例如BDT1,的N+主動區域7直接透過 節點B1與N·井2(或是幫浦單元中選擇電晶體的基底)中的 N+區相連,N-井未與任何偏壓耦合,爲浮置的狀態。因爲 沒有任何的偏壓加於幫浦單元中選擇電晶體的Ν·井基底 中,下列兩個水平接和面,例如在Ρ+區4和Ν+區9之間以 及在Ρ+區4和N+區9之間,就極有可能產生寄生二極體 Dls和Did的結構,其中陽極透過節點Β1以及BDT1的 通道連接至VBDp。眾所周知的,二極體會保持起始電壓 的既定値與由源極和汲極區至B1通過的電流,如此可以 保持住二極體自身陰極和陽極的電位差。 請參照第7圖,其所繪示的是根據第5圖(或第6圖), 充電幫浦操作時的時序示意圖。PGMB爲一啓動編程動作 的訊號’ PGMvfB則允許編程確認模態(program verifying mode)。首先,當先行充電訊號φΕΝΟϋ爲低時,Vpp+的輸 出端與正的充電幫浦電路140,會因爲先行充電電晶體I42 本紙張尺度適用中國國家榡準(CNS ) A4規格(2丨0X297公釐〉 (請先閲讀背面之注意事項再填寫本頁)40627Q ί \ 7 Β7 V. Description of the invention (q) First pump node) N1. The transistor MP1 is connected to the power supply voltage Vcc, its source is connected to the base node B1 through the parasitic diode Dls, and its drain is connected to the base node B1 through the parasitic diode Did. In addition, the base of the electric crystal MP1 is also connected to the base node B1. Parasitic diodes Dls and Did are formed on the interface between the source region and the drain region and the base region, respectively, as shown in FIG. 6. Discharge node B1 is connected to VBDp through transistor BDT1. With the pump unit UP2 as the next pushing unit of UP1, the capacitor C2 will correspond to the pump timer ΦP2, and it has a phase complementary to the pump timer ΦP1. The other electrode of the capacitor C2 is connected to the gate of the PMOS selection transistor MP2, and this gate is further connected to the drain node (or the second pump node) N2. The drain node N2 is connected to the base node B2 through a parasitic diode D2d. The source of the transistor MP2 is connected to the base node B2 through the parasitic diode D2s. As in the transistor MP1, the substrate of the transistor MP2 is connected to the substrate node B2, and the parasitic diodes D2d and D2s are formed on the source region and the drain region, respectively, on the interface with the substrate region. Base node B2 is connected to VBDp through transistor BDT2. Pump units of similar structure are repeatedly arranged on the line to form a series of pump unit chains. Pump timers ΦP1 and ΦP2 are alternately applied to capacitors Cl-Cn, timer ΦP1 is applied to even-numbered capacitors C1, C3 ..., and timer ΦP2 is applied to odd-numbered capacitors C2, C4 ... As shown in Fig. 4, the leading charge transistor 142 of the PMOS FET is connected to the output terminal of Vpp +. Please refer to Figure 6, which shows a schematic diagram of the structure of the charging pump according to Figure 5. The formation of parasitic diodes can be understood from this Figure 6. The transistors MP1-MP2 are selected, and they are formed in their own paper. The paper standards are applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm). ^ -----. # '.-(Please read the notes on the back before filling this page) 2832pif.doc / 002 406270 A7 B7 V. Description of the invention (y) N · (micro-doped N · type ion ) In well 2, this N · well 2 is used as the base of the transistor. The N · well 2 is formed in a ρ · (micro-hybrid ρ · type ion) semiconductor substrate 1 ', and the semiconductor substrate 1 is connected to Vss (for example, 0V) through a P + region (concentively doped ρ + type ion). In N · well 2, the P + active region is formed to be used as a source and the drain, and the N + active region 9 is formed to receive a well bias when necessary. The N + active regions 7 and 8 are formed in the semiconductor substrate 1 for the drain and source of a discharge transistor, such as BDT1. The N + active regions 8 of this discharge transistor are all connected to the discharge voltage VBDp in common. The N + active region 7 of a discharge transistor, such as BDT1, is directly connected to the N + region in N · well 2 (or the base of the transistor selected in the pump unit) through node B1, and the N-well is not coupled with any bias voltage. It is floating. Because no bias voltage is applied to the N · well base of the transistor selected in the pump unit, the following two are connected horizontally, for example, between P + region 4 and N + region 9 and between P + region 4 and Between the N + regions 9, it is highly likely that a structure of parasitic diodes Dls and Did is generated, in which the anode is connected to VBDp through the channels of nodes B1 and BDT1. It is well known that the diode will maintain the predetermined voltage of the starting voltage and the current passing from the source and drain regions to B1, so that the potential difference between the cathode and the anode of the diode can be maintained. Please refer to FIG. 7, which shows a timing diagram of charging pump operation according to FIG. 5 (or FIG. 6). PGMB is a signal that initiates the programming action. PGMvfB allows a program verifying mode. First, when the pre-charging signal φΕΝΟϋ is low, the output of Vpp + and the positive charging pump circuit 140 will be based on the pre-charging transistor I42. The paper size applies to the Chinese National Standard (CNS) A4 specification (2 丨 0X297 mm). 〉 (Please read the notes on the back before filling this page)

'IT 4齊年矢—各苟負1·肖奢合卞:i.fft. ^06270 2832pif.doc/〇〇2 A7 B7___ 五、發明説明(^ ) 被打開,而設至VOD的先行充電電壓大小(Vcc的電壓大 小)。然後,假如PGMB在時間tl時爲低,此時電路開始 推動VPP+到約10V的電壓,用以做編程電壓,對應至幫浦 計時器ΦΡ1與ΦΡ2,其彼此互相交替連續振盪著。 在此同時,電晶體MP1的基底節點B1電壓,變爲 Vcc-Vtd(Vtd爲Dls的起始電壓),節點N1被充電至Vcc-Vtpl(Vtpl爲MP1的起始電壓)。因爲Vtpl —般比Vtd還 高,Did無法被導通,而節點N1的電壓保持在目前的電 壓,以致於ΦΡ2爲高,ΦΡ1爲低,使得MP1會被關閉。在 時間t2時,ΦΡ1爲高而ΦΡ2爲低,MP1與MP2分別會被 關閉與開啓。其條件可產生下列方程式:'IT 4 Qi Nianya-each bears one's share 1. Xiao Shehe: i.fft. ^ 06270 2832pif.doc / 〇〇2 A7 B7___ 5. The description of the invention (^) is opened, and the pre-charging voltage set to VOD Size (the voltage of Vcc). Then, if PGMB is low at time t1, the circuit starts to push VPP + to a voltage of about 10V for programming voltage, corresponding to the pump timers ΦP1 and ΦP2, which oscillate alternately and continuously with each other. At the same time, the voltage at the base node B1 of transistor MP1 becomes Vcc-Vtd (Vtd is the starting voltage of Dls), and node N1 is charged to Vcc-Vtpl (Vtpl is the starting voltage of MP1). Because Vtpl is generally higher than Vtd, Did cannot be turned on, and the voltage at node N1 remains at the current voltage, so that ΦP2 is high and ΦP1 is low, so that MP1 will be turned off. At time t2, ΦP1 is high and ΦP2 is low, MP1 and MP2 will be turned off and on respectively. The conditions can produce the following equations:

Vnl=(Vcc-Vtpl) X α X Vcc [ 1 ] 此處,a=Cl/(Cl+Cl’),Cl’爲節點Ν1的電容。節點 N2的電壓,以及Vn2的關係式如下:Vnl = (Vcc-Vtpl) X α X Vcc [1] Here, a = Cl / (Cl + Cl ′), and Cl ’is the capacitance of the node N1. The relationship between the voltage at node N2 and Vn2 is as follows:

Vn2=(Vcc-Vtpl) X a X Vcc-Vtp2 [2] 此處,Vtp2爲MP2的起始電壓。當時間從t2到t3時, 電晶體MP1的關閉可以保護避免電流由節點N1流至 Vcc。而基底節點B2被充電至電壓Vnl-Vtd,VB2則爲從 Vnl減少的電壓: VB2=(Vcc-Vtpl)X aX Vcc-Vtd [3] 如第[2]式和第[3]式所示’在幫浦操作時,在井 2(ΜΡ1的基底)與Ρ+區4(ΜΡ1的源極)之間的電壓VSB維持 在寄生二極體Dis(i意指1到η任意値)的起始電壓Vtd。 所以,正高電壓Vpp+根據下式而得: (請先閲讀背面之注意事項再填寫本頁) %Vn2 = (Vcc-Vtpl) X a X Vcc-Vtp2 [2] Here, Vtp2 is the starting voltage of MP2. When the time is from t2 to t3, the turn-off of transistor MP1 can protect the current from flowing from node N1 to Vcc. The base node B2 is charged to the voltage Vnl-Vtd, and VB2 is the voltage reduced from Vnl: VB2 = (Vcc-Vtpl) X aX Vcc-Vtd [3] As shown in equations [2] and [3] 'At the time of pumping operation, the voltage VSB between well 2 (the base of MP1) and P + region 4 (the source of MP1) is maintained at the start of the parasitic diode Dis (i means 1 to n arbitrary 値). Starting voltage Vtd. Therefore, the positive high voltage Vpp + is obtained according to the following formula: (Please read the precautions on the back before filling this page)%

*1T 本紙張尺度適用中國國家標準(CNS>A4規格(210Χ297公釐) 2832pif.doc/002 406270 A7 B7 五、發明説明(丨i?)* 1T This paper size applies to Chinese national standard (CNS > A4 specification (210 × 297 mm) 2832pif.doc / 002 406270 A7 B7 V. Description of invention (丨 i?)

Vpp+=Vcc + nX(aXVcc-|vtp|) [4] (請先閲讀背面之注項再填寫本頁) 在編程的時候,放電訊號ΦΕΝΒϋ以及先行充電訊號 0>ENODB係分別保持在預定的高壓Vpp與低壓。假如編程 操作利用約10V的正高電壓Vpp+係在超過時間t4時,先 行充電訊號ΦΕΝΟϋΒ則爲低,用以拉下電壓Vpp+到電壓 VOD,Vcc用以做充電電壓Vpp+,目的爲準備後續的編程 確認操作。而ΦΡ1與ΦΡ2在編程確認開始前是無作用的。 即使在推動操作時,電壓VSB被設至在大小爲電壓 Vtd,不可避免的基底電壓仍是會增加,因爲1ST井基底的 電壓會隨著選擇電晶體的源極電壓而增加,電壓差約爲 Vtd。此種較高的基底電壓,會超過選擇電晶體的源極電 壓,而降低電路的推動效率(boosting efficiency)。因此, 當時間由t4到t5時,放電訊號ΦΕΝΒϋ會變成高電壓Vpp, 用以開啓電晶體BDTl-BDTn,而電晶體基底的推動電壓會 降低到VBDp,例如基底電壓會放電至使VBDp成爲0V。 it· 假若PGMvfB在時間t6時變低,開始進行編程確認操 作,從正充電幫浦電路140中,ΦΡ1與ΦΡ2再次被啓動, 而Vpp+產生至約6V。我們知道Vpp+的大小可以調整,藉 由控制操作幫浦計時器ΦΡ1與ΦΡ2的次數。結束編程確認 操作之後,爲了先行充電Vpp+的輸出端,ΦΕΝΟΒϋ會在 編程過程中變低·。如果需要的話,<DENBD在編程操作中 也會以同樣的方式啓動。 另一方面,第8圖和第9圖所繪示的是根據本發明之 第二較佳實施例,用於第4圖的一種負充電幫浦電路150 本紙張尺度適用中國國家標準(CNS ) Α4規格(210X297公釐) 2832pif.doc/002 406270 at B7 一 —— _______ ____ ______ _ 五、發明説明(丨I ) 的示意圖。其可以產生負高壓Vpp·,用以進行抹除與抹除 確認動作。其中每個幫浦單元UN1到UNn串連於基底電 壓Vss和Vpp·的輸出端之間,且每個幫浦單元由一個耦合 電容、一個NMOS選擇電晶體和兩個寄生二極體所組成。 幫浦單元UNl-UNn均有相同的結構,且連接至放電電壓 VBDn,此VBDn通過NMOS選擇電晶體BDTl-BDTn的閘 極耦合至放電訊號OENBD。舉例來說,在第一幫浦單元 UN1中,電容C1的一個電極耦合至幫浦計時器ΦΡ1,而 另一個電極連接至NMOS選擇電晶體MN1的閘極,此MN1 更連接至汲極節點N1(或是第一幫浦節點)。選擇電晶體 MN1的源極連接至Vss,其亦透過寄生二極體Dls連接至 基底節點B1,且選擇電晶體MN1的汲極透過寄生二極體 Did連接至基底節點B1。除此之外,選擇電晶體MN1的 基底亦連接至基底節點B1。寄生二極體Dls與Did分別 形成在源極區和汲極區,與基底區的交界面上,如第9圖 所示。放電節點B1透過選擇電晶體BDT1連接至VBDn。 而以幫浦單元UP2作爲UP1的下一個推動單元,電容C2 會對應到幫浦計時器ΦΡ2,而其具有與幫浦計時器ΦΡ1互 補的相位。電容C2的另一個電極連接至NMOS選擇電晶 體MN2的閘極,此閘極更連接到汲極節點(或是第二幫浦 節點)N2。汲極節點N2透過寄生二極體D2d連接至基底節 點B2。電晶體MN2的源極透過寄生二極體D2s連接至基 底節點B2。就如在電晶體MN1中一樣,電晶體MN2的基 底會連接到基底節點B2,而寄生二極體D2s與D2d分別 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公釐) (請先閲讀背面之注意事項再填寫本頁)Vpp + = Vcc + nX (aXVcc- | vtp |) [4] (Please read the note on the back before filling this page) During programming, the discharge signal ΦΕΝΒϋ and the pre-charge signal 0 > ENODB are kept at the predetermined high voltage respectively Vpp and low voltage. If the programming operation uses a positive high voltage Vpp + of about 10V when the time t4 is exceeded, the pre-charging signal ΦΕΝΟ 则 Β is low, which is used to pull down the voltage Vpp + to the voltage VOD, and Vcc is used as the charging voltage Vpp +, in order to prepare for subsequent programming confirmation operating. However, ΦP1 and ΦP2 have no effect until the programming confirmation starts. Even during the push operation, the voltage VSB is set to the voltage Vtd, and the inevitable substrate voltage will still increase, because the voltage of the 1ST well substrate will increase with the selection of the source voltage of the transistor, and the voltage difference is about Vtd. Such a high substrate voltage will exceed the source voltage of the selected transistor and reduce the boosting efficiency of the circuit. Therefore, when the time is from t4 to t5, the discharge signal ΦΕΝΒϋ will become high voltage Vpp to turn on the transistors BDTl-BDTn, and the driving voltage of the transistor substrate will be reduced to VBDp. For example, the substrate voltage will be discharged to make VBDp 0V . it · If PGMvfB goes low at time t6, the program confirm operation is started. From the positive charge pump circuit 140, ΦP1 and ΦP2 are started again, and Vpp + is generated to about 6V. We know that the size of Vpp + can be adjusted by controlling the number of times the pump timers ΦP1 and ΦP2 are operated. After finishing the programming confirmation operation, in order to charge the output of Vpp + in advance, ΦΕΝΟΒϋ will go low during the programming process. If necessary, < DENBD will be started in the same way during the programming operation. On the other hand, Figures 8 and 9 show a negative charging pump circuit for Figure 4 according to the second preferred embodiment of the present invention. The paper size is applicable to the Chinese National Standard (CNS). Α4 specification (210X297 mm) 2832pif.doc / 002 406270 at B7 I-_______ ____ ______ _ 5. Schematic diagram of the description of the invention (丨 I). It can generate negative high voltage Vpp · for erasing and erasing confirmation. Each of the pump units UN1 to UNn is connected in series between the output terminals of the base voltage Vss and Vpp ·, and each pump unit is composed of a coupling capacitor, an NMOS selection transistor, and two parasitic diodes. The pump units UN1-UNn have the same structure and are connected to the discharge voltage VBDn. This VBDn is coupled to the discharge signal OENBD through the gate of the NMOS selection transistor BDT1-BDTn. For example, in the first pump unit UN1, one electrode of the capacitor C1 is coupled to the pump timer ΦP1, and the other electrode is connected to the gate of the NMOS selection transistor MN1, and this MN1 is further connected to the drain node N1. (Or the first pump node). The source of the selection transistor MN1 is connected to Vss, which is also connected to the base node B1 through the parasitic diode Dls, and the drain of the selection transistor MN1 is connected to the base node B1 through the parasitic diode Did. In addition, the base of the selection transistor MN1 is also connected to the base node B1. Parasitic diodes Dls and Did are formed on the interface between the source and drain regions and the base region, respectively, as shown in Figure 9. The discharge node B1 is connected to VBDn through a selection transistor BDT1. With the pump unit UP2 as the next driving unit of UP1, the capacitor C2 will correspond to the pump timer ΦP2, and it has a phase that complements the pump timer ΦP1. The other electrode of the capacitor C2 is connected to the gate of the NMOS selection transistor MN2, and this gate is further connected to the drain node (or the second pump node) N2. The drain node N2 is connected to the base node B2 through a parasitic diode D2d. The source of the transistor MN2 is connected to the base node B2 through the parasitic diode D2s. Just as in transistor MN1, the base of transistor MN2 is connected to the base node B2, and the parasitic diodes D2s and D2d respectively apply the Chinese National Standard (CNS) A4 specification (210X297 mm) for this paper. (Read the notes on the back and fill out this page)

、1T 1«TUT UK c f .1 ρΐ* 2832pif.doc/002 4〇627〇 A7 _____B7_ 五、發明説明(11) 形成在源極區和汲極區,與基底區的交界面上。基底節點 B2透過電晶體BDT2連接到VBDn。相似結構的幫浦單 元’重複的安排於線路上,形成一連串的幫浦單元鏈。幫 浦計時器ΦΡ1與ΦΡ2則交替的應用於電容C1_Cn,計時器 ΦΡ1應用於偶數的電容C1,C3...,計時器ΦΡ2應用於奇數 的電容C2,C4.··。如第4圖所示,NMOS FET的先行充電 電晶體I52連接在Vpp·的輸出端與Vss之間。 請參照第9圖,其所繪示的是根據第8圖,充電幫浦 的部分結構示意圖。根據此第9圖可以瞭解寄生二極體的 形成。選擇電晶體MN1-MN2,係分別形成於他們自己所 屬的P·(微摻雜P_型離子)井12中,此P·井12係用以做電 晶體的基底用。P·井12係形成於N·井11中,而N·井11 在F半導體基底10中,且此N_井11透過N+區連接至Vcc。 在?_井12中’ N+主動區域14和15係形成用以做源極和 汲極用,而P+主動區域19係形成用以接受井偏壓(well bias) 用。至於P+主動區域17和18形成於Ν·井11中,用以做 放電電晶體,例如BDT1,的汲極和源極用。此放電電晶 體的Ρ+主動區域18均共同連接至放電電壓VBDn。放電電 晶體,例如BDT1,的P+主動區域17直接透過節點B1與 卩_井12(或是幫浦單元中選擇電晶體的基底)中的P+區相 連,P_井未與任何偏壓耦合,處於浮置的狀態。因爲沒有 任何的偏壓加於幫浦單元中選擇電晶體的P·井基底中,下 列兩個水平接和面,例如在P+區19和N+區14之間以及在 P+區19和N+區15之間,就極有可能產生寄生二極體Dls 本紙張尺度適用中國國家榡準(CNS ) A4規格(2丨Ο X 297公釐) ---------i------IT------.# - - (請先閱讀背面之注意Ϋ項再填寫本頁) 406S70 2832pif.doc/002 A7 B7 五、發明説明(丨)) .和Did的結構,其中陽極透過節點B1以及BDT1的通道 連接至VBDn。眾所周知的’二極體會保持在p-區丨2與N+ 主動區域14和15之間起始電壓的既定値,如此可以保持 住二極體自身陰極(基底)和陽極(源極和汲極)的電位差。 當負充電幫浦電路150進行幫浦操作時,其與正充電 幫浦電路140相似,但是電壓値有些不同,例如用以進行 抹除的電壓Vpp_約爲-12V ’而電壓Vpp+約爲10V。如第7 圖所示,當進行抹除操作時,編程訊號PGMB和編程確認 訊號PGMvfB將會分別對應到抹除訊號(例如ERSB)和抹 除確認訊號(例如ERSvfB)。而先行充電訊號OENOD提供 到電晶體152的閘極,會在抹除動作之前或之後設定到高 値,至於抹除確認動作則與編程確認動作時相 同。此ΦΕΝΟϋ以及放電電晶體BDTl-BDTn仍然扮演有用 的角色,可以加強推動效率。如前所述,VBDn可用以放 電以減少電晶體基底增加的電壓。舉例來說’ VBDn較佳 的是爲Vcc。 綜上所述,本發明的優點在於降低由於基底對源極電 壓增加所造成的一些缺點或是整體效應。以及在幫浦單元 中加強後續幫浦單元鏈的推動效率。 綜上所述,雖然本發明已以較佳實施例揭露如上’然 其並非用以限定本發明’任何熟習此技藝者’在不脫離本 發明之精神和範圍內’當可作各種之更動與潤飾’因此本 發明之保護範圍當視後附之申請專利範圍所界定者爲 本紙張尺度適用中國國家標準(CNS ) A4規格(210X297公楚) I— ^ H 訂 I 鍵 1 - . (锖先閱讀背面之注意事項再填寫本頁)1T 1 «TUT UK c f .1 ρΐ * 2832pif.doc / 002 4〇627〇 A7 _____B7_ 5. Description of the invention (11) It is formed on the interface between the source region and the drain region and the base region. Base node B2 is connected to VBDn via transistor BDT2. Pump units of similar structure are repeatedly arranged on the line to form a series of pump unit chains. The helper timers ΦP1 and ΦP2 are alternately applied to the capacitors C1_Cn, the timer ΦP1 is applied to the even-numbered capacitors C1, C3 ..., and the timer ΦP2 is applied to the odd-numbered capacitors C2, C4. As shown in Figure 4, the NMOS FET's leading charge transistor I52 is connected between the output of Vpp · and Vss. Please refer to FIG. 9, which shows a schematic diagram of a part of the charging pump according to FIG. 8. The formation of parasitic diodes can be understood from this Figure 9. The transistors MN1-MN2 are selected and formed in the P · (micro-doped P_-type ion) wells 12 which belong to them, and this P · well 12 is used as the substrate of the transistor. The P · well 12 is formed in the N · well 11, and the N · well 11 is in the F semiconductor substrate 10, and the N_well 11 is connected to Vcc through the N + region. in? _ Well 12 'N + active regions 14 and 15 are formed for source and sink, while P + active region 19 is formed for receiving well bias. The P + active regions 17 and 18 are formed in the N · well 11 for the drain and source of a discharge transistor, such as BDT1. The P + active regions 18 of this discharge transistor are all connected in common to the discharge voltage VBDn. The P + active region 17 of a discharge transistor, such as BDT1, is directly connected to the P + region in the 卩 _well 12 (or the base of the transistor selected in the pump unit) through the node B1. The P_ well is not coupled to any bias voltage. In a floating state. Because no bias voltage is applied to the P · well base of the transistor selected in the pump unit, the following two are horizontally connected, for example, between P + region 19 and N + region 14 and between P + region 19 and N + region 15 There is a high probability that parasitic diodes Dls will be produced. This paper size is applicable to China National Standard (CNS) A4 (2 丨 〇 X 297 mm) --------- i ----- -IT ------. #--(Please read the note on the back before filling in this page) 406S70 2832pif.doc / 002 A7 B7 V. Description of the invention (丨) Connect to VBDn through the channels of node B1 and BDT1. The well-known 'diode will be maintained at a predetermined threshold of the starting voltage between p-region 2 and N + active regions 14 and 15, so that the diode's own cathode (base) and anode (source and drain) can be maintained. The potential difference. When the negative charging pump circuit 150 performs a pumping operation, it is similar to the positive charging pump circuit 140, but the voltage is slightly different. For example, the voltage Vpp_ for erasing is about -12V 'and the voltage Vpp + is about 10V. . As shown in Figure 7, when the erase operation is performed, the programming signal PGMB and the program confirmation signal PGMvfB will correspond to the erase signal (such as ERSB) and the erase confirmation signal (such as ERSvfB). The gate of the advance charging signal OEOND provided to the transistor 152 will be set to high before or after the erase operation. The erase confirmation operation is the same as the programmed confirmation operation. This ΦΕΝΟϋ and the discharge transistor BDTl-BDTn still play a useful role, which can enhance the driving efficiency. As mentioned earlier, VBDn can be used to discharge to reduce the increased voltage on the transistor substrate. For example, 'VBDn is preferably Vcc. In summary, the present invention has the advantage of reducing some disadvantages or overall effects caused by an increase in substrate-to-source voltage. And strengthen the promotion efficiency of the subsequent pump unit chain in the pump unit. In summary, although the present invention has been disclosed in the preferred embodiment as above, it is not intended to limit the present invention. 'Any person skilled in the art' can make various changes and modifications without departing from the spirit and scope of the present invention. "Retouching" Therefore, the scope of protection of the present invention should be defined as the scope of the patent application attached to this paper. The paper size applies the Chinese National Standard (CNS) A4 specification (210X297). I— ^ H Order I key 1-. (锖 先(Read the notes on the back and fill out this page)

Claims (1)

2832pif.doc/002 ^|Qg27〇 g8g D8 六、申請專利範圍 1. 一種充電幫浦電路,該電路包括: 一第一導電型基底; (請先閲讀背面之注意事項再填寫本頁) 複數個井,設在該第一導電型基底中,該些井係爲第 二導電型與浮置情況;以及 複數個第一導電型電晶體,每個該第一導電型電晶體 在該井中具有複數個主動區,以及每個該第一導電型電晶 體具有一閘極;該閘極對應至一電容耦合。 如申請專利範圍第1項所述之記憶體,其中 2. —種充電幫浦電路,該電路包括: 複數個幫浦單元,連接於一第一電壓源極和一輸出端 之間,該幫浦單元包括一電晶體,該電晶體設於一浮置的 基底上,該基底具有複數個主動區;以及 複數個放電電晶體,該些放電電晶體設於一基底上, 將該基底連接至一第二電壓源極,對應一第一控制訊號。 3. 如申請專利範圍第2項所述之電路,其中更包括一 電晶體,該電晶體連接於該輸出端與一第二電壓源極之 間,對應一第二控制訊號。 4. 一種非揮發性記憶體,該非揮發性記憶體具有一記 憶單元陣列,一行解碼器與一列解碼器,包括: 一第一充電幫浦電路,該第一充電幫浦電路包括複數 個幫浦單元,連接在一第一電壓源極和一第一輸出端之 間,該第一輸出端連接至該行解碼器,該幫浦單元包括一 電晶體,該電晶體設於一浮置的基底上,且該基底中有複 數個主動區,以及複數個放電電晶體,該些放電電晶體設 太紙掁尺唐逋用中國國家標準(CNS) A4規格(210X297公嫠) A8 2832pif.doc/002 406270 B8 D8 六、申請專利範圍 於一基底上,將該基底連接至一第二電壓源極,對應一控 制訊號;以及 (請先閲讀背面之注意事項再填寫本頁) 一第二充電幫浦電路,該第二充電幫浦電路包括複數 個幫浦單元,連接在一第三電壓源極和一第二輸出端之 間,該第二輸出端連接至該行解碼器,該幫浦單元包括一 電晶體,該電晶體設於一浮置的基底上,且該基底中有複 數個主動區,以及複數個放電電晶體,該些放電電晶體設 於一基底上,將該基底連接至一第四電壓源極,對應一控 制訊號。 5.如申請專利範圍第4項所述之記憶體,其中該第一 充電幫浦電路更包括一電晶體,該電晶體連接該第一輸出 端至一第五電壓源極,對應一第一先行充電控制訊號,以 及該第二充電幫浦電路更包括一電晶體,該電晶體連接該 第二輸出端至一第六電壓源極,對應一第二先行充電控制 訊號。 18 太板恢尺疳诮用中國國家樣準(CNS ) A4規格(210X297公釐)2832pif.doc / 002 ^ | Qg27〇g8g D8 6. Scope of patent application 1. A charging pump circuit including: a first conductive substrate; (please read the precautions on the back before filling this page) Wells provided in the first conductivity type substrate, the well systems being the second conductivity type and floating conditions; and a plurality of first conductivity type transistors, each of the first conductivity type transistors having a plurality in the well Each active region and each of the first conductive transistors have a gate; the gate corresponds to a capacitive coupling. The memory according to item 1 of the scope of the patent application, wherein 2. a kind of charging pump circuit, the circuit includes: a plurality of pump units connected between a first voltage source and an output terminal, the battery The pump unit includes a transistor provided on a floating substrate having a plurality of active regions; and a plurality of discharge transistors provided on a substrate and connecting the substrate to A second voltage source corresponds to a first control signal. 3. The circuit according to item 2 of the scope of patent application, further comprising a transistor connected between the output terminal and a second voltage source, corresponding to a second control signal. 4. A non-volatile memory having a memory cell array, a row of decoders and a column of decoders, comprising: a first charging pump circuit, the first charging pump circuit including a plurality of pumps A unit connected between a first voltage source and a first output terminal, the first output terminal being connected to the row decoder, the pump unit including a transistor, the transistor being disposed on a floating substrate And the substrate has a plurality of active regions and a plurality of discharge transistors. The discharge transistors are set on a paper ruler. The standard is Chinese Standard (CNS) A4 (210X297). A8 2832pif.doc / 002 406270 B8 D8 6. Apply for a patent on a substrate, connect the substrate to a second voltage source, corresponding to a control signal; and (Please read the precautions on the back before filling this page) A second charging help Pump circuit, the second charging pump circuit includes a plurality of pump units connected between a third voltage source and a second output terminal, the second output terminal is connected to the row decoder, the pump single The transistor comprises a transistor, which is arranged on a floating substrate, and the substrate has a plurality of active regions, and a plurality of discharge transistors. The discharge transistors are arranged on a substrate, and the substrate is connected to A fourth voltage source corresponds to a control signal. 5. The memory according to item 4 of the scope of patent application, wherein the first charging pump circuit further includes a transistor, the transistor is connected to the first output terminal to a fifth voltage source, corresponding to a first The advance charge control signal, and the second charge pump circuit further includes a transistor connected to the second output terminal to a sixth voltage source, corresponding to a second advance charge control signal. 18 Taipan Hui ruler uses China National Standard (CNS) A4 specification (210X297 mm)
TW087103901A 1997-04-11 1998-03-17 A charge pump circuit includes a plurality of pumping units connected TW406270B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR19970013426 1997-04-11

Publications (1)

Publication Number Publication Date
TW406270B true TW406270B (en) 2000-09-21

Family

ID=19502561

Family Applications (1)

Application Number Title Priority Date Filing Date
TW087103901A TW406270B (en) 1997-04-11 1998-03-17 A charge pump circuit includes a plurality of pumping units connected

Country Status (1)

Country Link
TW (1) TW406270B (en)

Similar Documents

Publication Publication Date Title
US8339183B2 (en) Charge pump with reduced energy consumption through charge sharing and clock boosting suitable for high voltage word line in flash memories
US5553030A (en) Method and apparatus for controlling the output voltage provided by a charge pump circuit
US6522559B2 (en) Low voltage charge employing optimized clock amplitudes
TWI391805B (en) Efficient charge pump for a wide range of supply voltages
US7512007B2 (en) Data processing device
US5841703A (en) Method and apparatus for removal of VT drop in the output diode of charge pumps
US6456541B2 (en) Booster circuit for raising voltage by sequentially transferring charges from input terminals of booster units to output terminals thereof in response to clock signals having different phases
JPH10302493A (en) Charge pumping circuit and non-volatile memory device circuit having it
JPH1027480A (en) Bootstrapping circuit utilizing ferroelectric capacitor
JPH06261538A (en) High-voltage charge pump
TW201801075A (en) Non-volatile semiconductor memory device and driving method for word line thereof
US6160440A (en) Scaleable charge pump for use with a low voltage power supply
US6633494B2 (en) Low voltage charge pump employing distributed charge boosting
EP0750314B1 (en) Negative word line voltage regulation circuit for electrically erasable semiconductor memory devices
JP2003249088A (en) Charge pump circuit
JP2002261239A (en) Step-up circuit of nonvolatile semiconductor storage device
EP0830686B1 (en) Negative voltage switching circuit
US6738292B2 (en) Nonvolatile semiconductor storage device
US5986935A (en) Semiconductor memory device with high voltage generation circuit
JPH0748310B2 (en) Semiconductor integrated circuit
US6532177B1 (en) Low voltage charge pump apparatus and method
US6191642B1 (en) Charge pump circuit
TW406270B (en) A charge pump circuit includes a plurality of pumping units connected
JP2723946B2 (en) Circuit for charging the word line of EEPROM
TW512347B (en) Method and circuit of erasing a flash memory

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MK4A Expiration of patent term of an invention patent