TWI236217B - Apparatus and method for adjusting the impedance of an output driver - Google Patents

Apparatus and method for adjusting the impedance of an output driver Download PDF

Info

Publication number
TWI236217B
TWI236217B TW93107376A TW93107376A TWI236217B TW I236217 B TWI236217 B TW I236217B TW 93107376 A TW93107376 A TW 93107376A TW 93107376 A TW93107376 A TW 93107376A TW I236217 B TWI236217 B TW I236217B
Authority
TW
Taiwan
Prior art keywords
impedance
bias
output
value
logic
Prior art date
Application number
TW93107376A
Other languages
Chinese (zh)
Other versions
TW200520381A (en
Inventor
James R Lundberg
Original Assignee
Ip First Llc
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
Priority claimed from US10/730,425 external-priority patent/US6949949B2/en
Application filed by Ip First Llc filed Critical Ip First Llc
Publication of TW200520381A publication Critical patent/TW200520381A/en
Application granted granted Critical
Publication of TWI236217B publication Critical patent/TWI236217B/en

Links

Abstract

An output impedance bias compensation system for adjusting output impedance of at least one output including a reference impedance generator, an impedance matching controller, at least one output impedance generator, and a programmable bias controller. The reference impedance generator develops a reference impedance based on a reference value. The impedance matching controller continually adjusts an input of the reference impedance generator to match the reference value within a predetermined tolerance. Each output impedance generator is coupled to a corresponding output and is controlled by an output impedance control input. The programmable bias controller combines a bias amount with the value of the input of the reference impedance generator to provide the output impedance control input. The bias controller is programmable to provide a bias amount to compensate for any process variations between the reference impedance generator and each output impedance generator.

Description

1236217 五、發明說明(1) 【相關申請案之交互參考】 [0001]本申請案主張案號60/434284,申請日為2002 年12月17日之美國暫時申請案(Provisional Application),以及案號為10/730425,申請日為2003年 12月5日的美國正式申請案,二件的優先權,其實際上在 此會併入做為參考。 [0 0 0 2 ]本申請案係與以下共同等候判定的美國專利申 請案有關,其與本申請案的申請曰相同,其具有共同受讓 人及至少一個共同發明人,並且其實際上會全部併入做為 參考。 序號“ 文件编號> 名Φ 10/730389- OUTPUT DRIVER IMPEDANCE CNTR.2116 CONTROLLER. 輸出驅動程式電路阻抗之控制器f 10/730169. APPARATUS AND METHOD FOR CNTR.2117, PRECISELY CONTROLLING TERMINATION IMPEDANCE.· 精喊控制终端阻抗之裝置及方法p 【發明所屬之技術領域】 [0003]本發明係有關於一種積體電路(1C)輸出驅動電 路,尤指一種調整輸出驅動程式電路的阻抗之方法及裝 置,用以補償遍及1C各處的製程變化。 【先前技術】1236217 V. Description of the invention (1) [Cross-reference to related applications] [0001] This application claims the case number 60/434284, the application date is the Provisional Application of the United States on December 17, 2002, and the case The number is 10/730425, and the application date is December 5, 2003. In the United States, the two applications of priority are actually incorporated herein by reference. [0 0 0 2] This application is related to the following U.S. patent applications that are jointly awaiting judgment, which is the same as the application of this application, which has a common assignee and at least one co-inventor, and that it will actually All are incorporated for reference. Serial number "File number> Name Φ 10 / 730389- OUTPUT DRIVER IMPEDANCE CNTR.2116 CONTROLLER. Controller of output driver circuit impedance f 10/730169. APPARATUS AND METHOD FOR CNTR.2117, PRECISELY CONTROLLING TERMINATION IMPEDANCE. · Device and method for controlling terminal impedance [Technical field to which the invention belongs] [0003] The present invention relates to an integrated circuit (1C) output driving circuit, and more particularly, to a method and device for adjusting the impedance of an output driver program circuit. To compensate for process variations throughout 1C. [Prior Art]

五、發明說明(2) 電路=4f]i;早的積體電路(ic)設:十中,⑽以動程式 置為推挽式(push-pul 1)兀件。因此,輸出匯流 所看到的雜訊會回應各種因素(包括電路溫度、供應 =壓、製程差異、匯流排上的元件數目等)而明顯地變 =變化(scaung),所以已迫使設計者積極地寸處理電;;的 匯μ排上的雜訊問題,以使系統内的電路之運作速度 :::5流排ϋ t包括群集地位☆系統板或類似之物:的一 /、或夕條訊號線,其中每條訊號線可模型化 (例如,反射,串音(cr〇ss talk)等)支配的傳輸^雜°孔 [00 05]更近的輸出驅動程式電路解決方式中的一 已使工業從推挽式輸出配置改變為差動接收器配置。在差 動接收器配置t,差動接收器的一側係以參考電壓來供 L而側係由開路汲㈣通道元件來驅動。開路汲極N 兀件係位於晶片上,而匯流排上拉終端阻抗 (tennination) —般係位於外部,通常位於系統主機板或 Ϊ =亡二ΪΪΠ端阻抗位於主機板上,可使系統設 的匯流排雜訊問題時,具有-定程 變的二。』6」在ΐ述之輪出驅動程式電路的型式已 Π Χ86微處理器(英特爾公司的產==子係由Pentium 係使用開路汲極N通道輸出元件:&、不。Pentlum 1 1 參考臨界電壓m.5V匯流排件關/驅,動具有h0伏特⑴ 机排關於此處理器的主機板一般 1236217V. Description of the invention (2) Circuit = 4f] i; The early integrated circuit (ic) is set: Ten, and the moving program is set as push-pul 1 element. Therefore, the noise seen by the output bus will obviously change = scaung in response to various factors (including circuit temperature, supply = voltage, process difference, number of components on the bus, etc.), so the designer has been forced to actively The ground handles the noise problem on the μ row, so that the operating speed of the circuit in the system ::: 5 current row ϋ t includes cluster status ☆ system board or similar: 1 /, or evening Signal lines, where each signal line can be modeled (for example, reflection, cr0ss talk, etc.) dominated transmission ^ Miscellaneous holes [00 05] One of the more recent output driver circuit solutions The industry has changed from a push-pull output configuration to a differential receiver configuration. In the differential receiver configuration t, one side of the differential receiver is provided with a reference voltage for L and the side is driven by an open-drain channel element. The open drain N element is located on the chip, and the bus pull-up terminal impedance (tennination) is generally located externally, usually on the system motherboard or Ϊ = Ϊ 二 ΪΪΠ terminal impedance is located on the motherboard, which allows the system to set When the bus noise problem, there are two-fixed range changes. 』6 The type of driver circuit that has been described in the wheel of the description has been Π86 microprocessor (product of Intel Corporation == sub-series by Pentium series using open-drain N-channel output elements: & no. Pentlum 1 1 Reference Threshold voltage m.5V busbar switch on / off, drive with h0 volts. The motherboard of this processor is generally 1236217

端阻抗。雖然還未指定特定的下拉 換及時序規格。然而,未對製:式 行補償:會使開路沒極n通道輸出驅動程式電路皿的 :之可月匕變化範圍約從4歐姆變化到8〇歐 處理器的設計者只能預先考慮製程 且田 的可接受範圍,所以已迫使Pentium_n相容主二= 將2-3奈秒(ns)等級的變動率(slew rate)控制2寿到 號緣,以降低輸出匯流排上的雜訊。 會使用 阻抗, 流排切 變化進 通道電 因為微 度變動 的設計 輸出訊 [ 00 07 ]在Pentium-ηι中,英特爾引進一種機制, 此會提供可用來設定匯流排上的輸出驅動程式電路之曰 的參考阻抗給設計者。處理器封裝上的接腳(稱為 几 NCHCTRL)係經由精密的14歐姆電阻(最大指定電阻值為 歐姆)而連接至匯流排電壓(稱為νττ)。精密電阻係外接於 微處理器晶片,因此與晶片上的輸出驅動程式電路所看到 的溫度及電壓變化無關。因此,外接電阻係用來當作設定 開路汲極Ν通道輸出驅動程式電路的下拉阻抗之參考。 [0 0 0 8 ]此外,相容配置的上拉終端阻抗 (termination)係位於晶片上,而不是位於系統的主機板 上。對於上拉終端阻抗而言,稱為RTT的另一個接腳係用 以使精密電阻R可連接於此接腳與接地之間。跨接於精密 電阻的阻抗係表示所有上拉終端阻抗之需要的阻抗。因 此,系統設計者能經由一個外接電阻,而設定部件上的戶 有訊號之匯流排上拉阻抗。根據規格書,電阻的範圍Terminal impedance. Although specific pull-down and timing specifications have not been specified. However, the system is not compensated: the line compensation: will make the open-circuit n-channel output driver circuit circuit board: the range of the month can be changed from about 4 ohms to 80 ohms. The designer of the processor can only consider the process in advance and The acceptable range of the field has forced Pentium_n to be compatible with the main two = control the slew rate of 2-3 nanoseconds (ns) level to 2 years to the edge to reduce noise on the output bus. Will use the impedance, the bus cuts into the channel due to the slight change in the design output signal [00 07] In Pentium-ηι, Intel introduced a mechanism that will provide the output driver circuit that can be used to set the bus The reference impedance is given to the designer. The pins on the processor package (referred to as a few NCHCTRL) are connected to the bus voltage (referred to as νττ) through a precision 14 ohm resistor (the maximum specified resistance value is ohms). The precision resistor is external to the microprocessor chip and is therefore independent of the temperature and voltage changes seen by the output driver circuit on the chip. Therefore, the external resistor is used as a reference for setting the pull-down impedance of the open-drain N-channel output driver circuit. [0 0 0 8] In addition, the pull-up termination impedance of the compatible configuration is located on the chip rather than on the system board. For the pull-up termination impedance, another pin, called RTT, is used to allow a precision resistor R to be connected between this pin and ground. The impedance across the precision resistor is the impedance required for all pull-up termination impedances. Therefore, the system designer can set the pull-up impedance of the bus of the user signal on the component through an external resistor. Range of resistance according to specification

1236217 五、發明說明(4) 4 0到1 3 0歐姆,因此使系統設計处、 流排上的上拉終端阻抗,來補此N通道開路汲極匯 [ 000 9 ]本發明所引用參辟雨二二 描述了用以精確控制N通道開路1件美^國專利申請案,係 祖、、、鈿阻抗的裝置與方法,而本上拉或下 進位陣列的N通道或p通道元件當作終之例係使用二 數係由用以監測參考阻抗的邏輯之片$…。導通的元件 來決定。然:而’若遍及晶粒各處有製‘蠻:寫二進位陣列 寫二進位陣列與輸出驅動程式電路$ 姓廷會導致複 差異,則由參考陣列所決定 數::二:間的顯著 實最佳數目。 数不會疋輸出陣列的真 【發明内容 [0010] 償系統,用 抗產生器、 以及可程式 考阻抗控制 以持續調整 配的偏差在 一對應輸出 壓控制器係 生輪出阻抗 [0011] 根據本發 以調整至 阻抗匹配 偏壓控制 輸入所控 參考阻抗 預定容忍 ,並且由 用以將偏 控制輸入 明的一實施 少一個輸出 控制器、至 器。參考阻 制之參考阻 控制輸入, 度内。每個 輸出阻抗控 壓量與參考 例之一種輸 的輸出阻抗 少一個輸出 抗產生器係 抗。阻抗匹 以使參考阻 輸出阻抗產 制輸入來控 阻抗控制輸 出阻抗 ,包括 阻抗產 用以產 配控制 抗與參 生器係 制。可 入結合 偏壓補 參考阻 生器、 生由參 器係用 考值匹 耦接至 程式偏 ’而產 在另一實 施例中’偏壓控制器包括輸出偏壓邏 第10頁 五、發明說明(5) 輯及偏壓調 量,而偏壓 結合’而產 發性邏輯可 記憶體等。 值,其可被 為一百分比 抗控制輸入 [0012] 路(1C),具 阻與至少一 阻抗產生器 生器係由輸 腳。阻抗匹 出調整邏輯 邏輯係用以 阻與可程式 度内。輸出 整值結合, 整邏輯。 調整邏輯 生輸出阻 適用於將 偏壓量可 加入到參 ’其代表 可加上或 根據本發 有輸出阻 個輸出接 、以及阻 出阻抗控 配邏輯包 控制輸入 週期性地 參考阻抗 調整邏輯 而產生輸 輸出偏壓 係用以將 抗控制輸 偏壓量程 包括符號 考阻抗控 參考阻抗 減去此百 明的再一 抗調整, 腳的參考 抗匹配邏 制輸入來 括由參考 之可程式 調整參考 產生器的 係用以將 出阻抗控 邏輯可 偏壓量 入。任 式化, 或極性 制輸入 控制輸 分比。 實施例 包括用 接腳、 輯。每 控制, 阻抗、參考阻 阻抗控 值之間參考阻 制輪入 程式化 與參考 何型式 如保險 的位元 〇另一 入的百 ’而產生偏壓 阻抗控制輸入 的可程式非揮 絲’非揮發性 ,或可為符號 種是,偏壓量 为比,參考阻 所提供之一種 以耦接至外部 至少一個可程 個可程式輸出 並且耦接至一 比較器邏輯、 抗產生器、。 制輸入,以使 的差距位於預 抗控制輸入與 積體電 參考電 式輪出 阻抗產 輸出接 以及輸 比較器 參考電 定容忍 偏壓調 考 將 阻 與 [^01 3]根據本發明之更一實施例所提供之一種基於夂 阻抗而調整1C之至少一個輸出的輸出阻抗之方法,包ς f考電壓施加到參考阻抗及參考阻抗產生器,調整參 t產生器的參考阻抗輸入,以使參考阻抗產生器的阻 參考阻抗的差距在預定容忍度内,量測參考阻抗與至^ 12362171236217 V. Description of the invention (4) 40 to 130 ohms, so the pull-up terminal impedance on the system design and on the bus is used to make up the N-channel open-drain sink [000 9] Yu Er'er describes a device and method for accurately controlling an open circuit of the N channel in a US patent application, and the impedance of the ancestral,, and 钿 are used. The N channel or p channel element of the pull-up or down-carry array is used as The final example is the use of two by the logic piece $ ... to monitor the reference impedance. The conducting element is determined. However: 'If there is a system throughout the die', it is: writing a binary array. Writing a binary array and the output driver circuit $ The surname will cause a complex difference, which is determined by the reference array. Real best number. [0010] Compensation system, using an anti-generator, and programmable impedance control to continuously adjust the deviation of a matching output impedance of a corresponding output voltage controller system [0011] according to In the present invention, the reference impedance controlled by adjusting to the impedance matching bias control input is predetermined tolerant, and one implementation is used to clarify the bias control input and one output controller is omitted. Reference resistance reference resistance control input, within degrees. The output impedance of each output impedance is less than the output impedance of one of the reference examples. The impedance matches the reference impedance output impedance production input to control the impedance control output impedance, including the impedance production to produce the control reactance and the generator system. The reference bias device can be combined with a bias bias, and the reference device is coupled to the program bias with the reference value. It is produced in another embodiment. The bias controller includes an output bias logic. Page 10 V. Invention Explain (5) to edit and adjust the bias voltage, and the combination of bias voltage and memory can be generated by logic. Value, which can be a percentage of the anti-control input [0012] circuit (1C), with resistance and at least one impedance generator. Impedance Matching Logic Logic is used for resistance and programmability. Output Integer value combination, whole logic. Adjusting the logic output resistance is suitable for adding a bias amount to the parameter. Its representative can be added or based on the output. There is an output resistance, and the impedance control logic package control input periodically refers to the impedance adjustment logic. The generation of the output and output bias voltage is used to reduce the range of the impedance of the control input bias, including the symbol, the reference impedance, and the reference impedance. The generator is used to bias the output impedance control logic into the input. Arbitrary, or polarity input control input ratio. Examples include using pins and series. For each control, between the impedance and the reference resistance, the reference resistance is programmed in turn and the reference is made to a type such as a safe bit. Another one hundred is used to generate a programmable non-swinging wire for the impedance impedance control input. Volatility, or a symbolic type, is the ratio of the bias voltage. The reference resistor provides one that is coupled to at least one programmable external programmable output and coupled to a comparator logic, anti-generator, and so on. Control input so that the gap is between the pre-impedance control input and the integrated electrical reference output wheel output impedance output connection and the input comparator reference electrical setting tolerance bias voltage adjustment will be blocked by [^ 01 3] according to the present invention. An embodiment provides a method for adjusting the output impedance of at least one output of 1C based on chirped impedance, including applying a reference voltage to a reference impedance and a reference impedance generator, and adjusting a reference impedance input of a reference generator so that The difference between the impedance of the reference impedance generator and the reference impedance is within a predetermined tolerance. Measure the reference impedance to ^ 1236217

一個輸出阻抗 揮發性元件程 整值與參考阻 少一個輸出阻 之間的任何差 式化,以補償 抗輸入結合, 抗產生器的輸 異,以偏壓調 任何量測差異 而產生耦接至 出阻抗輸入。 整值將I c上的非 ;以及將偏壓調 一對應輸出的至 【實施方式】 [〇 0 2 4 ]以下說明係使一般的熟習此項技術者能完成及 使用本發明’如特定應用及其需求的本文内所提供的。然 而s對於熟習此項技術者而言,對較佳實施例的各種修改 將疋,然可知的,並且在此所定義的一般原則可應用於其 他實^例。因此’本發明不意欲受限於在此所顯示及說明 的特定實施例’而是符合在此所揭露的原則及新穎性之最 廣的範圍。 [0 0 2 5 ]本申請案的發明人已瞭解需補償參考阻抗產生 器與包括上拉或下拉元件的輸出阻抗產生器之間,遍及晶 粒各處(across-die)之製程變化。因此,其已發展一種調 整輸出驅動程式電路的阻抗之裝置及方法,茲將配合附圖 1 - 7之圖示針對本發明之技術特徵作進一步說明。 [0 0 2 6 ]圖1係包括用以精確地控制傳輸線的終端阻抗 之一系統的積體電路(1C) 1〇1之簡化方塊圖。其中ic 101 包括一些可用於外接的輸入/輸出(I/O)接腳,包括參考電 阻接腳RTT及多個輸出接腳,個別地顯示為0UT1、〇UT2、 …、0UTN,其中Ν為正整數。除非有其他的指定,一個接 腳及其傳送的訊號會以相同名稱來稱呼。1C 101會產生電Any difference between the output impedance of the volatile component and the reference resistance is reduced by one output resistance to compensate for the anti-input combination, anti-generator output variation, and bias any measurement difference to produce coupling Out impedance input. The integer value will be negated on I c; and the bias will be adjusted to correspond to the output. [Embodiment] [〇0 2 4] The following description is to enable a person skilled in the art to complete and use the present invention, such as a specific application And their requirements are provided in this article. However, it will be apparent to those skilled in the art that various modifications to the preferred embodiment will be apparent, and that the general principles defined herein may be applied to other examples. Thus, the invention is not intended to be limited to the specific embodiments shown and described herein, but is to be accorded the broadest scope consistent with the principles and novelty disclosed herein. [0 0 2 5] The inventors of the present application have understood the need to compensate for process variations across the die between the reference impedance generator and the output impedance generator including a pull-up or pull-down element. Therefore, it has developed a device and method for adjusting the impedance of the output driver circuit, and will further explain the technical features of the present invention with reference to the drawings of FIGS. 1-7. [0 0 2 6] FIG. 1 is a simplified block diagram of an integrated circuit (1C) 101 which includes a system for precisely controlling the termination impedance of a transmission line. Among them, ic 101 includes some input / output (I / O) pins that can be used for external connection, including the reference resistance pin RTT and multiple output pins, which are individually displayed as OUT1, OUT2, ..., OUTN, where N is positive Integer. Unless otherwise specified, a pin and its transmitted signal will be referred to by the same name. 1C 101 will generate electricity

第12頁 1236217 五、發明說明(7) 壓參考訊號,或接收供應電壓訊號〇])。〇D可從相對於接 f(GND)接腳(未顯示)的外部接腳來提供。在所顯示的實 施例中,以虛線顯示的的參考電阻R係外接於接腳rtt與接 地之間2雖然本發明並未限制任何特定值、範圍或電阻型 式,但是根據規格書,電阻R各處的範圍係介於4〇到13〇歐 姆之間,並且可以是精密電阻或類似的電阻(例如,1%電 阻)。Page 12 1236217 V. Description of the invention (7) Press the reference signal or receive the supply voltage signal 0]). 〇D can be provided from an external pin relative to the f (GND) pin (not shown). In the embodiment shown, the reference resistance R shown in dotted lines is externally connected between the pin rtt and the ground. 2 Although the present invention does not limit any specific value, range or resistance type, according to the specification, the resistance R The range is between 40 and 13 ohms, and can be a precision resistor or similar (for example, a 1% resistor).

[0027]IC 101更包括有阻抗匹配邏輯103,其用以接 收VDD訊號’以及用以監測參考電阻r及内部阻抗產生器 2 0 7的阻杬(如圖2所示)。本實施例中,阻抗匹配邏輯丨〇 3 係用以監測RTT接腳的電位,並且會將6位元内部匯流排 1 0 5上的6位元數位值SUM [ 5 : 0 ]傳送到多個偏壓控制邏輯元 件1 0 6,其依序麵接至位於I c 1 〇 1上的對應終端阻抗或上 拉邏輯元件1 0 7 (個別地從1到n予以編號)。每個個別的上 拉邏輯元件107會接收VDD訊號,並且耦接至輸出接腳〇UTx (其中「X」為從1到代表特定輸出接腳的Ν之任何整數)中 的對應一個。在每個上拉邏輯元件丨07内,調整過的SUM [5:0]值(亦即,PSUM —X[5:0])中的每個位元會致能/除能 一對應群組的一陣列匹配p通道元件,其係可具有共汲極 點並係可用來上拉及終止對應的〇UTx接腳之。PSUM_X [5:0]值會指定每個上拉邏輯元件1〇7内之欲導通的p通道 元件之數目,以上拉及終止在指定容忍度内之對應的〇UTx 訊號。在所顯示的實施例中,PSUM__X[ 5 : 0 ]值會以64個相 等間隔的步驟,而可調整上拉邏輯元件1 〇 7的阻抗。[0027] The IC 101 further includes impedance matching logic 103, which is used to receive the VDD signal ' and to monitor the resistance of the reference resistor r and the internal impedance generator 207 (as shown in FIG. 2). In this embodiment, the impedance matching logic 丨 〇3 is used to monitor the potential of the RTT pin, and transmits the 6-bit digital value SUM [5: 0] on the 6-bit internal bus 105. The bias control logic element 106 is sequentially connected to the corresponding terminal impedance or pull-up logic element 10 7 (individually numbered from 1 to n) located on I c 1 〇1. Each individual pull-up logic element 107 receives a VDD signal and is coupled to a corresponding one of the output pins OUTx (where "X" is any integer from 1 to N representing a particular output pin). Within each pull-up logic element 07, each bit in the adjusted SUM [5: 0] value (that is, PSUM —X [5: 0]) will enable / disable a corresponding group An array of matched p-channel elements can have a common sink point and can be used to pull up and terminate the corresponding OUTx pin. The PSUM_X [5: 0] value specifies the number of p-channel components to be turned on in each pull-up logic element 107, which pulls up and terminates the corresponding UTx signal within the specified tolerance. In the embodiment shown, the PSUM__X [5: 0] value is adjusted in 64 equally spaced steps, and the impedance of the pull-up logic element 107 can be adjusted.

第13頁 1236217Page 13 1236217

元件[的〇〇二1在ΐ作時,阻抗匹配邏輯103會保持匹配p通道 107中、的二進鱼位陣列’其實質上與每個上拉邏輯元件 、一進位陣列相同。每個陣列會配置或分成用於數 上:ff制的二進位群組’如底下進-步的說明。會 持續監測阻抗匹配邏輯1〇3内的局部二進位陣列之阻 f且會週期性地調升或調降SUM[5:〇]值,以致於跨接内部 陣列的電壓與位於跨接電阻R的電壓的差距係在一預定容 忍度内。在一實施例中,預定容忍度為約5〇毫伏特(mv)的 誤差電壓。匯流排時脈(INT Bcu)的選擇週期(例如,每 二個INT BCLK週期)會決定或週期性地更新上拉邏輯元件 107的最佳阻抗,並且SUM[5: 0]值會藉由匯流排1〇5,而傳 送到每個偏壓控制邏輯元件106,而使上拉邏輯元件1〇7明 顯地更新。 [0 0 2 9 ]偏壓控制邏輯元件丨〇 6係用以使位於匯流排 1 05上的SUM [5:0]值加上或減去偏壓。這可用於實施每個 上拉邏輯元件1 0 7的阻抗調整,來補償遍及晶粒各處的製 程變化。在一實施例中,偏壓控制器丨〇 6係用於每個上拉 邏輯元件1 0 7。在另一實施例中,偏壓控制器係用於上拉 邏輯元件107中的局部群組。 [0 0 3 0 ]圖2係阻抗匹配邏輯1 〇 3之一範例實施例的更詳 細方塊圖。阻抗匹配邏輯1 0 3包括阻抗控制器2 〇 1,其用以 接收INT BCLK、VDD及RTT訊號。阻抗控制器2〇1包括電壓 感測器203,其用以接收VDD訊號,並且用以監測RTT接腳 的電壓(局部地顯示為訊號I NP )。I NP訊號會傳送到阻抗產During the operation of the component [0021], the impedance matching logic 103 will keep matching the binary fish array of the p-channel 107, which is substantially the same as each pull-up logic element and the binary array. Each array will be configured or divided into a binary group for the number: ff as described below. The resistance f of the local binary array in the impedance matching logic 103 will be continuously monitored and the SUM [5: 〇] value will be increased or decreased periodically, so that the voltage across the internal array and the resistance R The voltage difference is within a predetermined tolerance. In one embodiment, the predetermined tolerance is an error voltage of about 50 millivolts (mv). The selection period of the bus clock (INT Bcu) (for example, every two INT BCLK cycles) determines or periodically updates the optimal impedance of the pull-up logic element 107, and the value of SUM [5: 0] is determined by the bus Row 105 is passed to each bias control logic element 106, and pull-up logic element 107 is significantly updated. [0 0 2 9] The bias control logic element 06 is used to add or subtract a bias to the value of SUM [5: 0] located on the bus 105. This can be used to implement impedance adjustments for each pull-up logic element 107 to compensate for process variations throughout the die. In one embodiment, the bias controller 106 is used for each pull-up logic element 107. In another embodiment, the bias controller is used for a local group in the pull-up logic element 107. [0 0 3 0] FIG. 2 is a more detailed block diagram of an exemplary embodiment of the impedance matching logic 103. The impedance matching logic 103 includes an impedance controller 201, which is used to receive INT BCLK, VDD, and RTT signals. The impedance controller 201 includes a voltage sensor 203 for receiving a VDD signal and for monitoring the voltage of the RTT pin (locally shown as a signal I NP). The I NP signal is transmitted to the impedance

1236217 五、發明說明(9) 生器207,其係基於6位元輸入控制值SUM[5 : 〇 ]而顯示VDD 與I NP訊號之間的阻抗。電壓感測器2〇3可有效地比較VDD 與I NP訊號的電壓,而產生送到阻抗控制邏輯2 〇 5的訊號η【 及LO,以試圖使跨接阻抗產生器20 7之電位,與電阻r之電 位相差在一預定容忍度内。阻抗控制邏輯205會回應HI/LO 訊號,而增加/降低SUM[ 5 : 0 ]值,以控制阻抗產生器20 7的 阻抗,直到(VDD — INP)與INP的差距不大於預定誤差電壓 (或以致於INP訊號的電壓與VDD電壓之一半的差距在預定 誤差電壓内)。換句話說,電壓感測器2 〇 3及阻抗控制邏輯 2 0 5會配合,以試圖使阻抗控制器2 〇 7的阻抗(藉由電壓)與 電阻R之阻抗(藉由電壓)相差在預定容忍度(藉由誤差電壓 量)内。 [003 1 ]VDD來源電壓係藉由電阻R及阻抗產生器207的 阻抗而分壓,而提供INP訊號的中間電壓。若INP訊號的電 壓太高(係表示阻抗產生器2 0 7的阻抗太低或低於電阻R的 電阻值),則電壓感測器203會使HI訊號致能,並且使LO訊 號無效。阻抗控制邏輯205會藉由降低SUM [ 5 : 0 ]值來回 應,以增加阻抗產生器207的阻抗。當INP訊號太低(係表 示相對於電阻R,阻抗產生器207的阻抗太高)時,電壓感 測器20 3會使LO訊號致能,並且會使HI訊號無效。阻抗控 制邏輯20 5會藉由增加SUM[ 5 : 0]值來回應,以降低阻抗產 生器2 0 7的阻抗。在所顯示及說明的實施例中,雖然也考 慮比例關係,但是SUM [ 5 : 0 ]值係與阻抗產生器2 0 7的阻抗 成反比。1236217 V. Description of the invention (9) The generator 207 displays the impedance between the VDD and I NP signals based on the 6-bit input control value SUM [5: 〇]. The voltage sensor 20 can effectively compare the voltages of the VDD and I NP signals and generate a signal η [and LO sent to the impedance control logic 2 05 to try to make the potential across the impedance generator 20 7 and The potentials of the resistors r differ within a predetermined tolerance. The impedance control logic 205 will respond to the HI / LO signal and increase / decrease the value of SUM [5: 0] to control the impedance of the impedance generator 20 7 until the difference between (VDD — INP) and INP is not greater than a predetermined error voltage (or So that the gap between the voltage of the INP signal and half of the VDD voltage is within a predetermined error voltage). In other words, the voltage sensor 2 03 and the impedance control logic 2 05 will cooperate to try to make the impedance (by voltage) of the impedance controller 2 07 differ from the resistance (by voltage) of the resistor R by a predetermined value. Within tolerance (by the amount of error voltage). [003 1] The VDD source voltage is divided by the resistor R and the impedance of the impedance generator 207 to provide an intermediate voltage of the INP signal. If the voltage of the INP signal is too high (indicating that the impedance of the impedance generator 207 is too low or lower than the resistance value of the resistor R), the voltage sensor 203 enables the HI signal and invalidates the LO signal. The impedance control logic 205 responds by reducing the SUM [5: 0] value to increase the impedance of the impedance generator 207. When the INP signal is too low (meaning that the impedance of the impedance generator 207 is too high relative to the resistor R), the voltage sensor 20 3 enables the LO signal and invalidates the HI signal. The impedance control logic 20 5 responds by increasing the value of SUM [5: 0] to reduce the impedance of the impedance generator 207. In the embodiment shown and described, although a proportional relationship is also considered, the SUM [5: 0] value is inversely proportional to the impedance of the impedance generator 207.

第15頁 1236217 五、發明說明(10) [0 0 3 2 ]在一實施例中,電壓感測器2 0 3包括一對感測 放大器(未顯示),其電壓設定係與VDD電壓的一半相隔一 預定誤差電壓。在此情況中,高感測放大器具有約為超過 1 / 2 V D D加上一半誤差電壓的設定點,用以控制η I訊號,低 感測放大器具有設定為低於1/2 VDD減去一半誤差電壓的設 定點,用以控制LO訊號。每個感測放大器會比較與其設定 點相關的I ΝΡ訊號之電壓。若I ΝΡ訊號的電壓上升到超過誤 差電壓的一半,則會使ΗI致能,若I ΝΡ下降到低於誤差電 壓的一半,則會使LO致能,而若ΙΝΡ與1/2VDD的差距不超 過1/2誤差電壓,則不會使HI或LO致能,並且不會採取動 作。在一更特定的實施例中,預定誤差電壓約為5〇mV,以 致於高感測放大器係設定約為超過1/2VDD加上25mV,而 低感測放大器係設定約為低於1/2VDD減去25mV。誤差電 壓的間隙可設定用於較高精確度的嚴格容忍度,或者是設 定為相當寬的容忍度,以節省功率。 [ 0 033 ]在一實施例中,阻抗控制邏輯2〇5為由INT BCLK訊號所控制的數位電路,並且在INT BCLK訊號的選擇 週期期間(如每個時脈週期或每隔一個時脈週期等)會調整 (例如,增加或降低)SUM[5:0]值。 [0034]現在請參閱圖3,所顯示的是附圖1中的偏壓控 制邏輯106的更詳細之方塊圖。偏壓控制邏輯1〇6具有非^ 發性邏輯302,其耦接至輸出偏壓邏輯3〇1。輸出偏壓邏輯 3W係經由訊號PADD[3:0]及psuBEN,而耦接至偏壓調整邏 輯3 03。讯唬INT BCLK及SUM[5: 0]會傳送到偏壓調整邏輯Page 15 1236217 V. Description of the invention (10) [0 0 3 2] In one embodiment, the voltage sensor 2 0 3 includes a pair of sense amplifiers (not shown), the voltage setting of which is half the VDD voltage Separated by a predetermined error voltage. In this case, the high-sense amplifier has a set point of approximately more than 1/2 VDD plus half the error voltage to control the η I signal, and the low-sense amplifier has a setting lower than 1/2 VDD minus half the error. The voltage set point is used to control the LO signal. Each sense amplifier compares the voltage of the I NP signal relative to its set point. If the voltage of the I NP signal rises to more than half of the error voltage, it will enable ΗI. If the value of I NP drops to less than half of the error voltage, it will enable the LO, and if the difference between INP and 1 / 2VDD is not Exceeding 1/2 error voltage will not enable HI or LO and will not take action. In a more specific embodiment, the predetermined error voltage is about 50 mV, so that the high-sense amplifier system is set to approximately more than 1 / 2VDD plus 25mV, and the low-sense amplifier system is set to approximately less than 1 / 2VDD. Subtract 25mV. The gap of the error voltage can be set to tight tolerances for higher accuracy or to a fairly wide tolerance to save power. [0 033] In an embodiment, the impedance control logic 205 is a digital circuit controlled by the INT BCLK signal, and during a selection period of the INT BCLK signal (such as each clock cycle or every other clock cycle) Etc.) will adjust (for example, increase or decrease) the SUM [5: 0] value. [0034] Referring now to FIG. 3, a more detailed block diagram of the bias control logic 106 in FIG. 1 is shown. The bias control logic 106 has a non-volatile logic 302, which is coupled to the output bias logic 301. The output bias logic 3W is coupled to the bias adjustment logic 303 via the signals PADD [3: 0] and psuBEN. INT BCLK and SUM [5: 0] are sent to the bias adjustment logic

1236217 五、發明說明(11) 303,其會產生對應訊號PSUM—χ[5··〇],如圖1所顯示。 [ 0035 ]在運作時,在時脈訊號INT BCLK的選擇週期期 間(如每隔一個時脈週期或類似的時脈週期),偏壓調整邏 輯303會基於PADD[3:0]的值及訊號PSUBEN的狀態,來調整 (例如,增加或降低)PSUM—χ[5:〇]值。4位元值PADD[3:〇] 係從輸出偏壓邏輯301傳送到偏壓調整邏輯303,以識別 SUM[5:0]值要加上或減去的數量。符號或極性訊號psuBEN 會由輸出偏壓邏輯301傳送到偏壓調整邏輯3〇3,以決定是 要加上(當PSUBEN未致能時)或減去(當PSUBEN致能時)此數 量。PSUBEN訊號及PADD[ 3 ·· 0]值會共同地構成訊號偏壓調 整值。在一實施例中,SUM[5:0]值會直接加上(例如,當 PSUBEN為邏輯0或未致能),或直接減去(例如,當psuBEN 為邏輯1或致能)PADD[ 3 : 0 ]值。在此情況中,pADD[ 3 : 〇 ]值 係表示固定量的偏壓達到SUM[5:〇]值的1/4範圍。在另一 實施例中’ SUM[ 5 : 0 ]值係根據PADD [3 : 〇 ]及psuBEN訊號而 成比例地增加或降低。例如,若pADD[ 3 ·· 〇 ]係設定於丨〇〇〇b (二進位),並且PSUBEN未致能,則SUM[5:〇]會以百分之5〇 增加。 [0 0 3 6 ]在另一特定實施例中,輸出偏壓邏輯3 〇1包括 包含於ic ιοί上的可程式非揮發性邏輯元件3〇2,或係由 其所程式化。可考慮任何型式的非揮發性可程式元件,如 任何型式的非揮發性記憶體或一組保險絲或類似元件。例 如,可燒斷一條或多條保險絲,或者是設定或程式化非揮 發性記憶體元件的數個位元,而產生pADD[ 3 ·· 〇 ]值及1236217 V. Description of the invention (11) 303, which will generate the corresponding signal PSUM_χ [5 ·· 〇], as shown in FIG. 1. [0035] During operation, during the selection period of the clock signal INT BCLK (such as every other clock cycle or similar clock cycle), the bias adjustment logic 303 will be based on the value and signal of PADD [3: 0] PSUBEN state to adjust (eg, increase or decrease) the value of PSUM_χ [5: 〇]. The 4-bit value PADD [3: 〇] is transferred from the output bias logic 301 to the bias adjustment logic 303 to identify the amount to be added or subtracted from the SUM [5: 0] value. The sign or polarity signal psuBEN is passed from the output bias logic 301 to the bias adjustment logic 303 to determine whether to add (when PSUBEN is disabled) or subtract (when PSUBEN is enabled) this number. The PSUBEN signal and the PADD [3 ·· 0] value collectively constitute the signal bias adjustment value. In one embodiment, the SUM [5: 0] value is directly added (for example, when PSUBEN is logic 0 or disabled), or directly subtracted (for example, when psuBEN is logic 1 or enabled) PADD [3 : 0] value. In this case, the value of pADD [3: 〇] indicates that a fixed amount of bias voltage reaches a range of 1/4 of the value of SUM [5: 〇]. In another embodiment, the value of 'SUM [5: 0] is proportionally increased or decreased according to the PADD [3: 0] and psuBEN signals. For example, if pADD [3 ·· 〇] is set to 丨 00b (binary) and PSUBEN is not enabled, SUM [5: 〇] will increase by 50%. [0 0 3 6] In another specific embodiment, the output bias logic 3 01 includes, or is programmed by, a programmable non-volatile logic element 302 included in the IC. Any type of non-volatile programmable components can be considered, such as any type of non-volatile memory or a group of fuses or similar components. For example, one or more fuses can be blown, or several bits of a non-volatile memory element can be set or programmed to produce a pADD [3 ·· 〇] value and

IBM 第17頁 1236217 五、發明說明(12) PSUBEN訊號。元件302的程式化狀態係藉由以一部份一部 份為基礎的測試程序或類似的程序來決定。在一實施例 中’元件302中的幾乎每一個位元係對應於SUM[5:0]值的 較低位元。以此方式,將元件3〇2程式化可使設計者增加 或降低SUM[5:0]值。因此,輸出偏壓邏輯3〇1為一種使設 計者能補償遍及I C 1 0 1各處的製程變化。 [〇 〇 3 7 ]附圖4係阻抗產生器4 0 0的一範例實施例之概 要圖’其可用來實施阻抗產生器207,及/或實施上拉邏輯 元件107中的任一個。阻抗產生器4〇〇包括二進位陣列的63 個P通道元件PI-P63(或P63:P1)。在一實施例中,每個p通 道元件P 6 3 : P1會匹配,以致於汲極到源極的阻抗實質上會 相同。每個元件P63:P1的源極係耦接至VDD,而汲極係耦 接至上拉訊號PUP,其代表阻抗產生器2〇7的INP訊號,或 任何上拉邏輯元件1〇7的對應OUTx訊號。元件Ρ63··Ρ1係二 進位地分組,以對應二進位阻抗值XSUM[5 : 〇](當實施阻抗 產生器20 7時,其代表SUM[5:〇]值,或當實施任何上拉邏 輯元件107時’其代表PSUM_x[5:〇]值)之六個位元中的每 一個。第一陣列群組為單一元件ρι,其具有用以接收訊號 PS0的閘極,第二陣列群組401包括二個元件P2及P3 (P3:P2),每個具有用以接收訊號PS1的閘極,第三陣列群 組4 03包括四個元件p4 —p7(p7:p4),每個具有用以接收訊 號PS2的閘極,第四陣列群組4〇5包括八個元件p8-pi5 (P15:P8) ’每個具有用以接收訊號pS3的閘極,第五陣列 群組40 7包括16個元件1^16-1)31(?31:1>16),每個具有用以IBM Page 17 1236217 V. Description of the invention (12) PSUBEN signal. The stylized state of the element 302 is determined by a part-by-part test procedure or the like. In one embodiment, almost every bit in the 'element 302 corresponds to the lower bit of the SUM [5: 0] value. In this way, programming component 302 allows the designer to increase or decrease the SUM [5: 0] value. Therefore, the output bias logic 301 is a way for the designer to compensate for process variations throughout I C 101. 4 is a schematic diagram of an exemplary embodiment of the impedance generator 400, which can be used to implement the impedance generator 207, and / or to implement any one of the pull-up logic elements 107. The impedance generator 400 includes 63 P-channel elements PI-P63 (or P63: P1) in a binary array. In one embodiment, each p-channel element P 6 3: P1 is matched so that the impedance from the drain to the source will be substantially the same. The source of each component P63: P1 is coupled to VDD, and the drain is coupled to the pull-up signal PUP, which represents the INP signal of the impedance generator 207, or the corresponding OUTx of any pull-up logic component 107 Signal. The components P63 ·· P1 are grouped in binary to correspond to the binary impedance value XSUM [5: 〇] (when the impedance generator 20 7 is implemented, it represents the value of SUM [5: 〇], or when any pull-up logic is implemented Element 107 'represents each of the six bits of PSUM_x [5: 0] value). The first array group is a single element p1 having a gate for receiving the signal PS0, and the second array group 401 includes two elements P2 and P3 (P3: P2), each having a gate for receiving the signal PS1 The third array group 403 includes four elements p4-p7 (p7: p4), each having a gate for receiving the signal PS2, and the fourth array group 405 includes eight elements p8-pi5 ( P15: P8) 'Each has a gate for receiving the signal pS3, the fifth array group 40 7 includes 16 elements 1 ^ 16-1) 31 (? 31: 1 > 16), each with

第18頁Page 18

1236217 五、發明說明(13) 接收訊號PS4的閘極,以及第六陣列群組3〇9包括32個元件 P3 2_P63(P6 3 : P32),每個具有用以接收訊號PS5的閘極。 [0038]PS5-PSO訊號會共同構成由緩衝器4ΐι(其用以 接收XSUM[5 : 0 ]值)所致能的二進位值。pS5-pS〇訊號中的 母個為緩衝形式之XSUM [ 5 : 0 ]值中的對應位元。例如,會 緩衝SUM5位元,而產生NS5位元,會緩衝SUM4位元,而產 生NS4位元等等。因此,當SUM[5:0]值提升或增加時,介 於VDD與PUP之間的阻抗會降低,且反之亦然。例如, 100000b的XSUM[5:0]值會使耦接約並聯p通道元件中的一 半(或32個)之陣列群組409致能,而l〇〇〇〇lb的XSUM[5:0] 值會使耦接並聯P通道元件中的3 3個之陣列群組p 1及4 〇 9致 能,而1 000 1 0b的XSUM[ 5:0]值會使耦接並聯p通道元件中 的34個之陣列群組403及409致能等等。〇〇〇〇〇〇b的XSUM [5 ·· 0]值會關閉高阻抗狀態的所有P通道元件,而丨丨丨丨丨lb 的值會使最低阻抗準位的所有63個P通道元件致能。在一 實施例中,元件P63 : P1的陣列會按尺寸排列及分組,而產 生範圍從約2 0到1 5 0歐姆的上拉阻抗,用以使操作溫度及 匯流排電壓情況在預期的範圍,以及會遺留預先考慮製程 變化的邊限。 [0039]圖5係包括輸出驅動程式電路阻抗之控制器的 另一 1C 501之簡化方塊圖。1C 501與1C 101類似,並且若 想要的話,二者的邏輯及電路可位於相同1C上。1C 501也 包括多數可用於外接的I/O接腳,包括匯流排電壓輸入接 腳VTT、N通道控制接腳NCHCTRL、以及多個輸出接腳1236217 V. Description of the invention (13) The gate for receiving the signal PS4, and the sixth array group 309 includes 32 elements P3 2_P63 (P6 3: P32), each having a gate for receiving the signal PS5. [0038] The PS5-PSO signal collectively constitutes a binary value enabled by the buffer 4 (which is used to receive the XSUM [5: 0] value). The parent in the pS5-pS〇 signal is the corresponding bit in the buffered XSUM [5: 0] value. For example, SUM5 bits are buffered, NS5 bits are generated, SUM4 bits are buffered, NS4 bits are generated, and so on. Therefore, when the value of SUM [5: 0] is increased or increased, the impedance between VDD and PUP will decrease, and vice versa. For example, an XSUM [5: 0] value of 100000b enables an array group 409 coupled to about half (or 32) of the p-channel elements in parallel, while an XSUM [5: 0] of 1000b The value will enable coupling of 33 array groups p 1 and 4 009 in the parallel P-channel element, and the value of XSUM [5: 0] of 1000 1 0b will make the coupling in the parallel p-channel element 34 array groups 403 and 409 are enabled and so on. The XSUM [5 ·· 0] value of 〇〇〇〇〇〇〇b will turn off all P-channel components in the high impedance state, and the value of 丨 丨 丨 丨 丨 lb will cause all 63 P-channel components at the lowest impedance level to cause can. In one embodiment, the array of components P63: P1 will be arranged and grouped according to size, and a pull-up impedance ranging from about 20 to 150 ohms will be generated to make the operating temperature and the bus voltage condition in the expected range. , As well as margins that pre-consider process changes. [0039] FIG. 5 is a simplified block diagram of another 1C 501 including a controller of an output driver circuit impedance. 1C 501 is similar to 1C 101, and the logic and circuits of both can be located on the same 1C if desired. 1C 501 also includes most of the external I / O pins, including bus voltage input pin VTT, N-channel control pin NCHCTRL, and multiple output pins

第19頁 1236217 五、發明說明(14) 〇UT1、0UT2.....OUTN或OUTx,如先前所述。產生參考匯 流排電位的外部電壓訊號VTT係位於接腳VTT上,如例如是 1.5伏特(V)。在一實施例中,以虛線顯示的外部參考電阻 REXT係耦接於接腳VTT與]vCHCTRL之間。若未提供ΚΕχτ電 阻,則會使用内部電阻RINT來取代,當作預設參考電阻。 假設會提供電阻REXT,以及電阻RI NT的說明係完全地敘述 於相關揭露中,而在此將不會進一步說明。在一特定實施 例中,REXT電阻為14歐姆,並且可為精密電阻或類似的電 阻。在此特定實施例中,RINT電阻係實施為矽擴散。 [ 0040 ]IC 501也包括阻抗匹配邏輯5〇3,其係以與阻 抗匹配邏輯103類似的方式來運作,用以監測及比較^ ?議了與内部阻抗產生器的阻抗。在所顯: 中,阻抗匹配邏輯503係用以監測νττ及仳此丁礼接腳J 位,並且會將6位元内部匯流排5〇5上的6位元數 電Page 19 1236217 V. Description of the invention (14) 〇UT1, OUT2 ......... OUTN or OUTx, as described previously. The external voltage signal VTT that generates the reference bus potential is located on the pin VTT, such as 1.5 Volts (V), for example. In one embodiment, the external reference resistor REXT shown in dashed lines is coupled between the pins VTT and VCHCTRL. If a KE × τ resistor is not provided, the internal resistor RINT will be used instead as the default reference resistor. It is assumed that the resistor REXT is provided, and the description of the resistor RI NT is fully described in the relevant disclosure, and will not be further explained here. In a particular embodiment, the REXT resistor is 14 ohms, and may be a precision resistor or a similar resistor. In this particular embodiment, the RINT resistor is implemented as a silicon diffusion. [0040] The IC 501 also includes impedance matching logic 503, which operates in a similar manner to the impedance matching logic 103 to monitor and compare the impedance with the internal impedance generator. In the displayed :, the impedance matching logic 503 is used to monitor νττ and the J-bit of the Dingli pin, and it will transmit the 6-bit number on the 6-bit internal bus 505.

[5:0]傳送到多個偏壓控制邏輯元件5〇6,其 M 位於IC 5〇1上的對應輸出驅動程式電路5〇7 輸輕接至 動程式電路507係個別地從i到N予 —加/ 出驅動程式電路507係輕接至輸出接腳〇υΤχ中的J二輸 固。每個輸出驅動程式電路5〇7與每個 ; 似,以及也包括一陣列的匹配 、科兀仵1(Π類 N i甬道,而X曰P、s、者 * 抗70件,除了這些元件為 N通道’而不疋P通道,而運作為下拉元件, :為 件°特別而言’在每個輸出驅動程式電路5Q7内 ^ OSUM —X[5:0]中的每個位元會使呈 、值 動對應的OUTx接腳之對庫群έ的、八汲極點並且用來驅 應群組的一陣列匹配Ν通道元件被 1236217 五、發明說明(15) 致选/ 除能。〇SUlW Y「c; γμ 雷敗夕去饮ΛΑ Μ — Χ [ 5 : 0 ]值係藉由具有根據輸出驅動程式 央ρρ — 、輪出狀態而設定的狀態之輸出致能訊號0ΕΝ =庫二Λ號係藉由元件邏輯(未顯示)…,用以 二τ 1 ί 接腳0UTx <輸出訊號的狀態。,對應的 〇=號”輯低準位時,咖_x[5:o]值會指定每個輸 、首-°从紅工電路507中’即將開啟(或致能)的開路沒極n通 ^ =之數目。在一實施例中,6位元匯流排5 0 5會以6 4個 相專間隔的步驟,而可調整輸出驅動程式電路的阻抗。 一 0041 ]在運作時,阻抗匹配邏輯5〇3會保持匹配n通道 局Λ二進位陣列,其實f上與每個輸出驅動程式電 、一進位陣列相同。每個陣列係以與先前所述之 * :二211:陣列之相似方式’而配置或分成用於數位輸 ^ 、一進位群組。會持續監測阻抗匹配邏輯5 〇 3 局η部二進位陣列之阻抗,並且會週期性地調升或調降 .]值,以致於跨接内部陣列的電壓係與跨接選擇參 2電阻REXT(或RINT電阻)的電壓相差在一預定容忍度内。 ^ 一實施例中,預定容忍度為約5〇毫伏特(mV)的誤差電 壓。再者,在一實施例中,輸出驅動程式電路5〇7的最佳 阻抗係由INT BCLK訊號的每二個週期來決定,而sum[5. 值係經由匯流排505,而傳送到每個偏壓控制邏輯元件. 5〇6,而顯然會更新輸出驅動程式電路5〇7。 [ 0042 ]就如同IC 101所述,IC 5〇1包括類似的偏壓控 制邏輯元件506,用以使位於匯流排5〇5上的讥1[5:〇]值 上或減去偏壓。這可用於實施每個輸出驅動程式電路5〇7 1236217 發明說明(16) 的阻抗凋整,來補償遍及晶粒各處的製程變化。在一實施 例中’、偏壓控制器5 〇 6係用於位於丨c 5 〇 1上的每個輸出驅 動程式電路5 0 7。在另一實施例中,偏壓控制器5 〇 6係用於 位於1C 501上的輸出驅動程式電路5〇7中的局部群組。 [0 0 4 3 ]圖6係阻抗匹配邏輯5 〇 3之一範例實施例的更 詳細方塊圖。在所顯示的配置中,阻抗匹配邏輯50 3實質 上與配合圖1所討論之阻抗匹配邏輯丨〇3的配置及運作類 似。在相關的揭露中,阻抗匹配邏輯5〇3係以稍微較複雜 的方式來實施’而包括感測電路。用以感測REXT電阻是否 連接’並且若無的話,會使另一個致能,與使用電阻R丨NT 當作參考電阻的阻抗控制器類似。 [ 0044 ]阻抗匹配邏輯50 3包括阻抗控制器6〇]1,其用以 接收INT BCLK訊號,並且其包括電壓感測器6〇3,用以監 測VTT及NCHCTRL接腳的電壓。所顯示的NCHCTRL接腳會產 生訊號I NP ’其會傳送到參考接地的阻抗產生器6 〇 7。阻抗 產生器607係輸入控制值SUM [5: 〇]而顯示INp訊號與接地之 間的阻抗。電壓感測器603可有效地將VTT與NCHCTRL接腳 之間的電壓,與從NCHCTRL接腳到接地的電壓進行比較, 而產生送到阻抗控制邏輯605的訊號HI及LO,以試圖使電 位位於預定容忍度之内。阻抗控制邏輯6〇5會增加/降低 SUM[5:0]值,以控制阻抗產生器6〇7的阻抗,直到(VDD — INP)與INP的差距在預定誤差電壓内(或以致於INp訊號的 電壓為νττ電壓之一半)。換句話說,假設外接REXT電阻, 電壓感測器6 0 3及阻抗控制邏輯6 〇 5會配合,以試圖使阻抗[5: 0] transmitted to multiple bias control logic elements 506 whose M is on the corresponding output driver program circuit 507 on IC 501, and the input and output program circuits 507 are individually from i to N The pre-add / output driver circuit 507 is lightly connected to the J-second output of the output pin 〇υΤχ. Each output driver circuit 507 is similar to each; and also includes an array of matching, rugged 1 (II class N i channel, and X is P, s, or *) resistant to 70 pieces, except for these components For N-channel 'instead of P-channel, it operates as a pull-down element: for the °° In particular, in each output driver circuit 5Q7 ^ Each bit in OSUM —X [5: 0] causes An array of OUTx pins that correspond to and move to the bank group, eight drain points, and an array used to drive the group to match the N channel element is 1236217 V. Description of the invention (15) Enabled / disabled. 〇SUlW Y "c; γμ to go to drink ΛΑ Μ — χ [5: 0] value is an output enable signal with a state set according to the output driver program ρρ —, turn-out state 0 En = library two Λ number Based on the logic of the component (not shown) ..., the state of the output signal of the two τ 1 ί pins 0UTx. When the corresponding 0 = number is low, the value of _x [5: o] will be Specify the number of each open, first-° from the red circuit 507 'the open circuit (or enable) to be opened (or enabled) n = ^ =. In one embodiment, the 6-bit bus Row 5 0 5 will adjust the impedance of the output driver circuit in 64 step-by-step steps. 0041] During operation, the impedance matching logic 503 will keep matching the n-channel bureau Λ binary array, in fact f is the same as each output driver and a carry array. Each array is configured or divided into a digital input ^ and a carry group in a similar manner to the *: 2 211: array described previously. The impedance of the binary array of the impedance matching logic 503 will be continuously monitored, and will be adjusted up or down periodically.], So that the voltage system across the internal array and the jumper selection parameter 2 REXT ( (Or RINT resistor) voltages are within a predetermined tolerance. ^ In one embodiment, the predetermined tolerance is an error voltage of about 50 millivolts (mV). Furthermore, in an embodiment, the output driver circuit 5 The optimal impedance of 〇7 is determined by every two cycles of the INT BCLK signal, and the sum [5. Value is transmitted to each bias control logic element via bus 505, which will obviously be updated. Output driver circuit 507. [0042] Just like IC 10 As described in 1, IC 501 includes similar bias control logic element 506 to add or subtract the value of 讥 1 [5: 〇] on bus 505. This can be used to implement each Output driver circuit 507 1236217 Description of the invention (16) The impedance is trimmed to compensate for process variations throughout the die. In one embodiment, the 'bias controller 5' is used in the 5c Each output driver circuit 507 on 〇1. In another embodiment, the bias controller 506 is for a local group in the output driver circuit 507 on 1C 501. [0 0 4 3] FIG. 6 is a more detailed block diagram of an exemplary embodiment of impedance matching logic 503. In the configuration shown, the impedance matching logic 50 3 is substantially similar to the configuration and operation of the impedance matching logic 03 discussed in conjunction with FIG. 1. In the related disclosure, the impedance matching logic 503 is implemented in a slightly more complicated manner 'and includes a sensing circuit. It is used to sense whether the REXT resistor is connected ’and if not, it will enable another one, similar to an impedance controller using the resistor R 丨 NT as a reference resistor. [0044] The impedance matching logic 503 includes an impedance controller 601, which is used to receive the INT BCLK signal, and includes a voltage sensor 603, which is used to monitor the voltage of the VTT and NCHCTRL pins. The NCHCTRL pin shown will generate a signal I NP ′ which will be transmitted to an impedance generator referenced to ground 607. The impedance generator 607 inputs the control value SUM [5: 〇] and displays the impedance between the INp signal and ground. The voltage sensor 603 can effectively compare the voltage between the VTT and NCHCTRL pins with the voltage from the NCHCTRL pin to ground, and generate signals HI and LO sent to the impedance control logic 605 to try to make the potential at Within the tolerance. The impedance control logic 60 will increase / decrease the value of SUM [5: 0] to control the impedance of the impedance generator 607 until the difference between (VDD — INP) and INP is within a predetermined error voltage (or even the INp signal The voltage is half of the voltage νττ). In other words, assuming an external REXT resistor, the voltage sensor 603 and the impedance control logic 605 will cooperate to try to make the impedance

1236217 五、發明說明(17) 控制器607的電壓與預定誤差電麼内的電阻r之電壓相等。 [ 0045 ]當REXT電阻為外接時,ντΤ來源電壓係藉由1236217 V. Description of the invention (17) The voltage of the controller 607 is equal to the voltage of the resistor r in the predetermined error voltage. [0045] When the REXT resistor is external, the ντΤ source voltage is determined by

REXT電阻及阻抗產生器607的阻抗而分壓,而提供ijyp訊號 上的對應電壓。若INP訊號的電壓太高(係表示阻抗產生器 607的阻抗太高(或大於REXT)),則電壓感測器603會使HI 訊號致能,並且使LO訊號無效。阻抗控制邏輯6〇5會藉由 增加SUM[5:0]值來回應,以降低阻抗產生器6〇7的阻抗 值。當阻抗產生器6 0 7的阻抗太低時,電壓感測器6 〇 3會使 LO訊號致能,並且會使ΗI訊號無效。阻抗控制邏輯6〇5會 藉由降低SUM[5 ·· 0]值來回應,以增加阻抗值。在所顯示及 說明的實施例中’雖然也考慮比例關係,但是sUM [ 5 ·· 〇 ]值 係與阻抗產生器6 0 7的阻抗成反比。在一實施例中,電壓 感測器6 0 3包括一對感測放大器(未顯示),其實質上係以 與上述的電壓感測器2 0 3類似的方式來配置。再者,在一 實施例中’阻抗控制邏輯6〇 5為由I NT BCLK訊號所控制的 數位電路,並且在INT BCLK訊號的選擇週期期間(如每個 時脈週期或每隔一個時脈週期等)會調整(例如,增加 低)SUM[5:0]值。 幸 [0046]現在請參閱圖7,所顯示的是圖5中的偏壓控制 邏輯50 6的更詳細之方塊圖。偏壓控制邏輯5〇6具有非^發 性邏輯元件702,其耦接至輸出偏壓邏輯7〇1。輸出偏壓邏 輯701係經由訊號ADD[3:0]&SUBEN,而耦接至偏壓調整 輯703。訊號INT BCLK及SUM[5:0]會傳送到偏壓調整邏輯 703,其會產生對應的訊號0SUM一χ[5:〇],如圖5中所顯 1236217 五、發明說明(18) TJn ° [ 0047 ]在運作時,在時脈訊號INT BCLK的選擇週期期 間(如每隔一個時脈週期或類似的時脈週期),偏壓調整邏 輯703會基於ADD[3:0]的值及訊號SUBEN的狀態,來調整 (例如,增加或降低)〇SUM-X[5:〇]值。4位元值ADD[3:0] 係從輸出偏壓邏輯7 0 1傳送到偏壓調整邏輯7 〇 3,以識別 SUM[5: 0]值要加上或減去的數量。符號或極性訊號SUBEn 會由輸出偏壓邏輯701傳送到偏壓調整邏輯703,以決定是 否要加上(當SUBEN未致能時)或減去(當SuBEN致能時)此數 量。SUBEN訊號及ADD[ 3 : 0 ]值會共同地構成訊號偏壓調整 值。在一實施例中,SUM [ 5 : 0 ]值會直接加上(例如,當 SUBEN為邏輯〇或未致能),或直接減去(例如,當SUBEN* 邏輯1或致能)PADD[3:0]值。在此情況中,ADD[3:0]值係 表示固定量的偏壓達到SUM [5:0]值的1/4範圍。在另一實 施例中’ SUM[5 : 0 ]值係根據ADD[3 : 0 ]及SUBEN訊號而成比 例地增加或降低。例如,若ADD [3·· 0]係設定於1〇〇 0b(二進 位),並且SUBEN未致能,則SUM[ 5: 0]會以百分之50增加。 [0048]在另一特定實施例中,輸出偏壓邏輯7〇1包括 或係由包含於I C 5 0 1上的可程式非揮發性邏輯元件了 〇 2 (如 非揮發性記憶體或保險絲或類似元件)來程式化。I C 5 〇 1 中的輸出偏壓邏輯701及可程式非揮發性邏輯元件7〇2實質 上係以與1C 101中的輸出偏壓邏輯301及可程式非揮發性、 邏輯元件302相同的方式來運作,而不需再作進一步說 明。 1236217 五、發明說明(19) [〇〇49]附圖8係阻抗產生器800的一範例實施例之簡要 示意圖’其可用來實施阻抗產生器607,及/或實施輸出驅 動程式電路507中的任一個。阻抗產生器8〇〇包括二進位陣 列的63個N通道元件\1一1^63(或N63 : N1)。在一實施例中, 每個N通道元件N 6 3 : N1會匹配,以致於汲極到源極的阻抗 實質上會相同。每個元件N 6 3 : N1的源極係耦接至接地,而 其沒極係耦接至訊號INP。元件N6 3:N1係二進位地分組, 以對應二進位阻抗值ZSUM [5: 0]之六個位元中的每一個。 第一陣列群組為單一元件N1,其具有用以接收訊號⑽^的 閘極,第二陣列群組8〇1包括二個元件?^2及们(1^312),每 個具有用以接收訊號NS1的閘極,第三陣列群組8〇3包括四 個元件N4-N7(N7:N4),每個具有用以接收訊號NS2的閘 極’第四陣列群組8〇5包括八個元件N8 —N15(N1 5 : N8),每 個具有用以接收訊號NS3的閘極,第五陣列群組8〇7包括1 6 個元件N16-N31(N31:N16),每個具有用以接收訊號NS4的 閘極’以及第六陣列群組8〇9包括32個元件n32-N63 (N63:N32),每個具有用以接收訊$NS5的閘極。 [〇〇50]NS5-NS0訊號會構成由緩衝器811(其用以接收 ZSUM[5 : 0]值)所致能的二進位值NS[5: 〇]。Ns[5 : 〇]值中的 每個位元為緩衝形式iZSUM[5:〇]值中的對應位元。因 此’當Z S U Μ [ 5 : 0 ]值提升或增加時,I n p訊號的阻抗會降 低,且反之亦然。例如,1〇〇〇〇〇b的ZSUM[5 : 〇]值會使耦接 約並聯N通道元件中的一半(或32個)之陣列群組8〇9致能, 而1 000 0 1 b的ZSUM[ 5 :0]值會使耦接並聯n通道元件中的33The REXT resistor and the impedance of the impedance generator 607 divide the voltage, and provide the corresponding voltage on the ijyp signal. If the voltage of the INP signal is too high (meaning that the impedance of the impedance generator 607 is too high (or greater than REXT)), the voltage sensor 603 will enable the HI signal and invalidate the LO signal. The impedance control logic 605 will respond by increasing the value of SUM [5: 0] to reduce the impedance value of the impedance generator 607. When the impedance of the impedance generator 607 is too low, the voltage sensor 603 will enable the LO signal and invalidate the ΗI signal. The impedance control logic 605 will respond by reducing the SUM [5 ·· 0] value to increase the impedance value. In the embodiment shown and described, 'although a proportional relationship is considered, the value of sUM [5 ·· 〇] is inversely proportional to the impedance of the impedance generator 607. In one embodiment, the voltage sensor 603 includes a pair of sense amplifiers (not shown), which are substantially configured in a similar manner to the voltage sensor 203 described above. Furthermore, in one embodiment, the 'impedance control logic 605 is a digital circuit controlled by the I NT BCLK signal, and during the selection period of the INT BCLK signal (such as each clock cycle or every other clock cycle) Etc.) will adjust (eg, increase low) the SUM [5: 0] value. Fortunately, referring now to FIG. 7, a more detailed block diagram of the bias control logic 506 in FIG. 5 is shown. The bias control logic 506 has a non-volatile logic element 702 which is coupled to the output bias logic 701. The output bias logic 701 is coupled to the bias adjustment logic 703 via a signal ADD [3: 0] & SUBEN. The signal INT BCLK and SUM [5: 0] will be transmitted to the bias adjustment logic 703, which will generate the corresponding signal 0SUM-χ [5: 〇], as shown in Figure 1236217 V. Description of the invention (18) TJn ° [0047] During operation, during the selection period of the clock signal INT BCLK (such as every other clock cycle or similar clock cycle), the bias adjustment logic 703 will be based on the value and signal of ADD [3: 0] SUBEN status to adjust (eg, increase or decrease) the 〇SUM-X [5: 〇] value. The 4-bit value ADD [3: 0] is transferred from the output bias logic 7 0 1 to the bias adjustment logic 7 03 to identify the amount to be added or subtracted from the SUM [5: 0] value. The sign or polarity signal SUBEn is transmitted from the output bias logic 701 to the bias adjustment logic 703 to determine whether to add (when SUEN is disabled) or subtract (when SuBEN is enabled) this number. The SUBEN signal and the ADD [3: 0] value together constitute the signal bias adjustment value. In one embodiment, the SUM [5: 0] value is directly added (for example, when SUBIN is logic 0 or disabled), or directly subtracted (for example, when SUben * is logic 1 or enabled) PADD [3 : 0] value. In this case, the value of ADD [3: 0] means that a fixed amount of bias voltage is in the range of 1/4 of the value of SUM [5: 0]. In another embodiment, the value of 'SUM [5: 0] is proportionally increased or decreased according to the ADD [3: 0] and the suben signal. For example, if ADD [3 ·· 0] is set to 1000b (binary) and SUBIN is not enabled, SUM [5: 0] will increase by 50%. [0048] In another specific embodiment, the output bias logic 701 includes or is comprised of a programmable non-volatile logic element included in the IC 501 (such as a non-volatile memory or fuse or Similar components). The output bias logic 701 and programmable non-volatile logic element 702 in IC 5 〇1 are substantially the same as the output bias logic 301 and programmable non-volatile, logic element 302 in 1C 101 Operation without further explanation. 1236217 V. Description of the invention (19) [0049] Figure 8 is a schematic diagram of an exemplary embodiment of the impedance generator 800, which can be used to implement the impedance generator 607, and / or the output driver circuit 507. Either. The impedance generator 800 includes a binary array of 63 N-channel elements \ 1-1 ^ 63 (or N63: N1). In one embodiment, each N-channel element N 6 3: N1 is matched so that the impedance from the drain to the source is substantially the same. The source of each element N 6 3: N1 is coupled to ground, and its non-pole is coupled to signal INP. Elements N6 3: N1 are grouped in binary to correspond to each of the six bits of the binary impedance value ZSUM [5: 0]. The first array group is a single element N1, which has a gate for receiving a signal ⑽ ^, and the second array group 801 includes two elements? ^ 2 and (1 ^ 312), each with Receiving the gate of the signal NS1, the third array group 803 includes four elements N4-N7 (N7: N4), each having a gate for receiving the signal NS2. The fourth array group 805 includes eight N8 — N15 (N1 5: N8), each having a gate for receiving the signal NS3, the fifth array group 807 includes 16 elements N16-N31 (N31: N16), each with The gates for receiving the signal NS4 and the sixth array group 809 include 32 elements n32-N63 (N63: N32), each having a gate for receiving the signal $ NS5. [0050] The NS5-NS0 signal will constitute a binary value NS [5: 〇] enabled by the buffer 811, which is used to receive the ZSUM [5: 0] value. Each bit in the Ns [5: 〇] value is the corresponding bit in the buffered iZSUM [5: 〇] value. Therefore, when the value of Z S U M [5: 0] is increased or increased, the impedance of the I n p signal will decrease, and vice versa. For example, a ZSUM [5: 〇] value of 100000b enables the array group 809, which couples about half (or 32) of the parallel N-channel elements, and 10001 b A ZSUM [5: 0] value will cause 33

1236217 五、發明說明(20) 個之陣列群組N1及809致能,而1 00 0 1 0b的ZSUM[5:0]值會 使耦接並聯N通道元件中的3 4個之陣列群組8 0 3及8 0 9致能 等等。0 0 0 00 0b的ZSUM[ 5:0]值會關閉高阻抗狀態的所有N 通道元件,而111111 b的值會使最低阻抗準位的所有6 3個N 通道元件致能。在一實施例中,元件N63 :N1的陣列會按尺 寸排列及分組,而產生範圍從約4到2 4歐姆的下拉阻抗, 用以使操作溫度及匯流排電壓情況在預期的範圍,以及會 遺留預先考慮製程變化的邊限。 [0051]雖然未顯示,但是每個輸出驅動程式電路5〇7 可以與阻抗產生器5 0 7類似的方式來配置。對於每個輸出 驅動程式電路507而言,31^[5:0]值係以(^1^[5:0]值來取 代,而額外邏輯(如一陣列的及閘或類似元件)係用以將 Ο E N訊號與每個N S 5 - N S 0訊號結合,如相關揭露中所述。以 此方式’ I C 5 0 1係以與IC 1 0 1類似的方式來配置,其中每 個輸出驅動程式電路5 0 7包括二進位陣列的匹配阻抗元 件’其會與位於阻抗產生器607中之參考二進位陣列的匹 配阻抗元件匹配。 [ 0052 ] 1C 101係使用匹配P通道上拉元件來說明,而 1C 501係使用匹配N通道下拉元件來說明。在任一種情兄 中,阻抗控制器(例如,201,601)會修改數位值(例二' SUM[5:0]),以試圖使阻抗產生器(例如,2〇7,6〇7)的阻 抗與參考元件(例如,r,REXT)匹配,其中,數位值秋 會用來設定位於耦接至相關IC的輸出接腳之輸出元件中 相似阻抗產生器之阻抗。阻抗匹配邏輯(例如,1〇3,5〇3)1236217 V. Description of the invention (20) array groups N1 and 809 are enabled, and the ZSUM [5: 0] value of 1 0 0 0 1 0b will couple 34 to 4 array groups in parallel N-channel components 8 0 3 and 8 9 enable and so on. A ZSUM [5: 0] value of 0 0 0 00 0b will turn off all N-channel components in the high impedance state, while a value of 111111 b will enable all 63 N-channel components at the lowest impedance level. In an embodiment, the array of elements N63: N1 will be arranged and grouped according to size, and a pull-down impedance ranging from about 4 to 24 ohms will be generated to make the operating temperature and the bus voltage situation in the expected range, and will Leaving margins for advance consideration of process changes. [0051] Although not shown, each output driver circuit 507 may be configured in a similar manner to the impedance generator 507. For each output driver circuit 507, the value of 31 ^ [5: 0] is replaced by the value of (^ 1 ^ [5: 0], and additional logic (such as an array and gate or similar) is used Combine the 0 EN signal with each of the NS 5-NS 0 signals, as described in the related disclosure. In this way, the IC 5 0 1 is configured in a similar manner to the IC 1 0 1, where each output driver circuit 5 0 7 includes a matching impedance element of a binary array 'which will match the matching impedance element of a reference binary array located in the impedance generator 607. [0052] 1C 101 is described using a matched P-channel pull-up element, while 1C The 501 is explained by using a matched N-channel pull-down element. In any kind of brother, the impedance controller (for example, 201, 601) will modify the digital value (Example 2 'SUM [5: 0]) in an attempt to make the impedance generator ( For example, the impedance of 20.7, 60.7) matches the reference component (for example, r, REXT), where the digital value is used to set a similar impedance generated in the output component that is coupled to the output pin of the relevant IC Device impedance. Impedance matching logic (for example, 103, 503)

12362171236217

中的參考阻抗產生器之阻抗係用以使輪出元件(例如, ,507)中的每個阻抗產生器之阻抗匹配。若遍及晶粒 各處有製程變化,這會導致複寫二進位陣列與輪出驅動元 件之間的顯著差異,則由參考陣列所決定的阻抗元件數不 會是輸出元件的最佳數目。 [ 0053 ]輸出偏壓邏輯(例如,3〇1,7〇1)及偏壓調整 邏輯(例如,303,703)會提供一種使設計者能藉由分配在 已知上拉元件或輸出程式驅動電路(或者是一群上拉元件 或輸出程式驅動電路)的局部之偏壓控制邏輯元件丨〇 6, 506内的這些元件,來補償遍及晶片各處的這些製程變 化。例如,在製造之後,設計者可將丨c放入測試裝置(未 顯示),其使1C辆接至參考阻抗(例如,r*rext)。測試裝 置會量測參考阻抗與結果阻抗(上拉或下拉)的輸出之間的 任何差異,來識別誤差或偏壓偏移。因此,設計者可將輸 出偏壓邏輯(例如,3 0 1,7 0 1)程式化,而去除誤差,以補 償晶片上的製程變化。 [ 0 054 ]附圖9係根據本發明的另一範例實施例之基於 參考阻抗來調整IC之至少一個輸出的輸出阻抗的方法之流 程圖。在第一方塊9 0 1,係表示將屬於測試的I c放入用以 執行測試程序的測試裝置或類似裝置,並且若有必要可以 會外接參考阻抗(例如,參考電阻R,REXT)。要注意的 是,若參考電阻R I NT位於内部,則可用來當作參考阻抗。 在方塊90 3,係表示在1C的電源開啟且開始運作之後,其 會將參考電壓施加到參考阻抗及參考阻抗產生器。在本實The impedance of the reference impedance generator in is used to match the impedance of each impedance generator in the wheel-out element (eg, 507). If there are process variations throughout the die, this will cause a significant difference between the replication binary array and the wheel-out drive element. The number of impedance elements determined by the reference array will not be the optimal number of output elements. [0053] Output bias logic (for example, 301, 701) and bias adjustment logic (for example, 303, 703) will provide a way for designers to drive by assigning to known pull-up components or output programs These components within the circuit (or a group of pull-up components or output program drive circuits) bias control logic components 06, 506 to compensate for these process variations throughout the chip. For example, after manufacturing, the designer may place c into a test device (not shown), which connects the 1C vehicle to a reference impedance (eg, r * rext). The test setup measures any difference between the output of the reference impedance and the resulting impedance (pull-up or pull-down) to identify errors or bias offsets. Therefore, the designer can program the output bias logic (for example, 301, 701) to remove errors to compensate for process variations on the chip. [0 054] FIG. 9 is a flowchart of a method for adjusting an output impedance of at least one output of an IC based on a reference impedance according to another exemplary embodiment of the present invention. In the first box 9 0 1, it means that I c belonging to the test is placed in a test device or similar device for performing the test procedure, and a reference impedance (for example, a reference resistance R, REXT) may be externally connected if necessary. Note that if the reference resistor R I NT is internal, it can be used as a reference impedance. At block 90 3, it means that after the power of 1C is turned on and started to operate, it will apply the reference voltage to the reference impedance and the reference impedance generator. In this real

第27頁 1236217 五、發明說明(22) 施例中,參考電壓可為電壓源(如VDD訊號或類似訊號), 其會跨接串接的參考電阻及參考阻抗來施加。 [0055] 在方塊905,係表示1C會調整參考產生器的參 考阻抗輸入,以使參考阻抗產生器的阻抗與的參考阻抗值 的差距在預定容忍度内。在本實施例中,電壓會在參考阻 抗與參考阻抗產生器之間的中間接面進行測量,並且會與 參考電壓(例如,VDD或VTT)的百分比進行比較。再者,會 週期性或持續地調整參考阻抗輸入,以保持在預定容忍度 之内。在方塊9 0 7,係表示參考阻抗產生器之二進位陣列 的匹配阻抗元件中的選擇數個會基於參考阻抗輸入,藉由 Ϊ C而内部地致能。在本實施例中,參考阻抗輸入為數位 值,其中每個位元會使選擇群組之一陣列的匹配阻抗元件 (可為N通道或P通道元件)致能。 [0056] 在方塊909,係表示連接至每個輸出阻抗產生 器(其耦接至對應輸出)之輸出阻抗輸入係基於參考阻抗輸 入,而藉由I C來控制。在方塊9 11,係表示每個輸出阻抗 產生器之二進位陣列的匹配阻抗元件中的選擇數個會基於 輪出阻抗輸入而致能。以此方式,I C會根據參考阻抗,而 試圖調整其輸出的阻抗。如先前所述,每個上拉邏輯元件 1〇7及/或每個輸出驅動程式電路5 07包括與參考阻抗產生 器的配置相同之匹配阻抗元件,以致於每個輸出的輸出終 端阻抗會基於參考阻抗。 [0 0 5 7 ]在方塊9 1 3,係表示會量測參考阻抗與至少一 個輸出阻抗之間的任何差異。這可藉由測試裝置而自動達Page 27 1236217 V. Description of the invention (22) In the embodiment, the reference voltage may be a voltage source (such as a VDD signal or a similar signal), which is applied across a series-connected reference resistance and reference impedance. [0055] At block 905, it means that 1C adjusts the reference impedance input of the reference generator so that the difference between the impedance of the reference impedance generator and the reference impedance value is within a predetermined tolerance. In this embodiment, the voltage is measured at the mid-plane between the reference impedance and the reference impedance generator, and compared with the percentage of the reference voltage (for example, VDD or VTT). Furthermore, the reference impedance input is adjusted periodically or continuously to stay within a predetermined tolerance. At block 9 07, the selected number of matching impedance elements representing the binary array of the reference impedance generator will be internally enabled by Ϊ C based on the reference impedance input. In this embodiment, the reference impedance input is a digital value, where each bit enables a matching impedance element (which may be an N-channel or P-channel element) in an array of a selection group. [0056] At block 909, it means that the output impedance input connected to each output impedance generator (which is coupled to the corresponding output) is based on the reference impedance input and is controlled by IC. At block 9-11, it means that the selected number of matched impedance elements in the binary array of each output impedance generator will be enabled based on the round-out impedance input. In this way, IC will attempt to adjust the impedance of its output based on the reference impedance. As mentioned earlier, each pull-up logic element 107 and / or each output driver circuit 507 includes a matching impedance element with the same configuration as the reference impedance generator, so that the output termination impedance of each output will be based on Reference impedance. [0 0 5 7] At box 9 1 3, it means that any difference between the reference impedance and at least one output impedance is measured. This can be achieved automatically by the test device

第28頁 1236217 五、發明說明(23) 成’或藉由測试操作者以手動的方式達成。另一種方式’ 若電阻R I NT在測試期間,係用來當作參考電阻,則量測的 輸出阻抗會與輸出上拉及/或下拉阻抗的已知值或希望值 進行比較。在下個方塊9 1 5,在I c上,與量測的輸出阻抗 對應的非揮發性元件會以偏壓調整值來程式化,以補償任 何里測的阻抗差異。在所述的特定實施例中,會設定非揮 發性記憶元件的位元,或燒斷包含於〗c的被選定保險絲, 以提供一種補償遍及I c各處的製程變化之控制機制。在方 塊917,係表示1C會將偏壓調整值與參考阻抗輸入結合, 以《周整輸出阻抗輸入。在所顯示的實施例中,偏壓調整邏 輯(303 ’703)會將ADD[ 3 : 0 ]值或PADD[ 3 : 0 ]值併入(加上、 減去、或結合)SUM[5:0]值,而分別產生OSUM-X[5:〇]值 或PSUM —X[ 5 ·· 0 ]值,其會傳送到每個輸出元件(例如,上拉 邏輯元件107或輸出驅動程式電路5 〇7)。測試程序會再次 重複’以確保適當補償,或對於對應於丨c 1 〇丨,5 〇丨的不 同區域之上拉邏輯元件1〇7或輸出驅動程式電路5〇7之 PSUM —X[5:0]及OSUM 一 X[5:0],會產生不同補償值。 [0 0 5 8 ]在運作期間,阻抗控制器會以前述之明確易解 的方式,而持續調整1C之每個選擇輸出元件(或元件群組) 的輸出阻抗(輸出驅動程式電路的匯流排下拉阻抗,或上 拉終端阻抗)。根據本發明的一實施例之調整輸出驅動程 式電路的阻抗之裝置及方法使系統設計者能調整阻抗,以 補償遍及1C各處的製程變化。可程式非揮發性邏輯元 (如非揮發性§己憶體或保險絲或類似元件)係位於晶片上,Page 28 1236217 V. Description of the invention (23) It can be achieved manually or by the test operator. Another way, ’if the resistance R I NT is used as a reference resistance during the test, the measured output impedance is compared with the known or desired value of the output pull-up and / or pull-down resistance. On the next box 9 1 5, on I c, the non-volatile components corresponding to the measured output impedance will be programmed with bias adjustment values to compensate for any measured impedance differences. In the specific embodiment described, the bits of the non-volatile memory element are set, or the selected fuse contained in [c] is blown to provide a control mechanism that compensates for process variations throughout Ic. In block 917, it means that 1C will combine the bias adjustment value with the reference impedance input, and output the impedance as the whole round. In the embodiment shown, the bias adjustment logic (303 '703) incorporates (adds, subtracts, or combines) the value of ADD [3: 0] or PADD [3: 0] SUM [5: 0] value, and respectively generate OSUM-X [5: 〇] value or PSUM —X [5 ·· 0] value, which will be transmitted to each output element (for example, pull-up logic element 107 or output driver circuit 5 〇7). The test procedure will be repeated again to ensure proper compensation, or to pull up the logic element 107 or the output driver circuit 507 PSUM-X [5: for different areas corresponding to 丨 c 1 〇 丨, 5 〇 丨0] and OSUM-X [5: 0] will produce different compensation values. [0 0 5 8] During operation, the impedance controller will continuously adjust the output impedance of each selected output element (or component group) of 1C (the bus of the output driver circuit) in the clear and easy-to-understand manner described above. Pull-down impedance, or pull-up termination impedance). An apparatus and method for adjusting the impedance of an output driver circuit according to an embodiment of the present invention enables a system designer to adjust the impedance to compensate for process variations throughout 1C. Programmable non-volatile logic elements (such as non-volatile memory or fuses or similar) are located on the chip.

第29頁 1236217Page 121236217

而能將每個偏壓調整值程式化。每個偏壓調整值係藉由其 對應的偏壓調整邏輯元件來使用,而產生希望的補償。八 [0059 ]雖然本發明已配合某些較佳實施例形式,而相 當詳細地說明,但是其他的形式及變化是可行的且可以考 慮的。例如,可考慮使可程式化阻抗產生器與參考電阻相 4之各種變化的方法’如目前技術或類似的技術。再者, 任何型式的非揮發性可程式裝置可考慮用來將補償程式 化。此外,雖然本發明考慮到使用金氧半導體(M〇s)型式 元件(包括互補MOS元件及類似元件,如例如是NM〇s &pM〇s 電晶體)之一種實施方式,但是也可以用類似的方式應用 於不同或類比型式的技術或相似元件,如雙載子元件或類 似元件。 [0 0 6 0 ]最後’凡是熟習此項技術者應該了解到的 是,在不脫離後附的申請專利範圍所定義之本發明的精神 及範圍之下,為了進行與本發明相同的目的,其可立即使 用揭露的概念及特定的實施例,來當作設計或修改其他的 結構之基礎°It is possible to program each bias adjustment value. Each bias adjustment value is used by its corresponding bias adjustment logic element to produce the desired compensation. [0059] Although the present invention has been described in detail in conjunction with certain preferred embodiments, other forms and variations are possible and can be considered. For example, various methods of making the programmable impedance generator and the reference resistance phase 4 can be considered, such as current technology or similar technology. Furthermore, any type of non-volatile programmable device may be considered for programming the compensation. In addition, although the present invention contemplates an embodiment using a metal-oxide-semiconductor (MOS) type element (including a complementary MOS element and the like, such as, for example, a NMs & pMOS transistor), it may also be used. A similar approach applies to different or analogous types of technology or similar elements, such as bi-amplitude elements or similar elements. [0 0 6 0] Finally, those skilled in the art should understand that, in order not to depart from the spirit and scope of the present invention as defined by the scope of the appended patent application, in order to carry out the same purpose as the present invention, It can immediately use the disclosed concepts and specific embodiments as the basis for designing or modifying other structures.

第30頁 1236217 圖式簡單說明 【圖式簡單說明】 [0014]配合以下的說明,以及附圖來揭露本發明的優 點、特徵、以及功效將會使 貴審查委員更易於了解, 其中·[ 0 0 1 5 ]圖1係包括用以精確地控制傳輸線的終端阻 抗之一範例系統的積體電路(IC)之簡化方塊圖; [0 0 1 6 ]圖2係圖1中的阻抗匹配邏輯之一範例實施例的 更詳細方塊圖; [0 0 1 7 ]圖3係圖1中的偏壓控制邏輯的更詳細方塊圖;Page 1236217 Brief description of the drawings [Simplified description of the drawings] [0014] With the following description and accompanying drawings to disclose the advantages, features, and effects of the present invention will make it easier for your reviewers to understand, of which [[0 0 1 5] FIG. 1 is a simplified block diagram of an integrated circuit (IC) including an example system for accurately controlling the termination impedance of a transmission line; [0 0 1 6] FIG. 2 is a diagram of impedance matching logic in FIG. 1 A more detailed block diagram of an exemplary embodiment; [0 0 1 7] FIG. 3 is a more detailed block diagram of the bias control logic in FIG. 1;

[0 0 1 8 ]圖4係阻抗產生器的一範例實施例之更詳細的 概圖,其可用來實施圖2中的阻抗產生器,及/或實施圖1 之上拉邏輯元件中的任一個; [0 0 1 9 ]圖5係包括輸出驅動程式電路阻抗之控制器的 I C之簡化方塊圖; [0 0 2 0 ]圖6係圖5中的阻抗匹配邏輯之一範例實施例的 更詳細方塊圖; [0 0 2 1 ]圖7係圖5中的偏壓控制邏輯的更詳細方塊圖;[0 0 1 8] FIG. 4 is a more detailed overview of an exemplary embodiment of the impedance generator, which can be used to implement the impedance generator in FIG. 2 and / or implement any of the pull-up logic elements in FIG. 1 A; [0 0 1 9] FIG. 5 is a simplified block diagram of an IC including a controller of an output driver circuit impedance; [0 0 2 0] FIG. 6 is a modification of an exemplary embodiment of the impedance matching logic in FIG. 5 Detailed block diagram; [0 0 2 1] FIG. 7 is a more detailed block diagram of the bias control logic in FIG. 5;

[0 0 2 2 ]圖8係圖6中的阻抗產生器之一範例實施例的概 圖,其也可稍做修改,而用來當作位於圖5中的輸出驅動 程式電路内之阻抗產生器;以及 [0 0 2 3 ]圖9係根據本發明的一範例實施例之基於參考 阻抗,來調整I C之至少一個輸出的輸出阻抗的方法之流程 圖。 圖式標示說明:[0 0 2 2] FIG. 8 is a schematic diagram of an exemplary embodiment of the impedance generator in FIG. 6, which can also be slightly modified and used as the impedance generation in the output driver circuit in FIG. 5. [0 0 2 3] FIG. 9 is a flowchart of a method for adjusting an output impedance of at least one output of an IC based on a reference impedance according to an exemplary embodiment of the present invention. Schematic description:

第31頁 1236217 圖式簡單說明 101 ,501 積 體 電 路(1C) 103 ,503 阻 抗 匹 配 邏 輯 105 ,505 内 部 匯 流 排 106 ,506 偏 壓 控 制 邏 輯元 件 107 :上拉 邏輯元件 201 ,601 阻 抗 控 制 器 203 ,603 電 壓 感 測 器 205 ,605 阻 抗 控 制 邏 輯 207 ,607 阻 抗 產 生 器 301 ,701 輸 出 偏 壓 邏 輯 302 ,702 非 揮 發 性 邏 輯元 件 303 ,703 偏 壓 調 整 邏 輯 400 ,800 阻 抗 產 生 器 401 ,801 第 二 陣 列 群 組 403 ,803 第 三 陣 列 群 組 405 ,805 第 四 陣 列 群 組 407 ,807 第 五 陣 列 群 組 409 ,809 第 六 陣 列 群 組 411 ,811 緩 衝 器 507 •輸出 驅動程式電路1236217 on page 31 Brief description of the diagram 101, 501 Integrated circuit (1C) 103, 503 Impedance matching logic 105, 505 Internal bus 106, 506 Bias control logic element 107: Pull-up logic element 201, 601 Impedance controller 203 , 603 voltage sensor 205, 605 impedance control logic 207, 607 impedance generator 301, 701 output bias logic 302, 702 non-volatile logic element 303, 703 bias adjustment logic 400, 800 impedance generator 401, 801 No. Two array groups 403, 803 Third array group 405, 805 Fourth array group 407, 807 Fifth array group 409, 809 Sixth array group 411, 811 Buffer 507 • Output driver circuit

第32頁Page 32

Claims (1)

12362171236217 種輪出阻抗偏壓補償系統,係 至少一個輸出 的輸出阻抗,包括有: 一參考阻抗產生器,用以產生由一參考阻抗控制輸入所 控制之一參考阻抗; 阻抗匹配控制器,用以持續調整該參考阻抗控制輸 入’以使該參考阻抗與一參考值的相差在一預定容忍 度内; 至少一個輸出阻抗產生器,每個係耦接至一對應輸出, 並且由一輸出阻抗控制輸入來控制;以及 一可程式偏壓控制器,用以將一偏壓量與該參考阻抗控 制輸入結合,而產生該輸出阻抗控制輸入。 2 ·如申請專利範圍第1項所述之輸出阻抗偏壓補償系統, 其中該偏壓控制器,包括有: 輸出偏壓邏輯,其可程式化,而產生該偏壓量;以及 偏壓調整邏輯,耦接至該輸出偏壓邏輯及該阻抗匹配控 制器,用以將該偏壓量與該參考阻抗控制輸入結合, 而產生該輸出阻抗控制輸入。 3·如申請專利範圍第2項所述之輸出阻抗偏壓補償系統, 其中該輸出偏壓邏輯包括複數條保險絲。 4 ·如申請專利範圍第2項所述之輸出阻抗偏壓補償系統, 其中該輸出偏壓邏輯包括一可程式非揮發性邏輯元件。 5·如申請專利範圍第2項所述之輸出阻抗偏壓補償系統, 其中該偏壓量包括一加上符號偏壓值,其會加入該參考 阻抗控制輸入。An output impedance bias compensation system is an output impedance of at least one output, and includes: a reference impedance generator for generating a reference impedance controlled by a reference impedance control input; an impedance matching controller for continuously Adjust the reference impedance control input to make the difference between the reference impedance and a reference value within a predetermined tolerance; at least one output impedance generator, each of which is coupled to a corresponding output, and is controlled by an output impedance control input Control; and a programmable bias controller for combining a bias amount with the reference impedance control input to generate the output impedance control input. 2 · The output impedance bias compensation system according to item 1 of the scope of patent application, wherein the bias controller includes: output bias logic, which can be programmed to generate the bias amount; and bias adjustment Logic, coupled to the output bias logic and the impedance matching controller, for combining the bias amount with the reference impedance control input to generate the output impedance control input. 3. The output impedance bias compensation system according to item 2 of the scope of patent application, wherein the output bias logic includes a plurality of fuses. 4. The output impedance bias compensation system according to item 2 of the scope of patent application, wherein the output bias logic includes a programmable non-volatile logic element. 5. The output impedance bias compensation system as described in item 2 of the patent application scope, wherein the bias amount includes a sign bias value, which is added to the reference impedance control input. I2362i7I2362i7 申靖專利範圍 以使該參考電阻與該可程式參考阻抗產生器的 差距在一預定容忍度内;以及 别,調整邏輯,其會將該參考阻抗控制輸入與一偏壓 11 ·如周整值結合,而產生該輸出阻抗控制輸入。 調申清專利範圍第1 〇項所述之積體電路,其中該輪出 1 2 /楚邏輯包括複數條可程式保險絲。 t申睛專利範圍第1 〇項所述之積體電路,其中該輪出 1 3·=楚邏輯包括一可程式非揮發性邏輯元件。Apply the patent scope so that the gap between the reference resistor and the programmable reference impedance generator is within a predetermined tolerance; otherwise, adjust the logic, which will adjust the reference impedance control input with a bias voltage 11 Combined to generate the output impedance control input. The integrated circuit described in item 10 of the patent scope is adjusted, wherein the round-out 1 2 / Chu logic includes a plurality of programmable fuses. The integrated circuit described in item 10 of the patent application scope, wherein the round out 1 3 · = Chu logic includes a programmable non-volatile logic element. 申Μ專利範圍第1 〇項所述之積體電路,其中該可程 紅參考阻抗產生器及每個該可程式輸出阻抗產生器包 括—二進位陣列的匹配元件。 14 ·如申請專利範圍第丨〇項所述之積體電路,其中該參考 阻抗控制輸入會加上或減去該偏壓調整值。 1 5 ·如申凊專利範圍第1 〇項所述之積體電路,其中該偏壓 調整值包括一百分比值,其代表該參考阻抗控制輸入 的一百分比,該參考阻抗控制輸入會加上或減去該百 分比值。The integrated circuit described in claim 10 of the patent scope, wherein the programmable red reference impedance generator and each of the programmable output impedance generators include a matching element of a binary array. 14 · The integrated circuit as described in the scope of the patent application, wherein the reference impedance control input adds or subtracts the bias adjustment value. 1 5 · The integrated circuit as described in item 10 of the patent scope of the patent, wherein the bias adjustment value includes a percentage value, which represents a percentage of the reference impedance control input, and the reference impedance control input is added with or Subtract this percentage value. 16· —種基於一參考阻抗而調整一積體電路之至少一個輸 出的輸出阻抗之方法,包括: 將一參考電壓施加到該參考阻抗及一參考阻抗產生 器’該參考阻抗產生器具有一參考阻抗輸入·, 調整該參考阻抗輸入,以使該參考阻抗產生器的阻抗 與該參考阻抗的差距在一預定容忍度内; 量測該參考阻抗與至少一個輸出阻抗之間的任何差16. · A method for adjusting the output impedance of at least one output of an integrated circuit based on a reference impedance, comprising: applying a reference voltage to the reference impedance and a reference impedance generator ', the reference impedance generator has a reference impedance Input ·, adjust the reference impedance input so that the difference between the impedance of the reference impedance generator and the reference impedance is within a predetermined tolerance; measure any difference between the reference impedance and at least one output impedance 第35頁 1236217Page 1236217 六、申請專利範圍 異; 以一偏壓調整值將該積體電路上的一非揮發性元件程 式化,以補償任何量測差異;以及 將該偏壓調整值與該參考阻抗輸入結合,而產生至少 一個輸出阻抗產生器的一輸出阻抗輸入,每個輸出 阻抗產生器係耦接至一對應輸出。 1 7·如申請專利範圍第1 6項所述之方法,其中該程式化一 偏壓調整值包括燒斷至少一條保險絲。 1 8·如申請專利範圍第1 6項所述之方法,其中該程式化一 偏壓調整值包括將一非揮發性記憶體的至少一個位元 程式化。 1 9·如申請專利範圍第1 6項所述之方法,其中該偏壓調整 值包括一加上符號偏壓值,並且其中該將該偏壓調整 值與該參考阻抗輸入結合包括加上該加上符號偏壓 值。 2 0 ·如申請專利範圍第1 6項所述之方法,其中該偏壓調整 值包括一百分比值,其代表該參考阻抗輸入的一百分 比,以及一極性訊號,並且其中該將該偏壓調整值與 該參考阻抗輸入結合包括基於該極性符號,使該參考 阻抗輸入加上或減去該百分比。6. The scope of the patent application is different; a non-volatile component on the integrated circuit is programmed with a bias adjustment value to compensate for any measurement difference; and the bias adjustment value is combined with the reference impedance input, and An output impedance input is generated for at least one output impedance generator, and each output impedance generator is coupled to a corresponding output. 17. The method as described in item 16 of the scope of patent application, wherein the programming of a bias adjustment value includes blowing out at least one fuse. 18. The method as described in item 16 of the scope of patent application, wherein the programming a bias adjustment value includes programming at least one bit of a non-volatile memory. 19. The method as described in item 16 of the scope of patent application, wherein the bias adjustment value includes a plus sign bias value, and wherein combining the bias adjustment value with the reference impedance input includes adding the Add the sign bias value. 2 0. The method as described in item 16 of the scope of patent application, wherein the bias adjustment value includes a percentage value, which represents a percentage of the reference impedance input, and a polarity signal, and wherein the bias adjustment Combining the value with the reference impedance input includes adding or subtracting the percentage to the reference impedance input based on the polarity sign. 第36頁Page 36
TW93107376A 2003-12-05 2004-03-19 Apparatus and method for adjusting the impedance of an output driver TWI236217B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US10/730,425 US6949949B2 (en) 2002-12-17 2003-12-05 Apparatus and method for adjusting the impedance of an output driver

Publications (2)

Publication Number Publication Date
TW200520381A TW200520381A (en) 2005-06-16
TWI236217B true TWI236217B (en) 2005-07-11

Family

ID=34377727

Family Applications (1)

Application Number Title Priority Date Filing Date
TW93107376A TWI236217B (en) 2003-12-05 2004-03-19 Apparatus and method for adjusting the impedance of an output driver

Country Status (2)

Country Link
CN (1) CN100468969C (en)
TW (1) TWI236217B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102045054B (en) * 2009-10-13 2013-04-24 晨星软件研发(深圳)有限公司 Method and related device for correcting output/input circuit
WO2016077970A1 (en) * 2014-11-17 2016-05-26 Micron Technology, Inc. Apparatuses and methods for compensating for process, voltage, and temperature variation in a memory
CN111416603B (en) * 2019-01-04 2023-03-24 瑞昱半导体股份有限公司 Transmission gate circuit
TWI719891B (en) * 2020-04-22 2021-02-21 瑞昱半導體股份有限公司 Bias compensation device and method for operating a bias compensation device

Also Published As

Publication number Publication date
CN1553569A (en) 2004-12-08
CN100468969C (en) 2009-03-11
TW200520381A (en) 2005-06-16

Similar Documents

Publication Publication Date Title
US6949949B2 (en) Apparatus and method for adjusting the impedance of an output driver
JP4127743B2 (en) Driver circuit output impedance calibration circuit
US7254079B2 (en) Electrical fuse circuit
US7917875B1 (en) Clock tree adjustable buffer
KR100937951B1 (en) Calibration circuit, on die termination device, and semiconductor memory device
US6480798B2 (en) Method and apparatus for independent output driver calibration
JP4923048B2 (en) Pin electronics driver
US7847584B2 (en) On-die termination circuit and driving method thereof
JPH11340810A (en) Semiconductor device
JPH03185921A (en) Semiconductor integrated circuit
JPH09223955A (en) Parameter tuning method for integrated circuit after manufacture and tunable integrated circuit
JP2010166299A (en) Calibration circuit and method
US6483354B1 (en) PCI-X driver control
JP2006319965A (en) Output drive circuit and output drive method
JP2008537366A (en) Edge rate control circuit for I2C bus
TWI236217B (en) Apparatus and method for adjusting the impedance of an output driver
TWI280742B (en) ASIC architecture for active-compensation of a programmable impedance I/O
US6653882B2 (en) Output drivers for IC
US6998875B2 (en) Output driver impedance controller
US7268578B2 (en) Transmission circuit, data-transfer control device and electronic equipment
US6922076B2 (en) Scalable termination
US6985008B2 (en) Apparatus and method for precisely controlling termination impedance
TWI242703B (en) Output driver impedance controller
TWI324853B (en) Apparatus and method for precisely controlling termination impedance
Durgaryan et al. Pull-up/pull-down line impedance matching methodology for high speed transmitters