478平台无基准信号！望高人指点！

会员6881

今日修一块精英 L4VXA2 故障为：上U整块板无复位，上假负载有复位！经检查AGP 内存 CPU供电均正常，上假负载有时钟 PG 复位。而后上带灯测试座发现GTLREF灯不亮。无基准信号！对地阻值无穷大（正常的有60-100阻值），我估计哪里有开路现象！但没有修过基准信号出错的板子，不知其信号从哪里过来？望高人指点！

会员6881

查不出其线路，是不是直接从NB过来？

会员3847

看一下这个摘要，文献我搞不到，人家要米。
Method for adjusting the input switch-point level of the MCH using a pull-up resistor on the CPU package
Disclosed is a method for adjusting the input switch-point (GTLRef) level of the memory controller hub (MCH) using a pull-up resistor on the central processing unit (CPU) package. Benefits include improved functionality and improved performance.

--------------------------------------------------------------------------------

Method for adjusting the input switch-point level of the MCH using a pull-up resistor on the CPU package

Disclosed is a method for adjusting the input switch-point (GTLRef) level  of the memory controller hub (MCH) using a pull-up resistor on the central processing unit (CPU) package. Benefits include improved functionality and improved performance.

Background

      When new process-generation CPUs are plugged into previous-generation motherboards, the optimal GTLRef level of the MCH with the new CPU may not match the level hard-wired into the motherboard. This problem is conventionally solved by asking motherboard suppliers to change the option for the resistors on the motherboard that sets the GTLRef level for the MCH. As a result, motherboard suppliers must requalify their motherboards.

      Conventional motherboards set the GTLRef voltage for the CPU and MCH by placing resistor dividers on the motherboard to generate the appropriate voltage. Because future CPUs are required to plug into existing platforms, and the GTLRef level set in the past for previous CPUs may not be optimal for future platforms, a method is required that adjusts the GTLRef level of existing devices mounted on existing platforms already in the market to tune them to accept the latest CPUs (see Figure 1).

Description

      The disclosed method adjusts the motherboard chipset’s input/output (I/O) GTLRef through the use of a resistor placed on the CPU’s package. A pull-up resistor, ...

会员10650

鸟语看不懂！～！～！～！～！ :funk:

会员3847

值得注意的是，Gigabyte首次在BIOS中提供CPU GTLREF Ratio，據官方指出，CPU GTLREF Ratio主要用作設定處理器與北橋之間，資源傳送高低及低緣電壓值比例，以往主機板對GTLREG Ratio值是固定的，但Intel在3系列晶片組中建議把GTLREF作浮動調整，以增加主機板對未來處理器的兼容性，一般來說BIOS將會為GTLREF自行定義。

另有高階玩家認為GTLREF Ratio設定可提升超頻穩定性，儘管Gigabyte工程師不認同此項說法，但為滿足高階玩家需要，仍於新一代3系列主機板加入此項選項。

会员6881

谢谢张老的资料。此GTLREF我个人理解为就是一个参考电压，类似于反馈电路！好多提到的提高它的电压来改变CPU的VCC达到超频的目的。然而此点电压是由哪个产生呢（CPU的VCC还是NB还是其他）？相信好的菜鸟（一些老鸟）都不知道！

会员3847

Method for adjusting the input switch-point level of the MCH using a pull-up resistor on the CPU package
这句话就已经说清楚了：
主板为调整MCH（北桥）的输入开关点电平，在CPU座上使用了一个上拉电阻。

会员3847

看一下也许能了解一点
http://www.chinafix.com.cn/viewt ... e%3D1&frombbs=1

会员6881

谢谢张老了，经检测发现GTLREF是由CPU的主供电经一几十欧的电阻分压而来。位于CPU插座中间

会员753

GTLREF这个点好多好板是也不亮的,不要抓这个问题了,
你现在的问题是负载带不起来!
换电源IC或相关管子!

会员11992

呵呵

会员2906

上假负载有时钟 PG 复位。那上u供电是否正常，有时钟吗？座子没问题吧。

会员9777

我也遇到过类试的情况,就把座子加焊了一下还是一样,后来测别的好板有的是这样子,就只好改查方向了

会员12005

个人认为是NB的问题。