多个8594E型号未通过噪声边带测试失败

原始性能测试写得不正确。
在进行此测试时，OEM频率为1 GHz，设置显示在书中并且没有信号应用，您将看到噪声基底从-75到-77 db从满量程下降。
这是设备的对数转换器本底噪声，对于整个系列而言是正常的。
测试应该以这种方式进行，就像设置和设置功率一样，使标记等于0 dBm。
然后将您的量程设置为1 KHz / 1 KHz RBW滤波器。
根据本书输入上下偏移量（10K，20K，30K和130K）。
将分析仪上的参考电平设置为-30 dBm。
进行测量时，您希望实际噪声基底比您尝试测量的最小误差至少低20 dB。
当本底噪声接近您所需的测量值时，误差会显着增加。
当您以这种方式测试设备时，您将看到设备实际通过所有偏移至少10 dB或更多。
这是正确测试所需的方法，但由于产品已过时，是德科技不会提供书面更正程序。



     以下为原文

  The original performance test was written incorrectly.  When doing this test, OEM frequency is 1 GHz, with the set up they show in the book and no signal applied you will see your noise floor bounce around from -75 to -77 db down from full scale.   This is the log converter noise floor for the unit and is normal for the entire series.   The test should be performed this way, as you would set up and set the power so the marker to equals 0 dBm.  Then set your span to 1 KHz/ 1 KHz RBW filter.   Enter in the upper and lower offsets (10K,20K,30K and 130K) as per the book.  Set your reference level on the analyzer to -30 dBm.  When doing measurements you want the actual noise floor to be at least 20 dB lower than what you are trying to measure for minimum error.  As the noise floor is closer to your desired measurement the error increases significantly.   When you test the unit this way you will see the unit actually passes all offsets by at least 10 dB or more.   This is the way it should have been done for proper correct testing, but since the product is obsolete, Keysight will not provide a written corrected procedure.