如何解决34901A频率测量限制？

问候，我正在使用34901A / 34970A和一个自动程序来对LED颜色传感器的输出进行频率测量;
我发现它只有95％的时间有效。当它失败时它只返回'0'，虽然有时候我收到了9.9E37，我把它解释为超出范围的指示。
当测量成功时，它在12KHz范围内，因此它肯定不会太快。
设置为10V范围，5 1 / 2D，FAST。
信号是5V，每106uS下降到0 V约12 uS。
我想知道如果因为LO仅占周期长度的12％左右，那么我可能是测量算法能力的“边缘”吗？
有一些过冲/下冲，但我​​在探头上放了一个3.3 nF的帽;
它清除了过冲/下冲但没有帮助测量。
请从信号的oscope捕获中查看附图。建议？谢谢！



     以下为原文

  Greetings,

I am using a 34901A/34970A with an automated program to make a frequency measurement of an output from a LED color sensor; I find that it works only about 95% of the time.

When it fails it simply returns '0' although sometimes I have received 9.9E37 which I interpret as an out-of-range indication.

When the measurement succeeds it is in the 12 KHz range so it certainly isn't too fast.  The settings are 10V range, 5 1/2D, FAST.   The signal is a 5V that drops to 0 V for about 12 uS every 106 uS.  I am wondering if because the LO is only about 12% of the cycle length I may be 'on the edge' of the measurement algorithms capability?  There is some over/undershoot but I put a 3.3 nF cap across the probes; it cleared up the over/undershoot but didn't help the measurement.  Please see the attached image from an oscope capture of the signal.

Advice?

Thanks!   
附件

• LED_Sensor_Wave.bmp949.3 KB