This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DS15EA101: Loop filter capacitor

Part Number: DS15EA101

Hello, we are using the DS15EA101 part to capture 8b/10b differential signals after 150ft of cable. We are running at 800Mbps and have had some issues receiving good quality data. We added a 100K resistor in parallel with the 1uF loop filter cap. This addition has allowed the device to capture reliable data now. It seems very robust. 

The question is, why does this help and do you have any test data for a usage in this way?

Thank you

  • Hi David,
    First of all, we haven't seen or used resistor in series with the 1uF capacitor. Therefor this has not been characterized or there is not a test data with this configuration.

    1uF Capacitor is the integration capacitor used for equalizer adaptation. Voltage across this capacitor is proportional to the length of the cable being equalized. This also sets the RC time constant for adaptation completion time. Reducing the value of this cap, decreases adaptation time. On the other hand, increasing this cap value increases adaptation time.

    By adding a resistor, I believe, you are adding an offset in the loop. The end result is that you could equalize more or equalize less - this depends on whether this is connected to AEC+ or AEC- side. To see if adding this resistor provides more equalization or less equalization, you can do the following experiment:
    1). Operate the device as noted in the data sheet and measure voltage across 1uF cap - across AEC+ and AEC- pins respectively.
    2). Add the resistor as you've done, verify all is good, and again measure the voltage across AEC+ and AEC- pins.
    Higher value at #2 tells us we are equalizing more while lower value tells us we are equalizing less. By equalizing more it means we have more gain at fundamental frequency of your data pattern.

    Regards,,nasser
  • Thank you for the response Nasser, we are actually adding the resistor "across" or in parallel with the cap.
    From what you state above, this would slightly lower the voltage which likely means we are equalizing less.
    I will check the voltage with and without the resistor.
    Regards
  • Hi David,

    If this is the case, I was expecting lowering the 1uF cap value only could have the same effect as well.

    Regards,,nasser
  • We will try that also.
    Thank you for the information