Post Go back to editing

AD9545 Small Cell / IEEE1588 application more complex explanation

Category: Hardware
Product Number: AD9545

Dear Sir,

I've been struggling with understanding the whole concept of AD9545 working in the small cell application which is shown in the Fig. 117 in the datasheet and I was hoping you could give me more complex explanation of it. For the sake of simplicity let's assume that the input reference (entering for example REFA/REFAA input) signal is 156.25 MHz SyncE signal from the external network. Following the example frequencies (greyed out ones) given in the mentioned picture let's further assume that from the OUT1A/OUT1B outputs exit 156.25 MHz signals to the "Network Connection" block and further again to the "External Network" block (so basically in the loop). The other OUTPUTS (OUT0A, OUT0B, OUT0C) outputs have frequencies: 7.68 MHz, 122.88 MHz, 245.76 MHz. Could you please provide with the more complex answers for the following questions regarding this configuration?

1) What is the purpose of creating the loop of 156.25 MHz frequency when External Network acts as a main controller and is the node that sources the frequency to the AD9545 - why is this backloop coming out of the AD9545 to the External Network?
2) Because I believe it is not directly pointed out in this picture - since input/output (from PLL1) are the same frequency so are they are in ZDM mode? And also since there is a fractional relationship between output frequencies of PLL0 and reference (REFA) the phase aligment between them is not preserved. In such apllication as shown here this lack of phase synchronization for those signals is not a problem?

I'm also not sure of the application shown in the Fig. 118. (Using an AD9545 in an IEEE1588 Version 2 Application) and I have also some questions regarding this configuration:

1) Why there are two reference input simultaneously? What should be the frequency range of the signal entering the REFB/REFBB pins? Is this signal also supposed to be of SyncE standard?
2) Are these two 1 PPS signals going to the Mx pins used for closed loop compensation mechanism?
3) Why there is also loopback signal going out of the AD9545 (PLL0) and enetring this external timing device?
4) What are the phase dependencies between input signals and output signals in this application (REFA/AA and REFB/BB vs outputs of the PLL0 or PLL1)? Are any of them configured in ZDM?








Best regards

  • Hi,

    as you can see in the Revision History section, the AD9545 was released to manufacturing and sales in June 2017. It is hard for me to come up with the reasons behind the block diagrams from the data sheet. I do not have the system perspective a system engineer dealing with radio cells may have.

    Anyway:

    "1) What is the purpose of creating the loop of 156.25 MHz frequency when External Network acts as a main controller and is the node that sources the frequency to the AD9545 - why is this backloop coming out of the AD9545 to the External Network?"

    It just wants to show that if the SyncE is used as reference clock, that SyncE comes from the external network.

    "2) Because I believe it is not directly pointed out in this picture - since input/output (from PLL1) are the same frequency so are they are in ZDM mode? "

    It depends on the relation between the reference clock and the output frequencies. If the 1PPS is the reference clock, then all the outputs may be generated in ZDM. If there is no direct proportionality between the reference clock and the outputs, then you are forced to set the PLL in phase buildout mode.

    "And also since there is a fractional relationship between output frequencies of PLL0 and reference (REFA) the phase aligment between them is not preserved. In such apllication as shown here this lack of phase synchronization for those signals is not a problem?"

    I'm not s system engineer, so I cannot respond to this question.

    Regarding the IEEE 1588 implementation from figure 118. I usually send customers the attached paper that has more details of the implementation.IEEE1588_servo_implementation_20210126.pdf

    It should have the answers to your questions.

    The AD9546 rev 0 data sheet has a different architecture of an IEEE 1588 Servo (figure 131, page 195). It also has more explanations. The AD9545 could implement this architecture as well.

    Petre

Before You Switch


Switching languages will make ADI Explorer unavailable. Resume your session by switching back to English and reopening ADI Explorer.