CDCE925: Internal load capacitance in Xtal mode

Thank you for your quick reply.

My question is not how to set the internal capacitance, but why the clock frequency in the Xtal mode is the same as the clock frequency in the VCXO mode with Vctrl=0V.

Your document "VCXO Application Guideline for CDCE(L)9xx Family" says the internal capacitance can be changed from 0.3xCLon-chip to 1.3xCLon-chip in the VCXO mode.
According to my measurement, the internal capacitance is set to 1.3xCLon-chip in the Xtal mode.
Is this correct?

Thank you for your reply.

In our environment,the situaion of the capacitance is as follows.
  the external capacitance : not used
  the internal capacitance setting : 8pF
  the external Xtal capacitance : 10pF

And in our measurement, the output clock frequency is changed about 100 ppm by changing the Vctrl from 0 to 1.8V.

In this situation, is the internal capacitance set to 1.3xCLon-chip in the Xtal mode?
If possible, please give us the simple answer to this question at first.

Because of the holidays, TI E2E design support forum responses may be delayed from Dec. 22 through Jan. 2. Thank you for your patience.

Susumu,

So, to be clear - you have 18 pF total capacitance. Is this how much the XTAL datasheet calls for, or does the datasheet call for 10 pF total capacitance?

The varactor (variable capacitor) is able to be swung from 0.3x to 1.3x the selected load capacitance value by adjusting the tuning voltage. 1.3xCLon-chip is with 0V as the tuning voltage, as is shown by Figure 5 in the application note.

If you remove the 10-pF capacitance and only use the on-chip capacitance, is the pulling range increased?
Thanks,

Kadeem

Thank you for your reply.
I do appreciate your repeated explanations.

The datasheet of the XTAL we use calls for 10pF total capacitance.
So I set 8pF to the internal capacitance.
And we use only the internal capacitance.
There is no external capacitance from the beginning.

In this situation, the pulling range is about 100ppm.

What I want to know is the setting of the coefficient for CLon-chip at the Xtal mode (INCLK=00).
Not at the VCXO mode (INCLK=01).

Thank you for your reply!!

>> The coefficient for CLon-chip in XTAL mode is fixed at 1.

Thank you for your comment. This is the answer I want.
However, we think it contradicts our experiment I sent at first.

>> ・Measurement result
>> In Xtal mode (Vctrl=1.8V) 26.9992MHz
>> In VCXO mode (Vctrl=0 V) 26.9992MHz
>> In VCXO mode (Vctrl=0.9V) 27.0011MHz
>> In VCXO mode (Vctrl=1.8V) 27.0019MHz
>>
>> ・Fixed conditions
>> CDCE925 external reference crystal 25.000MHz
>> CDCE925 output clock frequency setting 27.000000MHz
>> CDCE925 internal load capacitance register setting 8pF
>> * Other than OFFSET=01h, bit3-2 INCLK settings are not changed.

We think that the output frequency at Xtal mode should be the same as the output frequency at VCXO mode (Vctrl=0.9V).
Because we understand that "Vctrl=0.9V" is the situation for the coefficient for CLon-chip is 1 in the VCXO mode.

Is this correct?
Please teach why the output frequency at Xtal mode is the same as the output frequency at VCXO mode (Vctrl=0V).

Thank you for your reply.

I understood that "Xtal Mode" can be understood as "VCXO mode with Vctrl=0V".
This conclusion matches our experiment.

I think this is last question.
Please teach actual equation for "CL" at Xtal mode.

The "VCXO Application Guideline for CDCE(L)9xx Family" says, CL = a x CLon-chip + Cs ( 0.3 < a < 1.3 : a is determined by Vctrl).

We understood that this is actual equation for "CL" at VCXO mode.

Please teach the equation of the CL at the Xtal mode.
We concern that "CL = 1.3 x CLon-chip + Cs" at Xtal mode.

Thank you.

Susumu,

I am discussing this with the designer. Please expect a response by EOD Monday.

Thank you,

Cris

Susumu,

We are looking into the design to determine the coefficient for the Xtal load capacitance. However I do want to reiterate that the architecture for Xtal mode is very different from VCXO mode. So upon even having the same CL, these two modes will almost always have a different frequency response. 

Best,

Cris