LTC6953 datasheet(51/56 Pages) AD

51 / 56 page

LTC6953

Rev 0

For more information www.analog.com

Preliminary Technical Data

Advance Product Information Subject to Change

Rev PrA

Figure 40. Example of Spurious Measurement Technique

and differential input resistance (RDIFF). Use Equation 14

to calculate R1 and Equation 15 to calculate R2.

R1 = RDIFF •

VCM

0.44 – 0.5

⎡

⎣⎢

⎤

⎦⎥

(14)

R 2 = RDIFF •

VDD – VCM

0.44 – 0.5

⎡

⎣⎢

⎤

⎦⎥

(15)

For receiver devices with internal 100Ω terminations, the

values of R1 and R2 can be very small and will affect the

overall termination impedance, leading to undesirable

impedance mismatch. For this reason, the use of pulsed

SYSREFs (MODEx = 3) AC-coupled into receiver parts

with internal 100Ω terminations is not recommended.

Settling time for pulsed SYSREF connections (Figure 39)

is approximately determined by the AC-coupling capaci-

tors (CAC), both the differential and common mode input

resistance of the receiver device (RDIFF and RCM), and

resistors R1 and R2:

tsettleP ≅ 10 •

RDEV • ROS

RDEV + ROS

⎛

⎝

⎜

⎞

⎠

⎟ • C AC

(16)

where:

RDEV =

2RCM + RDIFF

ROS = MINIMUM(R1,R2)

(17)

For pulsed mode SYSREFs to work correctly with

AC-coupling, tsettleP must be greater than 1000/fSYSREF,

where fSYSREF is the frequency of the SYSREF pulses.

Use the following procedure to achieve correct JESD204B

SYSREF behavior for AC-coupled pulsed SYSREFs.

These methods assume that the SYSREF outputs have

already been synchronized and that the SYSREF output

drivers have been disabled for power savings (PDx = 2).

Pulsed SYSREFs (MODEx = 3)

1. Enable the LTC6953 SYSREF output drivers by setting

PDx = 0 and set SRQMD = 1.

2. Wait for a settling period of at least tsettleP.

3. Set the receiver device to accept SYSREFs.

4. Set SSRQ or the EZS_SRQ inputs to “1” for at least

1ms, then set back to “0”.

5. Set the receiver device to stop accepting SYSREFs.

6. Disable the LTC6953 SYSREF output drivers by set-

ting PDx = 2 and set SRQMD = 0.

MEASURING DIFFERENTIAL SPURIOUS SIGNALS

USING SINGLE-ENDED TEST EQUIPMENT

Using a spectrum analyzer to measure spurious signals

on the single-ended output of a clock generation chip

will give pessimistic results, particularly for outputs that

approximate square waves. There are two reasons for

this.

First, since the spurious energy is often an AC signal

superimposed on the power supply, a differential output

will reject the spurs to within the matching of the posi-

tive and negative outputs. Observing only one side of the

differential output will provide no rejection.

Second, and most importantly, the spectrum analyzer will

display all of the energy at its input, including amplitude

modulation that occurs at the top and bottom pedestal

voltage of the square wave. However, only amplitude

modulation near a zero crossing will affect the clock.

The best way to remove this measurement error is to

drive the clock generator output differentially into a limit-

ing buffer on a separate clean power supply. One of the

differential outputs of the limiting buffer can then connect

to a spectrum analyzer to correctly measure the spurious

energy. An example of this technique using the LTC6953

as the clock generator and an LTC6955 as the limiter is

shown in Figure 40.

100Ω

50Ω

OUTx +

OUTx –

6953 F40

LTC6953

OUTx +

OUTx –

LTC6955

IN+

IN –

SPECTRUM

ANALYZER

APPLICATIONS INFORMATION

LTC6953 数据表(PDF) 51 Page - Analog Devices

LTC6953 数据表(HTML) 51 Page - Analog Devices

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

数据表下载