ADF5610 datasheet(21/48 Pages) AD

21 / 48 page

Preliminary Technical Data

ADF5610

Rev. PrE | Page 21 of 48

where:

NINT is the integer set in Register 0x03.

NFRAC is the fractional part set in Register 0x04.

The autocalibration state machine and the data transfers to the

internal VCO VSPI run at the rate of the FSM clock, tFSM, where

the FSM clock frequency cannot be greater than 50 MHz.

tFSM = tXTAL × 2m

(3)

where m is 0, 2, 4, or 5 as determined by Register 0x0A[14:13].

The expected number of VCO counts, V, is given by

V = floor (N × 2n)

(4)

The nominal VCO frequency measured, fVCOM, is given by

fVCOM = V × fXTAL/(2n × R)

(5)

where the worst case measurement error, fERR, is

fERR ≈ ±fPD/2n + 1

(6)

For example, the autocalibration time (tcal) for an 8-bit step

tuned VCO (where the total number of bits includes those

needed to switch VCO cores) is as follows. First a 20 * (k+1) bit

wait state occurs. The measurement has a programmable

number of wait states, k, of 128 FSM cycles defined by Register

0x0A[7:6] = k. Wideband VCO’s like the ADF5610 require

additional wait time (typically, k = 1) whereas narrowband

VCO’s may be fine setting k =0. The wait period is followed by

the measurement time (Tmmt) for calibration then VSPI data

transfers of 20 clock cycles each. This repeats 8 times since we

have 1 bit to select the VCO core and 7 bits to select the VCO

band for a total of 8 bits. Total calibration time, worst case, is

given by

tCAL = k128 tFSM + 8tPD 2n + 8 × 20 tFSM

(7)

or equivalently

tCAL = tXTAL (8R × 2n + (160 + (k × 128)) × 2m)

(8)

For guaranteed hold of lock, across temperature extremes, the

resolution must be better than 1/8th of the frequency step caused

by a VCO sub-band switch change. Better resolution settings

show no improvement.

VCO Autocalibration Example

The VCO subsystem must satisfy the maximum fPD limited by

the two following conditions:

N ≥ 16 (fINT), N ≥ 20.0 (fFRAC)

where:

N = fVCO/fPD. fPD ≤ 100 MHz.

fINT is integer mode.

fFRAC is fractional-N mode. The minimum N values changes

depending on the operating mode.

For example, if the VCO subsystem output frequency is to

operate at 7.01 GHz and the crystal frequency is fXTAL = 50 MHz,

R = 1, and m = 2 (see Figure 25), then tFSM = 80 ns (12.5 MHz).

When using autocalibration, the maximum autocalibration

FSM clock cannot exceed 50 MHz (see Register 0x0A[14:13]).

The FSM clock does not affect the accuracy of the measure-

ment; it only affects the time to produce the result. This same

clock clocks the 16-bit VCO serial port.

If the time to change frequencies is not a concern, the

calibration time for maximum accuracy can be set and,

therefore, the measurement resolution is of no concern.

Using an input crystal of 50 MHz (R = 1 and fPD = 50 MHz), the

times and accuracies for calibration using Equation 6 and

Equation 8 are listed in Table 7, where minimal tuning time is

1/8th of the VCO band spacing.

Figure 25. VCO Calibration

Table 7. Auto-calibration Example with fXTAL = 50 MHz, R = 1, m = 2

Control Value Register 0x0A[2:0]

tMMT (μs)

tCAL (μs)

fERR Maximum

0.02

33.44

±25 MHz

0.04

33.60

±12.5 MHz

0.08

33.92

±6.25 MHz

0.16

34.56

±3.125 MHz

0.64

38.40

±781 kHz

XREF

CALIBRATION WINDOW

REG 0x02

START

VCO

CTR

FSM

50MHz MAX FOR

FSM + VSPI CLOCKS

REG 0x0A[14:13]

m = [0, 2, 4, 5]

REG 0x0A[2:0]

n = [0, 1, 2, 3, 5, 6, 7, 8]

STOP

tMMT = tXTAL × R × 2n

tPD

÷ 2n

÷ 2m

÷ R

ADF5610 Datasheet(PDF) 21 Page - Analog Devices

ADF5610 Datasheet(HTML) 21 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

Datasheet Download

Link URL