kad5514p Kenet Inc., kad5514p Datasheet - Page 28

• Manufacturer Part Number: kad5514p
• Description: 14-bit, 250/210/170/125msps Adc
• Manufacturer: Kenet Inc.

Exposed Paddle
The exposed paddle must be electrically connected
to analog ground (AVSS) and should be connected
to a large copper plane using numerous vias for opti-
mal thermal performance.
Bypass and Filtering
Bulk capacitors should have low equivalent series re-
sistance. Tantalum is a good choice. For best per-
formance, keep ceramic bypass capacitors very
close to device pins. Longer traces will increase in-
ductance, resulting in diminished dynamic perform-
ance and accuracy. Make sure that connections to
ground are direct and low impedance. Avoid form-
ing ground loops.
LVDS Outputs
Output traces and connections must be designed for
50Ω (100Ω differential) characteristic impedance.
Keep traces direct and minimize bends where possi-
ble. Avoid crossing ground and power-plane breaks
with signal traces.
LVCMOS Outputs
Output traces and connections must be designed for
50Ω characteristic impedance.
Unused Inputs
Standard logic inputs (RESETN, CSB, SCLK, SDIO, SDO)
which will not be operated do not require connec-
tion to ensure optimal ADC performance. These in-
puts can be left floating if they are not used. Tri-level
inputs (NAPSLP, OUTMODE, OUTFMT, CLKDIV) accept
a floating input as a valid state, and therefore should
be biased according to the desired functionality.
Definitions
Analog Input Bandwidth is the analog input fre-
quency at which the spectral output power at the
fundamental frequency (as determined by FFT analy-
sis) is reduced by 3dB from its full-scale low-frequency
value. This is also referred to as Full Power Bandwidth.
Aperture Delay or Sampling Delay is the time re-
quired after the rise of the clock input for the sam-
pling switch to open, at which time the signal is held
for conversion.
Aperture Jitter is the RMS variation in aperture delay
for a set of samples.
KAD5514P
Preliminary
Clock Duty Cycle is the ratio of the time the clock
wave is at logic high to the total time of one clock
period.
Differential Non-Linearity (DNL) is the deviation of any
code width from an ideal 1 LSB step.
Effective Number of Bits (ENOB) is an alternate
method of specifying Signal to Noise-and-Distortion
Ratio (SINAD). In dB, it is calculated as: ENOB =
(SINAD-1.76) / 6.02
Gain Error is the ratio of the difference between the
voltages that cause the lowest and highest code
transitions to the full-scale voltage less 2 LSB. It is typi-
cally expressed in percent.
Integral Non-Linearity (INL) is the maximum deviation
of the ADC’s transfer function from a best fit line de-
termined by a least squares curve fit of that transfer
function, measured in units of LSBs.
Least Significant Bit (LSB) is the bit that has the small-
est value or weight in a digital word. Its value in terms
of input voltage is V
in bits.
Missing Codes are output codes that are skipped
and will never appear at the ADC output. These
codes cannot be reached with any input value.
Most Significant Bit (MSB) is the bit that has the largest
value or weight.
Pipeline Delay is the number of clock cycles between
the initiation of a conversion and the appearance at
the output pins of the data.
Power Supply Rejection Ratio (PSRR) is the ratio of the
observed magnitude of a spur in the ADC FFT,
caused by an AC signal superimposed on the power
supply voltage.
Signal to Noise-and-Distortion (SINAD) is the ratio of
the RMS signal amplitude to the RMS sum of all other
spectral components below one half the clock fre-
quency, including harmonics but excluding DC.
Signal-to-Noise Ratio (without Harmonics) is the ratio
of the RMS signal amplitude to the RMS sum of all
other spectral components below one-half the sam-
pling frequency, excluding harmonics and DC.
SNR and SINAD are either given in units of dB when
the power of the fundamental is used as the refer-
ence, or dBFS (dB to full scale) when the converter’s
full-scale input power is used as the reference.
Spurious-Free-Dynamic Range (SFDR) is the ratio of
the RMS signal amplitude to the RMS value of the
largest spurious spectral component. The largest spu-
rious spectral component may or may not be a har-
monic.
FS
/(2
N
-1) where N is the resolution
Page 28