Cryogenic Performance of NGC 35nm
InP Low Noise Amplifiers
Eric W. Bryerton
• Wafer run complete 1/15/2008
• Chip Dimension: 2.1mm x 0.8mm
EBLNA81
Cryogenic Performance of 35nm InP LNAs
On-chip Measurements
EBLNA81 (VG=+0.3V, VD=1.2V)
On-wafer, T=297K
0
5
10
15
20
25
30
35
40
65 70 75 80 85 90 95
Frequency (GHz)
Gain
(d
B)
R13C02M0R13C02M1R14C02M0R15C02M0Simulation
• Gm increased in simulation (from 1950 to 2500 mS/mm) to match
measured performance
Cryogenic Performance of 35nm InP LNAs
On-Chip Noise Measurement
Cryogenic Performance of 35nm InP LNAs
EBLNA81 (VG=+0.22V, VD=0.80V, ID=20mA)
on-wafer, T=297K
(+/-30K error bars)
0
50
100
150
200
250
300
350
400
450
500
60 65 70 75 80 85 90
Frequency (GHz)
No
ise
Te
mp
era
ture
(K
)
R13C02M0R13C02M1R14C02M0R15C02M0Simulation
On-chip Noise Measurement
Cryogenic Performance of 35nm InP LNAs
WR-12 LNA Module
• Same MDM-15 pinout as CDL LNAs
Cryogenic Performance of 35nm InP LNAs
WR-12 LNA Module
20pF
1K
Cryogenic Performance of 35nm InP LNAs
LNA Module Measurements
S21 Noise
Cryogenic Performance of 35nm InP LNAs
LNA Module Measurements
EBLNA81
T=297K, Vd=0.8V, Id=21mA
0
5
10
15
20
25
30
35
70 75 80 85 90 95 100
Frequency (GHz)
S2
1 (
dB
)
Cryogenic Performance of 35nm InP LNAs
EBLNA81
Vd=0.8V
0
5
10
15
20
25
30
35
70 75 80 85 90 95 100
Frequency (GHz)
S2
1 (
dB
)
T=297K,Id=21mA
T=17.5K,Id=9mA
LNA Module Measurements
Cryogenic Performance of 35nm InP LNAs
Amplifier Noise Temperature
0
50
100
150
200
250
300
350
400
450
500
60 65 70 75 80 85 90 95
Frequency (GHz)
No
ise
Te
mp
[K
]
WR-12 mixer
WR-10 mixer
Amplifier ID: eblna81
Timestamp: 7/9/2008 11:12 AM
Measured by:
Dewar Temp: 327
Ambient Temp: 327
Noise Diode:11833
Software Version: 1.3
Vd=0.8V, Id=7mA/stage
LNA Module Measurements
Cryogenic Performance of 35nm InP LNAs
35nm 3-stage LNA (EBLNA81)
(Vd1=0.5V,Vd2=0.8V,Vd3=0.8V)
(Id1=1.5mA,Id2=3.0mA,Id3=3.0mA)
0
10
20
30
40
50
60
70
80
90
100
60 65 70 75 80 85 90 95 100
Frequency (GHz)
No
ise
Te
mp
[K
]WR-12 mixer
WR-10 mixer
Simulation
• Simulation: Gm=2500mS/mm, Igs=0, Tdrain=1400
• Pdiss = 5.55mW
LNA Module Measurements
Cryogenic Performance of 35nm InP LNAs
35nm 3-stage LNA (EBLNA81)
0
10
20
30
40
50
60
70
80
90
100
60 65 70 75 80 85 90 95 100
Frequency (GHz)
No
ise T
em
p [
K]
• Blue trace shows noise for minimum power dissipation:
Vd1=0.35V, Vd23=0.5V, Id=4.5mA -> Pdiss = 2.06mW
LNA Module Measurements
Cryogenic Performance of 35nm InP LNAs
Pinchoff Characteristics
Average Tnoise from 83-87 GHz
Relative Gain at 85 GHz
Tamb=17.5K
0
10
20
30
40
50
60
70
80
90
100
0 1 2 3 4 5 6
Idrain (mA)
Tn
ois
e (
K)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Rela
tive G
ain
Tnoise
Gain
Average Tnoise over min 4GHz
Relative Gain at 81 GHz
Tamb=297K
0
100
200
300
400
500
600
700
800
900
1000
0 2 4 6 8 10 12 14 16
Idrain (mA)
Tn
ois
e (
K)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Rela
tive G
ain
Tnoise
Gain
Cryogenic Performance of 35nm InP LNAs
EBLNA81
(Vd1=0.5V,Vd23=0.8V,Id1=1.5mA,Id23=3mA)
0
50
100
150
200
250
300
350
400
450
500
65 70 75 80 85 90
Frequency (GHz)
No
ise
Te
mp
era
ture
(K
)
200K
40K77K
100K120K
150K180K
17.5K
240K
280K
Why the frequency shift cold?
Cryogenic Performance of 35nm InP LNAs
EBLNA81
(T=297K,Vd=0.8V,Id23=7mA)
0
100
200
300
400
500
600
700
800
65 70 75 80 85 90
Frequency (GHz)
No
ise T
em
pera
ture
(K
)
Id1=7mAId1=3mAId1=1.5mAId1=1mAId1=0.5mA
• Hypothesis: Cgs(Vgs) is causing frequency shift
Why the frequency shift cold?
Cryogenic Performance of 35nm InP LNAs
EBLNA81
Vd=0.8V
0
5
10
15
20
25
30
35
70 75 80 85 90 95 100
Frequency (GHz)
S2
1 (
dB
)
T=297K,Id=21mA
T=17.5K,Id=9mA
• Change in S21 also modeled well by decrease in Cgs
Revised cryogenic model for parasitics
Cryogenic Performance of 35nm InP LNAs
Amplifier Noise Temperature
0
50
100
150
200
250
300
350
400
450
500
60 65 70 75 80 85 90
Frequency (GHz)
No
ise
Te
mp
[K
]
Noise Temp (K)
Simulation
Amplifier ID: eblna1
Timestamp: 7/11/2008 4:59 PM
Measured by:
Dewar Temp: 393
Ambient Temp: 393
Noise Diode:11833
Software Version: 1.3
• Id1=1.5mA
• Simulation: gm1=33.6mS, gm23=66.6mS, Cgs1=12.6fF
(versus 17.5fF in original model)
Revised cryogenic model for parasitics
Cryogenic Performance of 35nm InP LNAs
35nm 3-stage LNA (EBLNA81)
(Vd1=0.5V,Vd2=0.8V,Vd3=0.8V)
(Id1=1.5mA,Id2=3.0mA,Id3=3.0mA)
0
10
20
30
40
50
60
70
80
90
100
60 65 70 75 80 85 90 95 100
Frequency (GHz)
No
ise
Te
mp
[K
]
WR-12 mixer
WR-10 mixer
Simulation
• Simulation: Gm1=64mS, Gm23=90mS, Td1=1400K,
Td23=2800K, Cgs1=12.6fF (compared to 17.5fF original)
Revised cryogenic model for parasitics
Cryogenic Performance of 35nm InP LNAs
EBLNA81B: Retuned for 67-90 GHz35nm 3-stage LNA (EBLNA81)
(Vd1=0.5V,Vd2=0.8V,Vd3=0.8V)
(Id1=1.5mA,Id2=3.0mA,Id3=3.0mA)
0
10
20
30
40
50
60
70
80
90
100
60 65 70 75 80 85 90 95 100
Frequency (GHz)
No
ise
Te
mp
[K
]
WR-12 mixer
WR-10 mixer
Sim:EBLNA81
Sim: EBLNA81B
L13: 65 to 34um
L14: 78 to 144um L14L13
Cryogenic Performance of 35nm InP LNAs
EBLNAW+: Cover full 68-116 GHz band
21Cryogenic Performance of 35nm InP LNAs
MMLN100: Cover full 68-116 GHz band
22Cryogenic Performance of 35nm InP LNAs
EBLNAW0: A “Tunable” MMIC LNA
23Cryogenic Performance of 35nm InP LNAs
Questions / Discussion Points
• Is the process repeatable?
• These results are with 70% channel, what do they look like with 100%
channel (higher gm)?
• Operation at 4K
• Ultimate limit for W-band LNA noise temperature
• Comparison to SIS development
Cryogenic Performance of 35nm InP LNAs 24