Top Banner
Warm Front End (2- meter) Wes Grammer NRAO March 15-17, 2012 EOVSA Preliminary Design Review 1
28

Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

Dec 22, 2015

Download

Documents

Ami Johnston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 1

Warm Front End (2-meter)

Wes GrammerNRAO

March 15-17, 2012

Page 2: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 2

Outline

• Design requirements• Block diagrams• Cascaded gain/noise analysis (FE + BE)• Component selection• Thermal management• Mechanical layout and enclosure• Interfaces (mechanical and electronic)• Production assembly and test• Costing and schedule

March 15-17, 2012

Page 3: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 3

2-meter Warm Front End Design Requirements and Specifications

• 1-18 GHz instantaneous BW, dual linear polarizations• Overall Tsys < 400K, at ambient (~298K)• Gain stability < 1%, phase stability < 1°, over TBC sec.• Receiver outputs modulated on SMF, range > 2 km• Active temperature control, < ±0.1C diurnal stability• Overall volume should fit within a 12” dia. x ~12” long cylinder,

excluding antenna feed and connectors• Total weight < 20 lbs (9.1 kg)• Sealed enclosure and bulkhead connectors to IP67, suitable for

outdoor installation• MTBF > 26,000 hrs (3 yrs)

March 15-17, 2012

Page 4: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 4

Front End Assembly Block Diagram

March 15-17, 2012

TE A s s e m b ly

F A N + F A N - T E C + T E C -

TE C 1L a irdD A -0 7 5 -2 4 -0 2 -0 0 -0 0

C P 1

2 0 d B C o u p le rK ry t a r1 8 0 1 2 0

C P 2

2 0 d B C o u p le rK ry t a r1 8 0 1 2 0

S P L 1

2 -wa y S p lit t e rW e in s c h e lW A 1 5 1 5

1

2

I N

D 1

3 0 d B E N RN o is e wa v eN W 1 8 G -3 0 -C S

R F +2 4 V

V RF In1-18 GHz

H RF In1-18 GHz

-12V

A 1

3 5 d B g a in , 1 d B N F

L N F -L N 1 _ 1 3 CL o w N o is e F a c t o ry+2 . 5 d B m P o _ 1 d B

A 3

2 8 d B g a in , 4 d B N F+1 8 d B m P o _ 1 d BH e ro t e kA 0 1 1 8 2 8 4 0 1 8 B

+24V

W D M 1F . O . D ip le x e rO p t ila b

P A D 1

-3 d B

B W -S 3 -2 W 2 6 3 +M in i-C k t s

+3 . 3 V

H / VFO Out

F I L T1

1 8 G H z L o wp a s s<M a n u f a c t u re r><M a n u f a c t u re r p / n >

A 5

3 1 d B g a in 5 d B N F+2 0 d B m P o _ 1 d BC ia o W ire le s sC A 1 1 8 -4 5 5

A 2

3 5 d B g a in , 1 d B N F

L N F -L N 1 _ 1 3 CL o w N o is e F a c t o ry+2 . 5 d B m P o _ 1 d B

M&C I/O

L N A _ B I A S 1

C P 3

2 0 d B C o u p le rK ry t a r1 8 0 1 2 0

1-Bit Digital Attenuator

A TN 1

-2 / -1 2 d BR H L a b o ra t o r ie s1 2 -2 0 8 5

I N O U T

L N A _ B I A S 2

D 2S c h o t t k y D e t e c t o rK ry t a r2 0 1 A P

RF

A 7

D u a l d e t e c t o r p re a m p

E O V S AE O V S A -F E -2 M -A U X-A

Ethernet I/O

TE A s s e m b ly

F A N + F A N - T E C + T E C -

TE C 2L a irdD A -0 7 5 -2 4 -0 2 -0 0 -0 0

A 4

2 8 d B g a in , 4 d B N F+1 8 d B m P o _ 1 d BH e ro t e kA 0 1 1 8 2 8 4 0 1 8 B

P A D 2

-3 d B

B W -S 3 -2 W 2 6 3 +M in i-C k t s

F I L T2

1 8 G H z L o wp a s s<M a n u f a c t u re r><M a n u f a c t u re r p / n >

A 6

3 1 d B g a in 5 d B N F+2 0 d B m P o _ 1 d BC ia o W ire le s sC A 1 1 8 -4 5 5

C P 4

2 0 d B C o u p le rK ry t a r1 8 0 1 2 0

1-Bit Digital Attenuator

A TN 2

-2 / -1 2 d BR H L a b o ra t o r ie s1 2 -2 0 8 5

I N O U T

D 3S c h o t t k y D e t e c t o r

2 0 1 A PK ry t a r

RF

A 8

Two -s t a g e H E M T b ia s

E O V S AE O V S A -F E -2 M -A U X-A

D C -A [ 6 . . 1 ]

TE _ C TR L [ 6 . . 1 ]

D C -B [ 2 . . 1 ]

E TH [ 8 . . 1 ]

F O O U T

DC PowerInputs A&B

L N A _ B I A S 1

O P T_ TX1F . O . Tra n s m it t e rO p t ila bL T-2 0

+V

I N

O P T_ TX2F . O . Tra n s m it t e rO p t ila bL T-2 0

+V

I N

L N A _ B I A S 2

V P O L _ P O U T

H P O L _ P O U T

+12V "A"

t

R T1TS E N S E

+2 4 V

U 2

E m b e d d e d M P UR M C 4 0 0 0R a b b it

G N D +3 . 3 V

I O [ 1 . . 1 5 ]

E TH [ 1 . . 8 ]

5 -B it D ig it a l A t t e n u a t o r

A TN 3

5 - 2 5 d B a t t e n . @ 1 8 G H zH M C 9 3 9 L P 4H it t it e

I N O U T

+12V "B"

5 -B it D ig it a l A t t e n u a t o r

A TN 4

5 - 2 5 d B a t t e n . @ 1 8 G H zH M C 9 3 9 L P 4H it t it e

I N O U T

H

VA N T1D u a l L o g -P e r.TE C O M8 0 5 9 2 0 -0 0 2

Front End Module Assembly Enclosure, 2-Meter Antenna

TE AssyConnect

Page 5: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 5

2-Meter Front End Optical Fiber, M&C and DC Supply Routing

March 15-17, 2012

S M F [ 1 2 . . 1 ]

A u x ilia ry E q u ip m e n t C a b in e t

A U XC A B

S M F [ 1 2 . . 1 ]

TE _ C TR L [ 6 . . 1 ]

D C _ I N -A [ 6 . . 1 ]

D C _ I N -B [ 2 . . 1 ]

F O _ A N L G

F O _ E TH [ 2 . . 1 ]

F R O N T E N D A S S Y

2 -M e t e r F ro n t E n d A s s e m b ly

D C -A [ 6 . . 1 ]

F O O U T

E TH [ 8 . . 1 ]

D C -B [ 2 . . 1 ]

TE _ C TR L [ 6 . . 1 ]

Antenna Control Cabinet

S W 1

E t h e rn e t S w it c h w / F O I n pE D S -2 0 5 A -S -S C -TM o x a

F X[ 1 . . 2 ]

E TH [ 1 . . 8 ]

2-Meter Antenna Apex

BuriedFO Cable

Page 6: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 6

System Cascade Analysis (1)

• Excel workbook created to perform stage-by-stage cascaded analysis of the following:– System gain, including mismatch loss– Noise temperature– Gain ripple and slope over IF bandwidth– Output spectral power density and total power– 1 dB compression point, and output margin– Third-order intercept point (IP3)– Output IMD level, assuming strong interference

March 15-17, 2012

Page 7: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 7

System Cascade Analysis (2)

• Component data entered in tables • Effect of cables and adapters estimated• Analyzed at four different input levels: -73, -60, -50 and

-35 dBm, at 18 GHz • Simplifications and assumptions:

– Worst-case parameters mostly used (min. gain, IP3 and P1dB; max. loss, VSWR and NF) @ 18 GHz

– Average of mismatch error used for cascade– Antenna conductor and dielectric loss unknown, a figure of

0.5 dB was arbitrarily assigned– Input level of optical TX is nominally +6 dBm

March 15-17, 2012

Page 8: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 8March 15-17, 2012

Page 9: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 9

System Cascade Analysis (3)

• Key results, after optimization:– ENR of 30 dB required with specified splitter and 20

dB coupler, in order to inject ~ 400K noise– LNA still ~6 dB below compression at -35 dBm max

input, other amps better. – Overall Tsys < 330K up to -50 dBm, but rises way over

400K for strongest input, from added atten.• Effect greatly reduced by driving optical TX at upper end of

its linear range (+11 dBm) for this case. Overall Tsys is still slightly higher than spec, ~420K

• May slightly complicate software to set atten. levels

March 15-17, 2012

Page 10: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 10

System Cascade Analysis (4)

• Results (cont):– Optical TX drive at lowest signal input:

• +4.3 dBm, at 18 GHz w/min. nom. setting -> -0.7 dB margin!

• Extreme case, due to steep gain slope of LNA at this end (~2 dB/GHz). Gain over most of band nearly 5 dB higher, thus total power from 1-18 GHz will be higher than worst-case above.

– Around 10 dB headroom in the digital attenuator, for reducing system gain, for non-worst case

March 15-17, 2012

Page 11: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 11

System Cascade Analysis (5)

• Limitations of analysis tool:– Cascaded gain ripple and slope are incomplete; lack of

component data, unknown dependence on phase– No broadband frequency dependence of component

parameters is modeled; lack of time and hard data– Output power is assumed linear; no attempt made to model

saturation or near-saturation behavior– Did not model change in solar input power across frequency, or

effect on total output power through components having significant gain slope (e.g., LNA)

– Analyzed only main signal path; noise source with splitter and coupler done separately and manually

March 15-17, 2012

Page 12: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 12

Component Selection Criteria

• Flat frequency response, where possible• Good VSWR, where possible, to limit mismatch loss and gain

ripple• Integration of multiple functions into a single package where

cost-effective, for better performance, fewer interconnects, saves critical space

• For amplifiers, lowest power dissipation that still meets output drive requirements

• Common bias voltages (e.g., +12V), where possible• Cost is critical: As there are 30 each of most Front End

component types in the array, an expensive item can have a large impact in the overall budget.

March 15-17, 2012

Page 13: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 13March 15-17, 2012

Page 14: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 14

Hittite 5-bit Digital Attenuator

March 15-17, 2012

Page 15: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 15

Optilab LT-20/LR-30 Link Loss

March 15-17, 2012

Page 16: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 16

Thermal Management

• Active cooling to be used, for following reasons:– Limited volume with high component packing density – Relatively large thermal dissipation (~64W max, est.)– External ambient temp can reach +45C or more, also has

direct solar exposure– Will allow best receiver gain stability w/reliability

• Requirements:– Robust, reliable system for harsh outdoor environment– Reasonable installation and operating costs– Compact and lightweight (portion mounted on antenna)

March 15-17, 2012

Page 17: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 17

Options for Front End Cooling• Liquid cooling

– Very compact, lightweight, quiet (on Front End side)– Works best when cooling a device to a temperature close to the

surrounding ambient. Supply line (cold side) needs to be insulated for most efficient operation.

– Requires a remote pump and chiller assembly – raises cost• Direct thermoelectric cooling

– Compact, efficient, and light enough for use on antenna– Works well over wide ambient temperature range– No coolant required, only DC power – easier to test and install– Reversible; can heat or cool as required, with suitable controller– Fewer, lower-cost system components– Requires external fans, which limit MTBF

March 15-17, 2012

Page 18: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 18

Laird 71W TE Assembly, Controller

March 15-17, 2012

Page 19: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 19

Internal layout considerations

• Cascade from feed outputs to LNAs should be as short and direct as possible, to minimize losses

• Optical TX modules drive overall placement, because of their large size and power dissipation.

• Good heat sinking for active components a must, for gain stability and long field life

• Interface connectors should be located at the opposite end of enclosure to the feed, for ease of access and minimal antenna obstruction.

• Design for easy assembly, testing, serviceability

March 15-17, 2012

Page 20: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 20

Enclosure Design Requirements

• Weather-tight to IP67 (no dust or non-submerged water ingress); one-piece replaceable seals

• UV-resistant material, rated for outdoor use• Clamshell design, for easy assembly and service• Well-insulated, to minimize ambient thermal

loading, for a stable internal environment• Base strong enough to support 9 kg max weight,

nor break under expected handling in the field• Low cost COTS catalog item, if possible

March 15-17, 2012

Page 21: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 21

Warm Front End AssemblyConceptual Mechanical Layout

March 15-17, 2012

Page 22: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 22

Front End Assembly Interfaces

• Hardware:– (2) SMA-M inputs from antenna feed– (1) SM fiber connector output; possibly LC/APC– (1) 10/100Base-T Ethernet I/O for all M&C– (3) MIL-DTL-26482 multi-pin connectors, for DC

power input to Front End electronics, TE coolers• Software:

– Refer to table in following slide

March 15-17, 2012

Page 23: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 23March 15-17, 2012

Signal Name Dir. Type Mode Range Min Range Max Unit Connector Freq/Rate PrecisionX Pol from Ant In RF Power Analog Elec -70 -35 dBm SMA 1-18 GHz N/AY Pol from Ant In RF Power Analog Elec -70 -35 dBm SMA 1-18 GHz N/AND Ctrl In Bool Digital Elec 0 24 V Ethernet 1 Hz 1 msLNA Vd Ctrl X-Pol In Float Digital Elec 0 1.5 V Ethernet 1 Hz 1 sLNA Vd Mon X-Pol Out Float Digital Elec 0 1.5 V Ethernet 1 Hz 1 sLNA Id Ctrl X-Pol In Float Digital Elec 0 50 mA Ethernet 1 Hz 1 sLNA Id Mon X-Pol Out Float Digital Elec 0 50 mA Ethernet 1 Hz 1 sLNA VG Mon X-Pol Out Float Digital Elec -4 0 V Ethernet 1 Hz 1 sLNA Vd Ctrl Y-Pol In Float Digital Elec 0 1.5 V Ethernet 1 Hz 1 sLNA Vd Mon Y-Pol Out Float Digital Elec 0 1.5 V Ethernet 1 Hz 1 sLNA Id Ctrl Y-Pol In Float Digital Elec 0 50 mA Ethernet 1 Hz 1 sLNA Id Mon Y-Pol Out Float Digital Elec 0 50 mA Ethernet 1 Hz 1 sLNA Vg Mon Y-Pol Out Float Digital Elec -4 0 V Ethernet 1 Hz 1 sAtten1 Ctrl X-Pol In Bool Digital Elec 0 10 dB Ethernet 1 Hz 1 msAtten1 Ctrl Y-Pol In Bool Digital Elec 0 10 dB Ethernet 1 Hz 1 msAtten2 Ctrl X-Pol In Integer Digital Elec 0 31 dB Ethernet 1 Hz 1 msAtten2 Ctrl Y-Pol In Integer Digital Elec 0 31 dB Ethernet 1 Hz 1 msTX PwrIn Mon X-Pol Out Float Digital Elec 0 20 dBm Ethernet 50 Hz 1 msTX PwrIn Mon Y-Pol Out Float Digital Elec 0 20 dBm Ethernet 50 Hz 1 msTemp Mon X-Pol Out Float Digital Elec -20 60 degC Ethernet 1 Hz 1 sTemp Mon Y-Pol Out Float Digital Elec -20 60 degC Ethernet 1 Hz 1 sLaserTX Alrm Mon X-Pol Out Bool Digital Elec n/a n/a n/a Ethernet 1 Hz 1 sLaserTX Alrm Mon Y-Pol Out Bool Digital Elec n/a n/a n/a Ethernet 1 Hz 1 sTemp Alrm Mon Out Bool Digital Elec 10 50 degC Ethernet 1 Hz 1 sTE Cooler M&C I/O RS-232 Digital Elec n/a n/a n/a DB9 male 1 Hz 1 sX / Y Pol from FE In RF on fiber Analog Fiber -30 -35 dBm SC/APC 1-18 GHz N/A

Page 24: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 24

Production Assembly, Testing

• Production process steps– Two-port VNA measurement of amplifiers, filters,

digital attenuators, couplers, and optical link sets– Assembly of receiver halves (w/o WDM, noise src)– Two-port VNA measurement of half-RX, w/Opt RX– Final assembly of full receiver– Test noise source functionality– Verification of M&C functionality and cooling system

function in a test chamber at 50°C– Documentation: Test results, configuration (s/n) list

March 15-17, 2012

Page 25: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 25

Production area at NTC Photonics Lab

March 15-17, 2012

Page 26: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 26

Component Costing, Delivery

• All RF components except LPF have been specified, price/delivery quotes received

• Good estimates or preliminary pricing on remaining electronic components

• Enclosure, connector and cable pricing are rough estimates, final TBD

• Longest lead times:– TECOM antennas (21 weeks); use existing OVSA units for prototype

Front Ends– 1-bit digital attenuator (18 weeks); can be shortened to 8 weeks for

30% extra– Many other have 12-16 week leads, but only for large qty

March 15-17, 2012

Page 27: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 27March 15-17, 2012

Page 28: Warm Front End (2-meter) Wes Grammer NRAO March 15-17, 2012EOVSA Preliminary Design Review1.

EOVSA Preliminary Design Review 28

Important Schedule Dates for Front End Prototypes

• Have all RF components and COTS support electronics on order by mid-April at the latest

• Complete mechanical CAD models and fabrication drawings by May 1, send out for quotes, begin fab in mid-May

• May 1 – June 30: Design, PCB fab and component ordering for custom support electronic boards

• July 1 – July 15: Order cables, connectors, wire, remaining components for prototype unit construction and site installation

• July 1 – August 1: RF component characterization• August – September: Assemble and test 3 prototypes• October 1: Ship 3 prototypes to California for installation• Front End embedded firmware and test software may need to be farmed

out, in the interest of saving time. This could happen during July and August, in parallel with assembly.

March 15-17, 2012