RF Ingress Measurements for STP Cables from Automotive ALSE Test

Larry CohenRamin Shirani

11/8/2017

11/8/2017

Overview• Measurement of the differential RF ingress voltage induced at the termination of an

automotive shielded cable (STP) based on standard automotive radiated immunity absorber-lined shielded enclosure (ALSE) test method described in ISO 11452-2

– Test results from 80 MHz to 1 GHz and 1GHz to 3GHz

– Direct measurement of RF ingress voltage at cable terminations; no PHY MDI port interface

– Results are for differential rms voltage vs frequency at cable termination (100 Ohms differential impedance)

– Test performed on STP with H-MTD connectors and STP with HSD connectors; both cable assemblies were 2 meters long with only end connectors

2

11/8/2017

3 ALSE Automotive Cable RF Ingress Measurement Test Setup

Ground plane table(90 cm above ground

plane floor)

1 m

1.5 m

DUT cable harness, totallength < 2 meters

Balun

Antenna height is 1mabove ground plane floor

Absorber-linedshielded chamber

(ALSE)

Insulating cable support(5 cm thick)

GPIBControl PC

Shielded TerminationBlock

(Sits on 5 cm ofinsulating material)

0.1 m

Shielded enclosurefor cable adapter

(Sits on 5 cm ofinsulating material)

CableAdapter

>1.5 m

Cableconnector

GPIB

RF Power Amplifier

RF SignalGenerator50 Ω 50 Ω

GPIB GPIB

50 Ω

Ferriteclamp

INOUT

50

DirectionalCoupler (40 dB)

FWDCPL

SpectrumAnalyzer / RFPower Meter

50 Ω

EMC Lab GPIBControl PC for

ALSE TestGPIB

The spectrum analyzer and directional coupler are used to monitorthe forward signal power to the antenna. At each frequency, thesignal generator output is adjusted to match the forward signalpower previously obtained during the calibration procedure.

R&S FSU-8SpectrumAnalyzer

(8 GHz BW)

GPIB

Balun

GPIB

Ferriteclamp

50Ω

Aquantia testequipment

EMC lab testequipment

24 ft. coaxialcable

ETSPI-102

Added 3 dBattenuators toimprove differentialport return loss

-3dB

-3dB

0.2 m

0.2 m 61 61

Split-ring

ferriteclamps(on 6"coax)

Split-ringferrite

clamps

Horn antenna aligned 10 cmoffset from DUT for testingfrom 1 GHz to 3 GHz

Test field strength is 50 V/mfrom 80 MHz to 3 GHz (CW).

Measured result scaled to100 V/m by adding 6 dB.

Bilog antenna aligned tomidpoint of cable span fortesting from 80 MHz to 1 GHz

Balun

The spectrum analyzerdetermines both the frequencyand power of the RF ingresssignal by monitoring themaximum peak signal (RFingress from the radiation) duringeach frequency sweep.

31 31

61 31

Cable adapter breaks out STP andSSQ 100Ω differential cablechannels to a pair of 50Ω SMA jacks

Split-ring ferrite clamps added to allcoaxial cables inside chamber tosuppress stray test fixture RFingress below 1 GHz and lower themeasurement noise floor.

Shielded enclosure has added coppertape on seams and EMI gaskets to

reduce stray ingress above 500 MHz

>2 m

3

11/8/2017

4 ALSE Automotive Cable RF Ingress Measurement Test Setup (80 MHz to 1 GHz)

Antenna

Shielded Cable (DUT)

Shielded Cable Adapter

24 ft Coaxial Cable

Copper Ground Plane Table

ETS PI-102

Shielded Termination Block

Split-ring Ferrite Clamps

4

11/8/2017

5 Test Setup • Measurement noise floor is the measured stray RF ingress from the stand-alone test fixture

– Measured under ALSE test conditions with the cable DUT disconnected from the cable adapter and all unconnected cable ports on the test fixture covered with copper tape

– Cable adapter enclosures have EMI gasket seals between the cable connectors and enclosure and copper tape on seams to reduce stray ingress above 500 MHz

– Added split-ring ferrite clamps to the exposed coaxial cables inside the chamber to reduce stray ingress coupling below 500 MHz; moderately effective below 200 MHz

• Spectrum analyzer operated in free running sweep mode; measures both frequency and level of maximum peak signal during each frequency sweep

– Proper measurement requires a spectrum analyzer measurement noise floor < -150 dBm/Hz to observe weak ingress signals, otherwise an additional pre-amplifier is necessary

• RF disturber field calibrated to 50 V/meter using a CW signal (no 80% AM)

– Reduced field strength to minimize harmonic distortion from the EMC lab RF amplifier

– Harmonic distortion can introduce measurement errors for this test setup by causing the spectrum analyzer to observe the wrong (harmonic) peak

– Test level a compromise between harmonic distortion and improved signal observability

– Used linear spaced frequency sweep step size(s) from ISO 11452-1 Table 2; dwell time set at 2 sec.5

11/8/2017

6

Port 2(Cable Adapter)100Ω Differential

Port 1(Spectrum Analyzer)50Ω Single-ended

ETSPI-102(Balun)

-3dB

-3dB

24 ft. Coaxial cable6 in Coaxial cables

ALSE Test Fixture

6

11/8/2017

7

7

11/8/2017

8

8

11/8/2017

9

9

11/8/2017

10

10

11/8/2017

11

Observations and Next Steps• Overall, the RF ingress in the STP cable with the H-MTD connectors was lower than the STP cable with the

HSD connectors

– The peak RF ingress in the STP cable with the H-MTD connectors (~1.5 mV rms) was lower than STPcable with the HSD connectors (~5 mV rms) over 80MHz to 1GHz

– The peak RF ingress in the STP cable with the H-MTD connectors (~3 mV rms) was lower than the level in the STP cable with the HSD connectors (~6 mV rms) over 1 GHz to 3 GHz

• There is a measurement discontinuity at 1 GHz because of the change in relative antenna position with the (slightly) different test setups for measurements above and below 1 GHz

• RF ingress measurements below 200 MHz may not be accurate because of a higher measurement noise floor from stray RF ingress into the test fixture cabling

– Some reduction in the measurement noise floor below 200 MHz obtained by adding split-ring ferrite clamps to the exposed coaxial cables inside the chamber

– Passing the 24 ft coaxial cable outside the chamber through a bulkhead connector at the chamber wall instead of a shielded aperture should provide some additional improvement

• Next steps– Extend measurements to 5 GHz; insertion loss notches reported in this range (possibly caused by helical

shield construction) may have important implications for radiated immunity

– Test on different cable types (e.g. SSQ) and cable spans with connector junctions

– Test on cable bundles 11