Sunday, January 28, 2018

PP792 Mark 2 Satcom Advanced Features

PP792 defines the form and fit characteristics of the Mark II Aviation Ku-band and Ka-band Satellite Communication (satcom) System, intended for installation in all types of commercial air transport aircraft. The satcom system described in PP792 represents the next evolution from the Mark I system defined in ARINC Characteristic 791, Part 1.

Significant advancements include:
  • Single, standard installation, no customization, fully interchangeable
  • Smaller: no KRFU, Virtual APM
  • Lighter: smaller connectors, fewer wires, no hyper-critical wiring, coax, or waveguide
  • Flexible lug pattern accommodates two apertures, each up to 42"
  • High power supply (2000 Watts) for large, solid-state, phased-array antennas
  • Integrated Position, Attitude and Heading sensor
  • High performance IF interface supporting NGSO networks

ARINC 791 Part 1 offers a flexible approach to accommodating a wide range of antenna and radio design approaches, with the downside of complicated and critical wiring.

ARINC 791 Part 1 Satcom Installation

Embracing the emergence of highly integrated flat-panel, phased-array antennas, PP792 strips away the ARINC 791 satcom installation to the bare bones.  The KRFU is removed. The APM becomes virtual.  The KANDU is reduced to a power supply: the KPSU.  The connectors are smaller, the wiring is a greatly simplified.

PP792 Satcom System

The standard PP792 provision is fully complete, there are no manufacturer-supplied interconnections, the equipment, including the antenna, is bolt-on, plug and play.  The critical RF paths are gone, leaving a simple IF interface for signal processing.

PP792 Satcom Installation
Interchangeability is a hallmark feature of ARINC standard installations.

PP792 evolved over the last three years: from discovery, discussion, debate and decision.  The cycle repeated when new information surfaced, based on changing assumptions, dead-ends, or new objectives.  With about five weeks left for final touches, here is a deep dive on what makes PP792 so great.

Flat-Panel Antenna Technologies

Aero antennas are designed for low-profile, to minimize drag.  Shaping, installation location, aspect ratio, and frontal area can contribute as much as 1% fuel burn increases.

Mobile, high-gain, Ku/Ka antennas typically use a horn array (for gain/directivity) on an azimuth/elevation pedestal. The design naturally leads to a relatively small aperture, equivalent to about an 18" parabolic dish.

Horn array beam-pattern can create skew angle issues in tropical regions.

Gogo is the first provider to field a flat-panel phased array, 2Ku (the ThinKom 3030 VICTS antenna). The tandem, 30" Transmit and Receive apertures enable a very-low-profile, high-aspect-ratio radome.  However, ARINC 791 Part 1 lugs are not designed to accommodate a second tandem aperture.  2Ku must be installed in a taller assembly than should be necessary when using ARINC 791 lugs.

NGSO operations require frequent handoffs (OneWeb is about every three minutes). Some phased-array antennas can create a second, independent, receive beam that can be used to acquire a rising satellite, while also tracking a setting satellite. The handoff can be seamless, as the receiver is already in sync, and the transmitter can rapidly acquire the rising satellite without mechanical-drive delay.

Having a second (or third!) receive beam can facilitate access to a wide-beam television service that is coupled to a spot-beam Internet access service.

Flat panel antennas suffer scan loss, in that their effective aperture shrinks as the target satellite moves towards the antenna horizon.

A 42" aperture should operate acceptably to a GEO satellite as low as five degrees above the horizon.  PP792 is designed to support two 42" apertures.

Reports from ThinKom relate success with Gogo 2Ku operating below 10 degrees elevation.

The good news: the larger the aperture, the higher the gain, the better the efficiency.  The better the efficiency, the higher the data rate, the lower the cost.

Passenger sessions are consuming more and more data as time goes by. Larger apertures are the best pathway to greatest efficiency and capacity.

Regional aircraft, and those solely using an NGSO constellation, may favor much smaller apertures. Some antenna technologies may utilize a single assembly for both transmit and receive.  These features will be explored once the initial version of PP792 is released, with a likely alternative lug layout for smaller airplanes.


The subcommittee has spent years contemplating a new lug arrangement that would be most accommodating.  Lugs are the part that is attached to the airplane.  Fittings are the part that attaches an adapter plate or other structure that takes the radome, the skirt, and all the outside antenna equipment.  Fittings are designed to take load on specific axis, and to "slip" on other axis.  The grip/slip arrangement is specifically engineering to apply loads surgically, and to accommodate any single failure. In summary all lugs take Z-loads, 3.4 take X-Loads, 1/5 or 2/6 take Y-loads.

A breakthrough from Chris Schaupmann and Markus Altmann, Airbus, revealed a new lug arrangement, with many possible configurations, based on a clever use of supporting structure.

It must be noted that Airbus has not made a commitment to install PP792 installation; that the proposals from Airbus engineers represent technical alternatives.  Boeing is similarly studying PP792 lug layouts, and makes no claim for installation at this time.

Aircraft structures include frames that form concentric rings about the fuselage.  These frames are designed to support the loads. Stringers run longitudinally between frames as part of the supporting structure for the skin. Only the frame offers a point for antenna lugs to connect.

If the lug sits more than about an inch from a frame, a supporting brace, an "intercostal", is required.  The intercostal spans between the adjacent frames and passes the loads to them.  An intercostal is riveted to the skin beyond the point where the fitting attachment penetrates.  There is much more to this than what is described here, particularly skin doublers for penetrations themselves, brackets/clips, bearings, or hardware.

One hardware point, we have agreed to upsize the bolt that attaches lug to fitting (through a spherical bearing) from 3/8" to 1/2" diameter.   In some cases, we had reports we were approaching limits for 3/8", and felt 1/2" would offer better headroom.  For reasons we all appreciate, no one wants to have to be concerned about bolts.

Frame spacing ranges from 20" to 26".  We have already compromised from ARINC 791 Part 1 to express locations assuming 25" spacing (we originally started with 21", 23" and 25" designs), and that other arrangements may need intercostal.

Using a base of 25" frame spacing, consider six lugs arranged 2 x 2 x 2.

The center lugs are placed over a reference frame, such that a simple bracket is used to pass the load between lug and frame.

The frame forward and aft are each 25" away.  The next is 50" away.  By provisioning an intercostal between the first and second frames away from the reference, a lug mounting position can be accommodated anywhere from one frame to the next, by reusing the holes already provided for when installing the intercostal.  The maximum span between lugs is 100".

A significant feature of this approach is that the rivets used to attach the skin to the intercostal can be "drilled out" where needed to attach the lug - basically without any additional structural cost.

While any position is possible, Chris and Markus propose an even more clever 1/3, 2/3 approach for the forward and aft intermediate position.  This allows for more unique combinations. 
In summary, while the middle two lugs are fixed over a reference frame, the forward and aft two lugs can be put into one of three positions, independently forward and aft (9 configurations).

Proving this point, a host of different aperture sizes up to 42" were "installed" showing how each lug location is positioned favorably.

Final refinement is underway on the region set aside for connector penetrations.  Latest proposals are shrinking the bulkhead zone a bit.  While the connector penetration zones do not overlap (791 Vs 792), the lugs are complementary (both can be provisioned in one airframe).

Integration under the Radome

A significant goal for ARINC 791 was to minimize the equipment under the radome, in favor of equipment inside the fuselage.  Solid-state phased array antennas distribute the high power amplifier to each individual element.  Taking that feature into account yields a system architecture with no discernable High Power Amplifier or KRFU.

Electronic phased-array antennas have no moving parts, and thus no motor drives or sensors. ARINC 791 provides for a 53 conductor cable to allow mounting the beam steering electronics inside the KANDU that is no longer needed with fully electronic antennas.

With all the RF components outside, the Modman IF interface comes through the bulkhead.

NGSO Tx IF requirements are driving a slightly higher performing (lightweight) coax, but the good news is the same coax can be used for any PP792 installation.  ARINC 792 Part 1 and Part 2 anticipate a performance on the IF coax at a breakpoint of 1450 MHz.

Rockwell Collins requested that the transmit coax breakpoint be raised to 4000 MHz.  This presents a new challenge for coax suppliers, where the goal is to retain the lowest weight.  Initial analysis revealed a weight gain of 1-2 lbs for 100 feet of coax.  While this seems like a lot, in fact that is a minimal increase compared to other options, and that many installations are much less than 100 feet (Modman to bulkhead penetration).

The receive coax may not need the higher performance, and may remain with a slightly lighter variant (final decision pending).

The difference in the coax is only necessary for OneWeb, at this time.  The subcommittee felt that other emerging satellite constellations may well also mandate the much higher IF bandwidth.  The challenge is moving away from DVB-S2X single carrier operation to multi-carrier LTE emissions.

DVB-S2X carrier bandwidth is currently not greater than about 417 MHz in any constellation.  Using a switchable Local Oscillator (LO) allows a minimal IF bandwidth (950-1450 Mhz) to be applied appropriately to any specific carrier frequency or polarization.

LTE offers an opportunity for 100% frequency reuse by applying many, many small carriers in a managed distributed manner across each beam. In this context, the full range of RF bandwidth must be preserved in IF.  OneWeb has about 4 GHz of available RF.  While only 250 MHz may appear in any given beam, the subcarriers may be anywhere within the RF bandwidth.

For those flat panel antennas needing an HPA, the subcommittee appeals to the marketplace to offer a cost-competitive fail-op HPA.  By and large, satcom should have fault-tolerance where possible given the passenger impact if service is lost due to a single failure.
 Teledyne Microwave

Frequency Reference

ARINC 791 Part 1 did not profess a location for the frequency reference.

ARINC 791 Part 2 includes specific list of frequency references.  These frequencies are negotiated using a special antenna/modem interface protocol (A791-AMIP).

General industry best-practice is to put the reference into the "indoor" unit, where temp variation may be minimized. Vibration may also be minimized in a typical industrial setting.  Aero installations are a bit more challenging, as the crown temps can get very hot. Qualification standards are more stringent for outside antenna equipment.  For these reasons, the Modman is the most suitable location for the frequency reference.  Furthermore, the Modman can inject the reference readily into the IF coax, as is already provided for.

Rockwell Collins requested an additional clock frequency, which will be accommodated in supplement 2 to ARINC 791 Part 2.

GPS disciplined frequency reference modules offer an opportunity to utilize less stringent absolute crystal references, with smaller enclosures and higher accuracy.  This attribute is welcomed when squeezing the reference crystal into the Modman.

An area still under analysis is the benefit from more stringent phase noise standards given the movement to higher and higher modulation encoding.

Airplane Personality Module (APM)

ARINC 791 provides for an APM to capture specific settings for the system related to the installation and configuration.  Support is provided for a wiring harness and a flange mount.  The signals are provided by the Modman, specific to each supplier.  The APM itself is a "memory" device mounted onto an assembly with a connector.

ARINC 791 resolved a great difficulty with APM, in general, as to managing its configuration, itself.  Traditionally, the APM was provided programmed by the supplier for each airplane, and bearing a unique part number to reflect its association.  Managing APM parts becomes significantly complex, while their reliability has not been a concern. ARINC 791 provides for a generic hardware APM that is programmed on the airplane as a part of installation.  Backup of the APM configuration (for APM replacement) offers an alternative to full installation procedures, when replacing the APM.

PP792 offers an alternative "virtual" APM. The virtual APM takes advantage of the distributed nature of the satcom system. The APM information is stored along with a signature of connected equipment (already communicated via SNMP or otherwise) in the Modman, the KPSU, and in the OAE. Removal and replacement of any individual LRU can lead to an automated process where the retained LRUs download the configuration as part of a power up procedure.  The LRUs can negotiate amongst themselves by comparing signatures of connected equipment in a deterministic manner to arrive at the suitable configuration.  As most, if not all, Ku/Ka satcom systems carry a cell modem for system maintenance, the cell modem can facilitate secure access to a remote server that can act as both an archive and arbitrator.  The cell modem, for example, has a unique SIM card, too.

The introduction of the virtual APM fulfills removal of one more LRU (the "physical" APM), its associated wiring, and mounting.  However, PP792 retains an optional physical APM, as provided for in ARINC 791.  Where ARINC 791 provisions provide the wire harness and connector to the APM, PP792 provisions provide space and mounting provisions only - the wire harness must be provided by the supplier and the connection to the ARINC 600 Modman connector completed as part of the optional work.

PP792 also offers a "dongle" style APM, where the APM is attached to a fixed receptacle.  The receptacle could be mounted on the Modman front panel, with a leash, but that was not viewed favorably.  This style APM would come with the harness and receptacle, just like the ARINC 719 Part 1 APM.

Position, Attitude, Heading (PAH)

ARINC 791 and PP792 provide for a connection to the onboard Inertial Reference System (IRS), using an ARINC 429 serial data receiver.  There has been continual reluctance to connect the IRS to a non-essential system.  While multiple IRS outputs are provided to segregate fault propagation as one option, most modern airplanes use data concentrators to bridge ARINC 429 to high speed time-triggered Ethernet backbone data networks.  The data concentrators add latency and jitter to the ARINC 429 data outputs.

IRS heading accuracy is advertised as 0.4 degrees, while the objective for beam steering is only 0.2 degrees (and there are other errors that take a chunk from the budget).  Utility of closed-loop steering based on receive signal strength is best applied for periodic calibration, not necessarily high-precision tracking. Furthermore, the IRS is installed close-by the aircraft center of gravity, or conversely, far from the satcom antenna.  The aircraft structure, between antenna and IRS, is elastic: it bends, twists, and distorts under pressure and load. The correlation between IRS and antenna boresight geometry can change with relative significance and no means to detect.

Some phased-array antenna designs use novel beam-steering methods. These may utilize rudimentary beam pointing coupled with inherent beam peaking based on a distributed signal processing, by aligning the phase of each element independently.  For these designs, the existing IRS interface is sufficient.  A better source of PAH would translate directly into more accurate beam steering for other antennas that rely primarily on open-loop tracking.

PP792 promotes installation of a PAH within the outside antenna equipment, under the radome. The combination of differential GNSS antennas separated by the extent of the radome, coupled with MEMS accelerometers and sophisticated data processing, provides an integral PAH with greater accuracy than the onboard IRS, and with absolute relevance.  Commercial issues have dogged this initiative, but price points below $5,000 are evident today, and below $1,000 in the coming years.

2000 Watts

ARINC 791 KRFU is the expected host for the HPA.  Generating 40 Watts of output power might take 500 Watts of electrical power (or 460 Watts of heat) (speaking simply). Fortunately, in most cases the KRFU was cooled by forced air.

By moving the HPA under the radome, the electrical power delivered outside is increased.

Much more significant is the need to power large solid-state phased arrays.  In particular, Phasor Solutions has negotiated two 1000 Watt power supplies. Unlike a passive radiating structure, these technologies power is applied both for transmit and receive, as a function of the number of integrated elements.

The number of elements scales with area and by wavelength, which is one reason Ku band is more suitable than Ka band for initial developments.  The lower elemental density is simpler to layout (pack the elements onto the circuit board); is easier to cool, as the heat density is similarly reduced; and allows more possibility for overlapping Tx and Rx arrays.

There are a number of tradeoffs with solid-state phased arrays, but notably they are cheap to manufacture once the foundry is cast.  It is likely their receive efficiency suffers by elevated noise floor, and that several suppliers, also including Rockwell Collins and others, are continually refining and developing better and better performance.

The enormous power budget has created a need for the PP792 KPSU to host a three-phase AC power supply.  A five-conductor, size 15 connector was added to the KPSU for power and bonding, separate from the common control connector used by both KANDU and KPSU.

The subcommittee has had to agree upon upper voltage limits that are adequate for the power supply and still safe to humans.

PP792 KPSU Power Characteristics

15-97 Standard Bulkhead Connector

ARINC 791 Part 1 provided for a 57-conductor control connector and an 8 wire power connector through the bulkhead.  PP792 phased array antennas are commonly formulated in a transmit assembly and a receive assembly.  Chris Schaupmann, Airbus, suggested a common connector be chosen to combine power and control wiring for each assembly.  Without spare contacts, an interface supporting one Ethernet data bus, 4 individual control conductors, and 4 power conductors was crafted which can be accommodated by a small 15-97 connector.  Using two size 15 connectors reduces the "decompression/inflation under radome" failure significantly from the size 19 ARINC 791 connector (the largest penetration is presumed to "blow-out", and size 15 is much smaller than size 19).
ARINC 791 Part 1 Bulkhead Control Connector (Size 19)

ARINC 791 Part 1 KANDU OAE Control Connector (Size 21)

PP792 KPSU and OAE Bulkhead Connector, Tx and Rx (size 15)

The KANDU used a size 21 connector and a special 4x Quadrax connector.  The KPSU does not use a quadrax insert, and utilizes the two 15-97 connectors discussed already.

By using the same pinout on the KPSU, a common interconnect cable can be used for both the Tx and the Rx connections.  Different keys inherent in the 15-97 connector design prevent cross wiring.  Bulkhead penetration connectors are an expensive component.  Use of a common connector design offers greatest economy and ease of manufacturing.

Gate-to-Gate (G2G)

Airline passengers are allowed the use of personal devices without regard, except they must be put into airplane mode typically while the doors are closed (on departure) until landing (on arrival).  While the Wi-Fi or seatback entertainment functions can be provided without interruption, many passenger connectivity solutions would only function while the airplane was above 10,000 feet.  In fact, for Line-of-Sight (LoS) networks, such as Gogo ATG, the minimum altitude plays directly into ground station density.

Satcom operation on the ground faces a few challenges.  A license must allow operation on the ground, or at low altitude.  This can be an issue based on terrestrial usage in the vicinity of the airport.  The airplane must be located at a point where the satellite is not blocked by buildings or other structures.  The onboard PAH must be available - noting that IRS is not left on indiscriminately. Power must be applied without interruption, as system restarts cause havoc to service utility. But beyond all of that, the one aspect that strikes fear into each subcommittee member is heat.  Moisture, particularly condensing water, is manageable. But heat, what do you do with it?

Radomes have breathable membranes that allow for air exchange.  Aircraft in flight will naturally bring some air through.  Adiabatic heating along the leading surfaces warms the radome, offsetting the cold ambient temperatures (in fact, with no equipment powered on, the radome temp does not go below about -15 deg C). There is no facility to embrace the fuselage itself as a heat sink, nor is it permitted.  In any case, operation inflight has not been viewed as problematic.

Operation on the ground removes the airflow.

Solar heating is a significant factor (1000 Watts per sq meter). Light colored radomes offer reflective benefits.

Along with the benefits of installing the HPA outside comes the additional challenge for dealing with its great inefficiency (80% or more goes to heat).  Solid state antennas are themselves a source of heat by their very nature.  PP792 delivers up to 2000 Watts of power to the OAE, and the majority of it might turn into heat under the radome.

Radome suppliers are encouraged to offer some form of forced air exchange under the radome, in a way that does not interfere with the required transmissivity, that accounts for dirt and ice, is quiet, and highly reliable.

Stay tuned!

Peter Lemme

peter @
Follow me on twitter: @Satcom_Guru
Copyright 2018 All Rights Reserved

Peter Lemme has been a leader in avionics engineering for 35 years. He offers independent consulting services largely focused on avionics and L, Ku, and Ka band satellite communications to aircraft. Peter chairs the SAE-ITC AEEC Ku/Ka-band satcom subcommittee, developing ARINC 791 and 792 characteristics and contributes to the Network Infrastructure and Interfaces (NIS) subcommittee developing Project Paper 848, standard for Media Independent Secure Offboard Network.

Peter was Boeing avionics supervisor for 767 and 747-400 data link recording, data link reporting, and satellite communications. He was an FAA designated engineering representative (DER) for ACARS, satellite communications, DFDAU, DFDR, ACMS and printers. Peter was lead engineer for Thrust Management System (757, 767, 747-400), also supervisor for satellite communications for 777, and was manager of terminal-area projects (GLS, MLS, enhanced vision).

An instrument-rated private pilot, single engine land and sea, Peter has enjoyed perspectives from both operating and designing airplanes. Hundreds of hours of flight test analysis and thousands of hours in simulators have given him an appreciation for the many aspects that drive aviation; whether tandem complexity, policy, human, or technical; and the difficulties and challenges to achieving success

No comments:

Post a Comment