Steve, I appreciate your effort in being as transparent as possible without exposing too much information to your competition.  I think all of us who have invested in this development would like to share in the excitement.  Is your plan to get the hardware qualified first, so you can gear up the factory to make those initial deliveries to people like me with the software qualification to follow?  Or will both the hardware and software be certified as a unit?

So Steve, if one was planning a cockpit upgrade around your 540 deliveries, hypothetically of course, when would be a good time to plan on taking one’s airplane being down?  Feb?  Mar? Apr?

Tomorrow is the first day of the second month of 2013; I thought this would be a good time for all of us to get a heading check. Steve would you please provide us status where we stand  in the certification process? Did we get all of our flight testing completed? Have all the reports been written and submitted?  What’s left to do?

IFD540 Status Update (31 Jan 2013):

We are making terrific progress but we are still a few months away from being done.   There are a lot of ways to slice this up but in general, here is a breakdown of the various efforts underway to get the IFD540 certified and installed in customer airplanes:

Hardware Qualification:

We are still undergoing the official for-credit hardware qual of the units.  This consists of three sub-categories.  We have to conduct all this testing on the fully assembled and representative units.  Accounting for potential snags, I bet we have 8ish weeks ahead of us on this effort.

1.       Environmental Qual (DO-160) – this is progressing well.  This part of testing in which the unit has be to subjected to all kinds of environmental testing to include extreme heat, and cold, altitude testing, humidity testing, RF emissions testing, RF susceptibility testing, vibration testing, salt spray, dust, dot, dot, dot.  This testing gets conducted in a special environmental test lab and the test results get submitted as part of the FAA approval process.  We’ve been in that testing since early December and have a few more weeks of that effort ahead of us.

2.       Complex Electronics Qual (DO-254) – we have conducted our own not-for-credit testing to ensure we have no problems but have not yet started the formal qual testing yet.  When we do, it should take a few weeks until completion.  For the techies in the audience, Complex Programmable Logic Devices (CPLDs) is where a lot of the embedded logic is conducted and we just have to show nothing unexpected or nefarious happens in that hardware/firmware.

3.       Minimum Operational Performance Standards (MOPS) testing – so far so good but not quite done.   The FAA specifies minimum performance requirements for hardware (e.g. GPS, VHF, etc) and each one of those standards comes with pages (and sometimes pages and pages and pages) of tests that are specified and must be run and shown to meet published pass/fail criteria.

Software Qualification:

I think the simplest way to think of this is a two-part effort – Development and Formal Qualification.  Both are governed by an FAA-written software process called DO-178.    This is a requirements-based process that involves all aspects of software development and cert starting with top level product definition and decomposing that into smaller and smaller and more detailed and more detailed written requirements, then the software coding to embody those requirements and then a lot of different kinds of testing designed to validate all the requirements are correct and fully embodied by the code, and conversely, the code doesn’t have anything in there that isn’t governed by the requirements.  A related type of testing is to show that every possible path the code can take results in a known or deterministic end state.  This is man-years of effort for a product of this size and complexity.   It is also an iterative process in that inputs from flight test and the FAA and sometimes the customer base, can be cause for rework and then a lot of those steps must be re-performed to re-validate the changed code.

We don’t run this in a serial manner.  In other words, we don’t wait until all the code is done before we start the validation testing.  We are well into the formal qualification of the code but some parts of the code are still changing as we fly it a lot and find things that simply must be changed.  The closer we get to the cert finish line, the higher the bar is that a potential code change must clear before we put it in.

This effort is running in parallel with the hardware qualification.   It’s all relative but flight test is still finding a number of areas that need to, or we want to, change in order to deliver a truly kick-ass product.

Past experience says it’s wise to be conservative on estimating the impact of these changes on schedule so I would estimate we have a few months of work still ahead of us before we declare the code to be 100% done.

Flight Testing:

Of course I find this to be the most enjoyable part of the effort by a long shot.   We have two dedicated airplanes for flight test and two other aircraft that we periodically use when we need to perform special tests or temporarily ramp up flight test intensity.

One thing that has slowed us down over the last 5 weeks is that one of our test airplanes has been undergoing a major modification in which a lot of sensors and 3^rd party boxes that the IFD540 integrates have been installed.  We did this to ensure we had a truly representative environment for testing in case our lab configurations couldn’t adequately represent certain flight conditions or corner conditions.   We can also fly the airplane regularly “in the system” by filing IFR and flying in real airspace dealing with real flight scenarios day in and day out.   We get that airplane back this weekend and we’ll be back into full intensity flight testing.

One side note, while I would love to show off our stuff and let people take their own personal evaluation flights in these planes, we have created a self-imposed “no flights for outside people” rule and we’re staying disciplined on adhering to it until the system is “done”.  Sorry.  (I will make periodic posts of pix on my Twitter feed however)

FAA Involvement:

For better or worse, this one has us worried.   The crew that we work with in the FAA are all good guys trying to do a good job.   However, I’m sure no one will find it surprising to note that it is a big organization with lots of disparate parts and heavy workloads, all leading to a burdensome bureaucratic environment.   There are two specific areas that will result in real schedule uncertainty.   The first is that the STC project has been put into “sequencing”.   This is an FAA term that means the level of FAA involvement in the cert process is large enough that the regional office is not permitted to slip it in to their normal work flow.   Until it is out of sequencing, no one in the FAA can work the certification approvals and the big unknown is no one in the FAA can tell when it will pop out of sequencing, and which regional FAA office will perform the certification approval tasks.   The other event that is very disruptive to knowing with high confidence when the unit will be certified/approved is an FAA task called a Multi-Pilot System Useability Evaluation (MPSUE).   This is a FAA-run evaluation of the system in the lab and in flight in which they conduct a human factors analysis of the safety and certifiability of the unit.  For anyone following the R9 SynVis cert effort thread, you have some sense of the MPSUE saga that is likely to unfold.  And unlike in the case of SynVis which was an “abbreviated” MPSUE, the FAA has informed us that the IFD540 will receive a “full” MPSUE.   That is likely to involve at least 10 FAA flight test pilots and flight test engineers and human factors folks (and maybe more) and coordinating all those schedules and viewpoints is a non-trivial task.  It is a safe assumption to say something will need to be changed in the code as a result of that FAA input and therefore we can’t declare we’re 100% code complete until the MPSUE has been conducted and any changes implemented.  And of course, they can’t even start scheduling the MPSUE until the project comes out of sequencing.

The rest of the FAA involvement is just par-for-the-course types of tasks and are in the noise when compared to the above two challenges.

I view this area as the single biggest schedule wild card, by far.

Launch Readiness/Pilot Program:

Let’s try to end on a high note.   Despite all that work noted above, we are also running a parallel launch readiness effort so we’re as ready for prime time as we can be.   This has all kinds of sub-parts to it like ramping up our production line at the factory, having ship kits all ready to go, having simulators and user documentation ready to go (we don’t plan to release those until just before certification) and a lot of other typical product launch types of tasks.

One big part of our plan is to run a “Pilot Program” that will start the day after we receive TSO and STC approval/certification.  This “Pilot Program” is intended to run long enough (I’d guess 4-10 weeks) for us to be confident there aren’t any must-fix gotchas before we open the flood gates to full-rate production and installations.   I am planning on installing TSO/STC units in 20-30 customer aircraft and collecting installation and flight behavior input as required.    I have many more details on the Pilot Program that we’ll share in the coming next few weeks in which I explain the types of aircraft configurations and patterns of flying we’re looking for and seeking volunteers to participate in that program.  I wouldn’t call it a “Beta Test” program since we’ll be fully certified by that time (and by the way, we have no plans to install pre-certified gear in people’s airplanes) but it will definitely be in the veins of an early adopter kind of phase.   That explanation and solicitation for volunteers will come in the form of an email to all those customers who have already signed up for one or more IFD540s but if you want to get a headstart on throwing your name into that hat, don’t hesitate to send an email to sjacobson@avidyne.com.

IFD540 Status Update (7 Mar 2013):

We've moved 6 weeks closer to cert since the last big update.   I'll use the same way to slice this up as the late Jan update, albeit in an abbreviated manner:

Hardware Qualification:

We are still undergoing the official for-credit hardware qual of the units.  We had one glitch with the VHF performance at low voltage testing and have subsequently tweaked the design to address that problem and are back in qual - that put a few week dent in the hardware qual schedule. 

1.       Environmental Qual (DO-160) – I didn't really mention this in the late Jan update but we front end loaded the tougher and riskier tests so we minimize the likelihood of late-in-the-game surprises.  As I noted above, the tests uncovered a design issue that we found and fixed and are back at it now.  I'd be willing to bet we have more of those ahead of us this spring.

2.       Complex Electronics Qual (DO-254) – we have conducted our own not-for-credit testing to ensure we have no problems but have still not yet started the formal qual testing.  When we do, it should take a few weeks until completion.  For the techies in the audience, Complex Programmable Logic Devices (CPLDs) is where a lot of the embedded logic is conducted and we just have to show nothing unexpected or nefarious happens in that hardware/firmware.

3.       Minimum Operational Performance Standards (MOPS) testing – so far so good still but not quite done.   The FAA specifies minimum performance requirements for hardware (e.g. GPS, VHF, etc) and each one of those standards comes with pages (and sometimes pages and pages and pages) of tests that are specified and must be run and shown to meet published pass/fail criteria.

Software Qualification:

We have a very specific punch list of items and we're systematically checking them off one-by-one.  The pace is definitely accelerating but there is still a small mountain of work ahead of us.  No show stoppers and we've hired up to augment our software testing crew.  Many of the formal software tests are written and more are done every day (there are almost one thousand tests to create/review/run/document).

I left this paragraph in from the Jan update since it's still relevant - Past experience says it’s wise to be conservative on estimating the impact of these changes on schedule so I would estimate we have a few months of work still ahead of us before we declare the code to be 100% done.

Flight Testing:

The airplane that I'm principally flying is a workhorse.  We upgraded it last week to ops test more of the systems we interface to in a real world situation.  The result is that the plane has a crazy antenna farm on the roof and I feel like Lily Tomlin in front of her switchboard as I change connections real-time in flight.  

I do have a minor fuel leak in a wing root which will require me to put the plane down for a little bit but aside from that, it's a reliable steed.

The FL-based test aircraft is equally as busy and that gets flown all the time by our Chief Pilot and our CEO.   We're bringing another test bird up in FL shortly.

FAA Involvement:

For better or worse, this one still has us worried.   We are still in sequencing and it's still not clear how the sequester-generated periodic FAA furloughs will impact the program.

Launch Readiness/Pilot Program:

I had a huge response in the solicitation email to all the pre-buy customers a few weeks back asking for volunteers to participate.  I haven't spent enough time pouring through all the data yet to down-select the best configured airplanes operated by eager to help owners.   I bet there are still a few weeks before that settles out.

Steve, Thank you for the update....

Steve, I'm not sure I share your optimism but it doesn’t matter;  it’s going to be what it’s going to be and its out of our control.  I know how its ilmost impossible to predict how long the FAA will take.  What I would be interested in hearing is the feedback from the “pilot program”.  Let’s call it a quick look from a customer’s point of view.  Would that be possible?

Steve, I think you will avoid issues down the road (hazardous misleading data) if you just remove the heading all together when you pan the map.  Then once you leave the pan mode and the ship's solution is centered on the display, put it back.  I'm assuming the CDI still indicates cross track error in pan mode and that’s what the pilot should be flying anyway.  Everything else is situational awareness.

Just my opinion

Are we seriously looking at October as the likely date for certification?  i was told when i purchased that the unit should be certified by June, I know that you've had to deal with the effects of sequestration but how has that caused a delay of 4 months?  I sold my Garmin 530W while I could still get something for it and am now renting one from a local avionics shop to use until the IFD 540 is available.  Any news on who or when the pilot program people will be selected and potentially have radios?   Thank you for your support!