Odds and Ends: CSeries EIS rescheduled, new deal announced; Airbus seeks unprecedented ETOPS rule

New CSeries EIS: Bombardier today announced a new delay in the entry-into-service of its CSeries, confirming what we reported Monday. BBD now says EIs will be in the “second half” of 2015, compared with the September 2014 planned EIS. We forecast a 6-9 month delay, so this now is a 9-15 month additional delay.

BBD’s statement: “The CS100 aircraft’s entry-into-service is now scheduled for the second half of 2015 and will be followed by the CS300 aircraft’s entry-into-service approximately six months afterwards.

“We are taking the required time to ensure a flawless entry-into-service. We are very pleased that no major design changes have been identified, this gives us confidence that we will meet our performance targets,” said Mike Arcamone, President, Bombardier Commercial Aircraft. “While the process has taken more time than we had expected, our suppliers are aligned with the program’s schedule and together, we will continue to work closely to move the program steadily forward. With the first flight of flight test vehicle 2 (FTV2) successfully completed on January 3, 2014, the CSeries aircraft program will continue to gain traction over the coming months.”

Below is what we published Monday.

BBD EMB EIS

We’re now looking at an 18-27 month delay for the originally planned EIS of December 2013.

Update: BBD tells us the the “software maturity” is behind the rescheduling. Basically, this means that all the various software systems have to completely and correctly talk to each other and avoid the issues that “have affected our competitors.” BBD declined to name the specific affected systems nor the suppliers, instead saying that this is a Bombardier “team” project and that the “team” is on board with the new timeline.

In addition, Bombardier and a Middle East airline announced an order for the CSeries.

Airbus seeks big ETOPS OK: The Wall Street Journal reports that Airbus is seeking regulatory approval for a 420 minute ETOPS for its A350. (Subscription may be required.) This means the airplane could fly seven hours on one engine if the other one fails. It opens trans-polar and trans-Antarctica route opportunities.

The Boeing 787 is restricted to 180 minute ETOPS. The Boeing 777 has 330 minute ETOPS.

Via Jon Ostrower, here’s an example of what works beyond 330 minute ETOPS.

Meanwhile, Boeing has begun high speed wind tunnel testing for the 777X.

Airbus v Boeing: The Seattle Times published a good graphic of the Airbus v Boeing order outcome for 2013.

33 Comments on “Odds and Ends: CSeries EIS rescheduled, new deal announced; Airbus seeks unprecedented ETOPS rule

  1. Hi Scott, referenced the Seattle Times link graphic, rather than the link itself. The link is here

  2. Regardless of the monthly stated production figures, Airbus is at 41 a month and Boeing is at 36.

    Airbus it pushing up through 56% of the real maker (deliveries not orders, orders are closer to 60%).

    That means Boeing is down to 44% of the actual market. Perilously close to the number you begin to get swamped by the economics of production scale. This is reflected in the number of airlines that have gone dual Airbus/Boeing and or Boeing has lost entirely.

    Airbus simply had to put new engines and winglets on the A320 to get a huge boost.

    Boeing has had to do major changes on the 737 which is simply outdated as it originally had tube engines. Nothing you can do with that setup other than do your best with an outdated configuration.

    If the GTF come through in spades then the 737 is in even deeper trouble as there would be no answer for that other than all new and there is nothing in the quiver (let alone the upper corporate will)

    • Airbus delivers at 11 months a year, hence 41 x 11 = 451.
      Boeing delivers at 12 months a year hence 12 x 36 = 432.
      (51% vs 49%)
      I believe Boeing is committed to 42 a month beginning in March, 2014 then 47 in 2017. I am sure AIrbus will follow suit as both have enourmos backlogs.

      • Well, Mike what ever your math reality trumps it 😉

        Airbus delivered 493 A320 in 2013
        that is 41.1 frames per 12 month year
        and is 44.8 frames per 11 month year

        Boeing delivered 440 737NG in 2013
        that is 36.6 frames per 12 month year
        and is 40.0 frames per 11 month year ( just for the fun of it )

        I would take announced rates with a bit of salt.
        Boeing allegedly produces the 787 @ 10/month now.
        but delivery of _recent frames_ is at ~5.5 frames per month on average
        and 1.5..2 reworked older frames. ( H2 2013, 2month grouped is : 16 17 15 16 )

        • I stand corrected! So 53/47% (493/933).
          ;0)
          Regardless, reasonably close with massive backlogs for both. Both will increase production methinks.

        • Uwe, out of the 65 787 frames that Boeing delivered in 2013 (over only 8 months due to the grounding), only 9 of them needed any rework. Therefore, of the 8.125 frames per month that Boeing delivered, only 1.125 per month were reworked, not 1.5 to 2. This means that Boeing delivered 7 “recent” frames per month in 2013, not ~5.5. So much for reality, right 😉

          Averaging the deliveries over 12 months gives 5.417 frames per month (close to your ~5.5) of which only 0.75 frames per month required any sort of rework. For this case your break from reality is only partial.

        • Hello MIke,
          plotting delivery dates over LN ( http://home.foni.net/~kuttersegeln/aerostuff/787deliveries.png ) you can see that there are a lot of
          stragglers beyond the Low LN reworked frames.
          Then with 65 delivered over 9 month I get an average of 7.2 frames.
          You may not like my numbers but they certainly are not “off”.

          When Boeing is able to keep deliveries (per LN order) in a tighter band
          they will have achieved something.
          There will always remain some out of order deliveries, obviously.

        • Uwe, I think you are getting too hung up over the delivery order. Customer requirements and ability to pay are huge factors. The firing order is determined well in advance and certain customers will find that their plans or cash flow situation changes on shorter timescales. Assembly loading dates and/or roll-out dates from the FAL give a better indication of production flow.

    • Hi Uwe,
      The values from AVITAS are probably accurate for resells, but not for the actual price received by the manufacturers. If you look at Airbus’s revenues over the first nine months (27.6 bn euros) at a 1.35 exchange rate it’s already pretty close to the 38bn USD on the chart. I haven’t checked the numbers for Boeing but I would think it’d be fairly similar.

  3. I think this is what we all expected for the BBD CS-100 EIS. I still think something isn’t right with flight testing. Maybe it is not meeting expected performance and fuel burn? Maybe the P&W GTF is the problem?

    I think Airbus is pushing the ETOPS envelope to far with wanting a 420 minute certification right out of the box, well before the A-350 matures in airline service. As Scott pointed out the B-777 has a 330 ETOPS, and the B-777 is the most reliable large twin flying.

    • BBD tells us the the “software maturity” is behind the rescheduling. Basically, this means that all the various software systems have to completely and correctly talk to each other and avoid the issues that “have affected our competitors.” BBD declined to name the specific affected systems nor the suppliers, instead saying that this is a Bombardier “team” project and that the “team” is on board with the new timeline.

      • Software is what drives and controls various systems, right? Yes, I understand the software needs to talk to other components and systems. But it can also make a system or component perform at less than optimal.

        • Performance shortfalls are not showstoppers for doing testflights.
          Spastic/autistic behaviour of the software invariably is.

      • FBW software for all systems on board exist for nearly 30 years now on military fighters which are by design very “unstable” and over 20 years on A 320.. Scott, can you explain why that old type of software cannot do the job for a commercial aircraft ..

    • Topboom, how do you come up with that assertion that Airbus wants 420 minutes out of the box. The article states clearly that the A350 , “is designed to eventually serve super-long-haul overwater or polar routes that take it up to seven hours away from the nearest diversion airport.” It also states that it is a long term goal, which apparently was not revealed with great fanfare when they submitted their documentation for this years ago. Not only does it not say anything about immediate 420 minute ETOPS but it states quite clearly that it is a long term goal.

      I admit that that the thought of flying on one engine for 7 hours, over what would most likely be rather unforgiving terrain, does not really make me scream for joy. But then again, 3 hours in an aircraft with any sort of problem isn’t exactly what I would call an occasion to be greeted with enthusiasm either.

      So, what’s an extra 4 hours among friends?! Right?

    • No idea what they want to investigate.
      The ship is said to be back in servide.
      Going over just the battery with a fine comb won’t reveal much.
      ( but it certainly looks proactive and busy )

    • I was never satisfied that the problems had been completely solved. But Boeing has managed the crisis remarkably well. We can expect ‘incidents” like this to occur once in a while, but with no grave consequences. As far as I can tell the aircraft is safe. If it was not the aircraft would be grounded by the FAA.

      The Boeing engineers did a fantastic job to solve an extremely difficult problem. They also did a remarkable job repairing what many considered a scrapped aircraft. Boeing’s reputation was at stake and they faced the challenge graciously and quietly. Only one year ago the situation seemed desperate. The least we will talk about the 787 the better we will understand what is going on. For emotional debates have a tendency to clog our minds.

  4. I expected the first half of 2015, not the second half. So the flight test campaign will last considerably longer than the original twelve month period. It will take at least one and a half year, possibly even two years, to certify the CS100.

    Obviously, Bombardier still has a lot to learn in regards to software integration on a large scale that a modern aircraft like the CSeries calls for. What they are going through is not due to bad planning or poor execution, but to a lack of experience with the hugely complex systems that we find on modern aircraft.

    I thought they had learned what they had to learn with the CRJ1000 when the programme was delayed nine months or so because of similar software problems. But this was just for a single system, the Rudder FBW. But the CSeries is FBW throughout. It’s not one system, it’s ALL the systems that are software driven. And on top of this those systems have to talk to each other harmoniously and without a glitch.

    What will come out of this is that CIASTA can mitigate those problems but cannot prevent them. I don’t think BBD could have done much better than they did so far, except perhaps be more careful which suppliers they want to work with.

  5. “software maturity” is nicer to say than “the thing is way more complex than anything we ever build and we can’t handle it”.
    I would assume that no sane customer will buy the aircraft before it achieves EIS. Fortunately, not all people with money in aviation are sane.

    • Yes, by that criterion most of the 787 customers were not sane, either — not to mention 707 customers, or A380 customers, or …

      • IMHO the madness associated with the 787 is a rather exceptional thing.
        Never before was a nonconsumer hard product successfully hyped beyond any reason. And my guess is we won’t see it again for some time.

  6. @Schorsch
    Who are you quoting when you wrote: “the thing is way more complex than anything we ever build and we can’t handle it”.
    Is it a Bombardier employee or one of their suppliers?

    • IMU Schorsch translated “software maturity” ( in corporate speak ) into “real speak”. Thus you only need the source for “lack of software maturity” and you’ve got that already.

  7. ETOPS rules now apply to all long-range aircraft, quads as well as twins. It’s more a matter of systems redundancy and reliability, not the number of engines. For example, one ETOPS requirement is lower lobe fire suppression, so 7-hr ETOPS would require larger lower lobe fire bottles, no matter how many engines there were.

    The wind tunnel story is a non-event. 777X design variations have probably been in and out of high- and low-speed tunnels for months if not longer. Otherwise how would Boeing have arrived at an offerable configuration?

  8. Real 787 production rates for 2013

    After taking a detailed look at the assembly loading and roll-out dates available at:
    https://docs.google.com/spreadsheet/pub?key=0AtfsHdXQ5rl9dFp4b1hETmNQNzN0b2dSUlRNWEFOOVE&single=true&gid=2&output=html
    the production rate out of the Everett FAL was 5/month from the beginning of April to about the middle of May. From the middle of May to the end of October the production rate was lowered to about 3/month. Then in November the rate was back up to 5/month followed by a slight taper off to 4.5/month in December.

    The first 787-9 was loaded into the Everett FAL on May 30 and the third 787-9 rolled out on November 1. It is clear that Boeing slowed 787-8 final assembly in Everett while the first three 787-9’s now flying were assembled. Boeing also implemented significant tooling changes on the 787 lines during those 5 months. This was done in preparation for finally introducing a 5th position in the Everett FAL about a week ago.

    The assembly loading data for the Charleston FAL only goes from June through the end of December. During this time the rate averaged out at 1.8/month.

    Thus, the total 787 production rate out of the FAL’s went as follows:
    (Date Range: Everett: Charleston: Total)
    January – March: no data: no data: no data
    April – May 15: 5/month: no data: no data
    June – October: 3/month: 1.8/month: 4.8/month
    November: 5/month: 1.8/month: 6.8/month
    December: 4.5/month: 1.8/month: 6.3/month

    Except for the slowdown due to 787-9 assembly and the tooling change, the total production rate seems very close the 7/month that was announced in May.

    Boeing said that the first 787-8 at 10/month was LN-152 which rolled out 3 days ago. I suspect that this aircraft was the first one assembled from parts produced by the supply chain at the new 10/month rate. Final assembly rate will naturally lag behind a bit. In a couple of months the real final assembly rate will become apparent.

    • ” .. by a slight taper off to 4.5/month in December”
      How do you deliver/roll out/whatnot half a frame per month if you look at single month?
      December is either 4 or 5 frames.
      Back to delivery dates: I was wondering about the strong seesawing in
      deliveries during the available 9 month from 2013. ( ever increasing from midyear to year end.)
      This IMHO looks like a significant part of the workforce is tasked with delivery in one month and the next month sees them elsewhere in the production process?

      • “How do you deliver/roll out/whatnot half a frame per month if you look at single month?”

        I made a plot of the inverse loading period (time between successive frame loading) vs time. Then I filtered the variations out to get an average rate vs time.

        • Then it is not the number of $something in $somemonth
          but the average $something centered about $somemonth.
          ( as your averaging only has a left companion you are a bit
          in hot water with your math imho.)

          Then I was careful to reference _nonrecent_ frames.
          That includes the longtime storage frames as well as those
          newer ones that seem to have timeconsuming issues.

          .. too hung up : maybe ( on occasion I can be quite the pest about it 😉

        • “as your averaging only has a left companion you are a bit
          in hot water with your math imho.”

          Not really. I differenced in both directions and either way gives the very similar loading period results. Then, I applied a non-causal numerical filter so there is no left or right bias. Easy-peasy.

          “That includes the longtime storage frames as well as those
          newer ones that seem to have timeconsuming issues.”

          What’s your definition of “time consuming issues”? Seems nice and arbitrary so you can come to any conclusion you want. The fact is that if one looks at loading dates and roll-out dates there are few instances where the frames went through the Everett FAL out of order. The Charleston frames appear out of order but that is because they consistently lag. Other than the vastly out of order frames that required rework, the out-of-orderness of the newer frames flight test and delivery is customer driven.

Leave a Reply

Your email address will not be published. Required fields are marked *