By Bjorn Fehrm
21 December 2015, ©. Leeham Co: Bombardier (BBD) received certification of the smaller CSeries, the CS100, by Transport Canada Thursday. Rob Dewar, vice president of CSeries, reflected on the journey to certification in an exclusive interview with LNC.
The interview was done against a backdrop of more than two years of delays, which in turn drained the coffers of BBD. To save the project and let it prove its game-changing character, management sold 49.50% of the CSeries program to the Province of Quebec for $1bn and 30% of its train division to Caisse de dépôt et placement du Québec, for an additional $1.5bn.
Dewar has managed the project from the program launch in 2008. The transcript of the interview follows.
LNC: First of all Rob, big congratulations. After all the delays, it must feel great to have got the certification when you said you would.
Rob Dewar: Thanks Bjorn, yes it feels great. Things have gone really well over the last nine months and we got certified when we said we would. We expect the CS300 to be certified as well within the next six months.
LNC: Looking at the last year, things have gone to plan. This is very different to the years before, where delay was added to delay. Tell us about it.
Flight testing and EIS
RD: Yes, things have gone well. We have had a 98.6% dispatch reliability of the test aircraft and an even better reliability of the first serial aircraft, P1 (for Production 1). This has been used for Function & Reliability (F&R) testing. During 60 days of testing of P1 with 270 flight hours and 120 flight legs, we have not had one delayed departure. We have had a 100% despatch reliability.
LNC: OK, that is with an engineering surrounding who knows the aircraft like its pocket with the full access to spares and a flight crew that can interpret and filter every spurious warning. Your next big challenge will be the Entry Into Service (EIS) with Swiss in Q2 next year. This has to go well; you can’t afford another entry into service like Q400.
RD: We have managed quite a few entries into service and we have good confidence that the CSeries will go well. As I said, the F&R has gone extremely well. We have a 100% dispatch reliability and we will continue this testing until Swiss gets their aircraft. We really trust the aircraft.
As an example, we flew P1 that is here for the press event from Orlando to Wichita Wednesday; Wichita-Atlanta yesterday afternoon and we started its flight home to Mirabel at 5 minutes past midnight today for a landing at 2AM in really bad weather. You just don’t do that ahead of such an important event (the Certification ceremony at 8 am) unless you have full confidence that the aircraft will get here.
We are not naïve. There will be problems to solve at EIS and we are setting up a comprehensive EIS support organization together with Swiss. But the testing has shown that there should be no big problems to expect. We are putting a very mature aircraft into production.
What took them so long
LNC: Let’s talk about how you got there. You had a long period when things where moving really slow. What was the problem?
RD: Apart for the engine problem that we had, there was no other hardware-based delay. All testing of hardware went well. For example, nothing broke during the static structural tests and we found margins in our wing that we will translate to future customer value.
What held us back was software. CSeries’ functionality is largely software based. Its Fly-By-Wire (FBW), Engine electronics and Avionics system contain some 11-12 million lines of code. This is a magnitude larger software system than our previous projects.
When we started our meetings with Transport Canada on how to qualify this large software base, it became clear that things had changed. The software validation would be vastly more elaborate than the last project. The authorities now required traceability of each realized aircraft function on a level not demanded before. It meant that our partners like Parker Aerospace for the FBW had to conduct much more work than planned and we and our partners issued thousands of documents beyond what we had planned.
Everything took much longer than anticipated. There were no special areas that gave problems, it was just vastly more work that we had foreseen. The positive out of all this is that the functionality of each subsystem got a much higher maturity level before entering our test rigs. The suppliers had to test to a very high maturity at their rigs, then in our rigs with other partners and when finally the software versions were released to flight test, they were of high quality. The test flights now were there to verify what the test rigs had shown, not tuning the software or finding faults.
We can see the result of the effort in our F&R flights on P1. The present software load, version 6, is very stable. For EIS we will have version 7, which is essentially a further cleaned-up version. Only 25% of the code is touched and part of that is to include optional functionality which will not be activated at EIS. An example would be steep approach for, e.g., London City Airport. The flight tests to certify these optional functions will be done during the spring, after Certification. Once Certified, we then activate these functions for the customers if they have selected them.
LNC: Time for some touchy questions: With the serious delays, you must have got a lot of pressure to take shortcuts and push nice-to-have functionality to be completed after EIS. What is missing from your original plans?
RD: It sure has been tough, quite a number of times. We have been lucky to have been able to convince top management to not force us to cut corners, to do it right first time around despite the costs. There is no functionality missing from the original plans for the aircraft at EIS.
LNC: What about rework? How much time has been planned for rework for aircraft P2, the first aircraft for Swiss?
RD: There is no planned rework for P2 or later aircraft. They will get delivered as they are built on the final assembly line. We have in total one week of contingency in our delivery plan to Swiss and that is for anything unforeseen that crops up. Part of no rework is that we have not needed to change any vital hardware as a result on flight tests.
As discussed at the Paris Air Show, the aircraft is clean, without vortex generators or other patches to fix issues. The aircraft will get delivered as designed. Any changes we had to make on internals or at system level we have been able to build in at production of the aircraft. The aircraft goes from production straight to flight test and then to delivery.
LNC: I have to give you credit for achieving that. You also managed the inevitable weight creep well; you are within 5%, a very good figure. With your aero and structural margins, you could increase weights to meet brochure performance and in some cases exceed these.
RD: Yes, weight management has gone well and we have further margins that can be used for future improvements. We discussed at Paris that we can standardize on the larger CS300 winglets later on and we have more to find in how we handle the open main landing gear wheels. We have also been conservative in how we have given our performance data. We have a passenger + bag weight of 225lb.
LNC: You are right, when I use the normal 210lb, our model gives 3,500nm as your comparative range to A319’s and 7MAX’s 3,800nm. That’s transcontinental range and even transatlantic range for London to US/Canada East Coast, for example.
We know that you claim around 10% better fuel burn than the re-engined competition, A319neo and MAX . You are also lighter, which gives lower landing/underway fees. What about maintenance costs?
RD: We are one generation ahead of the competition there as well. A-checks are at 850 hours and C-checks at 8,500 hours, with heavy structural checks at 12 years compared to our competitions’ eight years. We also don’t have daily/weekly checks,. The first check is at 100 hours and everything before that is on demand. In practice, an operator can fly the aircraft for over a week before needing to do the first check. During this time, he does not even need to fill oil; we have an extra capacity system for that purpose.
LNC: We will come back to fuel burn and maintenance costs in a future deep analysis of the CSeries, but one can see that you will be very competitive. Now everything will depend on the entry into service with Swiss. There, the aircraft will be flown by flight crews which can not filter your aircraft’s spurious warnings and maintained by operational airline personnel. Any last thoughts on that?
RD: I can assure you that we are not underestimating the challenge. We might have good reliability in our tests but we know there will be problems, for sure. Therefore we are really making sure we will be organized to take on the challenge. We will be ready to face any problem thrown at us and Swiss.
The CSeries program is one full of opposites:
The CSeries has the technical characteristics to be a game-changer. What remains to prove is the in-service reliability and then that the organization can sell the result. Should this be matching the technical capability of the aircraft, the CSeries will be the game-changer that Bombardier claims it has (and badly needs).
The market will wait for the news and be there for those that succeed.
I wonder a) how Bombardier could have failed to keep up to date with certification requirements before launching a new programme, and b) why Bombardier didn’t fess up in the media once it had discovered this problem. I don’t recall much transparency in reasons for delays (engine aside) during the programme (or is memory faulty?).
I think your memory might be faulty. For example, here is a link to a January 2014 Wall Street Journal article that said:
“A Bombardier spokesman said its planned 2,500-hour flight-testing phase will require more time than originally anticipated to ensure “all pieces of software are working well and synchronized” and systems are mature before delivery.”
“New jetliners are increasingly reliant on software to manage and integrate the complex systems aboard the aircraft and have produced headaches for early operators of the A380 and 787. Both aircraft faced challenges with software that often erroneously flagged problems on the jet that would delay flights.”
No, I do recall reading pieces like this and others where they “avoid(ed) pinpointing causes” and so on, but all of these seem to have a common line about getting the software working properly further into the programme, an issue that should be no surprise. None I’ve seen though mention that they’d been set back right at the start of development by failing to do something that they absolutely should have done, namely maintain awareness of the validation procedures they would be required to follow. One person having occasional chats with regulatory counterparts could have done that. Feels sloppy to me and reminiscent of Airbus’ failure to have a single, simple chat somewhere in the organisation about software versions before developing the A380.
Knowing the Canadian administration makes me understand BBD delays.
Recently I saw someone say its the software which has the highest risk and longest time in development for new aircraft, especially for military planes, yet the aerodynamics and shape is started first- which these days is fairly easy to get the correct configuration in the computers.
I hope the big manufacturers are using the time before the start of a new narrowbody to begin the software task, which could probably be tested in an existing flying laboratory, and be checked out against future certification requirements
Good interview. I always wondered on why the landing wheels were left open. Surely that must cause a bit more drag than if they were covered. Any thoughts?
Also, how long until FAA and EASA certification?
Landing wheels are exposed too on the CRJ. They must have felt comfortable with that and with the simpler mechanicals. It seems they are now thinking about adding doors, which doesn’t seem obvious considering tires are flush with fuselage.
I have read somewhere else (forgot where) that FAA and EASA certification should be obtained in January.
“I always wondered on why the landing wheels were left open.”
The 737 never had doors either. The idea is to save weight. But this solution should be less advantageous today with the use of composite material. The C Series is so refined aerodynamically, and doors could only make it better. I just hope it has been designed with the possibility of adding doors if needed.
Didnt the 737 have a sort of inflatable fairing around the outer wheel ring which smoothed off the airflow ?
The question I have in mind is this: what will Bob Dewar once after the structure of the commissioning will be up and running smoothly?As for me, is the last line of text below that intrigues me …
I hope all will work out and a sizeable operator will commit to a large fleet of CS300s soon. I you ignore all misty orders, something BBD will never do, the backlog is not big enough at this stage.
There’s a large 130-150 seat replacement market. But BBD has to offer a growth option, like Airbus and Boeing do.
Then airlines can commit to large fleets longer term, converting some to CS500’s when required.
I would bet the CS500 is already flying in the computer.
I wouldn’t be surprised if the upcoming federal loan of 1 B$ was in part allocated to development of the CS500.
Seeing the rest of the pack (Embraer, Airbus, Boeing) moving, the CS500 is becoming more a necessity, RATHER then an interesting future option.
If BBD had known before, CS300+ CS500 maybe Could have been a better start off. But at that stage (8 yrs ago) the priority was to not provoke A & B by suggesting they were not in the same marketplace.
Last week, in an interview with a Canadian newspaper, Alain Bellemare said about the partnership with the Quebec government, there is provided ago the beginning, there will be room for another partner. This is not the Canadian federal government. Who do you see ?
Boeing? Replacing the government of Quebec? Why did not he would undertake to develop and produce the CS700, CS900 and CS1000 (for the market 321neo vs 757) and Bombardier, the CS100, CS300 and CS500?
The British engineering firm GKN has been expanding its aviation interests, they have the Fokker structures business, the Airbus Filton business, Volvo Aero.
Could they be the bride in the wings ?
With the MRJ setback again it looks like C Series would have significant openings.
In the sales clause is ability to cancel if a program is late (2 years seems normal as I recall but could be different)