Interview: Kirk Pysher, president, International Launch Services
February 16, 2017
International Launch Services President Kirk Pysher views the engine-quality issues that have grounded Russia’s Proton rocket since June as a sign that Russia’s government and industry sector are serious about quality control. He’s hopeful of a return to flight by April or May.
For Reston, Virginia-based ILS, which is responsible for Proton’s commercial sales, the past eight months have not been easy. First there was a second-stage anomaly in June 2016 during the mission carrying Intelsat’s IS-31 telecommunications satellite.
The problem, attributed to a loss of material strength properties that was likely the result of an improper propellant line attachment, did not result in a failure but grounded Proton for several months.
Proton’s planned December return to flight, with EchoStar Corp.’s EchoStar 21 S-band satellite for European mobile services, was scrapped and a full-scale investigation was begun.
The global satellite communications market awaits return of SpaceX, ILS
The continued grounding of Russia’s workhorse rocket has come at a time when the SpaceX Falcon 9 vehicle has had its own issues following a September failure. That has put two of the three principal commercial launch service providers — the third is Europe’s Arianespace — on the sidelines.
The global commercial launch market being what it is, very few of the affected customers of SpaceX or ILS have been able to move elsewhere.
In a Feb. 15 interview, Pysher said ILS is cautiously optimistic of being able to return to flight, with the EchoStar 21, in April or May. Two further commercial launches, of Hispasat and AsiaSat satellites, are planned before the end of the year.
Pysher is asking customers to look beyond the pain of the launch delay to see just how seriously the Russian government and Russia’s industry takes the issue of engine quality.
What can you tell customers about the Proton rocket’s status?
Pysher: Let me start by saying that this is actually a quality- control success story.
As part of the standard quality process, and this may be unique to Russia, when they do engine builds, they actually have a periodic test program, where they will pull an engine from the production line once a year and do an extended duration test on it, typically at higher thrust levels than what is needed for flight.
The purpose is to validate the quality and production stability. They conducted that test on one of Proton’s second-stage engines in early December. The engine survived the test, typically run at twice the flight duration and at higher thrust value for at least part of the duration.
Following that test, they do a destructive inspection of the engine. They cut it open and do visual and metallurgical inspections. On one component, they found the brazing did not have the proper bonding characteristics.
Roscosmos organized an investigation and they determined that the solder used in that brazing operation had a higher melt point than what they expected.
But that solder issue did not affect the test firing?
Pysher: That’s right, the engine had survived the test. With the soldering, they lay sheets of the solder material and then put it in an oven and bake it. The solder then melts and flows throughout the engine. At a lower melt point, when it bakes in the oven at a set temperature it will flow better than a solder that has a higher melt point.
That leads to a decision point. Do they continue flying these engines or do they stop because of the uncertainties?
So the first point is the quality process did catch this before it got on a flight vehicle. It was found on the ground.
Second, from an engineering standpoint there was a decision to make.
You demonstrated the capability even with the higher-melt solder, but do you really want to take that risk? If you take the risk, how do you explain it to your customers?
Grounding Russia’s work horse
So the decision was made: Let’s ground the work horse of the Russian launch business and replace these engines with a component with the expected bonding characteristics.
This is part of the renewed emphasis of Roscoskos [Director-General Igor] Komarov and [Andrey V. Kalinovsky, director-general of Proton prime contractor Khrunichev Space Center] to ensure 100 percent mission success.
All this had no relation to the Proton anomaly in June?
Pysher: Correct. In fact, the paperwork has verified that the engines that had this higher-melt-point solder were not among the engines that were on the Intelsat IS-31 launch. There’s no relationship between the two.
What was the root cause of the IS-31 launch issue?
ILS Chief Operating Officer John L. Palme: That was traced to a likely quality lapse in an attachment point on a line heading to the gas generator on the engine. That line was subjected to higher-than-expected vibrations that led to a fatigue failure and eventually a failure of the line to the gas generator, leading to an earlier-than-planned shutdown of the engine.
Khrunichev was able to duplicate that exact scenario on the ground and traced it to a mechanical attachment that did not meet the specifications for tightening, or may not have been present.
And that accounted for the vibration in the gas generator line?
Yes. Somehow one of the degrees of freedom that would have been taken out by that attachment was not adequately taken out and that led to higher vibrations in the line and a leakage of the oxydizer.
When was this issue resolved?
Palme: In August, and we had a final briefing with customers in early September.
What happened between September and December, when the solder issue was discovered?
Pysher: We had to implement the corrective actions from that anomaly investigation, which led us to the start of the launch campaign in mid-November.
Palme: We basically had to pause the EchoStar hardware at the factory until the investigation by the Russian committee was completed. That took time as they wanted to see evidence of testing to duplicate the exact situation. So that ended up to being an investigation of three months or so. Then we had to out-brief it and then the corrective actions on that engine on the vehicle meant the vehicle itself had to be system tested.
So we started the launch campaign in mid-November, preparing for a launch in late December.
Then came the test of the engine with the solder issue.
Pysher: Yes. You may remember in 1999, a Delta 3 flight failed with an issue with the RL-10 engine that was linked to brazing operations. It was a very similar thing. You lay in this cold solder, put it in an oven and bake it and you didn't get the proper flow characteristics. The difference was that the RL10 issue was discovered after a launch anomaly, whereas this was discovered on the ground before it found its way into a flight.
Was the soldering issue at the Voronezh manufacturing plant due to someone wanting to save money by using below-spec components?
Pysher: The investigation is ongoing. They are at the point where they understand that the melt point was higher than expected. But how this found its way into the engine is not yet determined.
My impression is that this is not a case of a guy going to Home Depot and buyingsolder and installing it into the engines. The reality is there had to be a purchase order that was issued, someone had to have engineering behind the order, it’s pretty complex. We’ll probably find out over time what the story is.
I am guessing here, but the solder may be used on some other engine in the facility and there may have been some engineering which allowed that to be substituted. There have been reports that the technical documentation was confusing. But I don't think we’re at the point where we can say how it found its way into the engine.
What happens in the next couple of months?
Pysher: There is a launch opportunity in April or May. We’re not 100 percent there yet, but we understand there is a viable opportunity for that to occur. Komarov has said there were seven planned launches for Proton this year and they expect to conduct all of them. We plan three commercial missions.
So the return to flight in April or May will be the ILS mission for EchoStar?
Pysher: That’s our current expectation.
Proton’s uneven record in recent years gives an impression of quality-control breakdowns. But for the engines, quality assurance seems extensive — and costly.
Pysher: Here’s the engine test process for our third-stage engines, just to take an example. Each production lot is six engines. Two go into lot-acceptance testing. They pull out two of the six and do an extended-duration, extended-thrust level test firing. It’s not the same as the periodic test, which is more severe.
Then they cut up those engines and evaluate the quality and only then are the other four allowed to be put on the vehicle. That’s a very expensive process. And then you add the periodic-test engine, which is typically on an annual basis but is linked to time and the number of engines produced. So it could be conducted more or less frequently. But typically it’s an annual basis.
It’s a pretty extensive qualification test when it comes to engines. There has been discussion of eliminating that process to reduce cost. There is a lot of pressure on everybody to cut costs. But I don't think that’s going to happen.
Peter B. de Selding