Orbital ATK ready for Antares’ second life
Orbital ATK managers have finalized their procedures related to the newly upgraded Antares launch vehicle in readiness for upcoming missions to help deliver their Cygnus spacecraft on resupply runs to the International Space Station (ISS). The new Antares required a reworking of how to prepare the rocket for launch, such as moving away from sub-cooled liquid oxygen through to changes to the vehicle avionics and software.
After the CRS-3 mishap at Wallops in October 2014, Orbital ATK decided to re-engine the first stage of the launch vehicle with RD-181 engines from NPO Energomash.
The two RD-181 engines replaced Aerojet AJ-26 engines that were used in the first Antares launches in 2013 and 2014. The new vehicle configuration is designated as the Antares 200 series.
Cygnus flights to the ISS will use the new first stage coupled with Orbital ATK’s CASTOR 30XL solid motor as a second stage in the Antares 230 configuration.
The RD-181 engines are similar in design to the single RD-180 engine that is used on the Atlas V first stage. The two RD-181 engines will give Antares about 13 percent more thrust than the AJ-26 engines, according to Kurt Eberly, Orbital ATK Antares Deputy Program Manager.
“We’re using the same amount of propellant and we’ll get through it quicker because of that higher thrust,” Mr. Eberly noted to NASASpaceFlight.com. “The 130 (configuration) was about 234 seconds (in duration); this is going to be about 214, that’s the predicted cutoff.”
Mr. Eberly noted another significant change, where moving away from the AJ-26 meant the Antares would no longer use sub-cooled liquid oxygen.
“That was consistent with the design of the NK-33, to use sub-cooled LOX – here we use normal boiling point LOX, but the temperature constraints that we’ve been given from Energomash are fairly strict because they don’t have a lot of test history at different temperatures.”
The change from sub-cooled liquid oxygen to boiling point liquid oxygen also drove changes to the Antares pad at Wallops, Pad 0A. The original plan was to remove the sub-cooler needed for the AJ-26.
“We were going to take the sub-cooler out, but then we started looking back at the data for cross-country LOX temperature rise from the storage tank across an uninsulated line to the launch mount and we scared ourselves that on a worst-case day we could be too warm,” Mr. Eberly explained.
“So we traded off installing vacuum-jacketed line all across there which is very expensive and hard to implement and we made the decision with MARS (Mid-Atlantic Regional Spaceport) to add the sub-cooler back in, but implemented in such a way that it’s (only) there when we need it.
“We have a bypass line that goes through the sub-cooler so we can chill some of the LOX and mix it back into the main LOX. We’ve been testing that with them and so far it’s been working pretty well. It gives us fine temperature control of the LOX that is going into the vehicle.”
In addition to hardware changes to the first-stage propellant feedlines and thrust frame required by changing the engines, there were some significant changes to the vehicle avionics and software.
“There are a bunch of new interfaces,” Mr. Eberly added. “We had a Moog TVC (Thrust Vector Control) box that we communicated with digitally and then it closed the loop when we wanted to steer the AJ-26 engines.
“Now we’re closing the loop within our avionics, so Orbital ATK avionics are doing more of the job for the RD-181 configuration – and it’s just different, so we’ve had to adapt our avionics to interface with the engines.
“That lower level code is new and we’ve got a whole “bench-top” set up in Chandler, Arizona; that’s where we do most of our manufacturing for avionics and harnessing and the upper stack structures. So we’ve got non-flight valves set up ‘on the table,’ (and) we’re just running through all the different control algorithms and making sure they’re working right.
“Everything is going really well. We’re in a software validation phase, (where) we go down all the different logical paths at the code level, (doing) unit level testing, and then we’ll do systems level testing at Wallops when we get that code released.
“Then we’ll go through mission simulations where we (are) practicing with the rocket flying the mission.”
The way the RD-181 engines are gimballed is also different from the AJ-26.These engines can be gimbaled up to plus or minus five degrees.
“This is a Cardian joint on these engines, whereas with the AJ-26 there was a gimbal joint that gimballed the whole engine and you had to put a whole lot of flexibility in our feedlines,” Mr. Eberly continued. “(On) these, the thrust chamber and the turbopumps are all fixed and then there’s this Cardian joint, which is a basically a metal bellows that rotates the nozzle. The same joint is used on the RD-180.”
Another major difference is in engine operation and control.
“The ignition sequence for the RD-181 is very different from the AJ-26,” Mr. Eberly noted. “There’s a series of solenoid valves that have to open in a certain sequence when compared to the AJ-26 which had a lot of pyro valves.
“It’s a different design and so there’s a different sequence – we’ve worked through that with Energomash. We have a detailed document that prescribes all of that. After ignition there’s a health check; we already had the structure for that for the AJ-26, and one of the key parameters is pump speed.
“We’re going to mechanize that and practice it and then the hot-fire test – the stage test – will be the real verification of that. And then the shutdown sequence is similar, there’s a series of valves that have to be operated to shut down the engines and put on purges that are required after engine cut off.”
Currently, the hot-fire of an Antares first stage at Pad 0A is planned for mid-March.
The Antares 230 vehicle will be able to lift a Cygnus spacecraft loaded with as much as 3200 kg of cargo. However, depending on NASA’s needs for a given mission, some of that performance could be allocated in other ways besides cargo upmass.
“We can go to an instantaneous launch window, that would then mean we can devote more performance to Cygnus, but we try to strike a balance (with the length of the window),” Mr. Eberly added.
“Let’s say there’s a boat (on the range) – (it’s useful) if they (have time to) get a spotter plane so he can get his eyes on something to meet their probabilistic risk assessment. So there is some benefit to having a five or fifteen-minute launch window.
“So we try to balance the use of the excess performance – we could launch them to a higher orbit, we could have a longer launch window – it’s just a matter of how you spend the performance on the mission design.”
For the OA-5 mission on Antares, the current plan has a fifteen-minute long launch window. However, that remains subject to change until more details for the mission are finalized.
“Right now we’re protecting against loading 3200 kilograms of cargo, but we’re going through the process of what NASA intends to fly,” noted Dave Hastman, Orbital ATK CRS Deputy Program Director. “(But) they might not take advantage of that full 3200.
“If they came back with even less cargo, we may re-evaluate this approach to give us a bigger window. Typically, we finalize that manifest at about four months prior to launch with the customer – we have a cargo integration review, we call it.”
The launch day countdown will also be a little different for Antares. Mr. Eberly noted that Orbital ATK is restructuring the countdown and pre-countdown timeline.
“We had a long countdown for the 130 (original configuration) – we’re going to shorten that up (and) put a lot of stuff in the pre-count, things like filling the sub-cooler with liquid nitrogen,” he explained. “So we’re going to shorten that to around five hours.
“A lot of that (time) is spent in preparing the ground systems for the (propellant) loading. We start loading at about an hour and a half, we start with liquid oxygen and then about 30 minutes later we start with the RP (RP-1 refined kerosene fuel) and then we do them simultaneously from there on.
“It’s all automated process control with human intervention.”
Mr. Eberly also outlined the terminal countdown automatic sequence that will be employed with the new Antares.
“At three minutes we handover to the flight computer onboard the vehicle, but we can still stop (the sequence) from ground intervention.
“From three minutes, we pressurize the tanks, we switch the navigation to free inertial, and then inside three minutes we start the ignition sequence – there’s a purge of the main chamber and the turbopumps with heated nitrogen and then there’s a whole sequence that starts the engine. It’s similar to the RD-180.”
As noted earlier, the countdown culminates in the ignition sequence of the RD-181 engines, which involves the engines coming to life, but no vehicle movement for a few seconds.
“It sits on the ground for a while, this ignition sequence. It’s more of a heart attack situation for everybody involved where you see the flames come out but nothing really happens.
“At T-0 (we) start the ignition process, and we won’t lift off for over three seconds. There’s a number of automated health checks on a number of parameters on each engine. The main one being when we get up and running we check the turbopump speed, that’s the main indicator and that means everything is healthy if those speeds are in the right range.
“I think it is fifty-five percent is what we ignite to and (if) we pass the health check we throttle up the engines, which takes about a second, we release the hold-down mechanisms, and we rock back the transporter erector launcher (TEL) which serves as our umbilical mast. As thrust exceeds mass, we’re cutting commodity lines, which are pyro-released.”
As before, Antares will perform the “Baumgartner Maneuver” shortly after liftoff. Named after Antares Guidance Navigation and Control (GNC) lead Paul Baumgartner, the vehicle pitches slightly away from the TEL.
“That’s to save that transporter erector launcher and reduce the amount of impingement from the plume on that because we want to turn that around pretty quickly,” Mr. Eberly added.
There will also continue to be a relatively long coast period between the first-stage and second-stage burns when Antares returns to flight.
“We start closed-loop guidance during that coast phase and we figure out where we are, where we want to be, and we’ve got a whole algorithm. We use PEG, Powered Explicit Guidance, it was developed by NASA a while back,” Eberly explained.
“That’s really useful for when your last stage is a solid rocket motor. Basically, it divides up the burn of the solid rocket motor into a first half and second half – so you’re continuing to correct your guidance based on even the performance of the first half of the solid rocket motor burn and feed that into how you fly the second half.”
Cygnus still has one more ride atop the ULA Atlas V before the spacecraft joins forces with her intended ride to orbit.
Per the recent CRS2 contract award, Cygnus can now ride with either the Antares or Atlas V for future missions, with NASA’s focus on the upmass, as opposed to the launch vehicle tasked with lofting the spacecraft.
(Images via Orbital ATK, NASA and via L2’s Antares/Cygnus Section – Containing presentations, videos, a vast set of unreleased hi-res images, interactive high level updates and more).
(Click here: http://www.nasaspaceflight.com/l2/ – to view how you can support NSF and access the best space flight content on the entire internet).