Juno Experiences Engine and Computer Problems
Juno Enters Safe Mode And Then Regains Normal Operations
“NASA’s Juno spacecraft entered safe mode Tuesday, Oct. 18 at about 10:47 p.m. PDT (Oct. 19 at 1:47 a.m. EDT). Early indications are a software performance monitor induced a reboot of the spacecraft’s onboard computer. The spacecraft acted as expected during the transition into safe mode, restarted successfully and is healthy. High-rate data has been restored and the spacecraft is conducting flight software diagnostics. All instruments are off and the planned science data collection for today’s close flyby of Jupiter (perijove 2), did not occur.”
Engine Problems Delay Juno Engine Burn at Jupiter, SpaceRef
“Telemetry indicates that two helium check valves that play an important role in the firing of the spacecraft’s main engine did not operate as expected during a command sequence that was initiated yesterday,” said Rick Nybakken, Juno project manager at NASA’s Jet Propulsion Laboratory in Pasadena, California. “The valves should have opened in a few seconds, but it took several minutes. We need to better understand this issue before moving forward with a burn of the main engine.”
NASA’s Juno Team to Discuss Jupiter Mission Status, Latest Science Results, NASA
“Team members of NASA’s Juno mission to Jupiter will discuss the latest science results, an amateur imaging processing campaign, and the recent decision to postpone a scheduled burn of its main engine, during a media briefing at 4 p.m. EDT Wednesday, Oct. 19.”
Disappointing if they can’t fix it. It reduces the number of passes, and there are too few spacecraft in the outer solar system as it is.
Of more concern is why they have failed. They are perhaps the simplest unit in the propulsion system. We have had failures in the past where thermal conditions allowed liquid propellants to condense upstream of these valves and to detonate taking out a mission. We learned that lesson. This is a worrying echo. They would have been functionally checked multiple times prior to launch and validated during qualification. To fail now is suggestive of a mission related event maybe even propellant accretion. A potential mission killer, not good… not good at all. Mind you the Japanese lost the main engine on Akatsuki due to one of these sticking so such failures are not unknown.
Frozen propellant may have accumulated without detonating. The fact that they opened after a few minutes sounds like a good sign.
‘Tis liquid penetration of the valve seat material that is the bug bear here.
I remember an interesting paper on why hypergolic propellant valves still are not reliable after so many decades of use, and that was at least ten years ago. The conclusion was that hard seats are vulnerable to contamination by particles that keep them from sealing and soft seats are vulnerable to deterioration from exposure to the hypergols. If we can put a man on the moon…
All of the liquid valve seats that I know of are soft seats and are variations on PTFE (teflon) with different fillers. MON can displace some of this filler material over time but it rarely leads to a significant external leak. All the RCT’s on the western catalogs are good for 15 years nominal with a typical use to 18 in this passing generation of com sats. Why the soft seat NRVs buck this trend repeatedly is a little bit of a mystery to me. Perhaps it is the relatively low poppet-seat pressure. NRV cracking pressure is, by design, very low whereas the seat pressure on some of these dual seat RCT valves can be much much higher than the tank pressures due to the back pressure relief of the upstream valve. You can’t beat physics. The only way to eliminate this risk is to eliminate the need for NRVs: Separate the pressure control assemblies. Not an attractive mass proposition with mechanical regulators.. but with electronic regulation……The IT revolution marches on. YMMV
You make an excellent point that low seating pressure in check valves may be the culprit. NRVs are ancient technology. Electronically controlled/electrically powered positive-action isolation valves make a lot more sense.
But does this actually make sense? Wouldn’t the hypergolics be extremely pure and clear of fines? Or – I’m decades out of chemistry here – do they over time form tiny internal solids, perhaps driven by radiation?
Hypergolic fuels really should be free of contaminants. One of the reasons hydrogen peroxide got a bad reputation was poor purity when it was first used as a rocket fuel. Contaminants can act as a catalyzer and cause things to go boom.
But post-launch contamination is possible. Cassini has been running on its B side reaction control thrusters since 2009 (I think, but I may be off by a year.) The A side started showing anomalous behavior, so the project switched to the redundant system and adopted some life-extending practices. One theory for the A side problem was contamination. On that spacecraft, the tanks are kept under pressure with helium, and a polymer bladder is used to keep the helium and hydrazine from mixing. Over time, the polymer can, in theory, degrade and produce contaminants in the hydrazine. (And, just for clarity, the A side problem wasn’t with valves. It’s believed to be involve flow through the catalyst bed, but it’s still, potentially, a contamination issue.)
I did hear something second hand. Instead of spreading rumors, I’ll ask a leading question. Does anyone know anything about recent performance of the LEROS 1b engine on other spacecraft?
I would be very careful of moot that about. The LEROS 1B is a truck and is a go to engine for NASA missions (off the top of my head NEAR and Mercury Messanger come to mine as well as Juno. and others.). It has relatively poor performance when compared to either the hi-pat dual mode or their new AMBR. However it has masses of thermal margin and the higher thrust gains propellant budget by minimizing the gravity losses. It is a classic materials engine developed in the early 90’s N2H4/MON so it has no carbon reactions and consequently no stability issues. The only credible failure modes outside injector contamination (unlikely, as it is acceptance tested and verified clean repeatedly) are in the valves. However they are simple solenoids with a lot of inherent robustness. It is not unheard of for a valve inlet filter to unseat and skew the mixture ratio, however I haven’t heard of such an event getting past the acceptance testing. I would be surprised if the engine was found to be at fault. Although due diligence requires the engine to be examined it has yet to fall foul of the fault tree. It’s close cousin the LEROS1C has flown on many many com-sats and is on GPS2.
Well, without saying anything I was told in confidence, I think I can add to my comment. About a week before the decision to defer the period reduction maneuver, three different people involved with the Juno mission told me they were considering that option. This was before the odd behavior from the valves. According to those people there were two, recent incidents of anomalous behavior from the same model engine on two other spacecraft. When I asked for details, they didn’t know and one said the spacecraft involved were either military or national security assets (so don’t hold your breath for more information.) I was also told the mission was planning to go ahead with the maneuver. As I said, this was all a few days before they saw the off-spec performance from the valves. I guess that changed their minds.
I think there is some confusion going on in your information stream which is why I am cautioning making too much of it. The LEROS1B is only used for NASA deep space missions. Its cousin the LEROS1C is the preferred engine for any orbits from GEO down. So I am pretty sure the military don’t use the 1B. The 1C is a higher performer with a little less margin and is preferred for any earth orbit applications. These engines have never had an on-orbit failure to my knowledge. What has happened is that they have been subject to off nominal conditions due to upstream failures. Now while these engines do meet the mixture ratio specifications with margin they do not respond well to out of box conditions. Most engines don’t but these tend to fall off a cliff outside their allowable operating box. Whereas some do degrade more gracefully and others are even worse, some even within a nominal operating box. It is perhaps this issue that is driving the chatter you are perceiving. Now while I can chat freely and in detail about this, I am not subject to ITAR and the engines are European, other contributors to this site do not enjoy the same latitude. So rather than temp them into indiscretions. I would draw a veil over this conversation and allow NASA to conclude their ARB. We will all find out in due course.
Actually, that seems to fit very well. I heard this from scientists and a managers, and I think the distinction between a LEROS 1b and 1c would be lost on them. The project heard about a problem with a related (mistakenly called “same”) engine, were worried (enough so that they were already thinking about deferring the burn), and when they saw anything close to twitchy in the telemetry, called off the burn.
But I really don’t like the way information on this sort of thing comes out. I think it’s very unprofessional for people to avoid or minimize discussion of problems. We need to be improving the state of the art, and that means getting all the details out. Trying to protect a mission’s (or a person’s) image by limiting the flow of information doesn’t work, and can get misleading information out there. One of the comments I heard was made to a room full of about 150 scientists (a project science meeting) and some future Discovery and New Frontiers PIs were probably in the room. I hope the LEROS 1b didn’t get an undeservedly bad reputation.
Hmmm… just read some more on this. Don’t think it was the check valves. I wonder what regulator they are using… This looks like the flow limiter activated.
This should not affect the number of periapsis passes. Juno’s lifetime is limited by radiation, and almost all the radiation dose comes from periapsis. So they are good for the planned 37 passes, whether or not the first few are 54 days apart instead of 14.
There are some second-order effect: Over the course of the mission, the orbit does precess and the dose per periapsis goes up. Staying on the 54 day orbit will change that, but not (if memory serves) significantly. It ends up being about 1 deg. per orbit, with only a weak dependence on period. Also, the magnetometer investigation may take a slight hit. The orbits were designed to give a very regular spacing in the longitude of each periapsis, which is optimal for magnetic mapping. Adding an extra 54 day orbit will mess that up a little. But the mission was planned for an extra orbit or two (i.e. more than required for the magnetic mapping), just in case. So I doubt this will be a major problem.
If valves are in the proper states, and pressues are as expected, what is the problem? Command timing not as expected? If so that might be a FSW or staored command issue that definitely should be solved before main engine burn.
Is this the first main engine burn?
I think it would be the fourth. They did two deep space maneuvers on the way to Jupiter (effectively one, but they decided to do half the duration, wait a few days, and then the second half), and orbital insertion. That would make the period reduction maneuver the fourth use of the main engine. I think. That’s assuming they used RCS thrusters rather than main engines for trajectory corrections. That’s the usual practice unless the corrections are large.