OMI V1202 – The Main Propulsion Helium Signature Test

Back to the Space Shuttle documents page. If you are trying to follow the voice tapes below you will also want the OIS channel list open in another tab – it tells you who is talking.

The short version

V1202 is an end to end leak check of the Main Propulsion System, run with the orbiter vertical on the pad. You seal the aft fuselage, pressurise every MPS cryogenic and high pressure system with helium, and let the ECS aft purge air carry any leakage out through a vent door to the Hazardous Gas Detection System. Helium concentration goes up, you do arithmetic, and you get a total system leak rate in scim. Six scim was the maximum allowable.

The aft compartment IS the instrument. That is the whole idea. You are not chasing individual joints, you are treating the entire aft as one control volume and asking a single question: does everything in there, added together, leak less than the number.

Why it exists – and it isn't the accident

Before OV-099 Challenger arrived at KSC, leak checks were done point to point. You went joint by joint, fitting by fitting, and you signed each one off.

Then troubleshooting after Challenger's Flight Readiness Firings turned up a serious hydrogen leak that the point to point method had never found. Post test analysis concluded the undetected leak would have exceeded the ascent flammability limit. The flammability limit for H2 in air is 4%, which works out to an ascent limit of roughly 57,000 to 59,000 scim.

Read that again. The old method said the vehicle was good. The vehicle was not good. Confidence in point to point did not survive that, and a new total system check was needed that could GUARANTEE the number rather than infer it from a stack of individual signatures.

To be clear, because someone will get this wrong: this is the 1983 Flight Readiness Firings, not 51-L. Different event, three years before the accident.

IFLAA

We continued to learn that the test was not perfect. There is an acronym for the limitation: IFLAA – It Doesn't F***ing Leak At Ambient.

V1202 pressurises with helium at room temperature. Flight loads hydrogen at 20 Kelvin. Metal shrinks, seals move, and a joint that passes clean at ambient can open up when the cryo hits it. The signature test tells you the system is tight at ambient. It does not and cannot tell you the system is tight at tanking.

The summer of 1990 is what that costs. Everything passed. Then we tanked, and it leaked. OV-102 got LOTS of miles going between the Pad and the VAB, just none of it vertical ;-)

Which was not an argument against running V1202 – a test that catches the ambient leaks catches a lot of leaks, and the alternative is catching none of them. It is an argument against reading any single test as a statement about flight. Every test has a boundary condition, and the useful engineers are the ones who can tell you where theirs is without being asked.

V1201 and V1202

The same concept got built two ways. V1201 was the horizontal version, run in the OPF. V1202 was the vertical version, run in the VAB or at the pad. Same idea, different problems.

V1201 never worked as well, for reasons that are all fundamental rather than fixable:

Put those together and the conservatism plus the introduced error made all but a zero reading suspect. Which is a polite way of saying the test could reliably tell you when nothing was wrong and not much else.

So V1201 got dropped. NOT because it was a bad test, but because it was a redundant test whose results were less accurate than its vertical sibling – and you were going to run the vertical one anyway. The pad version had the real HGDS instead of a drifting portable, and it had the aft closer to flight configuration. That is the whole argument and it is a good one.

Preop, test, postop

Every OMI of any size splits into three phases, and people outside the business generally assume the middle one is the whole job. It isn't. The middle one is the short part.

Preops – configure for test

You put the vehicle and the ground systems into the exact state the procedure requires. For V1202 that meant jumpering the topping and hi-point bleed hydrogen sub-system to the LH2 manifold, setting up one aft PVD vent door with its HGDS sense port, connecting the aft delta pressure gauge, sealing any aft opening left over from missing hardware, running PVD purge ducts into the Orbiter/ET disconnect areas where mixing is poor, setting up the helium injection panel and test hoses, configuring the SSMEs to lockup, and installing blanking plates. The valves / system was configured so that when the fill/drain line was pressurized from the MLP (Mobile Launch Platform) no valves had to be opened or closed. The valves were operated by ... HELIUM (pneumatic) so bringing up MPS 750 Helium to open or close a valve added extraneous helium into the aft which is what you were trying to detect and would mask any leaks.

Then QC verifies it. Then MPS engineering walks the aft down and verifies the configuration independently. Two sets of eyes that are not the eyes that did the work, because the person who installed it is the worst possible person to confirm it is installed correctly. Everybody knows this and everybody does it anyway, every time, and that is not bureaucracy – that is the entire reason the thing worked.

Preops ran on second shift. They had to. The test needs the aft cleared of everybody and then sealed, and on day shift the aft compartment is contested real estate with several groups wanting access at the same time. Third shift is when you can actually have the aft to yourself to actually run the test. For the audio tape testing and preops ran late and some of the preops were pushed to third shift. The start of the test was about 4 hours late, which is why OTC was a little irritated.

Test

Before anybody pressurises anything:

Then you run it. Baseline helium background, pressurise each system, watch the aft concentration delta, calculate leakage. In the early version this involved pre-test injection runs at 6, 12 and 25 scim on three separate test hoses to build a calibration curve, then plotting the system delta against that curve to read off a worst case leak rate.

Postops – deconfigure

Everything you did in preops, backwards. Blanking plates out, flexhoses off, jumpers removed, purge ducts pulled, panels disconnected, vent door restored. And then QC verifies THAT, because a test rig left installed in the aft is a flight hazard and it does not care that the test passed.

Deconfiguration gets treated as an afterthought by people who have never done it. It isn't. More hardware has been left where it should not be during a cleanup than during a setup, because setup has everyone's attention and cleanup happens when everybody is tired and the interesting part is over.

Who is on the loop

Full list on the OIS channel page. For V1202 specifically:

ChFunctionCallsignWhy they are here
132Orbiter Test ConductorOTCMonitors / Runs the procedure. Calls to stations, gives the go, directs the below groups
168MPSCMPS / CMPLRuns the procedure and the test, declares sucess or not
146SSME / SSMECCSME / CCMEEngines configured to lockup, hot gas pressure supplemented, ball valve leak checks fold into this test
177HGDSCHGDWhere the actual measurement comes from. No HGDS, no test
242ECS-2 (Purge)CPVDThe aft purge is the transport medium carrying leakage to the sense port
166LH2CLHYHydrogen side
156LO2CLOXOxygen side
232NASA Test DirectorNTDAbove the OTC. Owns the pad
118Quality The buyoffs

S0017 and why V1202 always rode along with it

OMI S0017 is the Terminal Countdown Demonstration Test – the launch dress rehearsal. Flight crew at the pad, countdown team on console, the real S0007 sequence worked all the way down to a simulated T-0. Simulates everything a launch countdown does except load cryo's and light engines.

Three separate things get exercised in the same run, and it is worth keeping them apart in your head because they have nothing to do with each other:

The run terminates a few seconds before T-0 and usually picks up some contingency work in the same window – a simulated abort, a hold and recycle, sometimes a mode-specific bailout case.

The pairing

V1202 and S0017 were ALWAYS run together. The preops (see above) for S0017 were run on second / third shift, the crew loaded in at the beginning of first shift. Not coincidence, not scheduling convenience – MPS work is disruptive. When the aft has to be sealed and cleared and the whole compartment turned into a control volume, everybody else is out. That is a large hole punched in the pad schedule and you do not punch it twice if you can punch it once. Sometimes the preops and testing for S0017 and V1202 testing ... collided and had to be worked out.

So the disruptive things got stacked into a single window. TCDT already clears the vehicle of anybody who does not need to be there, because you have a crew aboard and hazardous operations running. V1202 needs an empty, sealed aft. Same requirement wearing a different hat. Pair them and you clear and plan for disruption of work on the vehicle once instead of twice.

If you are listening to these tapes, this is the thing to know: S0017 and S0007 sound nearly identical on the loop. Same procedure structure, same callouts, same console voices, same channels. The tells that you have a TCDT and not a launch are the absence of tanking traffic, a run that is suspiciously clean, and a T-0 that produces precisely nothing. Somebody who finds a mislabelled S0017 tape and thinks they are listening to a launch countdown will draw a whole set of wrong conclusions from it, and will have no reason to suspect they are wrong.

How it got better

Over roughly twenty five years V1202 went from about 36 hours to about 6, and got MORE accurate while doing it. That is not the usual direction and it is worth saying out loud.

Note what actually drove that. Not a clever idea. Twenty five years of running the same test on the same vehicle and paying attention to the results. Test repeatability is what let them delete steps safely, because they had the data to prove the steps were not buying anything.

The voice tapes

These are recordings of the loop while the procedure was being worked. Not documentaries, not narrated, no context provided – just people doing a job and talking to each other about it.

Documents

What I don't have: the V1202 OMI itself. I have the tapes of us working it, the engineering papers describing it, and the figures – but not the procedure document. Same for S0017 Volume 1, the preops. I have S0017 VL2 (which contains the Call To Stations) and S0017 VL3 If you have a copy of V1202 or S0017 VL1, or know where one survived, get in touch. A recording of people working a procedure is a great deal more useful when you can read along.

Figures

Scanned V1202 figures. Straight off the paper, no cleanup.

V1202-001 V1202-001A V1202-002 V1202-003 V1202-004
V1202-005 V1202-006 V1202-007A V1202-007B V1202-008
V1202-009 sheet 1 V1202-009 sheet 2 V30-15358 F sheet 2

Related procedures

Written by Ken Hollis, who worked these procedures at Kennedy Space Center. Where this page states something that comes out of the documents linked above, it comes out of the documents. Where it comes out of my head – the third shift scheduling, the QC and MPS walkdown sequence, what happens on which loop – I have said so, and you should treat it as one man's recollection thirty years on rather than as a citation. Corrections welcome.

If you share a link to any of the documents above, right click and copy the link FROM THIS PAGE. Internet Archive will redirect you to whichever server currently holds the file and that address changes. Every link you pass on should start with 'https://archive.org/download/GandalfDDI-SpaceShuttleDocuments/'.



Shuttle Page