Nope. NASA's engineers are. Devices used beyond the manufacturer's Safe Operating Area are not guaranteed to work properly. There are multiple levels at which this should have been spotted and corrected.
Mechanical switch contacts like the thermostats in question can overheat, arc, and/or pit when used beyond spec. ...now add in a liquid oxygen environment.
There was a political cartoon at the time of the Apollo 1 fiasco. It showed a wrench dripping blood. The wrench was labeled "American craftsmanship". Same deal there: 100% oxygen atmosphere. You'd think those guys would have learned some respect for O2.
Contacts to "blame"? No. Contacts failure, yes. The fact that is that people on the ground operated the contacts outside their design parameters. This damaged them. They then failed under flight conditions. Which is pretty much what the wiki article states.
** Dunno exactly what NASA used for a thermostat - but the usual kind that operate via a bi-metal disk and attach to things like heatsinks have contacts that open about 1mm.
These will reliably switch 240 volts AC at up to 10 amps OR 24 volts DC at the same current.
Increase the DC voltage to 65 volts and at 10 amps a * continuous arc * will form across the contacts as soon as they open - every time they open.
To reliably switch 65 volts DC at 10 amps requires a rather large open gap OR a strong magnetic field to pull the arc apart OR that a physical barrier be interposed between the contacts.
Contacts are rated for arc quenching capacity, so the main design factor is the DC Voltage rating. The poles of them are almost always capable of handling 5 times what the contact rating is for current,at least, so the rating is all about the span of the gap that is created when the contacts are "open", and the size of the plasma arc that occurs when opening such a switch at full rated capacity. The AC rating will always be far higher than the DC rating for voltage, and current is usually along the lines of being double the capacity as AC crosses zero often enough to quell the arc immediately, even with a tighter gap. DC arcing can transfer contact material from one contact face to the other under certain conditions. Lots of little X-ray emissions with each make/break too... heh!
Complex switches and circuit breakers have non-conductive 'blades' which pass between the contacts as they open. Some blow a wisp of gas across the gap as the switch opens.
There were, in fact two thermostats, one in series with each heater, each heater circuit protected with a 15A circuit breaker, and supplied from 28 volt main DC bus B.
The thermostats were of the bimetallic disk type, with a .015 inch to .020 inch gap when opened.There's a drawing (Fig. D3-14) in Appendix D of the report.
The manufacturer's rating for the devices used was 100,000 operations at
1 amp, 240 volts AC, or 5,000 operations at 3 amp, 240 volts AC.
"No requirements are imposed on the manufacturer for acceptance test verification of the operational characteristics of the thermostat with respect to current-carrying capability or ability to open under load at any of the several voltages (65 V dc, 65 V ac, or 28 V dc) to which the thermostat will normally be subjected." (Appendix D, Page 38)
Tests subsequent to the incident showed that the contacts would fuse closed at 1.5A, 65V DC.
The report concluded that the contacts fused closed during an abnormal procedure for detanking, during preflight checks, necessitated by earlier displacement of internal tank components, allowing the temperature to rise to the point where insulation of the stirrer motors (115V 400Hz
3-phase, directly immersed in supercritical oxygen) was degraded. When the stirrers were run, one or both arced and ignited.
--
"For a successful technology, reality must take precedence
over public relations, for nature cannot be fooled."
(Richard Feynman)
The thermostat was designed to open at 80 +/- 10 degrees F, and close at
60 +/- 7 degrees F. That may account for the small gap.
The thermostats appear from the drawing to be hermetically sealed, and are stated as such. There are substantial glass-to-metal seals around the terminals, and the whole unit is enclosed in a metal cup, welded to the base. A similar construction to what we see in hermetically sealed relays. There is no mention of what gases, if any, were introduced prior to sealing.
However, the whole thermostat was immersed in supercritical oxygen at 900 PSIG. No mention was made of the possibility of oxygen ingress to the interior of the thermostat, or of deformation of the thermostat structure under pressure.
During the abnormal detanking operation,at KSC, each heater was drawing approximately 6 amps at approximately 65 volts DC. The heaters were operating for 8 hours and 10 minutes. the fans were operating for all but the first hour and 20 minutes. During this time, there is "no conclusive evidence" that the thermostats operated to shut off the heaters, despite the fact that the ullage (effluent) temperature went offscale (84F), and was still rising.
--
"For a successful technology, reality must take precedence
over public relations, for nature cannot be fooled."
(Richard Feynman)
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.