Is it possible to convert a HB?

[dwight]

Oh, I have a beauty of an anecdote which appeared just last night. In a fit of boredom, I decided to visit a few YT Apollo discussion threads. I am a parrot because I used the word "pertaining", and apparently the accuser HB also used the word previous to my usage.

I wish I was joking.

[Luke Pemberton]

Oh, I have a beauty of an anecdote which appeared just last night. In a fit of boredom, I decided to visit a few YT Apollo discussion threads. Apparently I am a parrot because I used the word "pertaining". That is because the accuser HB also used the word.

So let me get this right. An 'HB' used the word, then you used the word, and they said you were a parrot? So now it's against the rules to repeat words they use?

[dwight]

You got it on the nosie there Luke.

Best of all I used it because it was appropriate to the discussion. I didn't even know he had used it already.

[dwight]

Just to add: It is like I say. I don't really need to do anything. The HB crowd make themselves look foolish all by themselves, especially with whoppers like this.

To top it all off, the HB was accusing everyone and ther dog of being paid shills. I offered my standard "What employment benefits do shills get?" routine, and the guy just ran with it. He claimed that I must be naive to think the CIA, SIS etc would be transparent enough to list employee benefits. I decided to check and lo and behold both the CIA and SIS specifically list benefits on their homepages. Our HB friend didn't like that owning, much.

[Dr.Acula]

Just to add: It is like I say. I don't really need to do anything. The HB crowd make themselves look foolish all by themselves, especially with whoppers like this.

To top it all off, the HB was accusing everyone and ther dog of being paid shills. I offered my standard "What employment benefits do shills get?" routine, and the guy just ran with it. He claimed that I must be naive to think the CIA, SIS etc would be transparent enough to list employee benefits. I decided to check and lo and behold both the CIA and SIS specifically list benefits on their homepages. Our HB friend didn't like that owning, much.

After reading the nonsense from youtubers IDW and/or humanmonkeyrace and several others... ehm... say mental challenged, I don't wonder about any stupidity 

[ka9q]

AIUI that is not what SCE to Aux did. My understanding is that the auxiliary signal conditioning circuits were designed to be less sensitive to low voltage conditions, and switching to them allowed mission control to get their telemetry back.

I checked the handbook again to verify my understanding. There's only one SCE but two power supplies, normal and aux. The switch in question is in a group labeled "Power", and according to the text it manually changes power supplies. Automatic switching can also happen if the normal power supply outputs go out of tolerance.

Since automatic switching did not occur, it seems reasonable to assume that the normal SCE power supply didn't actually fail. It simply went out of tolerance briefly because of the glitch in +28V supply power, putting the SCE into a strange state. The NORM/AUX power supply switch is the only SCE switch available to the crew, so Aaron must have thought that switching to the aux power supply would interrupt power to the SCE long enough to let it go through its proper power-on reset sequence when power was reapplied from the aux supply. Or at least that's what I would have thought had I been in his position and not simply panicked.

Lots of electronic hardware (especially, but not limited to computers) includes a "power on reset" circuit designed to force the hardware into a known starting state. Typically this is a simple analog RC (resistor/capacitor) timer that provides a reset pulse lasting for some number of milliseconds after power is applied. A short glitch in supply power is not always long enough to trigger this reset timer even though it may be long enough to scramble the state of the electronics. This is why turning power off and on again so often works. The idea is to keep it off long enough to ensure that the power-on reset timer fires reliably when powered back on.

Why the glitch in +28V supply power? Although CM entry batteries (A&B, I think) were on the bus, they were not supplying power; that was all coming from the fuel cells. When all three fuel cells tripped together because of falsing of their overcurrent protective relays by the lightning pulse, the entire CM load (something like 70 amps) was abruptly thrown onto the batteries. Silver-zinc batteries cannot ramp up current instantaneously, so the bus voltage plummeted for a few milliseconds before recovering to a low but acceptable value. This is why the computer reset and the SCE got scrambled even though the bus voltages read normal when Bean checked them a few seconds later. It certainly explains his confusion.

[smartcooky]

When all three fuel cells tripped together because of falsing of their overcurrent protective relays by the lightning pulse, the entire CM load (something like 70 amps) was abruptly thrown onto the batteries. Silver-zinc batteries cannot ramp up current instantaneously, so the bus voltage plummeted for a few milliseconds before recovering to a low but acceptable value. This is why the computer reset and the SCE got scrambled even though the bus voltages read normal when Bean checked them a few seconds later. It certainly explains his confusion.

So they expected the computer to reset and when it didn't, they suspected it was fried, but Aaron figured that switching SCE to Aux would manually restart it? This is what he saw that triggered the memory of something he had seen happen in a single simulation a year earlier?

With all the hundreds of failure and multiple failure simulations these crews and controllers did, Aaron remembered this one?

[Sus_pilot]

And, IIRC, like the 1200-series errors, they hadn't done it very often.

[JayUtah]

Lots of electronic hardware (especially, but not limited to computers) includes a "power on reset" circuit designed to force the hardware into a known starting state.

Yes, a substantial number of CSM and LM systems had this, which is why cycling breakers was often the first recommended action upon the indication of failure.  A few key systems did not, leading to the Caution and Warning system going nuts for a few seconds when the CM was first powered on.  Engineers never did figure out how to prevent this, but the crews easily got used to it.  One of them said something to the effect that cars at the time did this and no one cared.  You powered on the CM and then after a few seconds cleared the C&WS with MASTER RESET.

[ka9q]

So they expected the computer to reset and when it didn't, they suspected it was fried, but Aaron figured that switching SCE to Aux would manually restart it? This is what he saw that triggered the memory of something he had seen happen in a single simulation a year earlier?

No, the SCE and computer were entirely separate. The SCE, signal conditioning electronics, is part of the telemetry system. I haven't seen the details of the Apollo SCE, but if it's like others I've seen it contains a bunch of analog amplifiers and multiplexors that takes native signals from many different sensors, selects one for transmission, and converts it to a standard voltage range (typically 0 to +5V) so it can be digitized and sent to earth in the PCM (pulse coded modulation, i.e., digital) telemetry stream.

Those sensors seemed to be mostly independent of those feeding displays in the cockpit, giving the crew and the ground the ability to back each other up. In general they did not go to the computer, as it was used mainly for guidance tasks, not systems monitoring. (Today of course there wouldn't be just one onboard computer but many, each assigned a specialized function that Apollo did without them. And there wouldn't be a single "SCE" box, as its functions would be farmed out to the individual systems that generate the signals to be monitored. They'd communicate over a shared bus with the communications system.)

The Apollo 12 CSM computer was not expected to reset during launch. It did for the same reason that the SCE got confused: a momentary steep drop in +28V primary supply voltage when the fuel cells dropped offline and the entry batteries had to assume the entire CSM load. Apparently the computer's power-on reset did work properly, or like the SCE it would have required manual intervention.

With all the hundreds of failure and multiple failure simulations these crews and controllers did, Aaron remembered this one?

I can easily see this kind of failure happening many times during testing, and Aaron was around long enough to see and be familiar with it.

[ka9q]

Yes, a substantial number of CSM and LM systems had this, which is why cycling breakers was often the first recommended action upon the indication of failure.

The classic example being the descent radar on Apollo 14. The landing was saved almost at the last second by asking Mitchell to cycle its power. The radar had two range settings, one for high altitude and another for lower altitude just prior to landing, and it seems to have incorrectly powered up in the low range setting. Pulling and pushing the circuit breaker sent it back to the intended power-on state, the high altitude mode.

So, once again, somebody at NASA becomes a steely-eyed missileman for giving the same advice that IT "help" desks give thousands of times a day. :-)

[ka9q]

leading to the Caution and Warning system going nuts for a few seconds when the CM was first powered on.  Engineers never did figure out how to prevent this, but the crews easily got used to it. 

Seems to me you'd want all the alarm lights to go on at power-up so you can easily see if any bulbs are burned out. Just like today's cars light all their idiot lights when first turned on.

Burned out bulbs are much less of an issue with LEDs, but it still seems like a good idea to build a self-test function into the alarm system.

[JayUtah]

Seems to me you'd want all the alarm lights to go on at power-up so you can easily see if any bulbs are burned out.

That's usually a separate lamp-test button or switch.  Not only do you want that at power-up, but you want it any any time during the mission to see whether indicators are working.  That's what Pinball (AGC user interface software) was first programmed to do.  Its more sophisticated functions arose later.

[Kiwi]

In a fit of boredom...

But I thought you were newly married.  How could that possibly happen? 

[Kiwi]

This forum is marvellous. It's great how a small remark like...

There's one fascinating bit of minutiae above that's probably not very well known.

...can produce so much wonderful information from knowledgeable members. Thank you all.

Yesterday a series of links put me in YouTube and part of a documentary "Failure Is Not an Option", where John Aaron describes what happened regarding "SCE to Aux." It finishes with the sound of Pete Conrad elatedly cackling all the way into orbit once the problems were fixed. A documentary I saw years ago played even more of his laughter, so if it was this one, they've edited out some of it.

John Aaron appears much more on the brilliant CD-ROM "Apollo XIII -- A Week to Remember" by Odyssey Interactive, 1995 -- for IBM and MAC.  It would be great if they could release it again in a modern format with higher-res graphics on DVD, but I guess the market isn't all that big. I would want a better version of it. Must see if the CD-ROM still works in my current computer.