Magnetic Tape Drives Are Back In Style

Magnetic tape to the rescue, Economist
“The need for mass storage is reviving a technology which, only a few years ago, seemed destined for the scrapheap: magnetic tape. Tape is the oldest computer storage medium still in use. It was first put to work on a UNIVAC computer in 1951. But although tape sales have been falling since 2008 and dropped by 14% in 2012, according to the Santa Clara Consulting Group, tape’s decline has now gone into reverse: sales grew by 1% in the last quarter of 2012 and a 3% rise is expected this year. Alberto Pace, head of data and storage at CERN, says that tape has four advantages over hard disks for the long-term preservation of data. The first is speed. Although it takes about 40 seconds for an archive robot to select the right tape and put it in a reader, once it has loaded, extracting data from that tape is about four times as fast as reading from a hard disk.”

Lunar Orbiter Image Recovery Project (LOIRP) Status 27 November 2013

Dennis Wingo: Status, Wednesday morning November 27, 2013
Back at it today. I only did three tapes yesterday as they were the ones (M5-063, M5-100, and M5-101) that we had been baking for several days to help reduce the head clogs. That failed and it took several hours to transfer the image data without too many gaks in the images.
That is done and so I can start back with M5-116 today and move forward.
Status Wednesday afternoon, November 27, 2013
Things have gone well today. Lots of long captures that help to close the gap in our images. We are well past 75% of the Madrid tapes (running M5-122 out of 158 right now), and well past 90% of the total Lunar Orbiter V captures. Just today we have captured almost a thousand frame lets. Here is the list since my last update. Including the ones from before as well from the baked tapes to complete the update.
M5-063, partial capture LOV-201H (completes image with overlap on M5-060)
M5-100, complete capture LOV-113M, 114H
M5-101, complete capture LOV-112M, 113H
M5-102, complete capture LOV-111M, partial capture M5-112H (completes image with overlap with M5-034 and M5-035)
M5-103, partial capture LOV-111H
M5-104, partial capture LOV-111H (completes image with M5-103, and G5-037), complete capture 109M, 110H
M5-105, complete capture LOV-108M
M5-106, partial capture LOV-108H (overlaps with M5-033), partial capture M5-106M
M5-107, partial capture LOV-106M (two frame lets missing, partial overlap with M5-106), partial capture 107H
M5-108, partial capture LOV-107H (completes capture with M5-107)
M5-109, complete capture LOV-105M
M5-110, complete capture LOV-104M (duplicate to G5-111) partial capture LOV-105H (duplication and overlap with G5-111 that completes image)
M5-111, partial capture LOV-102M (completes image with overlap of G5-113)
M5-112, partial capture LOV-101M (completes image with overlap of G5-114)
M5-113, partial capture of a partial framelet of LOV-087M (completes image with an overlap of W5-138
M5-114, Not captured, duplicate of W5-138
M5-115, partial capture LOV-087H, 2086M (completes images with overlap of G5-035)
M5-116, partial capture LOV-086H (completes image with overlap of G5-035)
M5-117, complete capture LOV-084M, 085H
M5-118, complete capture LOV-083M, partial capture 085H
M5-119, partial capture LOV-085H (completes image with previous tape)
M5-120, complete capture LOV-082M, 083H
M5-121, complete capture LOV-081M, 082H,
M5-122, complete capture LOV-080M, 081H, partial capture 079M (completes image with overlap of G5-122)
M5-123, not needed, duplicates G5-122
M5-124, not needed, overlaps with W5-032
M5-125, not needed,duplicates G5-124
That ends the captures for November as everyone is off tomorrow until Monday. We don’t make our people work the holidays!

Lunar Orbiter Image Recovery Project (LOIRP) Status 25 November 2013

Dennis Wingo: Status, end of day November 25, 2013
Back at it today!
I moved over the past few days and I was on travel last week so did not get any tapes done last week.
Today I started back again and have completed M5-115. This leaves us 43 tapes + the three tapes that were baked to go to finish LOV image captures.
Tomorrow I will catch everyone up on the image count but suffice to say that we are still making progress.
Control track head update.
We found a control track head that looks fairly good and it is on the machine and working well right now. We have no idea how long it is going to hold out but what we will do is run it as long as we can while working on a backup plan to convert and Ampex Mark 15 head to one that we can use. It will require some machine shop work and a bit of design but if that works, we will get through our the rest of our captures.
A rough estimate is that we have about 330 tapes from LO-IV, and 290 tapes from LO-I to go. However, the LO-I tapes we will only be capturing medium resolution frames which are only 10 minutes per tape, thus it will be a lot faster. There are 11 high resolution frames that are good on LO-I and we will get those but the failure of the V/H sensor (image motion compensation) led to the loss of almost all high res images.
Onward Sancho!

Rediscovered Apollo Data Measures How Fast Moon Dust Piles Up

The Lunar Dust Detector, attached to the leftmost corner of this experiment package left by the Apollo 12 astronauts, made the first measurement of lunar dust accumulation. As the matchbox-sized device’s three solar panels became covered by dust, the voltage they produced dropped.
When Neil Armstrong took humanity’s first otherworldly steps in 1969, he didn’t know what a nuisance the lunar soil beneath his feet would prove to be. The scratchy dust clung to everything it touched, causing scientific instruments to overheat and, for Apollo 17 astronaut Harrison Schmitt, a sort of lunar dust hay fever. The annoying particles even prompted a scientific experiment to figure out how fast they collect, but NASA’s data got lost.
Or, so NASA thought. Now, more than 40 years later, scientists have used the rediscovered data to make the first determination of how fast lunar dust accumulates. It builds up unbelievably slowly by the standards of any Earth-bound housekeeper, their calculations show — just fast enough to form a layer about a millimeter (0.04 inch) thick every 1,000 years. Yet, that rate is 10 times previous estimates. It’s also more than speedy enough to pose a serious problem for the solar cells that serve as critical power sources for space exploration missions.

Continue reading “Rediscovered Apollo Data Measures How Fast Moon Dust Piles Up”

Lunar Orbiter Image Recovery Project (LOIRP) Status 15 November 2013

Dennis Ray Wingo Status November 15, 2013
It’s dead Jim.
Another control track head that we tried is essentially dead. Without a control track there is no way to synchronize the capstan to the head to the reel motors, thus no tape running even with good head tips. The Capstan head is a separate head from the main head and no one makes them anymore. We are looking for alternate sources of control track heads while mixing and matching our other control track heads to see if we can get one that will allow us to continue running tape.
No tapes were captured on Friday.

Lunar Orbiter Image Recovery Project (LOIRP) Status 14 November 2013

Dennis Wingo: Status, late afternoon, November 14, 2013
Lots of frustration today. The first couple of tapes had head clogs so bad they have to go into the oven to bake for a couple of days. Almost all of the tapes today have had some anomaly. Right now Ken has the head off the machine as the current tape (M5-110) dropped a piece of crap on the control track head and now it has to be disassembled and cleaned…
Well that was the good news, the bad is that the control track head on this head is dead and we have to start cannibalizing other heads to see if we can keep this head going…
Stepping back to explain.
There are actually four separate heads on this machine. One is the expensive one for reading video data that we have to refurbish often at great expense. Two more are for the audio track and for the timecode track. The fourth head is the control track head and it reads a signal off the tape that allows the servo control system to synchronize with the tape. This head is worn down to the nubbin on this head and Ken is cannibalizing another head to get this one going….

Lunar Orbiter Image Recovery Project (LOIRP) Status 13 November 2013

Dennis Wingo: Not too productive today, lots of details to get right when we are at this point in the process. Made it to tape M5-099. Wanted to do 100 but it is head clogging so I said heck with it, deal with it in the morning.
Have not done a tape log in a while but here ya go. Many of these that are “completes image” are from tapes previously captured that were partial captures. This is the nice thing about LOV, we are going to get probably over 99% recovery of images from this mission, including dozens that are not on the LPI site.
M5-058, not needed, overlaps with W5-085
M5-059, partial capture LOV-202H (completes image with overlap W5-086)
M5-060, partial capture LOV-201H, complete capture 200M
M5-061, not needed, overlap with W5-085
M5-062, not needed, overlap with W5-086
M5-063, head clog, tape baking in progress
M5-064, complete capture LOV-199M, 200H, partial capture 198M
M5-065, partial capture LOV-198M (completes image), complete capture 199H
M5-066, complete capture LOV-197M, partial capture 198H (completes image with overlap of M5-056)
M5-067, complete capture LOV-196M
M5-068, partial capture LOV-194M
M5-069, partial capture LOV-194M (completes capture) complete capture 195H
M5-070, complete capture LOV-194M, partial capture 194H (completes capture with overlap of G5-071
M5-071, partial capture LOV-192M (completes image with overlap of G5-072)
M5-072, partial capture LOV-191M (completes image with overlap of G5-073
M5-073, not needed, overlap of W5-098
M5-074, complete capture LOV-173M (overlap and duplicate of W5-063), partial capture of 172H (completes image with overlap of W5-063), complete capture 171M
M5-075, not needed, overlap with W5-063
M5-076, not needed, overlap with W5-064
M5-077, partial capture LOV-170H (completes image with overlap of W5-065)
M5-078, complete capture LOV-168M, partial capture 169H (completes image with overlap of G5-056
M5-079, partial capture of LOV-167M (completes image with overlap of G5-057)
M5-080, partial capture of LOV-166M (completes image with overlap of G5-058)
M5-081, partial capture of LOV-165M (completes image with overlap of G5-059)
M5-082, partial capture of LOV-164M (completes image with overlap of G5-053), partial capture of 165H, (completes image with overlap of G5-053)
M5-083, partial capture of LOV-163M (completes image with overlap of G5-054)
M5-084, partial capture of LOV-162M (completes image with overlap of G5-055)
M5-085, not needed duplicate of G5-086
M5-086, not needed duplicate of G5-087
M5-087, complete capture LOV-142H
M5-088, complete capture LOV-140M (overlaps with G5-049)
M5-089, not needed overlap of G5-044
M5-090, partial capture of LOV-137M (completes image with overlap of G5-045)
M5-091, partial capture of LOV-138H (completes image with overlap of xxx (don’t have info woomera image)
M5-092, complete capture LOV-136M
M5-093, not needed, overlaps with M5-040,041
M5-094, not needed, overlaps with M5-041
M5-095, not needed, overlaps with G5-098
M5-096, not needed, overlaps with G5-099
M5-097, partial capture LOV-132M (completes image with overlap of G5-100)
M5-098, not needed, overlap of G5-101, M5-037
M5-099, not needed, overlap of G5-102…
Whew, that is a lot of work over the last few days, even with the ones that we did not need to digitize as we have to figure out what is on the tapes and the only way to do that sometimes is to compare times with the other ground stations or play them.
Tomorrow we should pass the half way point with the Madrid tapes, slightly ahead of schedule.

Madrid Ground Station Crew Says Hello After 46 Years

Dennis Wingo just discovered a message written inside one of the Madrid ground station Lunar Orbiter 5 tapes today – tape M5-87 – recorded on day 233 of 1967 (21 August). It is blurred with time but says “Greetings from DSS-62 (indecipherable) in SPAIN (indecipherable)” DSS stands for “Deep Space Station”. Some people put messages in a bottle and throw the bottle into the ocean. These guys used a tape container lid to send us this message. No one has opened this container since 1967.
The DSS 62 antenna was on a polar mount and was 85 feet across. It operated from 27 December 1966 to 30 November 1981. DSN’s Deep Space History webpage notes “After NASA declared the station surplus to its operational needs in 1983, it was donated to the Spanish government”.
This photo posted by Robert Estalella shows several FR-1400 tape drives (but not the FR-900 drives).

Listening to the Deep Space Music Network

Dennis Wingo: A funny story from today. I was running a Lunar Orbiter tape today and all of a sudden I started hearing music coming from the audio speaker. It was really nice, staring out with a piano solo and then a couple of other pieces then a full on concert by Herb Alpert and the Tijuana Brass.
It was really cool to hear this old sixties music coming across the deep space network. Even funnier is that about ten minutes after it started, Ken Zin wandered over and asked if that was coming from the speaker. We laughed about it and then a little while later our new student Marco came out and was wondering where the hell the music was coming from.
We all had a good laugh today listening to the Deep Space Music Network!

Lunar Orbiter Image Recovery Project (LOIRP) Status 11 November 2013

Status, near end of day November 11, 2013
Christopher reminded me with his question to note something to the group. We literally have dozens of Lunar Orbiter V images that are not in the LPI database at I, like most everyone else that deals with lunar images loves their dataset and have used it ever since it came on line in the 1990’s. Without their work, our work would have been ten times harder, especially during our learning period.
This may be our payback to them in that we welcome them, when we get our images done, to mirror the ones that we are capturing now. There is a lot of interesting stuff in these images. I have no idea why they are not in the LPI database but soon we all can see these marvelous 45 year old images!
We had some computer problems this morning and we have had some pretty long captures so have not done a ton of tapes today. We will end at M5-069. We are officially on the final readout and Austin has finished our handy dandy notebook that incorporates all of the frame lets captured so far from Goldstone and Woomera that makes my job of avoiding a lot of overlapping captures easier.
Today we have…..
M5-058, not needed, overlaps with W5-085
M5-059, partial capture LOV-202H (completes image with overlap of W5-086)
M5-060, complete capture LOV-200M, partial capture 201H
M5-061, not needed, overlaps with W5-085
M5-062, not needed, overlaps with W5-086
M5-063, head clogs, will rerun after baking the tape
M5-064, complete capture LOV-199M, 200H, partial capture 198M
M5-065, partial capture LOV-198M (completes image), complete capture 199H
M5-066, complete capture LOV-197M, 198H
M5-067, complete capture LOV-196M
M5-068, partial capture LOV-194M
M5-069, partial capture LOV-194M (completes image), complete capture 195H
Status, morning of November 11, 2013
Here is the tape recap for the last few days.
M5-023, not needed, overlaps with W5-145,146
M5-024, not needed, overlaps with W5-143
M5-025, not needed, overlaps with W5-143,144
M5-026, not needed, overlaps with W5-134,135
M5-027, not needed, overlaps with W5-135,136
M5-028, not needed, overlaps with W5-139, G5-032,033
M5-029, partial capture, LOV-106H (completes image, overlaps with G5-111)
M5-030, not needed, overlaps with G5-111
M5-031, partial capture LOV-109H
M5-032, partial capture LOV-109H (completes image)
M5-033, complete capture LOV-107M, partial capture 108H
M5-034, partial capture LOV-112H
M5-035, partial capture LOV-112H (completes image), complete capture 110M, partial capture 111H (overlaps with G5-037 [I think])
M5-036, partial capture LOV-132M, complete capture 133H (overlaps with G5-100)
M5-037, partial capture, LOV-131M (completes image with an overlap with G5-101)
M5-038, partial capture LOV-130M (completes image with an overlap with G5-102)
M5-039, complete capture LOV-137H
M5-040, complete capture LOV-135M,136H
M5-041, complete capture LOV-134M, 135H
M5-042, not needed, overlaps with G5-098
M5-043, not needed, overlaps with G5-044
M5-044, captured earlier in 2013, Copernicus megaimage
M5-045, captured earlier in 2013, Copernicus megaimage
M5-046, captured earlier in 2013, Copernicus megaimage
M5-047, captured earlier in 2013, Copernicus megaimage
M5-048, captured earlier in 2013, Copernicus megaimage
M5-049, captured earlier in 2013, Copernicus megaimage
M5-050, complete capture LOV-162H (overlaps with G5-086)
M5-051, partial capture LOV-160M (completes image with an overlap with G5-087)
M5-052, partial capture LOV-160H (completes image with an overlap of two G5 tapes G5-088,089)
M5-053, not needed, overlaps with G5-077
M5-054, not needed, overlaps with G5-078
M5-055, partial capture LOV-185M (completes image with an overlaps of G5-079)
M5-056, complete capture LOV-196M, 197H
M5-057, complete capture LOV-195M, 196H
As those who read this page often can see, my surmise was right about being able to skip many Madrid tapes. Many of the tapes that I played that are partial captures of one medium image only required the capture of a few frame lets as many times those were missed on the other ground station captures.
All of the tapes above for Madrid were from the priority readout, that is when the film is moving from the lower numbered to the higher numbered images as further images were being taken. Today with the tapes we will start reversing this to the final readout where the image numbers will decline. This means that there are about 100 tapes for the final readout, meaning that there will be a lot of overlaps not only with Woomera and Goldstone, but from previous Madrid tapes as well. Knowing the mentality of contractors this was by design only because they figured that they had so much money and time left on the contract and that this would be a good way to maximize the acquisition of funds from the government. I would bet quite a lot that this is why LOV has 150% of the tapes of the previous missions as it is easy to see that there is a lot of redundant data here. Not to complain from a historical context, but interesting to note nonetheless.

Lunar Orbiter Image Recovery Project (LOIRP) Status 8 November 2013

Dennis Ray Wingo Status, end of day November 8, 2013
Had a lot of visitors today so got tapes done but will get the report done on Monday. We finished M5-057 at the end of the day so we are almost 40% of the way through LOV tapes!
On Monday we will be fished with the priority readouts and will be doing the final readout for Madrid tapes. Austin finished the handy dandy reference book for what frame lets we have left to go. They don’t have the Madrid captures in there though so we will get that done on Monday. We also have ordered another 16 Terabytes of hard drive in order to upgrade our RAID array that was out of space.
LOTS of frame lets, 25,000 just for Woomera and Goldstone for LOV.
Have a nice weekend!