Larry – Ampex, in Redwood City, has shrunk considerably. I don’t
know where all of the “old” equipment and documents are. My only
contact with the company is through Debbie Fuentes of Human
Resources. Her phone number is: 650-367-3013. You might try
calling her about locating some of the engineers that worked on the
different tape recorder system that you are interested in. A couple of
names come to mind: Hal Wright (mechanical engineer) and Ernie
Sorenson (electrical engineer). Debbie might know how they may be
contacted. Availability of NASA technical publications can be obtained
from: National Aeronautics and space Administration, Code ATU,
Washington, D.C. 20546. (1966 info)
Clay Hedman

Michael, funny you should mention Pournelle. I know Jerry fairly well and was with his son Rich this past weekend in Arizona. I also know many of the players in the book “fallen angel” as well, including Gary Hudson.

Yep, we are familiar with a lot of that info. I was a test engineer on those older Cray super computer disk drives and have storied to tell about getting the old heads back from TLA’s.

We are in contact with some of them as we have relationships there. Right now we probably have some of the best capability for recovering the type of data that we have in the world.

Thanks for that. As I replied before, we are kinda funding limited here so we have to use what we have in the best way that we can. One thing that we have done is to fix up the variable timing of many of the circuits in the system that have discrete transistor flip flops. We have been substituting 1970′s CMOS chips in many of the timing circuits with good results. There is also a subtle feedback loop between the RF system and the servo system so we have to have both of them functioning well in order to get the head swath to tape timing correct.

We are using PCIe digitizers on a Mac Pro workstation to capture our data so we are able to get high data rate captures up to 180 megasamples per second.

We can do a lot with that but we would have to do a LOT of digital post processing including doing the demodulation in the digital domain, which is NOT going to be easy. Much easier to do in hardware, especially since we have spent the money doing so.

I will be answering some of you guys this week as I am back in the lab now.

Thanks again

Dennis

“. . .version II of the PCIe bus which increases the bus speed to 10GBps.”

PCI Express SDR boards feature direct FPGA connectivity
SuperDSP.com
Apologies

Following is some thinking “outside the box” that might help your data recovery.
In recovering the images, each step in the process adds noise and/or changes the signal by some function. While very good for their day, the Ampex F900′s read and amplification electronics add noise and some distortion to the signals recorded on the video tape.
Tim Channon above notes above:

The image is 37,134×16,200 at 16 bit greyscale, a tad larger than most home monitors.

I’ll presume you are working with or would like 16 bit gray scale per RGB channel. Modern digitizers readily handle 16 bit per channel at video speeds. (At slower speeds A/D’s are available with 24 bit.)

Video A/Ds May I recommend digitizing the video signal directly at the tape read heads with as little intermediate amplification as needed to feed the A/D. This would help by eliminating most headaches of trying to maintain the playback read electronics. Following are some quick sketches of some potentials for using modern digital electronics to replace the Ampex read electronics.

PCI Express Bus

PCI Express supports 1x [2.5Gbps], 2x, 4x, 8x, 12x, 16x, and 32x bus widths [transmit / receive pairs]; 2.5Gigabits/second per Lane per Direction. The 8B/10B changes the data transfer numbers to 250MBps per lane, raw data [B= Bytes, b=Bits]. The reduction in throughput is accounted for under the protocol section.”

“. . .version II of the PCIe bus which increases the bus speed to 10GBps.”

Some examples of 16 bit video data acquisition. E.g. for a PCIe bus Pentek’s Quad 200 MHz, 16-bit A/D Data Converter

· Four 200 MHz, 16-bit A/Ds · . . .Multi-module synchronization capability · Variety of IP cores to match specific requirements · Built-in data capture modes”

PCI Express SDR boards feature direct FPGA connectivity
Alternatively see ADLink’s 16 bit video A/D on a PXI bus: PXI-9816/9826/9846

4-CH 16-Bit 10/20/40 MS/s Digitizer with 512MB SDRAM
. . .

Such higher frequency sampling would allow immediate signal processing to remove some noise and improve recovery of the signal on the tape. E.g. sample at 200 MHz and four fold average down to 50 MHz (4 fold sum and shift two bits.)

At slower tape speeds, you could use 24 bit A/Ds with more averaging. E.g., Innovative Integration X3-SDF

PCI Express XMC Module, (4) 24-bit, Fast Sigma-Delta A/D >110 dB, 1M FPGA, 4 MB Memory

Analog Devices AD7760: 2.5 MSPS, 24-Bit, 100 dB Sigma-Delta ADC with On-Chip Buffer

# 112 dB SNR at 78 kHz output data rate, # 100 dB SNR at 2.5 MHz output data rate

Capstain Rotary EncoderThe corresponding improvement on tape speed would be to add an encoder to the capstan and synch the A/D to the actual capstan speed. That would reduce motor speed frequency control issues. It would only require achieving a “fairly” stable tape drive. The rest would be handled by the encoder and signal processing.

Processing: The rest would be “just” a matter of some signal processing.
E.g., See SuperDSP for 100 GFlop DSP cards with 50 GB/s bandwidth. In 2008, a single XP-100 card (100 Gflop) listed for $20,000 and a single XP-30 (30 GFlop) listed for $12,000. They note that the PCIe bus can handle (x8, 1.5 GB/s).

There are probably a few instrumentation experts and software types around who could help with such data acquisition and signal processing.

Such a combination of high resolution video data conversion, capstain rotary encoders, precision frequency generator, and signal processing would eliminate most electronics issues in the Ampex recorder, while improving the accuracy of reading the lunar images.

Happy hunting.

PS If you haven’t already seen it: Preliminary Guide to the Ampex Corporation Records

I would also contact some of the TLA agencies … (Three Letter Acronym… you know, NSA, CIA, FBI, …) and ask if they could recover the data… they have ‘capabilities’ in a lot of media…

One of the “odd stories” our Cray support guy told us was of entering a TLA agency site where you stripped and put on their clothes and used there tools. Nothing other than bare body in, or out. Old disk drives were simply pushed to a corner. More trouble to assure no data could be recovered than to just warehouse them… If one of these went in, the odds would be good that it never came out. Getting access might be problematic, but they might read the tape for you…

(Our sales guy had stores of leaving a flat bed truck with a Cray on it at a designated location and picking up a different flatbed with money on it the next day… don’t know if he was just pulling our leg or not, but he seemed serious. And a TLA agency would not want folks to know what ‘capabilities’ they had or what they were buying, so it makes sense.)

Finally, on Shoreline / Amphitheater (drive?) in Mountain View is the Tech Museum with things like one of the original Apple I machines. They might have one (or have a network of contacts.)

Mike

Send your request through Anthony or Charles the moderator guy. Send me your address and I will get it.

BillP

Have all the h-r images been seen in the NASA labs? In what format?

Bill

No, they have not, ever. In the Apollo era, no computer could display the images at full resolution. They digitized 6 bit 800 x 800 “chits” that were used to do the final Apollo site selection. As far as I know, NONE of these captures ever made it into the public domain.

We will be doing more soon, next week as a matter of fact.