Click HERE for the usual bland content!
I guess I've been somewhat negligent in my blogging, but the scope of my retrochallenge this year is pretty wide and I've certainly gone off on a number of wild tangents. I'm putting together my scatter-brained notes into a single document in the hopes of presenting a coherent overview of the topic. Instead of trying to keep it all in the "blog format", I'm actually writing it up as a personal essay that you can read HERE.
It's very much a work-in-progress, but if you're a machochist you're welcome to read my scribbles... At least the first few pages!
Having tired of old computers, magnetic tape, and grungy biomedical telemetry gear, I'm taking a break to look at, play with, and drool over some truly vintage stuff at the Pioneer Steam and Gas Engine Society annual show in Saegertown, PA. Youtube video will follow...
It seems that, in the absence of modern DSP technology, any little sojourn into the world of analog data representation becomes an exercise in the construction of filters from discrete components. In this case, I'm not only having to build audio-frequency filters to process my analog telemetry, I'm failing miserably in the black art of tuning an RF duplexer!
The block-diagram of the ECG telemetry decoder is deceptively simple: a phase-locked loop is wired to produce an error signal proportional to the frequency shift of an audio tone -- the 'error' signal in this case is simply a voltage representing the electrical state of the human heart. The simple version works fine on a test signal generated by my computer's sound card, but it doesn't seem to like the (minimal) noise and distortion introduced in the radio system -- particularly when combined with simultaneous voice transmission.
Adding insult to injury, the APCOR is hopelessly out-of-whack, having sat on a shelf for a few decades and not giving me much to work with. As a portable full-duplex system, it has a particularly troublesome duplexer that allows the simultaneous transmission and reception of voice and ECG telemetry. Running this particular unit in full-duplex mode produces some interesting (and troubling noise) that makes me think that the duplexer isn't working as advertised. Life would be much simpler if I had appropriate test equipment! Right now, I'm just fiddling around in the dark!
As promised, I've been looking into the issue of analog telemetry of the sort used during the early space program. Much of the information was conveyed in the form of an audio signal that could easily be transmitted over a simple voice-grade radio channel. While various forms of pulse-code modulation were also used, particularly during the Apollo program, earlier missions required a simple and robust system that was compatible with the communication technologies of the day -- typically off-the-shelf aeronautical communication systems using HF and VHF radio.
One of the easier systems to replicate using available period hardware is basic biotelemetry in the form of an electrocardiogram (ECG). Although other vital signs were easy to measure and transmit electronically, the ECG is a good baseline indicator of astronaut health. It's also visually appealing, immediately understood by the layman, and a pretty good test of a communication system.
An early spin-off of the Apollo program was the biotelemetry used by American Paramedics in the late 1960's through the mid-1980's. In this sytem, where paramedics work closely with "online" medical control, the patient's vital signs and ECG are relayed to medical control. There, a physician decides on appropriate treatment and relays orders to the paramedics by radio. From a technological standpoint, the keystone was the Motorola APCOR -- the Advanced Portable Coronary Observation Radio.
As the acronym suggests, the APCOR was a "portable" (sort of...) radio set that allowed you to literally wire up that cardiac patient and transmit his ECG to the receiving hospital. The radio also included a full-duplex (capable of receiving and transmitting at the same time) voice link that allowed the paramedic and physician to communicate in real-time as the ECG was transmitted and received. I personally had to carry one of these radios for the first five years of my career in EMS and would dispute the use of the term "portable!"
Anyway, the signalling method was pretty basic: electrical signals from the patient's heart were amplified and used to modulate the frequency of an audio oscillator running at around 1350 Hz (plus or minus 50 Hz to indicate the 1 mv change in potential of an average ECG.) This audio signal was transmitted over a voice-grade FM radio channel and received at a hospital telemetry station. At the receiver, the ECG telemetry was decoded using a phase-locked-loop to generate an error signal in proportion to the frequency shift. This error signal was displayed on an oscilloscope, creating an analog representation of the patient's ECG at the receiving hospital. The physician would take the paramedics verbal report over the same radio, review the ECG, and issue orders for the patient's care that were relayed, again, over the same radio link.
For purposes of this exercise, I've obtained a non-functional APCOR that I am attempting to restore to operating condition. Once this is accomplished, I intend to use a standard UHF portable radio to receive the signal, and hopefully decode that ECG telemetry with a vintage computer to display the transmitted ECG. In keeping with my theme of "fun with magtape", I will continue using appropriate vintage tape equipment to store programs, data, and recorded telemetry.
I couldn't do much in the way of retrocomputing over the last two days, I had to work a day shift on Thursday and that resulted in a bad experience with some anhydrous ammonia. It's really nasty stuff that finds occassional use in refrigeration and certain agricultural applications. I'm also told it gets used in the cooking of methamphetamine.
This particular incident involved a gentleman that was scrapping on old absorption-type refrigeration unit -- he managed to break a pressurized fitting that contained a nasty mix of anhydrous ammonia, hydrogen gas, and ammonia water. He took a face full of the stuff and inhaled a bunch that resulted in severe chemical burns to his eyes, face, and respiratory tract. I was dispatched simply to a "burn victim" without any explanation as to the mechanism of injury. The pungent smell alerted me to the danger when I arrived, but you still have a job to do. I had appropriate personal protective equipment, but that only offers a modicum of safety. The equipment is also cumbersome and makes the job ten times more difficult. The victim was rescued and carried to the hospital without incident, but the effort took a lot out of me and put me in a rotten mood. I'll try to get back into the swing of things over the weekend.
The following reference materials may be of use, or at least of interest:
Apollo Spacecraft Unified S-Band Transponder
I sit dejectedly in my hot and humid electronic scrap-heap of a house, sweating uncontrollable in my LOIRP tee-shirt and drinking cold, bitter coffee from my Wallops Flight Facility NASA mug, and wondering how it is that I've been upstaged as both a computer geek and a space cadet by Urbancamo...
How can I up the ante? He demonstrated a DECTAPE drive and made passing reference to LOIRP, I'm slowly piddling around with an Apple II... How about 9-track? Nah, been there & done that. The gear would technically qualify as "vintage" but it pretty much still works as described...
Okay, so maybe it's time to hook the 1" Ampex up to something and demonstrate how ANALOG telemetry worked back in the day. First, we'll do some SSTV and then maybe some biotelemetry. Yeah, that's the ticket!
I'll be back in a couple of days, folks!
Heck, if the USA could get as good a sideshow out of a head-of-state as good-old Brittania, it might be worth a few tax dollars! If you folks across the pond aren't going to use Bonnie Prince Charley, could we buy him?
Happy Independence Day, America! Canada, this is your chance -- HRH is visting Winnipeg this morning... No offense intended, England!
Okay, I've put the tape recorder back together and it "sorta" works, despite the high degree of corrossion. The only thing that ultimately needed replace was a single transistor that combined the bias signal with the audio from the microphone preamplifier.
The GE Cassette Recorder - Circa 1975
Okay, for the less literate among you, I present the following pretty pictures of the Apple II+ I'm working with this year. I need to put the non-functional tape deck back together before taking a photograph of my "vintage" data storage device.
The Requisite Shot of the Apple II Plus Nameplate
The Complete System (in dire need of cleaning)
The electronics on the GE tape deck appear to have suffered from some manufacturing defect that caused severe corrosion to the metal traces on the circuit board and to the leads of the individual components... Something caustic in the flux, perhaps? At any rate, repair of this tape recorder will involve new parts, a new circuit board, and probably a re-design to replace devices that have become obsolete since the middle 1970's. A shame really, since the mechanical parts appear to work flawlessly.
In the meantime, I'm trying to work out the specifics of the Apple II Cassette Interface -- if I get around to it today, I'll shoot some video of the oscilloscope display of cassette I/O.
Day #2 already!
I'm still working on the GE Tape Recorder, trying to figure out WHY the damn thing won't record or reproduce sound. Photos to follow (assuming the camera works!)
I wasn't picking on any particular resident of the British Isles, particularly Scotland, when I commented about the preference for Wax Cylinders to either vinyl records or magnetic tape. Rather, I was referring to this charming story from the BBC:
Sorry if your browser doesn't support flash! (Nudge, Nudge, Wink, Wink)
Well, the topic I've selected for this retrochallenge is "fun with magnetic tape." As in most things "retro", "fun" is in the eye of the beholder: magnetic tape for data storage is generally slow, frequently unreliable, and characteristically sequential in a world that expects fast, reliable, random access to all things. Still, magnetic tape of all kinds has a historical relevance that belies its reputation. When historians begin to seriously consider the second half of the 20th century, many of the records on which they base their work will be found in the form of magnetic tape. Census records, taxes, and the dirty little economic secrets that make history "interesting" will be found on magnetic tape. Early computer software, "modern" music, motion pictures, and television currently exist almost exclusively on magnetic tape.
If you were born after 1948 (when Ampex began largescale production of audio tape recorders) and before 1999 (when widespread fear of the Y2K bug motivated unheard-of spending to replace "legacy" systems), magnetic tape formed the basis of your (western) culture. That's what I want to explore in this retrochallenge.
Some audiophiles prefer vinyl. I suppose that some (esp. in Britain) actually prefer wax cylinders. In fact, though, most never owned a record lathe. On the other hand, almost everybody has owned (or at least used) a tape recorder.
Our first subject then, is the relatively common audio cassette and the machines that recorded and played them. We'll get to computers soon enough, but it was the ubiquity of the audio cassette recorded that made the first generation of "hobby" computers practical. Paper tape was my personal favorite form of data storage, but how many people actually owned a teletype or similar device compared to the numbers that owned a cassette recorder? There's no contest. Until Woz produced his low-cost Disk II in 1978, cassette tape was the King of data storage for microcomputers.
So here's our first artifact, a "Certron" LN60 Compact Cassette tape that probably dates from the early 1980's. When purchased, it was probably the cheapest brand to be found that day. It's relatively small, measuring four inches (10.2 cm) wide, two-and-a-half inches (6.4 cm) high, and around 13/32 of an inch (1.2 cm) "deep. It weighs 1.1 oz (30 grams.) Using an Apple II Plus to record one side at an average speed of 1500 baud, it could store a theoretical maximum of around 263K bytes (30 minutes x 60 seconds x 1500 bits per second = 2,700,000 bits / 10 (1-start & stop-bit per byte) = 270,000 bytes / 1024 = 263.671875 K bytes.) With multiple small files and their associated "leader" tones, it might practically store an eighth of that. In actual use, most cassettes held a single program that used less than 32k of memory. Assuming that everything worked (remember the reliability issue?), you could load that 32K program in around 5 minutes or so!
Before I start playing with tapes on the Apple II+, I need to write a simple program that I can record to an expendable tape for test purposes. My plan is to take a look at some of the gritty details of the Apple II tape interface, but I don't want to risk any vintage software in the process. It's been a while since I wrote anything in Applesoft, so please bear with me...
100 HGR2
110 HCOLOR= 3
130 LET R1 = 90
140 LET R9 = (22 / 7) / 180
150 REM DRAW A BOX
160 HPLOT 0,0 TO 279,0
170 HPLOT 279,0 TO 279,191
180 HPLOT 279,191 TO 0,191
190 HPLOT 0,191 TO 0,0
275 FOR T = 0 TO 135 STEP 45
300 LET RX = 10:RY = 90: GOSUB 2000
320 NEXT T
330 PRINT "ALL DONE!"
999 END
2000 REM DRAW A DASHED ELLIPSE
2001 LET F = 0
2010 REM
2020 FOR I = 0 TO 360 STEP 10
2040 LET X2 = RX * COS (I * R9)
2050 LET Y2 = RY * SIN (I * R9)
2060 X9 = X2 * COS (T * R9) + Y2 * SIN (T * R9)
2070 Y9 = Y2 * COS (T * R9) - X2 * SIN (T * R9)
2071 IF F1 = 0 THEN 2080
2072 HPLOT X5,Y5 TO 140 + X9,96 + Y9
2073 GOTO 2081
2080 HPLOT 140 + X9,96 + Y9
2081 F1 = 255
2082 LET X5 = 140 + X9
2083 LET Y5 = 96 + Y9
2090 NEXT I
2100 RETURN
Okay... I've got a nice and easy BASIC program that draws a highly stylized representation of an "atom" on the HIRES display. The wonky colors are just how I remember them! Now to make a tape!
] SAVE
Well, that didn't work very well... The old GE tape deck is giving me a bit of a problem; namely it's not putting any audio on the tape. Time to dismantle the tape recorder and see what's going on.
Still anxious to get started. From a cursory glance at the other participants, this looks to be a year for the Apple II! I'm starting with the Apple II tape I/O system but I hope to get a little beyond that and cover the other popular tape formats. If my work schedule cooperates, this could be the most fun I've had all year!
Picking through the computer junkyard to find appropriate gear for my magtape adventures. System number one is going to be based on the Apple II+ I was playing with on Sunday. The system needs some work, including a thorough cleaning and some keyboard repairs -- I suspect it was sitting in a dusty garage or attic for the last few decades since it's covered inside and out with a fine silica-rich substance that looks strikingly similar to fly ash.
Obviously, I'm going to start with the usual Phillips "compact cassette" type of recorder. The best period one I've been able to find is a GE model that dates from 1975. Preliminary tests show that while the motor runs, the machine does not reproduce sound from "known good" tapes. The cassette recorder will also need to be repaired before work can progress.
Not surprisingly, cassette tapes were not the only audio recording media used with microcomputers in the middle 1970's -- open reel tape was still widely available and the overall higher fidelity and dynamic range of open-reel tape made it ideal for computer applications. I have at least two reels of Ampex 1/4" tape that apparently were used to store programs or data upon. Time will tell if these programs can be recovered more than thirty years later.
Finding an open-reel machine to use with the Apple II might be more of an issue than typical retrocomputing endeavors. The remaining stock of high-fidelity open-reel tape decks is highly sought after by audiophiles and the eBay prices for open-reel decks get excessive in rather short order. Recording speed and format might also be an issue. There were at least 3 common speeds for open-reel decks during the "stereo" era and at least six different track arrangements that I know of for 1/4" tape. I'm assuming these tapes were recorded in a monaural format, but it remains to be seen if that assumption is correct.
Tomorrow, I'll begin research on some more preliminaries; including the Apple II recording formats.
Okay, the last post was pure bullshit inspired by the needs to get the words "hairy beaver" into the blog. This was in reply to the retrochallenge twitter post using the same expression. Pretty lame, I confess, but it was terribly warm and humid here. Being the summer solstice, ex-wife #1 was undoubtably painting herself blue to dance naked in the forest (neo-pagan thing!) Ex #2 wanted to go deer spotting -- a peculiar rural custom involving high-candlepower spotlights, unsuspecting members of the species Odocoileus Virginianus, and an automobile. 2/3 of the associated offspring were also present, leaving me little opportunity to play with technological fossils. In short, the attempt at humor was born of not-so-quiet desperation... Mea Culpa!
The need for a simple pointing device in the graphic user interface has been met using a number of devices over the years, beginning with the "light gun" used on the AN/FSQ-7 Intercept Computer (SAGE) in the 1950's. Track-balls, cursor control keys, separate X-, Y-, and Z-Coordinate dials, joysticks, and light-pens have all found service in particular applications. There have been graphics tablets, point-digitizers, teaching pendants, and physical cursors that resemble the pointer of a Ouija board. Tablet PC's and iPhone work-alikes have multi-touch displays, the IBM laptops had the infamous "g-spot" pointer. Clone laptops have a small touch-pad. Most commonly though, at least since the early 1970's, the ubiquitous "mouse" has been the default pointing device for microcomputers.
Although inexpensive and intuitive to use, the small rodent of yore suffers from problems of scale -- when attempting to use the mouse within the relatively large virtual workspace of a mainframe computer, it is often necessary to navigate the mouse over large areas of the display. This can be frustrating when insufficient (real) desktop space is available. Another issue is the potential for repetitive stress injury resulting from the frequent precision movement of the pointing device. From an ergonomic standpoint, a large but lightweight control is easier to manipulate than a small, relatively dense, control. This is why our automobiles have large steering wheels and high-end radio tuners still use a large tuning dial to manipulate the variable frequency oscillator (VFO.)
Clearly, any solution to the problem of scale must retain the intuitive operation of the mouse while addressing the risk of bodily harm. Most computer users are familiar with the "mouse", so we can safely stay within the rodent family as far as operability is concerned. The device must be rugged and reliable, particularly in industrial and commercial settings where large-scale computers might be found. A water-tight enclosure is desirable, given the possibility of exposure to beverages and other liquids in the workplace. Apple Computer has repeatedly proven that form and function are not mutually exclusive -- the design should appear somewhat "organic" if not downright "warm and fuzzy."
With these considerations in mind, I present my own solution to the problem of an efficient pointing device for large-scale computers...
A drum-roll, please!
The replacement for the mouse, retaining all the desirable qualities of the rodent family, a rugged industrial design incorporating water-proof construction, a streamlined appearance, and unquestionable utility...
This morning, instead of doing actual work that generates revenue I'm playing with the Apple II+ and working out some of the details for grabbing video output, cleaning it up, and posting it on the web. This is my first screen shot, taken with a Hauppage Video Input Card and converted to a jpg for display in the browser of your choice (excepting Lynx!)
