My Year at AMPEX

A first-hand account by Ron Schauer

I worked in the Ampex Tape Duplication facility in Elk Grove Village, Illinois from September of 1970 to June of 1971 during my early college career.  It was a lot of fun to try and remember everything, and in my mind I can still walk around the plant and visualize all the equipment. Our son has all my old cartridges and equipment and most of them still sound great. Sadly, I have no photos of the equipment used to manufacture the tapes because of course they didn’t allow cameras inside the plant.

Upon being newly hired in 1970, I had no real idea of how tapes were commercially duplicated, thinking   that they must be copied in banks of tape recorders similar to how someone would do it at home. Most people also harbor this same notion. Well, the same principle applied, but the equipment was totally different from home gear. All the tapes we produced there, whether open reel, cassette or 8-track, were duplicated in roughly the following manner: master tape machines each having a number of recording slaves, reel winding machines followed by one form of cutter or another, then leader/hub/trigger tape splicing, insertion into a case or box, labeling and finally packing.

 

---

 

A master tape, 1″ in width, was run in a continuous loop through a refrigerator-sized playback machine. It looked strikingly similar to a 1950’s computer tape drive, and had many of the same features, including vacuum chambers to take up the slack of the tape before and after it was fed through the playback tape head. The playback heads were specially made for us in Palatine, Illinois, and were very, very expensive. Since these master tapes ran at 10x “normal” speed, the tape heads wore out rather quickly. There was a similar problem with the recording heads, though 10x on the recording end was about 1/4 the master playback speed, so they lasted a lot longer.

A specially trained technician checked the alignment and condition of the master tape machines, and only he/she was allowed to work on them. The line operators would often plug headphones into the auxiliary outputs of these machines, and even at 10x or 20x playback speed, they could spot defects and developing problems. After some number of playbacks, the tapes developed shiny spots where the oxide wore thin, and sometimes they creased or wrinkled. The lead operators would catch these defects because the thin spots in the oxide would sound muffled and any creases would make a hair-raising grating noise at 20x playback. They would stop the line immediately if these were heard. I plugged into the machines only occasionally because I was the only electronics line equipment technician for the entire shift and was often quite busy.

The lead person on the line would carefully unload the master tapes from reels and splice them in the playback machines. When loaded, these tapes resembled a jumble of loosely looped Christmas ribbon in a glass-front case. It fed out from one half of the glass case and dumped into the other half. As I didn’t work on these machines myself, I didn’t get glimpses of the internal controls or mechanisms much, but they didn’t appear to be particularly complex. They were really neat to watch though, sort of like giant $20,000 Lava Lights with frantically moving loops of Mylar tape instead of hot wax.

The eight track master tape had all eight tracks recorded in the “forward” direction. When played for duplication, all eight channels were reproduced simultaneously. All the master tapes shared one common feature. Anyone with a good enough sound reproduction system and younger ears has heard the “trigger” or “pilot” tones at the end (and sometimes the start) of the audio tracks. They were recorded real-time at 20 hertz, and on most playback systems of the time were well below the reproduction range of the electronics and speakers. The purpose for these tones was to tell the recording and reeling machines when to switch hubs or when to stop for cuts or splices. If you listen to pre-recorded tapes very carefully, you will hear them as a very low pitch rumble just about the time the tracks are going to switch. It has been likened to running a car over a series of grooves in pavement. Watch your woofer cones and you’ll might see them vibrate even if you can’t hear them.

---

The master tapes had an interesting story in and of themselves. In a fairly sound-deadened back room off the side of the factory floor, was a studio setup including an Ampex (what else?) eight channel studio recording deck complete with all the amplifiers and patch cords and so forth. There the master tapes for the production lines were made. They were most often duplicated from other masters, and depending upon how popular a particular recording was, there could be eight or ten duplicate masters running on the production floor simultaneously (such as the sound track to the movie “Woodstock”).

These production master tapes were themselves copied from actual studio master copies. I had the wonderful experience of hearing much of Jimi Hendrix’s “Electric Ladyland” directly from one of these tapes as a production master was being made. It was the cleanest, highest fidelity sound reproduction I had ever heard at the time, and it pretty much spoiled pre-recorded music for me for years until CDs finally came along. When you consider that this room also had studio monitor grade speaker systems and super high fidelity amplifiers and mixers installed, well you get the idea. You really felt like the musicians were sitting there with you. Even nearly thirty years later all I can say is that it was awesome.

The technicians that worked in this room also had the capability to produce masters from vinyl records. There was a very ponderous looking and totally professional turntable with variable speed drive, a huge heavy turntable platter and a manual tone arm installed. It was mounted on a pedestal having some sort of compliant balloon or other cushion (I never saw it with the cover off), and a heavy granite or dark concrete slab beneath the turntable itself. Someone told me with a completely straight face that it contained a “lead” balloon, but I never took them seriously. Whatever it really was, its purpose was to isolate the turntable from floor vibrations and noises as much as possible. If a recording was very old, or if there were no surviving master studio tapes, they could clean up a record and reproduce it quite nicely. Of course there was no way to get rid of the worst of the scratches and pops. There are quite a few pre-recorded tapes from that era that sound like recordings of records because in fact that’s what they are.

Now for a sore subject for any truly dyed-in-the-wool 8-track owner/enthusiast. Why the !xBss#e!!$55@ did they chop up songs where they did for the track splits? Well, of course you know that the media itself doesn’t lend itself to a continuous recording. For one thing the sensor tape will cause a momentary dropout as it passes over the tape head, and for another as I have explained, there was the necessary 20-hertz channel switch tone that signaled the end of the loop for the winding machines. There also had to be some finite amount of “dead” tape on either side of the splice because the winding and splicing machines could only respond just so fast. These were some of the compromises necessary for mass production.

Most of the folks who did the masters did a really good job of it and tried hard not to chop up anyone’s favorite songs. You will notice that the later cartridges often times had extended blank spots on the track ends where the editor just flat out gave up trying to reorganize the record tracks into equal-length sections. If an editor was really considerate, they would warn you by putting a “2:13 silence” notation on the play list or something like that. However, in any such endeavor, there are those folks who really grooved on say, Bing Crosby or Hank Williams and who could give a green flying leap about the latest lyrical offerings from Mott the Hoople, The MC5, Aretha Franklin or The Nazz, much less expressing any recognizable form of sensitivity regarding the seemingly 15 second long fade outs and fade ins they put around the track switches. Remember this was close to 30 years ago, and the record business was structured a lot differently. I would imagine that nowadays you’d have the artists, their agents, or their attorneys screaming bloody murder if someone chopped up their songs like that.

Now I like the Crooner and old Hank as much as the next guy, so please don’t flame me with e-mails, I only used them as an example of an apparent attitude. I still have some 8-tracks that irritate me to no end at how they were edited. I personally protested about the fades and splits whenever and to whomever I got a chance. After a while I think they didn’t want to talk to me anymore.

And let’s not forget the “cheap” factor. As the market for 8-track tapes declined, there was more and more pressure to get it done quickly (i.e., cheaply). Add this to a healthy dose of “who’ll be listening to this stuff a year from now anyway,” and it isn’t hard to see why they faded out and in again in the middle of Chicago’s “25 or 6 to 4” on one tape I have (a Columbia, not an Ampex) when they could have slightly rearranged the songs and avoided it altogether. Of course maybe the artists or producers nixed the shuffling of the song order, I never found out. I doubt if they were consulted. Sorry, it wasn’t my fault….. really!

The raw tape was loaded into the recorders in the form of sideless 12″ diameter 1/4″ wide tightly packed reels of tape on plastic drive hubs. There was a fair amount of skill and finesse required of the operators to remove these spools from the shipping boxes they came face-on stacked in, lay them down sideways onto the recorder feed spindles, and feed them through the machines.

Once the operator had laid the spool of tape onto the unwinding spindle, it was fairly safe because the spindle was made to be the same diameter as the tape spool and was flat, but getting it there could be a chore. Many a time I watched some poor operator pick up a spool and it would simply come apart in her (usually a “her”) hands. In the dumpster it went, it wasn’t worth salvaging. The Opelika (Ampex) tape was about the worst offender for this because they just couldn’t seem to tension it tightly enough at the factory, or perhaps it stretched in shipping, I don’t know. The Scotch was best and Memorex was in the middle. These reels of blank tape came stacked face-on in larger boxes, and were separated by nice foam pads that were discarded (I think I may still have a few at home salvaged for use as packing material).

OK, so the line operators punch the start buttons on the master playback machines, now what? Well, that varied a bit depending upon what form of tape was being produced. In all cases there were rows of recording decks (slimmer versions of the Ampex studio decks, what else?) which were slaved to the playback output. Well, sort of. When the operator pushed the start button on the master machine, the slave recorders did not start automatically. They would go up and down the lines and start each one individually. Why you may ask? Because then they wouldn’t all run out of tape at once giving the operators a chance to change the blank feed reels in a staggered manner. The first tape on a reel was always discarded, as was the last. The entire reel of tape was recorded from the continuously looping master. The end of track tones would be used as flags later to cut the reel of tape apart into the final products. There were 10 or 12 slave recorders per master unit, and there were five or six dedicated 8-track lines, four dedicated cassette lines, and two open reel lines. I might be +/-1 line each, it has been almost 30 years after all.

I know this is an 8-track interest website, but I want to explain each of the three different formats in order to convey a sense of just how different the 8-track production process was from the other two.

The open reel tapes were the simplest. The entire reel of blank tape (I believe it was 6000-8000 feet in length) was recorded and re-reeled onto a precision reusable reel (with sides, of course). The amount of tape on a reel varied because the thickness of the tape varied. It was then fed into machines that would cut and transfer the recorded reel onto individual “consumer” reels that the operators would mount into the machine one-at-a-time. The machine would transfer the recorded tape until it encountered the pilot tones, then stop and back up until they were found again. The operator would then splice on a length of leader, splice a leader onto the start of the next recording, change the output reel and go again. These machines ran at very high speeds and used pneumatic disc brakes to stop the reels. They were a real pain to balance so that the tape was neither stretched nor unwound as it stopped. They were also incredibly noisy, and you could hear them “screech” to a stop all over the plant. The individual reel of tape was then labeled, boxed, and packed. QC would sample these tapes on a random basis, often listening to the entire tape for defects. They could catch things most people wouldn’t even notice (like flutters, which often indicated that the tape had been stretched).

The cassette processes were similar, though a bit more complicated. The winding machines were made by Phillips, and had semi-automatic splicers. After recording, these reels were taken to another area where they were hand-inserted into partially assembled cassette halves (including the rollers and other hardware), the tops were placed on, and then the whole shebang was ultrasonically welded together. Other manufacturers used screws to hold the halves together, but the welder was extremely quick and positive (and cheap of course). Naturally you could NEVER get them apart again without destroying the case halves, and if the welder was misadjusted, it would collapse the case halves enough to jam the rollers and hubs. Since these machines were complex and finicky, you might guess that I spent a significant percentage of my time working on them. From there the cassettes were labeled either by applying adhesive labels or by ink transfer pad printers (also a major maintenance headache, but cheaper than dirt when it worked). They were then placed into their miniature cases with the inserts, shrink-wrapped, boxed and stocked. Finished tapes were sampled at random for defects including the same sorts of audible problems as the open reel tapes, packaging defects, and of course to make sure that the welded halves wouldn’t come apart short of total destruction.

And now, at last… the 8-tracks. First, about the tape: the tape we used mostly was Scotch, which unlike the earlier Scotch tape was pretty good. The Opelika, Alabama tape facility later developed a yellowish tape that I despised.  It flaked off and gummed up everything. The early Memorex (black oxide) wasn’t much better. The backside lubricant flaked off and looked a lot like disk brake dust with about the same bad effects (particularly when it stuck to the capstan).

My friend Ed from Ampex Corporate tape actually toured the Memorex factory (which used to be near San Jose). The Memorex tape started out as 4″ wide mylar roll stock and was fed into a building-length horizontal machine. The coating(s) was(were) applied to the tape and it was basically dried and baked (this is a “calendaring” operation). At the exit it was slit into whatever widths were required, from 2″ to 1/4″ (at that time). Ampex was always looking for a way to save a buck, and that was the reasoning he recalled for the transition to the all-plastic pinch rollers. Of course as it turned out they were ultimately more durable. They caused a lot of problems in players though because the amount of “retention” the cartridge saw when inserted into a player was more critical with them. If the side retention roller (or pawl) that detains the cartridge into the player is a bit weak, the tape will slip on the capstan. If a bit too strong, it will mar the oxide when starting and stopping, and can permanently deform the tape if the player is switched off with the cartridge left in.  The bearing surface of the roller on its support shaft can actually bind, skidding the tape on the pinch roller. Obviously that will do damage to the tape, if it doesn’t wear it through completely. Remember that the typical setup is a plastic roller or roller hub on a plastic shaft. Cheap, huh? The padded or molded rubber rollers have some “give” to them and are therefore less critical insofar as the retention pressure is concerned. Of course plastic on plastic is not a good formula for longevity.

Back to Page 2 | To Page 4