The Truth About Tape Recorders

[Pluto]

I wonder why they went digital instead of just using a different playback head.

If you looking for, say, 1 revolution of delay at 33rpm, you need about 1" delay, give or take. Running at 15ips, that's a lot of extra tape to manage on your deck plate. The last generation of tape transports went to some extraordinary lengths to minimize the problem of scrape-flutter, a problem that, broadly, increases with the length of the tape path.

What this comes down to is the need for a highly customized transport if you are looking for the ability to delay by more than a few hundred mS. The right solution was to use the best possible transport (i.e. the shortest possible tape path) in conjunction with the new-fangled digital delay.

 

[srrxr71]

If you looking for, say, 1 revolution of delay at 33rpm, you need about 1" delay, give or take. Running at 15ips, that's a lot of extra tape to manage on your deck plate. The last generation of tape transports went to some extraordinary lengths to minimize the problem of scrape-flutter, a problem that, broadly, increases with the length of the tape path.

What this comes down to is the need for a highly customized transport if you are looking for the ability to delay by more than a few hundred mS. The right solution was to use the best possible transport (i.e. the shortest possible tape path) in conjunction with the new-fangled digital delay.

Yes I wondered about how big that transport would be. I did not consider how much they were focused on flutter but that was a huge thing those days. I’m sure some engineer could have designed something for it but digital is cheaper and easier.

 

[Jim Shaw]

...My late dad was an electrical engineer and I have his wire recorder he made in the late 50s. He recorded Sputnik on it. I remember him playing it when I was younger...

Probably more like 40s than the 50s; the wire recorder disappeared quickly with the advent of analog tape in the early 50s. I had one that an older fellow gave me, years later. It worked, electronically, for speech. But avoiding wire snarls was a fool's errand. After WWII, some wire recorders were used by law enforcement, as I recall. But, oh the tangled life we lead when first we try to use wire.

 

[thorn]

Probably more like 40s than the 50s; the wire recorder disappeared quickly with the advent of analog tape in the early 50s. I had one that an older fellow gave me, years later. It worked, electronically, for speech. But avoiding wire snarls was a fool's errand. After WWII, some wire recorders were used by law enforcement, as I recall. But, oh the tangled life we lead when first we try to use wire.

Got me curious, I'll have to go listen and record those wire spools/reels to see what's on them. I hope it still works. It is large and he might of built it. The wire spools are kept in nice stainless steel containers. Sputnik 1 was orbiting in Oct 1957 so he had it then.

 

[Multicore]

It was not CD that killed cassette, it was CD players in the car, that killed cassette. Once the ubiquitous radio/cassette head unit was replaced with a radio/CD unit, that was the end.

Yes, the vibration resistance in the car and for the portable use was critical. I could ride my bike with a CD Walkman in my pocket. When was that relative to introduction of CDs?

Combined with computer based recording of CDRs and people had all the convenience of making their own recordings, but having a somewhat better sounding and resilient format in the car.

Looking back I don't think burning CDRs was as important. It was certainly a factor but I don't remember many people with collections of CDRs. This was the same time we learned to rip onto a computer and to use iTunes and iPods and Napster etc..

 

[atmasphere]

Interestingly, any LP that was half-speed mastered has had the audio passed through a digital delay line prior to it hitting the cutting head. So it would be A-(master tape) to D-(digital delay) to A-(lathe electronics). So some of the most cherished vinyl would have the dead hand of digital placed upon it, bad luck AAA guys!

This statement is incorrect and is very common myth.

The function of the preview head is to assign a signal level value. This value is then used as to influence the speed of the lead screws on which the cutter head assembly rides. The idea is with greater signal the cutter moves faster so as to provide space between grooves. If that space isn't there, with a loud sound that requires a wider groove, it might overcut into the groove just cut prior (about 2 seconds ago; 1 rotation of the platter).

The audio signal itself is not drawn from the preview head (which has to sit 2 seconds upstream from the playback head); it comes off of the playback head. The preview head signal is often converted to a digital value to run the lead screws but this can be done with an analog voltage as well.

I don't know how the idea that the playback head signal somehow got converted to digital ever got started; perhaps from the 'digital preview head' name used; which was IIRC the name used by the Compucut device that was used to run a lathe with variable groove spacing.

In our studio we used an Arduino to create a motion file by playing the project back to the Arduino and then storing the motion file. Then we would set up the lathe and start the project and the motion file after one platter rotation. This worked quite well and was a lot cheaper than having to find a preview head system.

 

[LtMandella]

Until last year when I bought a new MC cartridge and phono preamp, I used the unit you see in my avatar to transcribe my irreplaceable LPs so I could listen to the music as much as I wanted without wearing down the LPs. Worked great, I could play the Mobile Fidelity Beatles box set weekly for a decade and no added wear to the LPs!

Now I transcribe to DSD - a technology which was not available at the time I purchased 99% of my LPs...

 

[Punter]

This statement is incorrect and is very common myth.

The function of the preview head is to assign a signal level value. This value is then used as to influence the speed of the lead screws on which the cutter head assembly rides. The idea is with greater signal the cutter moves faster so as to provide space between grooves. If that space isn't there, with a loud sound that requires a wider groove, it might overcut into the groove just cut prior (about 2 seconds ago; 1 rotation of the platter).

The audio signal itself is not drawn from the preview head (which has to sit 2 seconds upstream from the playback head); it comes off of the playback head. The preview head signal is often converted to a digital value to run the lead screws but this can be done with an analog voltage as well.

I don't know how the idea that the playback head signal somehow got converted to digital ever got started; perhaps from the 'digital preview head' name used; which was IIRC the name used by the Compucut device that was used to run a lathe with variable groove spacing.

In our studio we used an Arduino to create a motion file by playing the project back to the Arduino and then storing the motion file. Then we would set up the lathe and start the project and the motion file after one platter rotation. This worked quite well and was a lot cheaper than having to find a preview head system.

Thanks for clearing that up and what a great Idea to use an Arduino as a lathe control! Excellent!

 

[JaccoW]

Daft Punk just dropped this short video for the 10th anniversary of Random Access Memories.
That's some big tape!

 

[Punter]

Yep, a gorgeous Studer multitrack! I could have picked a couple of these up for next to nothing when our production studios went digital....

 

[robwpdx]

Speaking of duplication, the process for duplicating pre-recorded reel to reel tape was a far more complex and expensive exercise than pressing an LP and another reason that the range of titles available was limited. The process involved master, slave and transfer decks and the process to cut and individually wind tapes for final packaging. The process started with a master deck running a master tape. However, the master tape was not on a reel but stored in a tape reservoir similar to that on the Marconi-Stille recorder.
View attachment 235623View attachment 235624
The master was spliced end to end to create a loop which would only be run forward. The master tapes were mostly duplicated direct from studio master tapes but sometimes they were created from an LP! The slave decks were loaded with blank tape 6000-8000 feet in length which was recorded and re-reeled onto a precision reusable reel. It was then mounted onto transfer machines that would wind the tape onto individual "consumer" reels that the operators would mount into the machine one-at-a-time. To add a modicum of automation to this process, there were pilot tones added to the bulk recorded tape which would signal the transfer deck to stop at the end of the recording. 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. The individual reel of tape was then labelled, boxed, and packed. A QC team would sample the duplicates on a random basis, listening to the entire tape to detect defects which would indicate problems with the slave recorders or transfer decks.
View attachment 235625
It’s no wonder that the cassette became the dominant tape format. Running an open reel duplication plant would have been a maintenance nightmare as well as being highly labour intensive. Cassette duplication, by contrast, was infinitely simpler.

Thanks for this article.

I worked in the recording department of a noted music school. They maintained a library of recordings at the school on acetate discs and magnetic tape. The library had 1/4" half track on 7 inch reels for the older ones and 10 1/2 inch pancakes ongoing. One of our background jobs was to spool off the 7 inch reels and insert paper tape lead in, lead out, and between the tracks. Now, of course, all recording is digital.

I was a student, it was my student job, essentially minimum wage. The food hall workers were better paid because that job was considered unpleasant. We were making several student recital recordings per day and all the ensembles and many guest artists were recorded.

The masters were copied when a performance went out for broadcast or for students as audition tapes, or their personal library. We had a custom built tape duplication system. There was one master and 2-3 slaves based on, I believe Ampex 300-series transports. I believe they had been modified to operate at 4x speed. That probably required custom heads and and custom tube electronics. There was a special Dolby A301 decoder. All the frequencies were multiplied by 4. All the equipment was custom built by Neil Muncy, probably in the late 1960s.

The funny part of the story is that the school got a big grant to move the tape library and this vacuum tube equipment to a new space with better climate control for the library and a Halon fire suppression system for the duplication room which was also electromagnetically shielded on all sides with copper mesh. The doors were heavy steel commercial doors with the mesh behind plexiglass and an electrically sealing gasket on all sides. It was a Faraday cage. It was on the top floor of the old building where no one had reason to go. The shielding was because of all the radio transmitters for taxis and contractors on adjacent buildings.

We were students, so only half trusted. All the recording studios had 2 locks. One lock the students and full time staff had the key, and the other lock, only full time staff and security had keys. I was working late one night in the new duplication room on the 5th floor. Unbeknownst to me, the security guard, doing his job, came by and locked the secure lock for which the architect and contractor had omitted an inside thumb turn. I finished my work, and discovered I was locked in. They also had not hooked up the telephone. There was no way I was going to camp overnight in a room full of tube equipment, even if I turned them off which was never done, with Halon fire suppression. And it was a pretty secure room.

I had a narrow time window before the whole building alarm would be activated and the building completely empty until morning. Fortunately there was a side storage room that was not shielded. I broke through the drywall celling of that room with my fist, climbed into the suspended ceiling over the tape library, then let myself down by removing a ceiling tile in the hall. I was able to get out before the alarm came on. Many jokes were made and some things were fixed.

 

[Will33]

I know I’m dredging up an old thread, but since this article is so detailed, I felt that some details about tape duplication should not be left uncorrected.

While the description of reel to reel duplication is accurate, it was not really more complicated than cassette duplication at manufacturing scale. In fact, the loop bin master pictured in the article is an Ampex that, with the correct modifications, could also be used for 8 track and cassette duplication.

For cassette duplication, tape came in large “pancakes”, reels without flanges. Slaves looked like horizontal reel to reel recorders capable of transporting 1/8” tape at high speed. Once a reel was recorded, it was put on a cassette loading machine which automatically spun the tape into a shell, cut and spliced it, and loaded the next shell until the reel was exhausted. The King Instruments model 790 was the workhorse of the industry for years.

At its limits, an Ampex duplicator could record at speeds of 64:1. 120 ips. Eventually, analog loop bins were replaced by digital playback masters most notably from Concept Design. Those could push speeds to 80:1 with good sound and no worn heads or master tape.

Here’s a 790 photo from the Internet:

 

[Punter]

Thanks for your contribution Will33 I remember tape pancakes as I used to have to use them to rewind broadcast cartridges. The tape was 1/4 inch and the back side of it had a graphite coating to allow the endless loop to feed from the center of the internal reel of the cartridge, pass over the heads and then wind back onto the outside. 8 track cartridges used the same type of tape and used the same internal construction with a single reel. The danger with a cartridge tape pancake was that if you didn't handle it with care, a single bump could make it collapse out of your hands and end up a useless birds-nest on the floor! Fortunately, it only happened to me once in my career.

 

[Will33]

I know I’m dredging up an old thread, but since this article is so detailed, I felt that some details about tape duplication should not be left uncorrected.

While the description of reel to reel duplication is accurate, it was not really more complicated than cassette duplication at manufacturing scale. In fact, the loop bin master pictured in the article is an Ampex that, with the correct modifications, could also be used for 8 track and cassette duplication.

For cassette duplication, tape came in large “pancakes”, reels without flanges. Slaves looked like horizontal reel to reel recorders capable of transporting 1/8” tape at high speed. Once a reel was recorded, it was put on a cassette loading machine which automatically spun the tape into a shell, cut and spliced it, and loaded the next shell until the reel was exhausted. The King Instruments model 790 was the workhorse of the industry for years.

At its limits, an Amped duplicator could record at speeds of 64:1.

Thanks for your contribution Will33 I remember tape pancakes as I used to have to use them to rewind broadcast cartridges. The tape was 1/4 inch and the back side of it had a graphite coating to allow the endless loop to feed from the center of the internal reel of the cartridge, pass over the heads and then wind back onto the outside. 8 track cartridges used the same type of tape and used the same internal construction with a single reel. The danger with a cartridge tape pancake was that if you didn't handle it with care, a single bump could make it collapse out of your hands and end up a useless birds-nest on the floor! Fortunately, it only happened to me once in my career.

Broadcast carts were indeed similar to 8 tracks with a few significant differences.
1. Carts ran at a higher speed for better fidelity at the cost of shorter lengths.
2. 8 tracks had a built-in pinch roller while cart machines had a pinch roller that popped up into the cart and engaged the tape against the capstan.
3. Two vs 8 tracks again trading quality for length.

 

[Petrushka]

I had a number of recorders: Teac, Sony, Uher. The format died for me when my Ampex tapes started squealing.

I recorded my wedding, including a nice pipe organ concert, on a Sony, and managed to digitize it before it deteriorated.

 

[atmasphere]

I had a number of recorders: Teac, Sony, Uher. The format died for me when my Ampex tapes started squealing.

I recorded my wedding, including a nice pipe organ concert, on a Sony, and managed to digitize it before it deteriorated.

You can often bake the tape to chase the moisture out of it and thus stop the squealing (tape shedding). Its also a good idea to store the tape with a packet of Silica Gel.

 

[dfuller]

You can often bake the tape to chase the moisture out of it and thus stop the squealing (tape shedding). Its also a good idea to store the tape with a packet of Silica Gel.

Once you do that, though, it should be transferred to new tape.

 

[atmasphere]

Once you do that, though, it should be transferred to new tape.

Or storage in an attic. Attics tend to be hot and dry so tapes last much longer there. Basements OTOH tend to ruin tapes.

 

[dfuller]

Or storage in an attic. Attics tend to be hot and dry so tapes last much longer there. Basements OTOH tend to ruin tapes.

Nah, once a tape has gone sticky it's pretty much cooked - transfer to new stock or to digital is a must.

 

[atmasphere]

Nah, once a tape has gone sticky it's pretty much cooked - transfer to new stock or to digital is a must.

If you bake a tape, if its not too far gone, it will no longer be sticky. If its 'gone sticky' IOW, it needs to be baked.

Once baked it can be good for quite a few months afterwards, since it takes a while to absorb the water molecules that caused the stickiness in the first place. That can be prevented by low humidity storage, such as the suggestions I just presented.

Naturally its not a bad idea to have a digital backup.