1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Speed of rotation of an audio record or audio tape

  1. Jan 11, 2007 #1
    My friend and I wondered this in junior high school: since a record (as in the kind played on a record player) or an audio tape turns at the same speed, doesn't the head reading the record or audio tape move faster on the outer edge?

    This means that the sound recording is more tightly condensed near the center of the record or audio tape, right?

    o| Hiram
  2. jcsd
  3. Jan 11, 2007 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The head doesn't rotate with the tape or record, it only moves radially. (Unless you have one of these new fangled record decks where the record doesn't spin :tongue:)
  4. Jan 11, 2007 #3


    User Avatar

    Staff: Mentor

    The relative speed of a phonograph record and the playback or recording stylus is indeed faster at the outer edge of the record and slower at the inner edge, because the angular speed is constant (at 33 1/3 RPM or whatever). The slower speed at the inner edge can produce "inner-groove distortion" unless the record is mastered carefully, because the wavelengths on the record become shorter.

    However, an audio tape records and plays back at constant linear speed, so the angular speeds of the tape spools change as you play through a tape.
  5. Jan 11, 2007 #4
    thanks jtbell

    Do you mean that the tape recorder actually speeds up and slows down its rotation of the cassette tape in order to compensate for the position on the cassette tape?
  6. Jan 11, 2007 #5


    User Avatar

    Staff: Mentor

    In a tape recorder, in playback or recording mode, the tape is pulled along by the pinch rollers in the playback/recording head. The feed and takeup reels simply follow this motion passively. I think the takeup reel is driven lightly by a motor, and the feed reel has a bit of resistance, so as to prevent slack in the tape.
    Last edited: Jan 11, 2007
  7. Jan 12, 2007 #6


    User Avatar
    Science Advisor
    Gold Member

    Indeed. Domestic cassette tape recorders run at a constant 4.75cm/second.
  8. Jan 14, 2007 #7
    Yup. For all reel-to-reel tape decks, the main speed control keeps the capstan at a constant RPM, such that the tangental velocity of the side of the capstan is constant. I recall working on some big Studer-Revox decks that had a detailed procedure for adjusting the motors that turn the hubs for both the take-up and supply reel, maintaining a small but constant tension. The take-up reel would exert more tension than the supply reel, but not so much that it caused the tape to slip through the capstan and pinch roller. The capstan actually served to turn the rubber pinch roller, which it contacted above and below the tape itself. The rubber pinch roller in turn pulled the tape through, so it was important that the rubber did not become dirty or "glazed". Motors that drove the reels also required separate tension adjustments for fast-forward, re-wind, braking, or "scrub" functions during which the pinch roller was not engaged.

    Cassette decks are more likely to have pullies to turn the take-up spindle, while the supply spindle would merely depend on friction to keep tension.

    Vinyl turntables typically either had a direct-drive central axle that was controlled for constant RPM, or else some drive-wheel that pushed the inner surface of the edge of the platter through an intermediate rubber wheel, or a belt-drive to a pulley on the centre axle. Some speed control schemes involved a pattern of stripes on the inside of the rim and a light bounced off them onto a sensor which would tell the control circuit what the tangential velocity was. Since the sensor does not travel inward with the tonearm, it's still aimed at keeping a constant RPM.

    VHS (Vertical Helical Scan) video tape drives also use a capstan and rubber pinch rollers to transport the tape slowly from one side toi the other, but the record/playback head is more exotic (than for audio tape), being on a spinning drum that is slightly tilted so that the head makes multiple diagonal passes across the tape. It has its own speed control circuits. It gives the same effect as running a reel-to-reel at much faster speeds without having to make a huge-diameter reel for the tape.
  9. Sep 1, 2008 #8
    [from now on, excuse me for my poor english]

    eehiram, I agree with your skepticism...

    I guess that a common, vulgar, typical cassette recorder or player, like a walkman, or the one built into an old hi-fi system, for tipical consumer music records, doesn't have the ability (control system) to do such precise work of controlling the linear speed of the tape passing in front of the head; I belive that the rotating speed of the motor is constant (as you can read at ..), so the tape linear speed in front of the head equals the tangent speed at the 'destiny' reel or capstan, so as time passes, the reel grows bigger in radius and so does the speed of the tape.

    As music is recorded de same way, as it is, first slower and then faster, in the same ammounts, because its a constant rotating speed, we hear no changes in speed or pitch in the record as we play the tape.

    And here it's i think the point of confussion we all reach at some point by reading several articles:
    I think that when we or other people ussually recall a type of cassette or tape by its so called playing speed, like 1 7/8 inch/second (for me 4,76 cm/s) what we're doing is just a simple check, and that's dividing the length of the tape in inches/meters by the playing/program time, like 60 or 90 min, etc, but it doesn't describe the actual speed of the tape in front of the head at different playing positions over the playing time.

    As planish says, and so I understand from some articles, big reel to reel tape desks do control linear speed at the front of the head, but that I dont think is the case for common cassette tapes. I might be wrong. If I am, someone please describe how the tape machines do control tape speed

    One of the possibilities I can recall is: recently I knew that tape machines ussually AC bias the tape while recording with a supersonic frequency (40khz to 150khz, so neither we can hear it or players/speakers can reproduce it) for lineality and noise reasons, so I guess this means that it superimposes (sums) this signal. I think we can think of it as a carrier. Anyhow, the point is that that could be enough to read that signal and drive the motors properly, by means of an electronic control loop. That's just an idea.

    But my hypothesis is: in the cassette tape (not reel-to-reel), the motor (or the capstan) rotates at a constant speed, varying the tape linear speed as the tape accumulates in the destiny reel, first playing slower and then speeding up as we reach the end of the music program.

    I will try to prove or check this hypotesis later.

    See you on the next post.
  10. Sep 1, 2008 #9
    Sorry, people, I made a mistake (so far),
    when I said capstan I actually mean 'spool', as Wikipedia says: "an unflanged plastic cylindrical hub on which magnetic tape is wound in a compact cassette"

    In fact, it seems that there is a 'capstan', and it's a rubber wheel that seems to pull the tape, probably at constant speed...

    By the way, I just tried to prove my hypotesis, but it seems that i'm wrong... I took an old tape, all wounded to the beginning of side A, and pulled the tape out through one of the little windows on the cassette case, like the one the head of the player enters to reach the tape, and carefully set 11 identificable marks spaced 10 cm (a mark every 4 in), like: 0,1,2...9,10
    Then I played the tape forward side A (with the destiny reel empty) by 10 seconds, stopped and the mark I saw was '5' (5cm/s sounds coherent with the nominal speed). Then I did FF to the mark 10 and played it from side B (now the destiny reel is full) for the same amount of time, and the mark I saw was '5' again (from 10 to 5 are 5 steps, the same length of tape, so it's the same speed).

    Of course, there must be errors by doing this manually, but those measures can't be so well matched by chance. I measured the reels diameters on the tape, and they are aprox. 5cm and 2,2cm when one reel is full and the other empty (a ration of more than 2). If the rotating speed of the motor (the spool motor) were constant, then the tape speed would change dramatically when the reel is full or empty (by the same factor, precisely), and that doesn't seem to happen...

    Even this is't scientific enough, I think that everything is falling in its place, and we can forget about 'varying speed at cassettes'. They probably play at a preety constant speed.

    Remember the case when the player went mad and crushed the tape around that so-called 'capstan'? Perhaps that was because the spool motor stopped by any cause, but the capstan continued pulling tape but the tape haven't got any place to go, and so acummulating it over there...

    Anyway, I can understand how the tape, by means of the constant rotating speed of the capstant, was pulled at a constant speed... what I can't recall is how the recorder/player manages to pull the tape and wind on the spool without pulling to hard with the spool motor (the 'winding' motor) to break or damage the tape, or too soft to let the tape crush around the capstan area...

    If I were to found something new on the topic, I promise I'll write it down. If someone has an answer or opinion, please comment.

    -By Leandro (Arg)-
  11. Sep 1, 2008 #10
    Leandro - I can not add anything to the discussion of tape speeds or how tape recorders work, but I would like to congratulate you for having the initiative to actually do an experiment with a cassette tape to find out the answer for yourself !

    So much info on the internet is people posting "I think this..." or "you are wrong..." It is refreshing for me, to see someone who can find an answer for themselves (and a real answer at that).

    Thanks for making my day a little brighter!
  12. Sep 2, 2008 #11
    I appreciate, gmax137, for your comment, I hope this brings a little bit of light on the subject for me and everybody interested

    I was remembering right now some details of the test that might be useful and some other ideas:
    - the cassette I mentioned about the reel dimensions was a 60 minute TDK TYPE I NORMAL POSITION (4,98cm and 2,18cm each reel, cassette rewinded).
    - The cassete I used for the test was a similar one, a BASF NORMAL POSITION - IEC I, FS 60.

    I guess this test was enough, but another idea could be: if the tape varies speed over time, if one could took a recorded, rewinded cassette to side A let's say, pull out of it a big amount of tape (half of it for example) and cut it out, and stick the remaining ends to form a 'new' cassette, with less tape, less playing time but the same music on it, as now the spool is empty it would pull tape at a slower speed for that part of the musci program, so it would sound pitched down, slowed, at least detuned. I didn't express my idea very well, but anyway I think the marks test is enough to probe at least that constant tape speed is the correct answer to the problem. So how do the spool motor manages to not to pull to much or to little? We'll see.

    For your information, my original question was about playing Tascam 414mkII portastudio recorded cassette tapes on a common cassette player to capture via soundcard on the PC, and how the music is altered by playing tape at probably half the speed on all four tracks, and playing tracks 3 and 4 in reverse, but how specially those tracks (3&4) were altered in the hypothetical case of a variable tape speed from beginning to end of the tape, which probably is not the case as we disscussed earlier.

    So in conclussion, if someone has the same issue as me (a lost portastudio and a lot of valuable recordings, tunes and memories on several tapes to digitalize and a common cassette deck/player) I recommend trying just capturing the audio, pitching up (altering the speed and duration) by aproximately an octave (2x) and for tracks 3 and 4 applying reverse.

    As soon as I learn to, I will post a couple of pics to show you the first test results.
  13. Sep 2, 2008 #12
    Hello again,
    here's a collage of pics depicting the tape speed experiment. You should track the pics from left to right, top to bottom. Hope it's clear.

    The cassette player was a Philips AZ1030 cd radio cassette recorder (nothing strange, just a common cassette player).


    Attached Files:

  14. Sep 2, 2008 #13


    User Avatar
    Science Advisor

    Leandro, I second gmax137. What a nice demonstration!
  15. Sep 2, 2008 #14
    Cassette tapes are capstan controlled to run at constant linear speed. They are not spool driven. The 1.875 in/sec (4.75 cm/sec) is the true linear speed constant from start to finisn.

    The variable linear speed you describe was used in the micro-cassette system. These were used in answering machines and background music applications. The Philips compact cassette used in Walkmans and other music equipment ran at constant 1.875 in/sec speed. It is well known.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook