Speed variation with unmatched reels

I received an interesting question from a European tape user with whom I frequently correspond:

“Is there any speed variation when playing a tape with different hubs on the supply reel and take-up reel?”

And I replied:

You raise an interesting question. The easiest answer to this is simple:
There is a risk of speed variations throughout the reel if the tape tension varies throughout the reel.
followed by the corollary:
This effect is made worse if the condition of the capstan / pinch-roller system is degraded.
and
Some capstan / pinch-roller / constant-torque systems handle this better than others.

This effect has been most often described in connection with high-tension, constant torque machines such as the Ampex 350 and 440 series (the latter without the optional TenTrol supply reel modification).

Obviously the unmatched reel sizes increase (depending on tape pack) the likelihood of this problem occurring, hence the reel size switches on the Ampex and ReVox machines. However, tape pack radius (the measure that is most directly related to the effect) will change the ratio of the two sides enough to show this effect even with matched reels.

During record/play, since the acceleration is relatively small, reel mass is less of an issue than during fast wind where reel mass differences may swamp the differential braking designed into every recorder, resulting in tape spillage upon stop.

It has been reported that with the Ampex-style machine (Scully and others made similar transports) in some instances it was impossible to match pitch from the end of a reel to the beginning of the next reel.

Somehow, Herr Studer and company seemed to make this less of an issue on the ReVox A77 as I made several well-received albums where we cut between takes made at different reel-radius ratios and no one ever complained about a pitch shift in these edits. I also kept these machines in good condition with frequent replacements of the pinch roller.

Now, all of my machines are constant tension–sort of. The Studer A80 and A810 as well as the Honeywell 101 and the Racal Store 4DS and Store 7DS instrumentation machines have tension sensors on both sides of the capstan. On the other hand, the Sony APR-5000, and Studer A807 have calculated take-up tension which is calculated from the tape speed and takeup reel rotation rate. By calculated, this could be an actual equation or a look-up table. There were four AES Convention papers and two papers in the Journal of the Audio Engineering Society by James Strickland of MCI discussing this in depth between 1971 and 1974. They are available online or in your local library.

A New Tape Transport Design

In approaching the design of a new tape transport, a decision had to be made at the outset whether to use the more conventional open-loop or one of the more exotic closed-loop geometries. The decision was made that the JH-10 transport would stick to proven open-loop design, but with refinements directly attacking certain commonly acknowledged deficiencies of the standard configuration.

Author: Strickland, James C.
Affiliation: MCI, Inc., Fort Lauderdale, FL
AES Convention:40 (April 1971) Paper Number:804-a

Speed, Pitch and Tension Revisited

A review of speed and tension control of professional audio transports is given. Systems are discussed with an eye toward helping the recording engineer understand the differences in basic methods and the specifics of how control is implemented. Phase-lock and several tensioning systems are discussed.

Author: Strickland, James C.
Affiliation: MCI, Inc., Fort Lauderdale, FL
AES Convention:49 (September 1974) Paper Number:968

Velocity Sensing-The Parameter for a Complete Tape Transport Motion Control

The use of reel-velocity sensing to improve the multimode handling of a two-inch tape transport is discussed. Operation in play, idle, runout, and fast modes is detailed. Also, considerations in using velocity and position information for non-overshoot automatic rewinds to a pre-selected location are detailed, with emphasis on approach-slope requirements for human-like operation.

Author: Strickland, James C.
Affiliation: MCI, Inc., Fort Lauderdale, FL
AES Convention:41 (October 1971) Paper Number:828

Velocity Sensing-The Parameter for a Complete Tape Transport Motion Control

The use of reel-velocity sensing to improve the multimode handling of a two-inch tape transport is discussed. Operation in play, idle, runout, and fast modes is detailed. Also, considerations in using velocity and position information for nonovershoot automatic rewinds to a preselected location are detailed, with emphasis on approach-slope requirements for human-like operation.

Author: Strickland, James C.
Affiliation: MCI, Inc., Fort Lauderdale, FL
JAES Volume 20 Issue 9 pp. 743-749; November 1972

Optimizing Tape Transport Motion Control Using Operational-Amplifier Analog Signal Processing

The use of monolithic operational amplifiers to process analog velocity signals in all motional modes of a modern tape transport is discussed. Circuits include summing, splitslope, and absolute value amplifiers; pulse-width modulators; and a new ac motor control method.

Author: Strickland, James C.
Affiliation: MCI, Inc., Ft. Lauderdale, FL
JAES Volume 22 Issue 3 pp. 172, 174, 176, 178; April 1974

Optimizing Tape Transport Motion Control Using Operational-Amplifier Analog Signal Processing

The ease with which low-level analog signals can be accurately processed with inexpensive, monolithic operational amplifiers provides new freedom for the design engineer in the pursuit of optimizing tape handling. The ability to process very low level DC signals with accuracy and stability is crucial in each of the four basic modes of a modern audio transport, e.g., Idle, Braking, Play, and Fast.

Author: Strickland, James C.
Affiliation: MCI, Inc., Fort Lauderdale, FL
AES Convention:46 (September 1973) Paper Number:932

EACH paper costs $20 for non-members, $5 for AES members and is free for E-Library subscribers via www.aes.org.