# Magnetic tape's head alignment: NRZI vs. PE/Manchester coding

• Jul 3rd 2011, 11:55 AM
courteous
Magnetic tape's head alignment: NRZI vs. PE/Manchester encoding
Quote:

The bit cells of each track of (say, 9-track) magnetic tape must be precisely vertically aligned when using NRZI encoding.

(a) Explain why such a requirement is necessary. What is the maximum allowed deviation from a vertical alignment when data density is 800 bpi.
(b) Does such a requirement exist also for PE/Manchester encoded tapes? What is the maximum allowed deviation in this case? What measures can be taken to increase the allowed deviation?
Would you accept the following answers? How would you correct/improve them?

(a) The NRZI encoding is not self-synchronizing, hence we must demand the vertical alignment to be as perfect as possible. The maximum allowed deviation from the vertical would be 1/2 the width of a bit cell in each direction ("left" and "right") of the tape? That is, $\frac{1}{2}*\frac{1}{800}\frac{\text{inch}}{\text{ bit}}$?

Even less sure about the (b) question. (Blush)

(b) No, there isn't such requirement for PE encoded tapes? The maximum allowed deviation is directly proportional to the number of bits "a buffer" can hold. (There is "a buffer" when using PE/Manchester encoding, right? (Wondering)) This "allowed deviation" can be increased by increasing the buffer?