[vcf-midatlantic] Zilog System 8000

[Dave McGuire]

On 11/17/2015 02:28 PM, Jonathan Gevaryahu via vcf-midatlantic wrote:
>>>> Keep in mind: it's the metal drive roller which determines tape
>>>> speed, not the roller.
>>> This is exactly why I know as well. It is just there to get the "speed
>>> on the tape" but it is not mission critical to get the dimensions 100%
>>> correct.
>>> Remember - the capstan dimensions change over time anyway because the
>>> material will be constantly removed during operating the tape. This is
>>> a normal process. IF the capstan dimensions would be vital for the
>>> correct function of the tape drive, it would have been unable to read
>>> the times not far after it was shipped - probably.
>>    Actual tape speed over the head is actually not that critical for many
>> types of tape drives.  Most modulation/encoding schemes are
>> self-clocking, either through separate clock tracks (DEC TU56) or
>> embedded clocks that are recovered via a phase-locked loop.  Getting
>> close will do it.
>>
>>    Speed variations, on the other hand, can be bad if a PLL is used for
>> clock recovery.  If the period of the speed variation falls outside of
>> the clock recovery PLL's loop bandwidth, the loop could lose lock,
>> resulting in the loss of the clock.  So, keep those replacement rollers
>> as round as possible, but don't worry so much about the exact diameter,
>> unless you know for a fact that there is an absolute tape speed
>> requirement, which is rare.
>
> These z8000 tapes are MFM, so they should be self-clocking, I think?

  I think that's a reasonable assumption.  MFM schemes usually use
PLL-based clock recovery.

             -Dave

-- 
Dave McGuire, AK4HZ
New Kensington, PA