Dale wrote,
> If we want to measure the remaining life on a head, all we
> need to do is measure the distance from the worn face to
> the jog.  But how do we do that?  The top and bottom of the
> head aren't worn!  One way is to use the technique
> described by John French - measure the distance at the
> unworn top of the head, and then subtract the depth of the
> wear groove.  Another technique is to re-contour and lap
> the head to cut the unworn top and bottom edges of the head
> down to the working surface at the cores, and then measure
> the distance from the new face to the jog.  But if the head
> is nearly worn out when we get to the final measurement, we
> have just 'wasted' the relap work.  Other techniques are
> available to measure the wear before relapping with
> specialized equipment, but you and I don't have those
> tools.


Measuring the depth of wear on a magnetic head is
much easier than the estimating the remaining life.  The
method that I mentioned in my last posting is just an
easy way to get some idea as to how deep the wear is
when you don't have any measuring equipment.  The
actual depth will depend on how consistant the wear is
top to bottom and left and right of gap.  This will vary
depending on how good the head was aligned and, the
initial contour of the head.  Incorrectly relapped heads
with contour error will produce some most unusual looking
wear patterns.  I have some diagrams of typical wear
patterns and head construction (both ferrite and metal
heads) on our website.

Our wear depth measurement techniques at JRF include
the use of microscopes, depth indicators, optical comparitors
and special guages.  As Dale mentioned,many heads are
manufactured with an undercut reference mark (usually at
both ends of the head).  These reference marks are very
useful and the most accurate way to establish where in the
lifecycle the head is.  We simply measure the initial tip
depth (outside tape contact area at both ends of the head),
then measure the depth of wear, evaluate the wear pattern
and determine how much additional, if any, material will need
to be removed to correct for any wear pattern errors, and
come up with the number.  If the head is relapped, we make
a final measurement of the tip to the depth of the undercut
reference.  This is an exact reading and, compared to new
head specifications, a percentage of life remaining can be
established.  Heads manufactured by Mincom, Otari (except
1/4-inch),AMC (MCI and Sony multi-track), Woelke
(except erase), Saki (metal), Flux(most), Grandy, B&A.
and our PLX line all have tip depth undercut reference marks.
I am sure there are others.

Unfortunatly, some heads are manufactured without this
undercut reference mark. This complicates the evaluation
process and requires some additional number crunching.These
manufacturers include Ampex, Studer, Nortronics, Vikron, DRS,
Tascam, Fostex, Saki (ferrite) and again, others I'm sure.  Over
the years, JRF has compiled a rather large data base of head
specifications which includes both new and wearout readings
of most analog heads.  As a head wears, we all know that the
inductance drops.  Knowing these new and wearout inductance
readings allows us to calculate where in the life cycle a specific
head is, based on its inductance reading.  Since most head
tolerances vary when new, this is just a starting point.  We then
evaluate the surface of the head to determine if the head has ever
been relapped. (we can usually tell).  If it has not, then we measure
the depth of wear, compare that number to a known (typical)
tip depth for that manufacturer and model, refer
to inductance readings for conformation, and come up with a figure.
If the head has been relapped, it takes a lot more inspection and
calculating.  With relapped heads, we relay on inductance readings
first and wear data to back up our estimates.  Some heads, such as
Studer, have edge relief slots machined to the tip depth reference
point.  We do not always use this as a reference but, it can
sometimes help in an analysis and, is very good for an eyeball estimate.
(something I almost never do)

 > For Ampexites, the problem is that many of the Ampex heads
> did not use the undercut technique.  Fortunately, on the
> very old heads the core tips are dangling out in thin air,
> allowing easy tip measurement.  The black epoxy heads,
> however, can be a problem.  On a multitrack head, for
> example, a metal shoe is bonded to the bottom of the head,
> blocking any possibility of monitoring what is happening at
> that end of the head.  And on heads for which Ampex might
> have employed undercutting, black Bakelite core holders and
> black epoxy don't give an easily visible jog.  John, how do
> you measure those heads?

Ampex was kind enough to slot between each track and shield
enabling us to view the tip depth.  We are able to do this by mounting
the Ampex head into a fixture, at a 45 degree angle, and inspect it under
a microscope (80 power).  Slots between the tracks to be
inspected always require a  thorough cleaning (we use a solvent)
and usually render a good visual reading.   We always measure a
track at both ends and at least one in the middle.  This process
is performed at both initial and final inspection  (after the head is
relapped).  Unless we can reference to an undercut reference mark,
we state that our expected life remaining estimate is acurate within 15%.


> Before the undercut technique was used, the individual
> tracks within a head could vary as much as a couple of dB
> in output level.  Undercutting reduces the total variation
> across a 2" head to just a few tenths of a dB.
>

True.  And, this makes lamination stacking for production of cores
a lot less critical since both surfaces (crown and undercut) are ground
to very accurate tolerances.   This is the best (and only) way to
manufacture a precision 2-inch head with a consistent tip depth end to end.

Sorry to be so long winded on this.  Believe it or not, this is my second
shot .  I was finished with a much better version this morning and
just prior to hitting send, my computer crashed!  I love talking about
this but, I really didn't want to write it again.   Fortunately, it is quiet
around here this week.  Happy Holidays.....

John French