Mark;
With reference to IBM's plan for a head foundry in China I am reasonably certain that it refers to an HGA/HSA assy plant. I suspect "sliders" will come from one of their three worldwide slider plants. (San Jose, S'Pore, and Mainz, Germany).
<<Do you find all head manufacturers make HGA's and HSA's?>>
Yes, and not necessarily happy to do so. In the early days Read-Rite resisted HSA assy citing low margin and high yield risk. But customers were forceful. Also, since a good deal of head manufacturing is vertical integration, then, of course, those head manufactures do the whole sub assy. I know for sure that SAE(TDK), Yamaha, and Kaifa do as well.
<<When MR heads were new, I had read that there was a lot of waste of suspensions because heads could only be tested once assembled.>>
Actually this is not true. Early MR (and still to this day) suffered yield loss due to electrostatic discharge. The MR element is very susceptive to thermal errors and until this was better understood and controlled in the process there was quite a bit of yield loss all the way through the pipeline (including at customer sights while the drive was being assembled). This has now largely been obviated against by tightening ESD controls and handling procedures. But one thing interesting about an MR element is that the magnetic phenomena is not dependent on rotation as in a thin film inductive head. This has actually led to a static MR head test. There is a private company called Phase Metrics Corp. (who probably would have been public by now were it not for the market turn) that has pioneered static MR head testing and they supply testers to head makers for this purpose. This means that the MR element can be tested not only at the HGA level but also at the row bar level (before sliders are sliced and diced) and even the wafer level (before wafers are sliced into row bars for final polish and edge blend). While all MR manufacturers still use a rotating HGA level test for final acceptance of the HGA and again at HSA levels, many have instituted a STATIC MR test as well, to catch bad elements before more value is added downstream. With Thin Film Inductive heads, you are correct, that for a read/write functional test, they had to wait until the slider was assembled into a full HGA before it could be mounted on a spin stand and passed over a reference surface (media with known magnetic characteristics) for a read/write function to test it. I do not know if manufacturers will ever depend solely on a static test for MR/GMR. My guess is that their customers will want to see results of a full spin stand test which, in a way, behaves like a big disk drive in that it supplies a pre-amplified signal, a spindle, and a positioner for a true read/write chain.
best,
Stitch
|
