Shugart 800-8 media centering problem

[tony duell]

> Not so on these drives.  On the Shugart SA-800 series, the spindle
> (driven by a belt from a motor) has a MALE cone on the end, and the
> clamp, from above goes around it (female if you like).

I have just pulled the case on one of my HP9885s, which uses an SA801 
chassis and HP electronics. The spindle clamping mechanism is as you
describe. I took the cover plate off the clamp arm and looked at the 
(female) cone, etc.

The cone is in 2 parts, with a bearing between them, the outer housing 
doesn't rotate in normal operation. There is also a spring ring round the 
outside of the nylon fingers. BUT I don't see how any of this has anything
to do with centering. If the disk is pushed down onto the male spindle end 
(and yes, the disk guides do move with clamp arm) it will centre. So all that
really affects things is the metal spinde.

I suppose gross errors in how the clamp comes down could do it, but 
unless the clamp arm is hanging on one pivot only, or something, I don't 
think that's it.

It virtually has to be a problem with the male cone end. I think you
need to measure that carefully and compare the measurements with
a good drive, look for burrs and wear. Remember that 10 thou is not
much. 

What if, due to some problem with the spindle cone, it clamps on one side
before the other? That could cause the disk to skew round and loose centering 
I think. 

-tony

The part of the spindle on the media of the side of the deck looks like
this:

           x        x
          x          x
         x            x
         y            y
   xxxxxxx            xxxxxx

At one point I thought, MAYBE, there was a little grouve at position "y"
that I could catch a blade of an Xacto or tiny screwdriver on on the bad
drive, but not the good one - but I am not at all sure of that.  If that
is true - if there really is a little groove there, then that would well
explain the symptoms.  It could get off center, and it  would NOT precess.

I would think that a spindle with a bearing issue would precess, so it
would never work during an entire floppy test - it would always exhibit
some wobble.

>
> That plastic part may be shown as one piece in the service manual as individual parts
> are not available as spares, but most of the time it's actually 2 bits with a spring inside.
>
>

I can see the entire nylon finger piece finger - it is completely
exposed except for a small bit under a screw and washer.  It has a bunch
of fingers attached to a disk - all one nylon piece during manufacture.
 The fingers can spread out.  The fingers have a flat part on them which
then presses the media against the flat part of the spindle.  They have
a spring around the outside of all of the fingers for tension.

Besides, as I have explained, I have switched out the entire assembly
with the fingers to no avail.

>> Haven't taken the spindle apart.  (Page 6, parts 2 through 10).  Not
>> sure I would want to - if it is that worn somewhere, requiring changing
>> out its bearing or the like, then for me the case is probably hopeless.
>
> If it's not the clamping cone then it pretty much has to be the spindle. A comment
> about pot metal got me thinking. If the cone has swollen due to oxidation of one
> component of the pot metal, it may be that the plastic cone is compressed too far too
> early and the media slips out of alignment.

The plastic fingers EXPAND on this design, AROUND the spindle, not compress.

>
> I would inspect the metal part of the spindle very carefully. I would check for excessive runout
> of the conical section. I would look for damage (burrs, etc). Do you have mechanical measuring
> tools (callipers, etc)?

No, but I could get some calipers.  They aren't expensive.  But if that
is the problem, it is a lost cause for me to fix anyway.

>
> The ball races are most likely standard parts and can be obtained. Other parts may have to
> be made.
>

And I'd have no way to make them.  And no point - it isn't like the
world is short of these drives.

>
>> My gut feeling about this remains that something interferes with the
>> media centering itself on the cone of the spindle before it gets clamped
>> down.
>
> My feeling is similar, that the clamping cone allows the media to slip about before it
> is finally clamped so that sometimes it works, sometimes it doesn't
>
>> What I am really looking for is someone who had a drive with similar
>> behavior that can say "hey, I had that too, and it was ......".
>
> Firstly I think it's a long shot that somebody else will have had this and secondly
> even if they have had the same symptoms it might be due to a different cause. Better
> to actually work out what is wrong with _your_ drive (sorry, but I have never gone along
> with the idea of stock faults and cures [1]).

Me neither, usually, but this one has me mystified, so I was curious of
someone else had ever seen similar symptoms.

>
> [1] May be OK if you have a lot of similar units to repair, say you are a repair shop and
> you keep seeing the same model of TV come in. Stock faults may then let you get 90% off
> the bench very quickly. But they are not much help with one-off machines like we tend to
> get on this list
>
> -tony
>
>
> JRJ
>