In article ,
"DoN. Nichols" wrote:

On 2008-03-23, Joseph Gwinn wrote:
In article ,
"DoN. Nichols" wrote:

On 2008-03-22, Joseph Gwinn wrote:
In article ,
"DoN. Nichols" wrote:

[ ... ]

Well ... Opera has been the best browser for my use (on a Sun
UltraSPARC Workstation running Solaris 10) with only one site regularly
crashing it on the first visit each time the page is changed. That is
a
web-based comic called Ctl-Alt-Del.

OK. Maybe they have solved the many problems I found.

I think so. At least on the Sun Solaris workstations, which is
the only place that I've tried it so far.

Perhaps I'll re-evaluate Opera. Nor do they seem likely to die now.
It has to have been at least 5 years.

O.K. I seem to remember that when I first got it, I would have
had to pay if I were a commercial user, but as a home users I was in the
clear. So that might be a bit after your $40.00 fee period.

Back when they had sugarplum dreams.


And someone who insists on publishing a web comic entirely in
Flash -- which I could not read for a while until the Flash plugin
version for Solaris finally caught up with his requirements.

Before, he had been running one which would work fine except
for not having a singing-dancing logo. Then he went to requiring the
latest Flash for the whole thing -- and thus lost me for several months.

If he does something like that again, he will lose me
permanently. :-)

Flash is a problem to be sure. On my Mac, with Firefox, flash windows
are simply absent, while Safari works OK with Flash. I would guess from
your comments that Flash will work OK on Opera as well.

Yes -- if you download the plug in and put it in the right
place. On a Mac, I would guess that the "put it in the right place" is
pretty automated. I tend to download the static libraries version of
the Solaris, with the gzip compression of a tar file so I can have more
personal control over where to put it than a ".pkg" version would give.
I tend to put each version in its own subdirectory in /opt so I can back
up to an earlier version if something breaks. (I do this with almost
everything which does not automatically install in /opt anyway. :-)

I've never been sufficiently motivated to figure out why Flash doesn't
work on Firefox MaxOS. Actually, crippling Flash is usually an
advantage.


Now there is a thought. Also, I wonder if DTF comes thinner.

I would have to go downstairs and do some digging to find out
which version I'm currently using on the Sanford surface grinder
spindle.

DTF24 is the lowest viscosity member, at ISO 32.

MSC lists two spindle oils, Mobil Velocite and Tru-Edge. Both are
thinner than DTF24, being no thicker than ISO 22, so it's probably a
good idea to get some real spindle oil.

For the DuMore toolpost grinder -- at least go for the lube
which they specify in the manual. You got a copy of my manual, didn't
you?

Yes. But I doubt that their recommended spindle oil is still available.


Well, it would become a permanent part of the setup. The BXA
toolpost
has a 3/8" (by eye) blind hole in the bottom, clearly intended for an
anti-rotation pin. So such a spacer plate would have a ridge below
and
a 3/8 hole to accept a short 3/8 dowel rod.

O.K. I've just verified that my Phase-II has the same hole. I
didn't bother measuring it (yet) to see whether it wound up at some
metric size close to 3/8". :-)

On the BXA, there is also a threaded hole on the blank side that faces
the tailstock.


I was thinking of milling a plate down so the thickness away from the
ridge is about 0.125". That's plenty for the pin to engage. The
alternative is to drill a hole in the compound.

O.K. I'm not sure whether 1/8" is thin enough. I would have to
check again. I think that I come out near 1/10" instead.

A 0.1" spacer is also thick enough for the index pin to be reliable.
The spacer is mostly in pure compression, and so need not be thick.

And -- I just measured, and find 0.550" clearance between the bottom of
the tool holder and the top of the compound, so one is certainly on the
project list. I might even make two plates -- one with the 3/8" pin and
one hole, and the other with an arc of holes to accept a pin at all of
the common compound settings (90 degrees, +/- 14 degrees, +/- 29.5
degrees,+/- 45 degrees etc) which I can shift without having to lift
the toolpost clear of the 3/8" pin. Put a large arc of holes on the side of
the compound away from the workpiece so I can get higher resolution and
easier access for changing. Perhaps tap the upper hole, and make a
dog-point screw to engage the lower hole.

Also, with a one-plate stack, you may need two plates with different
hole patterns to keep the holes far enough apart to not overlap. With
the two-plate stack, one can have two circles of holes, allowing
staggering and non-overlap.


My latest cycle with chatter while cutting off a steel bar is that
turning the test bar down to eliminate the rough surface did not solve
the chatter when the now trued bar was clamped in the 3-jaw chuck,
although a pry test now shows more or less equal displacement of chuck
and bar, implying that the bar is now firmly clamped in the chuck.

O.K.

Machining to reduce diameter and/or threading is quiet and uneventful,
although the threads are ratty, with cracks perpendicular to the surface
and to the direction of tool travel. This is true both for the original
test bar (not marked, but appears to be 1018 CRS) and for a 0.75"
diameter 1018 CRS bar bought new.

1018 and the like tend to machine poorly, and this could account
for the poor threads. If you want something lovely to machine, get some
12L14 (leaded free-machining steel). Its only disadvantage (other than
somewhat higher cost) is that it is very poor for welding.

I reread Moltrech (sp?) on threading, and I see one problem: I was
going straight in, cutting on both sides of the threadcutting tool,
instead of sliding in at a 30 degree angle, cutting on one side or the
other.


A 0.75" diameter CRS bar held in a collet was easily cut off, without
chatter or drama. One test I will make is to hold a piece of that same
0.75" diameter CRS rod in the 3-jaw chuck and try to cut it off.

It cut off without drama, if I recall.

I was also able to groove this CRS bar with the cutoff tool while the
bar was between centers, without drama. (I found the missing MT4.5 to
MT3 sleeve mixed in with the pile of 5C collets.)


And also see what happens when you try to thread that.

It threaded without drama, in chuck, in collet and between centers.

I don't actually know which steel alloy the big bar is. It came with
the lathe, as a rusty bit of rod.


I've also been assiduously cleaning and deburring the mating surfaces of
the L00 taper on headstock spindle (male) and on 3-jaw chuck (female).
I did see a stuck chip come off when I used a very fine file on the
headstock taper. Prior wiping with a shop towel had not dislodged this
chip, which was hard to get to, being up under the big threaded clamping
ring. Perhaps stuck chips were the root cause.

That could be. The 1018 may be another contributor.

I tried turning *very* slowly. It chattered, but very slowly.

Touching various gaps (between machine parts that slide with respect to
one another), the largest relative vibration found is between carriage
and rear hold-down plates. This has about 0.0015" clearance. Perhaps
it needs to be tightened. Limited by wear on the bed way undersides.

Front hold-down plates not tried, but will be. Harder to get at.

Then I tried going very fast, almost 2000 rpm. This worked, although it
still was on the verge of chattering, and made sliver chips rather than
curls. But I was able to cut the big bar off without shaking the house,
and the cut faces were very smooth, as were the bottoms of the grooves.


I also discovered that with the 3-jaw chuck running at 2000 rpm, just
stopping (set to 3 seconds) causes the VFD to trip from DC overvoltage
(the deaccelerating 2 HP motor acts as a generator). No harm done, but
time for an external resistor.

Joe Gwinn