[Emc-users] z drive probs

[Gene Heskett]

Greetings guys;

That <$200 2505 by 1475 screw I used for z drive is haunting me, again.

First, I never did get around to shimming some preload into the thrust
bearing, so a years messing around later, the far end of the screw is
moving around .0055" as it takes up the reversal slack. And obviously
the balls in the ballnut need refilled with decent balls about .005"
bigger.

Then today I made a cutting tool, .0495" wide, to cut those 10 tpi square
threads I'll need on the end of this barrel, and I am damned glad I
picked up a piece of pvc pipe to practice on. Setting g76 with a q0.000,
I found it was drifting left approximately .001" per cycle, and that a
q-0.5 helped but something wasn't slipping uniformly, so I added a
microsecond to the driver pulses, no diff, then started on the
mechanicals, so the first 5 or so passes did not result in a usable
thread. I wound up snugging up the setscrews in the motor pulley, and in
the pulley driving the screw before that seemed to settle. The backlash
seems to vary at random by about 2 thou, so the starting point of each
sweep to the left at cutting depth wanders around about 2 thou.

I can, its a pita to do, reynolds wrap the thrust bearings to get a bit
of preload, which would cut the backlash in half, but where the heck do
I find oversized balls for the nut that are that grossly oversized?
AFAIK, its a std Chinese 2505 ball and screw, but I don't have a clue
how big the std balls for one of those critters are.

Has anyone here tried to rebuild one of these, or are those rolled screws
and its equally cheap nut so soft I'd be wasting my time?

Thanks all.

Cheers, Gene Heskett

[Gene Heskett]

Post by Gene Heskett
Greetings guys;
That <$200 2505 by 1475 screw I used for z drive is haunting me, again.
First, I never did get around to shimming some preload into the thrust
bearing, so a years messing around later, the far end of the screw is
moving around .0055" as it takes up the reversal slack. And obviously
the balls in the ballnut need refilled with decent balls about .005"
bigger.
Then today I made a cutting tool, .0495" wide, to cut those 10 tpi
square threads I'll need on the end of this barrel, and I am damned
glad I picked up a piece of pvc pipe to practice on. Setting g76 with
a q0.000, I found it was drifting left approximately .001" per cycle,
and that a q-0.5 helped but something wasn't slipping uniformly, so I
added a microsecond to the driver pulses, no diff, then started on the
mechanicals, so the first 5 or so passes did not result in a usable
thread. I wound up snugging up the setscrews in the motor pulley, and
in the pulley driving the screw before that seemed to settle. The
backlash seems to vary at random by about 2 thou, so the starting
point of each sweep to the left at cutting depth wanders around about
2 thou.

Now, there is one problem with committing all this to gcode. The is an
extractor clearance that reaches across the back face, and which must be
timed to the end face that stops the barrel at the proper rotation angle
such that this relief is exactly centered on the left side. But the
drawings I can source, do not give a measurement! So one would have to
to test fit it, and trim the stop face forward by making the cutaway,
then measureing the angle to determine how much to face off of the stop
face to bring the recess to the proper location.

It could be rather handy to have a starting angle variable for the g76
cycle that we could use to delay the start of the threading pass in
relation to the index pulse in order to duplicate what we can measure on
the shot out barrel being removed.

The 800lb gorilla is: would that bring LinuxCNC back under the ITAR
restrictions?

Post by Gene Heskett
I can, its a pita to do, reynolds wrap the thrust bearings to get a
bit of preload, which would cut the backlash in half, but where the
heck do I find oversized balls for the nut that are that grossly
oversized? AFAIK, its a std Chinese 2505 ball and screw, but I don't
have a clue how big the std balls for one of those critters are.
Has anyone here tried to rebuild one of these, or are those rolled
screws and its equally cheap nut so soft I'd be wasting my time?

The only replacement I can find is another C7 grade, with freight, $165
on my front deck in 5 or 6 weeks. Another twin to this one with its
sloppy build doesn't pull my trigger. C5's and C3's add $500 or $700
more. If I can src the balls, I'll gladly re-stuff this one.

Post by Gene Heskett
Thanks all.
Cheers, Gene Heskett

Cheers, Gene Heskett

[jeremy youngs]

Try ebay seller cwjkman , he sells surplus screws reasonably

Post by Gene Heskett

Post by Gene Heskett
Greetings guys;
That <$200 2505 by 1475 screw I used for z drive is haunting me, again.
First, I never did get around to shimming some preload into the thrust
bearing, so a years messing around later, the far end of the screw is
moving around .0055" as it takes up the reversal slack. And obviously
the balls in the ballnut need refilled with decent balls about .005"
bigger.
Then today I made a cutting tool, .0495" wide, to cut those 10 tpi
square threads I'll need on the end of this barrel, and I am damned
glad I picked up a piece of pvc pipe to practice on. Setting g76 with
a q0.000, I found it was drifting left approximately .001" per cycle,
and that a q-0.5 helped but something wasn't slipping uniformly, so I
added a microsecond to the driver pulses, no diff, then started on the
mechanicals, so the first 5 or so passes did not result in a usable
thread. I wound up snugging up the setscrews in the motor pulley, and
in the pulley driving the screw before that seemed to settle. The
backlash seems to vary at random by about 2 thou, so the starting
point of each sweep to the left at cutting depth wanders around about
2 thou.

Now, there is one problem with committing all this to gcode. The is an
extractor clearance that reaches across the back face, and which must be
timed to the end face that stops the barrel at the proper rotation angle
such that this relief is exactly centered on the left side. But the
drawings I can source, do not give a measurement! So one would have to
to test fit it, and trim the stop face forward by making the cutaway,
then measureing the angle to determine how much to face off of the stop
face to bring the recess to the proper location.
It could be rather handy to have a starting angle variable for the g76
cycle that we could use to delay the start of the threading pass in
relation to the index pulse in order to duplicate what we can measure on
the shot out barrel being removed.
The 800lb gorilla is: would that bring LinuxCNC back under the ITAR
restrictions?

Post by Gene Heskett
I can, its a pita to do, reynolds wrap the thrust bearings to get a
bit of preload, which would cut the backlash in half, but where the
heck do I find oversized balls for the nut that are that grossly
oversized? AFAIK, its a std Chinese 2505 ball and screw, but I don't
have a clue how big the std balls for one of those critters are.
Has anyone here tried to rebuild one of these, or are those rolled
screws and its equally cheap nut so soft I'd be wasting my time?

The only replacement I can find is another C7 grade, with freight, $165
on my front deck in 5 or 6 weeks. Another twin to this one with its
sloppy build doesn't pull my trigger. C5's and C3's add $500 or $700
more. If I can src the balls, I'll gladly re-stuff this one.

Post by Gene Heskett
Thanks all.
Cheers, Gene Heskett

Cheers, Gene Heskett
--
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
------------------------------------------------------------
------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users

[Gene Heskett]

Post by jeremy youngs
Try ebay seller cwjkman , he sells surplus screws reasonably

I found him, but how do I do a search within what he has?

I was not able to find loose balls that are oversize. I /think/ they
are .125 and 3.25mm are hens teeth.

Also. I found some screws, a pair of them TBE, with dual ball nuts.

<http://www.ebay.com/itm/2-set-Anti-Backlash-Ball-Lead-screw-RM2505-1500mm-C7-Dual-Ball-nut-end-machine/152072443766?_trksid=p2046732.c100040.m2060&_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1%26asc%3D20140107095009%26meid%3Df4f402ce00d94286a67ccc7b7b3cee5e%26pid%3D100040%26rk%3D4%26rkt%3D4%26sd%3D232093037158>

Can anyone tell me if what I see in that set of pix, the coupling tabs
with a screw in only one end, is a way of adjusting the phase and
therefore the backlash of that nut style?

By making the coupling tabs with an offset for instance? Or perhaps some
Reynolds wrap shims between the 2 halves?

These are 50mm longer but that would allow me to remachine the ends for a
better fit as the 1450 I put there a year ago is a tad short, on the
driving end in particular. That short 10mm stub needs to be at least
15mm longer than shown in order to fully seat the timeing pulley on it.
And another 25mm would give me room to do a taperlock hub, which I've
grown very fond of over the last 60 years. They do not slip and do not
need a keyway to do it.

Post by jeremy youngs

Post by Gene Heskett

Post by Gene Heskett
Greetings guys;
That <$200 2505 by 1475 screw I used for z drive is haunting me, again.
First, I never did get around to shimming some preload into the
thrust bearing, so a years messing around later, the far end of
the screw is moving around .0055" as it takes up the reversal
slack. And obviously the balls in the ballnut need refilled with
decent balls about .005" bigger.
Then today I made a cutting tool, .0495" wide, to cut those 10 tpi
square threads I'll need on the end of this barrel, and I am
damned glad I picked up a piece of pvc pipe to practice on.
Setting g76 with a q0.000, I found it was drifting left
approximately .001" per cycle, and that a q-0.5 helped but
something wasn't slipping uniformly, so I added a microsecond to
the driver pulses, no diff, then started on the mechanicals, so
the first 5 or so passes did not result in a usable thread. I
wound up snugging up the setscrews in the motor pulley, and in the
pulley driving the screw before that seemed to settle. The
backlash seems to vary at random by about 2 thou, so the starting
point of each sweep to the left at cutting depth wanders around
about 2 thou.

Now, there is one problem with committing all this to gcode. The is
an extractor clearance that reaches across the back face, and which
must be timed to the end face that stops the barrel at the proper
rotation angle such that this relief is exactly centered on the left
side. But the drawings I can source, do not give a measurement! So
one would have to to test fit it, and trim the stop face forward by
making the cutaway, then measureing the angle to determine how much
to face off of the stop face to bring the recess to the proper
location.
It could be rather handy to have a starting angle variable for the
g76 cycle that we could use to delay the start of the threading pass
in relation to the index pulse in order to duplicate what we can
measure on the shot out barrel being removed.
The 800lb gorilla is: would that bring LinuxCNC back under the ITAR
restrictions?

Post by Gene Heskett
I can, its a pita to do, reynolds wrap the thrust bearings to get
a bit of preload, which would cut the backlash in half, but where
the heck do I find oversized balls for the nut that are that
grossly oversized? AFAIK, its a std Chinese 2505 ball and screw,
but I don't have a clue how big the std balls for one of those
critters are.
Has anyone here tried to rebuild one of these, or are those rolled
screws and its equally cheap nut so soft I'd be wasting my time?

The only replacement I can find is another C7 grade, with freight,
$165 on my front deck in 5 or 6 weeks. Another twin to this one with
its sloppy build doesn't pull my trigger. C5's and C3's add $500 or
$700 more. If I can src the balls, I'll gladly re-stuff this one.

Post by Gene Heskett
Thanks all.
Cheers, Gene Heskett

Cheers, Gene Heskett
--
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
------------------------------------------------------------
------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users

----------------------------------------------------------------------
-------- Check out the vibrant tech community on one of the world's
most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users

Cheers, Gene Heskett

[andy pugh]

Post by Gene Heskett
I was not able to find loose balls that are oversize. I /think/ they
are .125 and 3.25mm are hens teeth.

Are ny of these an improvement?
https://www.amazon.com/gp/search/ref=sr_nr_n_1?fst=as%3Aoff&rh=n%3A350657011%2Ck%3Atoolsupply+ballscrew&keywords=toolsupply+ballscrew&ie=UTF8&qid=1499025439&rnid=2941120011

[Gene Heskett]

Post by andy pugh

Post by Gene Heskett
I was not able to find loose balls that are oversize. I /think/ they
are .125 and 3.25mm are hens teeth.

Are ny of these an improvement?
https://www.amazon.com/gp/search/ref=sr_nr_n_1?fst=as%3Aoff&rh=n%3A350
657011%2Ck%3Atoolsupply+ballscrew&keywords=toolsupply+ballscrew&ie=UTF8
&qid=1499025439&rnid=2941120011

Could very well be usefull, bookmarked, thanks a bunch Andy.

I've spent the day ripping the thrust bearing out and putting some .004"
brass shimstock between the outer races, which cut the backlash quite a
bit, from .0095" to .0029". So now, with a tool .0444" wide, its
cutting, in white pvc water pipe, grooves about .0553" wide, with lands
about .0445 wide. As if the tool was almost 10 thou too wide. I've got
some new hss tooling But to get the exact OD's I am going to have to
solvent glue a 3/4" pipe into a 1" pipe else the outer pipe just peels
off in strips when the cutter hits the joint line. I'll have to build me
a press frame to put that combo together with. 1.5" at a time, before
the glue sets up solid. But my back is done for the day now. It took
about 2 hours to get the 40 tooth pulley off the end of the screw, the
setscrews had slipped and threw up a ridge on the screws stub, which cut
into the alu bore locking the alu pulley in place. I finally wound up
useing the adjustment nut as a jacking device, moving the pulley about
1mm at a time, then jamming something into the space to wash, rinse &
repeat till it finally wiggled off the last 1/8". Not quite matching my
personal definition of fun while sitting on a roller stool about 10"
high & still ducking under to see. Don't get old guys, its not fun.

Cheers, Gene Heskett