Re: [Emc-users] Chinese collet chucks [Was: Re: Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Gene Heskett 
On Sunday 14 January 2018 17:10:05 Chris Albertson wrote:

> If you buy the collets on eBay and you measure them, if they are out
> of spec you can return them as "not as described" and then seller is
> obligated to pay for the return postage.  You will get a prepaid
> shipping label. Many times sellers don't want to pay to return ship a
> bad part, after all what would they do with it?  So they jut resend a
> replacement.   It's a hassle to have to wait twice for you order but
> the up side is you might find a use for th defective parts they let
> you keep.
>
> I have four parts being replaced to me right now, One is a precision
> 8mm stainless bar that has rider marks on it and a linear
> recirculating ball bearing set that ha loose ball in the package. 
> They are seeing replacements.
>
> eBay is very safe this way.  If the seller does not handle the problem
> eBay can reverse the transaction and pull money out of the sellers
> account.
>
> So buy the Chinese part but make sure there is a spec on the web page
> and measure it after you unpack it   Most of the stuff is very good,
> better then spec'd   But I did just get some defective parts
>
This is a part that hard to measure.  The 50mm x1.5 threads have enough 
clearance that the nut falls from one side to the other of the threads 
clearance, giving you false high runout measurements, so chucking up a 
ground rod, putting the tension on the threads is about the only way to 
test it. I have some soft A2 thats centerless ground, but its someplace 
in the midden heap, stuck in a cardboard shipper tube, but I haven't 
seen it in 2 or 3 months. When the new ball bearing nut arrives, along 
with its wrench, I'll uncover it and use that to measure things with.

Then at some point I need to make a ring to fit the drawtube outside 
thruster pipe to center it in the left end of the spindle. As is, it 
wobbles about 1/8" and the weight of the hand wheel out on the end of 
it, limits the rpms to about 100 else it sets up a shake.  Lots of 
little piddly things to fine tune yet.

I was rather pleasantly surprised that the alu jig I made to hold that 
grizzly part worked so well. Remember I started out with about 7 thou of 
runout in the mouth of the spindle, which tells me it had a case of 
whiplash from the same fallover that wrecked the compound, long before I 
paid way too much for it. So my first operation to correct that was 
grinding a new mt5 in the spindle bore.

OTOH, expertise and experience like this has never been free. :-(
[...]

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Chinese collet chucks [Was: Re: Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Chris Albertson 
If you buy the collets on eBay and you measure them, if they are out of
spec you can return them as "not as described" and then seller is obligated
to pay for the return postage.  You will get a prepaid shipping label.
 Many times sellers don't want to pay to return ship a bad part, after all
what would they do with it?  So they jut resend a replacement.   It's a
hassle to have to wait twice for you order but the up side is you might
find a use for th defective parts they let you keep.

I have four parts being replaced to me right now, One is a precision 8mm
stainless bar that has rider marks on it and a linear recirculating ball
bearing set that ha loose ball in the package.  They are seeing
replacements.

eBay is very safe this way.  If the seller does not handle the problem eBay
can reverse the transaction and pull money out of the sellers account.

So buy the Chinese part but make sure there is a spec on the web page and
measure it after you unpack it   Most of the stuff is very good, better
then spec'd   But I did just get some defective parts



On Sun, Jan 14, 2018 at 11:44 AM, Les Newell 
wrote:

> I bought some Chinese BT30 and BT40 ER32 collet chucks off eBay a while
> back and was pleasantly surprised by how accurate they were. The finish
> looked good on all surfaces. I bought one before buying the rest and
> measured the bore when fitted in the spindle. My tenths dial gauge hardly
> moved. When I got the rest they were all pretty close. The nuts on the BT30
> chucks are a pain though. The bit that locates in the groove on the collet
> is concentric instead of being offset. This makes it a real pain to get the
> collets back out. Those ones are for my router and will spend most of their
> lives with 1/2" collets so I'm not that concerned.
>
> Les
>
>
>> There are a few reviews of Chinese collet chucks on youtube. The two
>> I've seen measure chuck internal taper runout similar to your
>> experience, Gene. (The outside's OK, but the bit that counts is poor.)
>> Once one of the Chinese collets is inserted, all bets are off, and it's
>> best for woodworking.
>>
>> One review showed one of the ostensibly ER32 collets in the set was over
>> 33mm in diameter, and couldn't be inserted in the nut in half a minute
>> of wrestling.
>>
>> Another video shows a Chinese collet chuck's internal taper being
>> ground. Nowhere have I seen anyone use one as-is.
>>
>> It appears to be best to remember how _much_ cheaper they are, and then
>> consider why. I wouldn't buy 'em, except as a kit to grind to tolerance,
>> chucking out the collets beyond rescue.
>>
>> To attempt a rescue, maybe skim the chuck internal taper, then hone it?
>> The chances of making it worse are slim. If it works well, there's a
>> very handy youtube video in it.
>>
>> Erik
>>
>
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Chinese collet chucks [Was: Re: Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Marcus Bowman 

On 14 Jan 2018, at 19:44, Les Newell wrote:

> I bought some Chinese BT30 and BT40 ER32 collet chucks off eBay a while back 
> and was pleasantly surprised by how accurate they were. The finish looked 
> good on all surfaces. I bought one before buying the rest and measured the 
> bore when fitted in the spindle. My tenths dial gauge hardly moved. When I 
> got the rest they were all pretty close. The nuts on the BT30 chucks are a 
> pain though. The bit that locates in the groove on the collet is concentric 
> instead of being offset. This makes it a real pain to get the collets back 
> out. Those ones are for my router and will spend most of their lives with 
> 1/2" collets so I'm not that concerned.
> 
> Les
> 
I have noticed that even good quality collets for expensive routers are tricky 
to get back out of a nut, once inserted. The ELU ones are held pretty securely, 
and the Kress ones are jolly difficult to get back out. Maybe that's why Kress 
try to sell each collet with its own nut.

Marcus

>> 
>> There are a few reviews of Chinese collet chucks on youtube. The two
>> I've seen measure chuck internal taper runout similar to your
>> experience, Gene. (The outside's OK, but the bit that counts is poor.)
>> Once one of the Chinese collets is inserted, all bets are off, and it's
>> best for woodworking.
>> 
>> One review showed one of the ostensibly ER32 collets in the set was over
>> 33mm in diameter, and couldn't be inserted in the nut in half a minute
>> of wrestling.
>> 
>> Another video shows a Chinese collet chuck's internal taper being
>> ground. Nowhere have I seen anyone use one as-is.
>> 
>> It appears to be best to remember how _much_ cheaper they are, and then
>> consider why. I wouldn't buy 'em, except as a kit to grind to tolerance,
>> chucking out the collets beyond rescue.
>> 
>> To attempt a rescue, maybe skim the chuck internal taper, then hone it?
>> The chances of making it worse are slim. If it works well, there's a
>> very handy youtube video in it.
>> 
>> Erik
> 
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> 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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Chinese collet chucks [Was: Re: Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Les Newell 
I bought some Chinese BT30 and BT40 ER32 collet chucks off eBay a while 
back and was pleasantly surprised by how accurate they were. The finish 
looked good on all surfaces. I bought one before buying the rest and 
measured the bore when fitted in the spindle. My tenths dial gauge 
hardly moved. When I got the rest they were all pretty close. The nuts 
on the BT30 chucks are a pain though. The bit that locates in the groove 
on the collet is concentric instead of being offset. This makes it a 
real pain to get the collets back out. Those ones are for my router and 
will spend most of their lives with 1/2" collets so I'm not that concerned.


Les



There are a few reviews of Chinese collet chucks on youtube. The two
I've seen measure chuck internal taper runout similar to your
experience, Gene. (The outside's OK, but the bit that counts is poor.)
Once one of the Chinese collets is inserted, all bets are off, and it's
best for woodworking.

One review showed one of the ostensibly ER32 collets in the set was over
33mm in diameter, and couldn't be inserted in the nut in half a minute
of wrestling.

Another video shows a Chinese collet chuck's internal taper being
ground. Nowhere have I seen anyone use one as-is.

It appears to be best to remember how _much_ cheaper they are, and then
consider why. I wouldn't buy 'em, except as a kit to grind to tolerance,
chucking out the collets beyond rescue.

To attempt a rescue, maybe skim the chuck internal taper, then hone it?
The chances of making it worse are slim. If it works well, there's a
very handy youtube video in it.

Erik



--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Jon Elson 

On 01/14/2018 11:39 AM, Gene Heskett wrote:

On Sunday 14 January 2018 12:23:51 Tom Easterday wrote:


On Jan 13, 2018, at 6:47 PM, andy pugh  wrote:

On 14 January 2018 at 00:27,  wrote:

But, as can be seen in the image below, the various retraction
paths (in blue) are not along a single driveline.  Why is that?

I believe that this is to that every pass is the same shape, so
ensure that everything is the same.
I am not sure that this makes any sense, given that the motion stops
and waits for the index.
(In fact I am not 100% certain that it does always stop, I think I
might have seen it start a path before stopping, and making a
slightly out-of-position cut when the index happens at just the
right point)

Why would it be important that the passes have the same shape?  What
is important is that the tool goes in and comes out in a known,
controllable location so you don’t hit the backside of your tool or
take the tops off your threads.  I don’t know if it always stops but
it better wait for index otherwise it isn’t spindle synchronized!

Your last sentence describes a bug doesn’t it?
-Tom

I think so, but I have yet to observe a damaged thread because of it.
However, I'd also recommend a g64p0.001 in the code just to make me feel
better. The curvature of the knee at l2 retract is affected and that can
be seen with enough magnification. IOW, it is not a show stopper.
Changing the spindle speed in the middle of the g76 run creates far
larger problems.

You do have to allow enough air pass for the Z to come up to 
speed while following the spindle synch.
Depending on the accel limits on the Z axis, the spindle 
speed and the thread pitch, this allowance for the synch-up 
will vary.  Coarse threads and high RPM require more allowance.


Jon

--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Gene Heskett 
On Sunday 14 January 2018 12:23:51 Tom Easterday wrote:

> > On Jan 13, 2018, at 6:47 PM, andy pugh  wrote:
> >> On 14 January 2018 at 00:27,  wrote:
> >>
> >> But, as can be seen in the image below, the various retraction
> >> paths (in blue) are not along a single driveline.  Why is that?
> >
> > I believe that this is to that every pass is the same shape, so
> > ensure that everything is the same.
> > I am not sure that this makes any sense, given that the motion stops
> > and waits for the index.
> > (In fact I am not 100% certain that it does always stop, I think I
> > might have seen it start a path before stopping, and making a
> > slightly out-of-position cut when the index happens at just the
> > right point)
>
> Why would it be important that the passes have the same shape?  What
> is important is that the tool goes in and comes out in a known,
> controllable location so you don’t hit the backside of your tool or
> take the tops off your threads.  I don’t know if it always stops but
> it better wait for index otherwise it isn’t spindle synchronized!
>
> Your last sentence describes a bug doesn’t it?
> -Tom

I think so, but I have yet to observe a damaged thread because of it.
However, I'd also recommend a g64p0.001 in the code just to make me feel 
better. The curvature of the knee at l2 retract is affected and that can 
be seen with enough magnification. IOW, it is not a show stopper. 
Changing the spindle speed in the middle of the g76 run creates far 
larger problems.
>
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Internal threading using external G76 tool path (back-tool lathe)

2018-01-14 Thread Tom Easterday 


> On Jan 13, 2018, at 6:47 PM, andy pugh  wrote:
> 
>> On 14 January 2018 at 00:27,  wrote:
>> 
>> But, as can be seen in the image below, the various retraction paths (in
>> blue) are not along a single driveline.  Why is that?
> 
> 
> I believe that this is to that every pass is the same shape, so ensure that
> everything is the same.
> I am not sure that this makes any sense, given that the motion stops and
> waits for the index.
> (In fact I am not 100% certain that it does always stop, I think I might
> have seen it start a path before stopping, and making a slightly
> out-of-position cut when the index happens at just the right point)

Why would it be important that the passes have the same shape?  What is 
important is that the tool goes in and comes out in a known, controllable 
location so you don’t hit the backside of your tool or take the tops off your 
threads.  I don’t know if it always stops but it better wait for index 
otherwise it isn’t spindle synchronized!

Your last sentence describes a bug doesn’t it?
-Tom

--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users