Re: [Emc-users] ER ATC tool idea

[Gene Heskett]

On Thursday 18 April 2019 20:02:45 Ken Strauss wrote:

> Weldon shank is round with a flat on the side. A setscrew prevents
> pullout.
>
> I haven't done personal measurements but I've often read that Weldon
> shank is worse than ER collets for low TIR when using small diameter
> tools which are usual in a high speed spindle.
>
In that case the TIR can't help but be worse as the setscrew will force 
it to the far side of any clearance.  So TIR's in the .002" range are 
not impossible.  In the case of an engraving v bit, the flat should be 
located such that this offset presents the cutting edge to the work as 
opposed to the heel, which likely will just burn the tool and the 
workpiece. We should be so lucky as to reliably get that timeing from 
the tool venders. While etching circuit boards, I've been known to stop 
if the cut is leaving a big burr, and rotate the tool a quarter turn in 
the ER and try again. Usually I have been able to get a cleaner cut.

A dead flat pallet, and a cut that does not dig into the glass makes a 
huge diff in the life of the tool, That dead flat bottom to the board 
pallet is one of the reasons I drill maybe 65% of the way thru the board 
and turn it over so the drill pattern does the same thing, with the 
holes meeting inside the thickness of the board, The drill never comes 
thru to damage the bottom of the pallet, but the etched side of the 
board does need to be scrubbed with $ steel wool so there are no 
burrs sticking up to warp the board. I drill and tap 0-80 flat headed 
screws to force the board to a consistent location since the pallets 
pocket is usually a couple thou oversized. Stiction in the z axis 
movement can be a board and bit killer if the Z ways aren't relubed and 
exercised to spread it before starting each side of the board. I've no 
clue if this 6040, running Z on rods, will be better or worse than the 
old HF I used to do that sort of work on.  The G0704 isn't noticeably 
better for movements that small. And of course thats as yet untested on 
this 6040.

Stiction was a huge problem on the old small HF, which is why it had 
extra 1/2" alu plates on each side of the sled, carrying 2 ball bearings 
that rode the face of the post, effectively extending its "wheelbase" 
from 2.5" to around 6". Gibs set just tight enough to make sure all 4 
bearings were hard to turn against the post. Wipe the dirt off the post, 
put a fresh squirt of vactra in the top of the gibs and the stiction 
disappeared for hours. It will be interesting to see how this 6040 
handles that when it gets its first pcb job. It did not come with lube 
instructions, just 40 pages of mach3 setup propaganda, and a demo copy 
of mach3 on a teeny cd. I'll never run it, no windows allowed on the 
property.

Thanks Ken, & take care.

[...]

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 



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

[Emc-users] Progress on 6040

[Roland Jollivet]

Hi Gene

I don't know how you're engaging the spanner on the spindle shaft for your
ATC, but you could use one of these;

https://www.ebay.com/itm/Durable-Car-Locking-System-2-Wire-Central-Door-Lock-Solenoid-Actuator-Motor-12V/131945015051?epid=511506243&hash=item1eb889270b:g:Ly0AAOSw4shX4NhW

Easier than air...
With linkages, you could mount it anywhere on the spindle head.

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

[Emc-users] ER ATC tool idea

[Roland Jollivet]

>
>
> Solid “collets” and off-the-shelf nuts seem so obvious that I am puzzled
> that I have never found them. It solves the tool length puzzle at a stroke.
>
>
>
Assuming you mean having a dedicated collet and nut per cutting bit;
You could just loctite the bit in the collet. Keep it tightened in a holder
until cured..

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

Re: [Emc-users] ER ATC tool idea

[Ken Strauss]

I have used some reasonable quality holders for Weldon shank tools -- Bison
R8 holders and Tormach TTS -- and none had angled setscrews. In all cases
the setscrew was very closely perpendicular to the axis of the tool.

> -Original Message-
> From: TJoseph Powderly [mailto:tjt...@gmail.com]
> Sent: Thursday, April 18, 2019 10:14 PM
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] ER ATC tool idea
>
> Greg thanks
>
> On 04/19/2019 08:36 AM, Gregg Eshelman via Emc-users wrote:
> >
> > The Weldon style has a set screw flat that's angled toward the cutting
> > end.
>
> > Tighten the set screw on it and the tool is forced upwards until the
> > deeper end of the flat contacts the side of the set screw.
>
> > That makes the too resistant to shifting axially, preventing loosening
> > or shifting of tool length.
>
> Thanks, I can see it working now and have used them on shank tool
> holders ( Bports etc )
>
> So, the set screw squeezes the angle to bottom out the shank in the bore.
> And, the end of shank and bottom of bore must be clean and solid.
>
> I recall doubting what I had used ,
> trying to jiggle and twist the tool with set screw just biting.
> They didnt 'feel' like the tool was constrained,
> The bore felt good, but the twist and length feature did not.
>
> I didnt chcek but would expect the one cheek of the Weldon to have marks
> from the screw.
>
> But, considering the shop owner,
> I was likely using cheap stuff from MSC ( who was down the street Elk
> Grove Illinois ).
>
> >
> > On Thursday, April 18, 2019, 5:59:29 PM MDT, Gene Heskett
> >  wrote: On Thursday 18 April 2019 19:49:10 Andy
> > Pugh wrote:
> >>> On 18 Apr 2019, at 23:29, Gene Heskett  wrote:
> >>> Cutting duties with 6mm end mills would still need a collets death
> >>> grip I think. YMMV of course.
> >> He did mention Weldon Shank.
> >
> > Ok, since I'm a new bee in terms of a weldon, what makes them magic?
> > Thanks Andy. Cheers, Gene Heskett
> > ___ Emc-users mailing
> list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> tomp
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users




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

Re: [Emc-users] ER ATC tool idea

[TJoseph Powderly]

Greg thanks

On 04/19/2019 08:36 AM, Gregg Eshelman via Emc-users wrote:


The Weldon style has a set screw flat that's angled toward the cutting 
end. 


Tighten the set screw on it and the tool is forced upwards until the 
deeper end of the flat contacts the side of the set screw. 


That makes the too resistant to shifting axially, preventing loosening 
or shifting of tool length.


Thanks, I can see it working now and have used them on shank tool 
holders ( Bports etc )


So, the set screw squeezes the angle to bottom out the shank in the bore.
And, the end of shank and bottom of bore must be clean and solid.

I recall doubting what I had used ,
trying to jiggle and twist the tool with set screw just biting.
They didnt 'feel' like the tool was constrained,
The bore felt good, but the twist and length feature did not.

I didnt chcek but would expect the one cheek of the Weldon to have marks 
from the screw.


But, considering the shop owner,
I was likely using cheap stuff from MSC ( who was down the street Elk 
Grove Illinois ).




On Thursday, April 18, 2019, 5:59:29 PM MDT, Gene Heskett 
 wrote: On Thursday 18 April 2019 19:49:10 Andy 
Pugh wrote:
On 18 Apr 2019, at 23:29, Gene Heskett  wrote: 
Cutting duties with 6mm end mills would still need a collets death 
grip I think. YMMV of course. 
He did mention Weldon Shank. 


Ok, since I'm a new bee in terms of a weldon, what makes them magic? 
Thanks Andy. Cheers, Gene Heskett 
___ Emc-users mailing list 
Emc-users@lists.sourceforge.net 
https://lists.sourceforge.net/lists/listinfo/emc-users

tomp


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

Re: [Emc-users] ER ATC tool idea

[TJoseph Powderly]

about weldon shank
https://tinyurl.com/yyxlreku

On 04/19/2019 06:52 AM, Andy Pugh wrote:
On 19 Apr 2019, at 00:47, Ken Strauss  wrote: 
there is no obvious way that the threads would achieve that level of 
concentricity. 
You can still register on the taper. Solid “collets” and off-the-shelf 
nuts seem so obvious that I am puzzled that I have never found them. 
It solves the tool length puzzle at a stroke. 

I dont understandwhat it solves about tool length.
I dont see the Weldon ensuring same tool length when tool bit is exchanged.
I do see it can repeat the tool length of a given pair of collet and 
tool-in-solid-collet.
___ Emc-users mailing list 
Emc-users@lists.sourceforge.net 
https://lists.sourceforge.net/lists/listinfo/emc-users

I have 10 collet-nut pairs eoc8a coming in post( for a Kress spindle )
5 with 3.175 collet, 5 with 6mm collet,
supposed to be .008mm concentricity.

BTW the Kress has an atc spindle addon ( a nose that fits over existing 
spindle )

https://tinyurl.com/y5zwlagb
*
*tomp

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

Re: [Emc-users] ER ATC tool idea

[Gregg Eshelman via Emc-users]

The Weldon style has a set screw flat that's angled toward the cutting end. 
Tighten the set screw on it and the tool is forced upwards until the deeper end 
of the flat contacts the side of the set screw. That makes the too resistant to 
shifting axially, preventing loosening or shifting of tool length. 

On Thursday, April 18, 2019, 5:59:29 PM MDT, Gene Heskett 
 wrote:  
 
 On Thursday 18 April 2019 19:49:10 Andy Pugh wrote:

> > On 18 Apr 2019, at 23:29, Gene Heskett  wrote:
> >
> > Cutting duties with 6mm end mills would still need a collets death
> > grip I think. YMMV of course.
>
> He did mention Weldon Shank.
>
>
Ok, since I'm a new bee in terms of a weldon, what makes them magic?

Thanks Andy.

Cheers, Gene Heskett  
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] ER ATC tool idea

[Ken Strauss]

Weldon shank is round with a flat on the side. A setscrew prevents pullout.

I haven't done personal measurements but I've often read that Weldon shank
is worse than ER collets for low TIR when using small diameter tools which
are usual in a high speed spindle.

> -Original Message-
> From: Gene Heskett [mailto:ghesk...@shentel.net]
> Sent: Thursday, April 18, 2019 7:58 PM
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] ER ATC tool idea
>
> On Thursday 18 April 2019 19:49:10 Andy Pugh wrote:
>
> > > On 18 Apr 2019, at 23:29, Gene Heskett  wrote:
> > >
> > > Cutting duties with 6mm end mills would still need a collets death
> > > grip I think. YMMV of course.
> >
> > He did mention Weldon Shank.
> >
> >
> Ok, since I'm a new bee in terms of a weldon, what makes them magic?
>
> Thanks Andy.
>
> 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 
>
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users




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

Re: [Emc-users] ER ATC tool idea

[Gene Heskett]

On Thursday 18 April 2019 19:49:10 Andy Pugh wrote:

> > On 18 Apr 2019, at 23:29, Gene Heskett  wrote:
> >
> > Cutting duties with 6mm end mills would still need a collets death
> > grip I think. YMMV of course.
>
> He did mention Weldon Shank.
>
>
Ok, since I'm a new bee in terms of a weldon, what makes them magic?

Thanks Andy.

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 



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

Re: [Emc-users] ER ATC tool idea

[Andy Pugh]

> On 19 Apr 2019, at 00:47, Ken Strauss  wrote:
> 
> there is no obvious way that the threads would achieve that level of
> concentricity.

You can still register on the taper. 

Solid “collets” and off-the-shelf nuts seem so obvious that I am puzzled that I 
have never found them. It solves the tool length puzzle at a stroke. 




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

Re: [Emc-users] ER ATC tool idea

[Andy Pugh]

> On 18 Apr 2019, at 23:29, Gene Heskett  wrote:
> 
> Cutting duties with 6mm end mills would still need a collets death grip I 
> think. YMMV of course.

He did mention Weldon Shank. 



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

Re: [Emc-users] ER ATC tool idea

[Ken Strauss]

Perhaps these would work for engraving since runout would be somewhat
non-critical. However, with small milling cutters you want TIR of 0.0002 or
so and there is no obvious way that the threads would achieve that level of
concentricity. I find that it is hard to get low TIR even using precision ER
collets.

> -Original Message-
> From: Gene Heskett [mailto:ghesk...@shentel.net]
> Sent: Thursday, April 18, 2019 6:30 PM
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] ER ATC tool idea
>
> On Thursday 18 April 2019 16:44:40 Roland Jollivet wrote:
>
> > Hi  All
> >
> > The current discussion on making an ATC for ER collet style spindles
> > has given me an idea.
> > As Gene aptly says, R8 and ER collets weren't designed with ATC's in
> > mind.
> >
> > In the application of routing, engraving and other light duty
> > machining, the assembly of ER collet, nut and tool could be reduced to
> > a simple screw-on tool holder.
> > The collet itself would be dispensed with and the nut itself becomes a
> > tool holder, with a Weldon style nose. Mostly it relies on the nut
> > mating with the face of the spindle.
> >
> > I have attached two pictures(not to scale) showing how it might look,
> > using a simple D bit, and where the second toolholder has a core that
> > engages with the cone of the spindle for heavier duty tooling.
> > A dozen of these could be machined by the user to take various tools.
> > Obviously this design is to used on the Hackaday or similar ATC
> >
> > https://ibb.co/6vpFs97
> > https://ibb.co/dpwVLkB
> >
> > Thoughts??
>
> Looks quite usable for engraving duties. But I'd be a bit spooky about
> the setscrews long term grip, and would favor a dimple in the side of
> the tool to foolproof it a bit.
>
> Cutting duties with 6mm end mills would still need a collets death grip I
> think. YMMV of course.
>
> 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 
>
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users




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

Re: [Emc-users] ER ATC tool idea

[Gene Heskett]

On Thursday 18 April 2019 16:44:40 Roland Jollivet wrote:

> Hi  All
>
> The current discussion on making an ATC for ER collet style spindles
> has given me an idea.
> As Gene aptly says, R8 and ER collets weren't designed with ATC's in
> mind.
>
> In the application of routing, engraving and other light duty
> machining, the assembly of ER collet, nut and tool could be reduced to
> a simple screw-on tool holder.
> The collet itself would be dispensed with and the nut itself becomes a
> tool holder, with a Weldon style nose. Mostly it relies on the nut
> mating with the face of the spindle.
>
> I have attached two pictures(not to scale) showing how it might look,
> using a simple D bit, and where the second toolholder has a core that
> engages with the cone of the spindle for heavier duty tooling.
> A dozen of these could be machined by the user to take various tools.
> Obviously this design is to used on the Hackaday or similar ATC
>
> https://ibb.co/6vpFs97
> https://ibb.co/dpwVLkB
>
> Thoughts??

Looks quite usable for engraving duties. But I'd be a bit spooky about 
the setscrews long term grip, and would favor a dimple in the side of 
the tool to foolproof it a bit.

Cutting duties with 6mm end mills would still need a collets death grip I 
think. YMMV of course.

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 



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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Gene Heskett]

On Thursday 18 April 2019 15:29:35 Roland Jollivet wrote:

>  > > Ditto if using an R8 directly, the R8 gets tightened with a 20
>  > > volt
> > >
> > > electric impact wrench driving an 8 point 10mm socket a lot
> > > tighter than I can draw it by hand with the supplied toy 10mm
> > > endwrench.  If not, the TTS in the R8 may slip and walk out. The
> > > life of that socket before it splits might be 50 times, its hell
> > > on good quality sockets.
>
> Hi Gene
>
> I found this video that may be food for thought.
>
I saw it even before I bought this thing.  And like you, either they 
can't write code that will do all that in one swell foop, or there was 
something wrong from the gitgo that I still haven't spotted.  That was 
also one honking big machine. I have not figured out what the piece 
bouncing around in the plate attached to the bottom of the motot to 
supply an anchor for the shaft lock.  It should be in or out, not 
bouncing around like a loose horseshoe.

I discounted the idea originally because of the cost of all those gears, 
and the ER32 nuts too.  Very well built, way too well built IMO. If I 
build it with a swing arm so it can stay off the table, I might consider 
such a set of relay gears to rotate the socket.  But that also would use 
up 1.5" of vertical move room this 6040 doesn't have to spare.  No way 
could I swap 2 drill bits half that long with the travel I have.

But with good driver code, it could work well on a machine you've already 
spent 10G's on.

> It's very slow because they were just testing out, and there's no
> follow up video either..  (still looking)
>
> https://www.youtube.com/watch?v=V0v_aP8hXR8
>
> From what I see;
> - this is a six station ATC
> - a motor drives a belt that drives two small gears
> - each small gear drives three large gears
> - each large gear carries a socket.
>
> While their sockets all rotate the same way, you could just cascade
> all the large gears in a straight line, and use opposite motion
> sequence for alternate pockets. With some mechanical loss as you go
> down the line.

I've not fine tuned it yet, but I finally did find the low speed torque 
boost controls in this VFD  So I may have that problem whupped. Maybe. 
Turns out the factory defaults place the frequency breakpoints for a 3 
point curve so low they have no effect at usable revs.  Nice way to keep 
a lid operator from burning up his motor, or the VFD, in the first 30 
minutes. 

Usefull settings TBD, but its progress.

Thanks & take care Rolland.

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 



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

[Emc-users] ER ATC tool idea

[Roland Jollivet]

Hi  All

The current discussion on making an ATC for ER collet style spindles has
given me an idea.
As Gene aptly says, R8 and ER collets weren't designed with ATC's in mind.

In the application of routing, engraving and other light duty machining,
the assembly of ER collet, nut and tool could be reduced to a simple
screw-on tool holder.
The collet itself would be dispensed with and the nut itself becomes a tool
holder, with a Weldon style nose. Mostly it relies on the nut mating with
the face of the spindle.

I have attached two pictures(not to scale) showing how it might look, using
a simple D bit, and where the second toolholder has a core that engages
with the cone of the spindle for heavier duty tooling.
A dozen of these could be machined by the user to take various tools.
Obviously this design is to used on the Hackaday or similar ATC

https://ibb.co/6vpFs97
https://ibb.co/dpwVLkB

Thoughts??

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

[Emc-users] Something to think about re the hack-a-day tool changer

[Roland Jollivet]

 > > Ditto if using an R8 directly, the R8 gets tightened with a 20 volt
> > electric impact wrench driving an 8 point 10mm socket a lot tighter
> > than I can draw it by hand with the supplied toy 10mm endwrench.  If
> > not, the TTS in the R8 may slip and walk out. The life of that
> > socket before it splits might be 50 times, its hell on good quality
> > sockets.
> >

Hi Gene

I found this video that may be food for thought.

It's very slow because they were just testing out, and there's no follow up
video either..  (still looking)

https://www.youtube.com/watch?v=V0v_aP8hXR8

>From what I see;
- this is a six station ATC
- a motor drives a belt that drives two small gears
- each small gear drives three large gears
- each large gear carries a socket.

While their sockets all rotate the same way, you could just cascade all the
large gears in a straight line, and use opposite motion sequence for
alternate pockets. With some mechanical loss as you go down the line.

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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[TJoseph Powderly]

 Forwarded Message 
Subject: 	Re: [Emc-users] Something to think about re the hack-a-day 
tool changer

Date:   Thu, 18 Apr 2019 23:22:19 +0700
From:   TJoseph Powderly 
To: Gene Heskett 



Hi Gene

On 04/18/2019 10:57 PM, Gene Heskett wrote:

On Thursday 18 April 2019 11:50:00 Ken Strauss wrote:
Google is your friend! ft-lb =Nm * 0.73756 So 75Nm = 55.3 ft lbs 
55 ft/lbs? Here the tool will walk out and break, very quickly at that 
tension. Doubtfull I have used less than 75 ft-pounds on the TTS 
adapters I use fairly universally on the G0704. 

on the web you can find suggested max torque for er collets
https://www.techniksusa.com/metal/torque_chart.htm
that url has torque wrench adapters as well ( $$$ )

( i have eoc8a collets tho, and have found no numeric values, only
descriptions of hand tightening )

you can find mfctrs saying use 75-80% of the max torque value.
you can find the caveats that the tool shank has to fully contact the
collet, not larger or smaller allowed.
you can find that the release is going to have stiction and release
requires more initial force.
other caveats about cleaning and air blasts are used in industry
( even collets covering the air blast holes and checking back pressure
in attempts to check good seating )

Andy's pragmatic checking of needed torque with a proper torque wrench
is very good ( what works is a good place to start)

also note the xatc ver 0.3 is already old and the google group on xatc
is cold since 2016.
you're kinda the lead developer now


-Original Message- From: Gene Heskett 
[mailto:ghesk...@shentel.net] Sent: Thursday, April 18, 2019 11:45 
AM To: emc-users@lists.sourceforge.net Subject: Re: [Emc-users] 
Something to think about re the hack-a-day tool changer On Thursday 
18 April 2019 07:47:51 andy pugh wrote:
On Thu, 18 Apr 2019 at 10:22, Roland Jollivet 

 wrote:
To recap; small steppers 'drive' the carousel around to tighten 
the motor. 
No, the motor drives round the carousel to tighten the collet. It 
is really rather clever. Assuming that it can't work based purely 
on guess-work seems silly. 
No, not silly, experience Andy, experience reaching into the bank 
account to replace tooling that broke by walking out of the holder 
when I thought it was tight enough.
I suggested a very simple test using the same torque wrench as Gene 
presumably uses now to tighten his collets to the correct torque to 
test the hypothesis. 
All 4 of my 5 torque wrenches are in foot/pounds. The smallest is in 
inch pounds. How much is 75Nm, in inch pounds.
But, assuming a 5mm pitch screw and 5:1 motor reduction: (0.5 Nm 
stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm arm 
that is 235Nm. Correct torque for an ER20 is 75Nm. 
That is no where near tight enough to keep it from walking out of 
the collet and breaking the tool because its digging 3/8" on a cut 
that should be 20 thou if it hadn't walked out. Here I use an end 
wrench about 15" long, ground thin enough to fit the flats on the 
TTS holder, and a 10 or 12" crescent wrench and a pull on the 
wrenches that is likely in excess of 100 lb-ft. I don't have a 
newtom-meters to pound-foot conversion handy but that certainly is 
tighter than 75Nm. Ditto if using an R8 directly, the R8 gets 
tightened with a 20 volt electric impact wrench driving an 8 point 
10mm socket a lot tighter than I can draw it by hand with the 
supplied toy 10mm endwrench. If not, the TTS in the R8 may slip and 
walk out. The life of that socket before it splits might be 50 
times, its hell on good quality sockets.
(Though I am not convinced that an ER17 can really lift a 300kg 
mass, so suspect I messed something up. ) 
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 
 
___ Emc-users mailing 
list Emc-users@lists.sourceforge.net 
https://lists.sourceforge.net/lists/listinfo/emc-users 
___ Emc-users mailing 
list Emc-users@lists.sourceforge.net 
https://lists.sourceforge.net/lists/listinfo/emc-users 

Cheers, Gene Heskett

tomp


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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Gene Heskett]

On Thursday 18 April 2019 11:50:00 Ken Strauss wrote:

> Google is your friend!
> ft-lb =Nm * 0.73756
> So 75Nm = 55.3 ft lbs
>
55 ft/lbs?  Here the tool will walk out and break, very quickly at that 
tension. Doubtfull I have used less than 75 ft-pounds on the TTS 
adapters I use fairly universally on the G0704.

> > -Original Message-
> > From: Gene Heskett [mailto:ghesk...@shentel.net]
> > Sent: Thursday, April 18, 2019 11:45 AM
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] Something to think about re the hack-a-day
> > tool changer
> >
> > On Thursday 18 April 2019 07:47:51 andy pugh wrote:
> > > On Thu, 18 Apr 2019 at 10:22, Roland Jollivet
> >
> >  wrote:
> > > > To recap; small steppers 'drive' the carousel around to tighten
> > > > the motor.
> > >
> > > No, the motor drives round the carousel to tighten the collet.
> > >
> > > It is really rather clever. Assuming that it can't work based
> > > purely on guess-work seems silly.
> >
> > No, not silly, experience Andy, experience reaching into the bank
> > account to replace tooling that broke by walking out of the holder
> > when I thought it was tight enough.
> >
> > > I suggested a very simple test using the same torque wrench as
> > > Gene presumably uses now to tighten his collets to the correct
> > > torque to test the hypothesis.
> >
> > All 4 of my 5 torque wrenches are in foot/pounds. The smallest is in
> > inch pounds.  How much is 75Nm, in inch pounds.
> >
> > > But, assuming a 5mm pitch screw and 5:1 motor reduction:
> > > (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a
> > > 75mm arm that is 235Nm.
> > > Correct torque for an ER20 is 75Nm.
> >
> > That is no where near tight enough to keep it from walking out of
> > the collet and breaking the tool because its digging 3/8" on a cut
> > that should be 20 thou if it hadn't walked out.  Here I use an end
> > wrench about 15" long, ground thin enough to fit the flats on the
> > TTS holder, and a 10 or 12" crescent wrench and a pull on the
> > wrenches that is likely in excess of 100 lb-ft. I don't have a
> > newtom-meters to pound-foot conversion handy but that certainly is
> > tighter than 75Nm.
> >
> > Ditto if using an R8 directly, the R8 gets tightened with a 20 volt
> > electric impact wrench driving an 8 point 10mm socket a lot tighter
> > than I can draw it by hand with the supplied toy 10mm endwrench.  If
> > not, the TTS in the R8 may slip and walk out. The life of that
> > socket before it splits might be 50 times, its hell on good quality
> > sockets.
> >
> > > (Though I am not convinced that an ER17 can really lift a 300kg
> > > mass, so suspect I messed something up. )
> >
> > 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 
> >
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> ___
> 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 



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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Stuart Stevenson]

Gene,
  One N/m is .737 ft/lb
  One hp is approx .745 watts
just think of the hp to watts conversion to mentally calculate the n/m to
ft/lb conversion - you will be close enough for almost any situation

in fact - using .75 for both conversions would work just fine

Stuart


On Thu, Apr 18, 2019 at 10:51 AM Ken Strauss  wrote:

> Google is your friend!
> ft-lb =Nm * 0.73756
> So 75Nm = 55.3 ft lbs
>
> > -Original Message-
> > From: Gene Heskett [mailto:ghesk...@shentel.net]
> > Sent: Thursday, April 18, 2019 11:45 AM
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] Something to think about re the hack-a-day tool
> > changer
> >
> > On Thursday 18 April 2019 07:47:51 andy pugh wrote:
> >
> > > On Thu, 18 Apr 2019 at 10:22, Roland Jollivet
> >  wrote:
> > > > To recap; small steppers 'drive' the carousel around to tighten the
> > > > motor.
> > >
> > > No, the motor drives round the carousel to tighten the collet.
> > >
> > > It is really rather clever. Assuming that it can't work based purely
> > > on guess-work seems silly.
> >
> > No, not silly, experience Andy, experience reaching into the bank account
> > to replace tooling that broke by walking out of the holder when I
> > thought it was tight enough.
> >
> > > I suggested a very simple test using the same torque wrench as Gene
> > > presumably uses now to tighten his collets to the correct torque to
> > > test the hypothesis.
> >
> > All 4 of my 5 torque wrenches are in foot/pounds. The smallest is in inch
> > pounds.  How much is 75Nm, in inch pounds.
> > >
> > >
> > > But, assuming a 5mm pitch screw and 5:1 motor reduction:
> > > (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
> > > arm that is 235Nm.
> > > Correct torque for an ER20 is 75Nm.
> >
> > That is no where near tight enough to keep it from walking out of the
> > collet and breaking the tool because its digging 3/8" on a cut that
> > should be 20 thou if it hadn't walked out.  Here I use an end wrench
> > about 15" long, ground thin enough to fit the flats on the TTS holder,
> > and a 10 or 12" crescent wrench and a pull on the wrenches that is
> > likely in excess of 100 lb-ft. I don't have a newtom-meters to
> > pound-foot conversion handy but that certainly is tighter than 75Nm.
> >
> > Ditto if using an R8 directly, the R8 gets tightened with a 20 volt
> > electric impact wrench driving an 8 point 10mm socket a lot tighter than
> > I can draw it by hand with the supplied toy 10mm endwrench.  If not, the
> > TTS in the R8 may slip and walk out. The life of that socket before it
> > splits might be 50 times, its hell on good quality sockets.
> > >
> > > (Though I am not convinced that an ER17 can really lift a 300kg mass,
> > > so suspect I messed something up. )
> >
> >
> > 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 
> >
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
>
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>


-- 
Addressee is the intended audience.
If you are not the addressee then my consent is not given for you to read
this email furthermore it is my wish you would close this without saving or
reading, and cease and desist from saving or opening my private
correspondence.
Thank you for honoring my wish.

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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Ken Strauss]

Google is your friend!
ft-lb =Nm * 0.73756
So 75Nm = 55.3 ft lbs

> -Original Message-
> From: Gene Heskett [mailto:ghesk...@shentel.net]
> Sent: Thursday, April 18, 2019 11:45 AM
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] Something to think about re the hack-a-day tool
> changer
>
> On Thursday 18 April 2019 07:47:51 andy pugh wrote:
>
> > On Thu, 18 Apr 2019 at 10:22, Roland Jollivet
>  wrote:
> > > To recap; small steppers 'drive' the carousel around to tighten the
> > > motor.
> >
> > No, the motor drives round the carousel to tighten the collet.
> >
> > It is really rather clever. Assuming that it can't work based purely
> > on guess-work seems silly.
>
> No, not silly, experience Andy, experience reaching into the bank account
> to replace tooling that broke by walking out of the holder when I
> thought it was tight enough.
>
> > I suggested a very simple test using the same torque wrench as Gene
> > presumably uses now to tighten his collets to the correct torque to
> > test the hypothesis.
>
> All 4 of my 5 torque wrenches are in foot/pounds. The smallest is in inch
> pounds.  How much is 75Nm, in inch pounds.
> >
> >
> > But, assuming a 5mm pitch screw and 5:1 motor reduction:
> > (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
> > arm that is 235Nm.
> > Correct torque for an ER20 is 75Nm.
>
> That is no where near tight enough to keep it from walking out of the
> collet and breaking the tool because its digging 3/8" on a cut that
> should be 20 thou if it hadn't walked out.  Here I use an end wrench
> about 15" long, ground thin enough to fit the flats on the TTS holder,
> and a 10 or 12" crescent wrench and a pull on the wrenches that is
> likely in excess of 100 lb-ft. I don't have a newtom-meters to
> pound-foot conversion handy but that certainly is tighter than 75Nm.
>
> Ditto if using an R8 directly, the R8 gets tightened with a 20 volt
> electric impact wrench driving an 8 point 10mm socket a lot tighter than
> I can draw it by hand with the supplied toy 10mm endwrench.  If not, the
> TTS in the R8 may slip and walk out. The life of that socket before it
> splits might be 50 times, its hell on good quality sockets.
> >
> > (Though I am not convinced that an ER17 can really lift a 300kg mass,
> > so suspect I messed something up. )
>
>
> 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 
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users




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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Gene Heskett]

On Thursday 18 April 2019 07:47:51 andy pugh wrote:

> On Thu, 18 Apr 2019 at 10:22, Roland Jollivet 
 wrote:
> > To recap; small steppers 'drive' the carousel around to tighten the
> > motor.
>
> No, the motor drives round the carousel to tighten the collet.
>
> It is really rather clever. Assuming that it can't work based purely
> on guess-work seems silly.

No, not silly, experience Andy, experience reaching into the bank account 
to replace tooling that broke by walking out of the holder when I 
thought it was tight enough.  

> I suggested a very simple test using the same torque wrench as Gene
> presumably uses now to tighten his collets to the correct torque to
> test the hypothesis.

All 4 of my 5 torque wrenches are in foot/pounds. The smallest is in inch 
pounds.  How much is 75Nm, in inch pounds.  
 
>
>
> But, assuming a 5mm pitch screw and 5:1 motor reduction:
> (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
> arm that is 235Nm.
> Correct torque for an ER20 is 75Nm.

That is no where near tight enough to keep it from walking out of the 
collet and breaking the tool because its digging 3/8" on a cut that 
should be 20 thou if it hadn't walked out.  Here I use an end wrench 
about 15" long, ground thin enough to fit the flats on the TTS holder, 
and a 10 or 12" crescent wrench and a pull on the wrenches that is 
likely in excess of 100 lb-ft. I don't have a newtom-meters to 
pound-foot conversion handy but that certainly is tighter than 75Nm.

Ditto if using an R8 directly, the R8 gets tightened with a 20 volt 
electric impact wrench driving an 8 point 10mm socket a lot tighter than 
I can draw it by hand with the supplied toy 10mm endwrench.  If not, the 
TTS in the R8 may slip and walk out. The life of that socket before it 
splits might be 50 times, its hell on good quality sockets. 

 
>
> (Though I am not convinced that an ER17 can really lift a 300kg mass,
> so suspect I messed something up. )


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 


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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Gene Heskett]

On Thursday 18 April 2019 05:30:13 TJoseph Powderly wrote:

> Hi Roland
>
> On Thu, Apr 18, 2019, 3:23 PM Roland Jollivet
> 
>
> wrote:
> > On Thu, 18 Apr 2019 at 01:22, Gene Heskett  
wrote:
> > > On Wednesday 17 April 2019 15:06:02 Chris Albertson wrote:
> > >
> > > [...]
> > >
> > > > But if maybe you lock the spindle and turn the nut.
> > >
> > > This is the case, except you are turnbing the nut by using the xy
> > > steppers to drive the carousel which is turning the nut.
> > >
> > > > Then your spindle
> > > > lock needs to have a torque gauge fitted.  The gauge is either a
> > > > spring and switch or a load cell.   The switch is much easier to
> > > > interface with. <
> >
> > https://lists.sourceforge.net/lists/listinfo/emc-users>
> >
> >
> > Hi Gene
> >
> > To recap; small steppers 'drive' the carousel around to tighten the
> > motor. Bad idea.
> >
> > Here's a different way of looking at it;
> > Keep the current stepper arrangement, but drive the carousel
> > instead, with a windscreen wiper motor.
> > You now have as much torque as you like in the carousel. Add a
> > reduction drive if you like. If you mount this under the bed you can
> > go to town on it.
> >
> > Gene might include a visual
>
> Gene could give a visual clue with each of these posts.

My movie camera is fairly hidef, a Sony handicam that works at 720x480, a 
1 minute clip is around 50 megabytes.  That would be abusing the server, 
so I don't. I  could also link to the movies on my web page, but my 
upload bw, at 175 kb sucks.0

> So search ' xatc ' in Google

I'll do that.
>
> Sorry I can't get this phone to copy the hackaday url ( stupid
> Newton!)
>
> You will see he needs to move the spindle about the carousel center.
> Not turn the carousel about it's center.
> ( The design as it is now)
> That needs an xy move of the carousel center Or an arc motion of the
> center or
> some lever motion or ( insert favorite idea here).
>
> The process would be to move into tool position, Z down, then either
> track
>
> > the motion of the carousel, or disable the stepper drives and let
> > them coast as the carousel takes the spindle with it as it
> > loosens/tightens. Once you've reached your target Amperage on the
> > motor, stop, and Z up. The X,Y is now a 'don't care', so home the
> > machine.
> > You then further rotate the carousel to home with a single switch so
> > it's always ready.
> >
> > Regards
> > Roland
>
> Hth tomp
>
It all helps, TomP. It gives me an assortment of ideas. Unanswered are 
questions that can only be determined by building it and testing it.

Like for instance using 'Merican tools because thats what I can 
(generally) buy here, and metric collets which are a lot smaller a jump 
between sizes, how many turns between using the spindle motor to "take 
up the slack", then driving the nut until a 4mm collet has a good grip 
on a 1/8" shank is needed. A 3mm collet just isn't big enough, but the 
collet crush in that 4mm case might be more than a full turn as the 
threads on an ER12 are finer than the threads on the TTS. I liked the 
idea of driveing the socket with a rack, until I calculated the needed 
rack length to get 2 full turns of the socket when its driving bull gear 
is a 3"er.  A good 12" of rack sticking out. Unforch, my imagination 
doesn't always come up with practical ideas, so I'm back to a chain 
drive, with a minimum small tooth sprocket of about 8 teeth, or a chain 
of idler gears, all on decent bearings or at least needle cartridges. 
Bring money in a little red wagon.  
   
>
Also is the problem I'm having with this YL-620-A VFD. I have found some 
of its config options, but low frequency boost escapes me so far. With 
an amprobe on a motor lead, I get max current at about 14 k revs of 
about 3.6 amps. At a requested 4000 revs, its actually turning about 200 
and I can stop it with one finger. At 3900, it hits maybe 500 at start 
up, but eventually fades to a stop with a very small pull occasionally 
while the H display claims something over 60 HZ but not much over an 
amp.

Take my finger away and it will eventually get back to maybe 100revs, 
drawing not much more than an amp.  No LF boost at all. The vendor 
hasn't replied to my help message other than saying he has asked his 
supplier, 2 days ago. Since, crickets. This is Thursday, and if no 
helpful info by Monday I think I'll ask ebay for a Haun Yang or a 
refund. I think I get the refund, extend my 250 volt single phase line 
to it and get a 220 volt motor and VFD of at least a 2.2KW rating. The 
one on the Sheldon is happy as a clam running that 1 hp motor at 10HZ. I 
think I've used the back gear once!  And its done it for half an hour at 
a time w/o heating the motor to the I can't touch it state.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-E

[Emc-users] Something to think about re the hack-a-day tool changer

[Roland Jollivet]

Disabling the X- Y drives sounds fine to me. Let the gantry get taken for
the ride.   (assuming it's easily moved by hand with drives disabled)
Once the nut is tight, reverse the carousel motor for a moment to undo the
load.
Then enable drives and do a Z up, then home the machine.

With a setup like this, homing after a toolchange would be prudent in any
case. And it would only add a few seconds per tool.

"Driving the carousel is an even worse idea.  Then the diameter of the
carousel becomes mechanical disadvantage"
I don't agree. Driving the carousel is more akin to tightening with a hand
wrench. Driving the gantry to do the same is probably beyond the ability of
most small systems.



On Thu, 18 Apr 2019 at 15:22, Todd Zuercher  wrote:

> Driving the carousel is an even worse idea.  Then the diameter of the
> carousel becomes mechanical disadvantage and the subsequent moving of the
> undriven x and y axis is guaranteed loss of position.
>
> Using the XY axis is probably the best option, for this kind of idea. But
> I can think of several reasons the concept on a whole is a bad one.
>
> 1st My experience with collets is they, the tapers they fit in, and the
> nut all need to be properly cleaned and inspected before reassembly and a
> simple blast of air usually isn't good enough, especially if you are
> milling wood products.
>
> 2nd Spindle bearing loading.  This kind of tightening load is going to
> directly load the spindle's bearings and will greatly reduce their service
> life (at least on high speed router spindles like we are talking about
> here).
>
> 3rd I think the potential of cross-threading or otherwise not starting the
> nut right are too high to rely on this for auto tool changes.
>
> Todd Zuercher
> P. Graham Dunn Inc.
> 630 Henry Street
> Dalton, Ohio 44618
> Phone:  (330)828-2105ext. 2031
>
> -Original Message-
> From: Roland Jollivet 
> Sent: Thursday, April 18, 2019 8:04 AM
> To: Enhanced Machine Controller (EMC) 
> Subject: [Emc-users] Something to think about re the hack-a-day tool
> changer
>
> Sorry, guys, I really do understand how it works. I thought I was quite
> clear..
>
> That's why I said the l steppers 'drive' the carousel around. Notice the '
> '. Yes, to spell it out the motion of the X-Y steppers in a circular
> motion results in the carousel rotating. The carousel rotates as a result
> of the XY stepper motion...
>
> The net result is that the carousel rotates while the nut is being
> loosened.
> The motive force for this to occur is the stepper motors. The carousel
> itself is passive.
>
> This is a bad idea because;
> - stiction can result in a much higher force to loosen the nut than
> tighten it
> - any lost steps in the stepper motors will result in a poor subsequent
> trajectory, it will be fighting all the way
> - there is no simple way to monitor the torque necessary, or applied.
>
> So... instead of using the steppers to effect the required action of
> loosening the nut, rather move... the motive force to the carousel.
> Visually the result is the same, but since the carousel is now DRIVEN
> instead of just coasting around, the benifits are;
> - geared DC motors of the required torque are readily available, or use a
> windscreen wiper motor with a further reduction
> - the motor current is easy to monitor
> - disable the stepper drives so the gantry becomes passive and is pulled
> around instead
> - indexing of the carousel only needs a single switch to home it.
>
> Regards
> Roland
>
>
>
> On Thu, 18 Apr 2019 at 13:50, andy pugh  wrote:
>
> > On Thu, 18 Apr 2019 at 10:22, Roland Jollivet
> > 
> > wrote:
> >
> > > To recap; small steppers 'drive' the carousel around to tighten the
> > motor.
> >
> > No, the motor drives round the carousel to tighten the collet.
> >
> > It is really rather clever. Assuming that it can't work based purely
> > on guess-work seems silly.
> >
> > I suggested a very simple test using the same torque wrench as Gene
> > presumably uses now to tighten his collets to the correct torque to
> > test the hypothesis.
> >
> >
> > But, assuming a 5mm pitch screw and 5:1 motor reduction:
> > (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
> > arm that is 235Nm.
> > Correct torque for an ER20 is 75Nm.
> >
> > (Though I am not convinced that an ER17 can really lift a 300kg mass,
> > so suspect I messed something up. )
> >
> > --
> > atp
> > "A motorcycle is a bicycle with a pandemonium attachment and is
> > designed for the especial use of mechanical geniuses, daredevils and
> > lunatics."
> > — George Fitch, Atlanta Constitution Newspaper, 1916
> >
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
> 

[Emc-users] Linuxcnc meeting 2019 in Stuttgart, Germany

[Rene Hopf via Emc-users]

Hi,

We are looking for a date for the next linuxcnc meeting in Stuttgart, Germany.
All info here:
https://doodle.com/poll/7aeg98zwbi6qrtvx

Rene



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

[Emc-users] Emc-users] Something to think about re the hack-a-day tool changer

[Roland Jollivet]

Disabling the X- Y drives sounds fine to me. Let the gantry get taken for
the ride.   (assuming it's easily moved by hand with drives disabled)
Once the nut is tight, reverse the carousel motor for a moment to undo the
load.
Then enable drives and do a Z up, then home the machine.

With a setup like this, homing after a toolchange would be prudent in any
case. And it would only add a few seconds per tool.

"Driving the carousel is an even worse idea.  Then the diameter of the
carousel becomes mechanical disadvantage"
I don't agree. Driving the carousel is more akin to tightening with a hand
wrench. Driving the gantry to do the same is probably beyond the ability of
most small systems.


On Thu, 18 Apr 2019 at 15:22, Todd Zuercher  wrote:

> Driving the carousel is an even worse idea.  Then the diameter of the
> carousel becomes mechanical disadvantage and the subsequent moving of the
> undriven x and y axis is guaranteed loss of position.
>
> Using the XY axis is probably the best option, for this kind of idea. But
> I can think of several reasons the concept on a whole is a bad one.
>
> 1st My experience with collets is they, the tapers they fit in, and the
> nut all need to be properly cleaned and inspected before reassembly and a
> simple blast of air usually isn't good enough, especially if you are
> milling wood products.
>
> 2nd Spindle bearing loading.  This kind of tightening load is going to
> directly load the spindle's bearings and will greatly reduce their service
> life (at least on high speed router spindles like we are talking about
> here).
>
> 3rd I think the potential of cross-threading or otherwise not starting the
> nut right are too high to rely on this for auto tool changes.
>
> Todd Zuercher
> P. Graham Dunn Inc.
> 630 Henry Street
> Dalton, Ohio 44618
> Phone:  (330)828-2105ext. 2031
>
> -Original Message-
> From: Roland Jollivet 
> Sent: Thursday, April 18, 2019 8:04 AM
> To: Enhanced Machine Controller (EMC) 
> Subject: [Emc-users] Something to think about re the hack-a-day tool
> changer
>
> Sorry, guys, I really do understand how it works. I thought I was quite
> clear..
>
> That's why I said the l steppers 'drive' the carousel around. Notice the '
> '. Yes, to spell it out the motion of the X-Y steppers in a circular
> motion results in the carousel rotating. The carousel rotates as a result
> of the XY stepper motion...
>
> The net result is that the carousel rotates while the nut is being
> loosened.
> The motive force for this to occur is the stepper motors. The carousel
> itself is passive.
>
> This is a bad idea because;
> - stiction can result in a much higher force to loosen the nut than
> tighten it
> - any lost steps in the stepper motors will result in a poor subsequent
> trajectory, it will be fighting all the way
> - there is no simple way to monitor the torque necessary, or applied.
>
> So... instead of using the steppers to effect the required action of
> loosening the nut, rather move... the motive force to the carousel.
> Visually the result is the same, but since the carousel is now DRIVEN
> instead of just coasting around, the benifits are;
> - geared DC motors of the required torque are readily available, or use a
> windscreen wiper motor with a further reduction
> - the motor current is easy to monitor
> - disable the stepper drives so the gantry becomes passive and is pulled
> around instead
> - indexing of the carousel only needs a single switch to home it.
>
> Regards
> Roland
>
>
>
> On Thu, 18 Apr 2019 at 13:50, andy pugh  wrote:
>
> > On Thu, 18 Apr 2019 at 10:22, Roland Jollivet
> > 
> > wrote:
> >
> > > To recap; small steppers 'drive' the carousel around to tighten the
> > motor.
> >
> > No, the motor drives round the carousel to tighten the collet.
> >
> > It is really rather clever. Assuming that it can't work based purely
> > on guess-work seems silly.
> >
> > I suggested a very simple test using the same torque wrench as Gene
> > presumably uses now to tighten his collets to the correct torque to
> > test the hypothesis.
> >
> >
> > But, assuming a 5mm pitch screw and 5:1 motor reduction:
> > (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
> > arm that is 235Nm.
> > Correct torque for an ER20 is 75Nm.
> >
> > (Though I am not convinced that an ER17 can really lift a 300kg mass,
> > so suspect I messed something up. )
> >
> > --
> > atp
> > "A motorcycle is a bicycle with a pandemonium attachment and is
> > designed for the especial use of mechanical geniuses, daredevils and
> > lunatics."
> > — George Fitch, Atlanta Constitution Newspaper, 1916
> >
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
> _

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Todd Zuercher]

Driving the carousel is an even worse idea.  Then the diameter of the carousel 
becomes mechanical disadvantage and the subsequent moving of the undriven x and 
y axis is guaranteed loss of position.

Using the XY axis is probably the best option, for this kind of idea. But I can 
think of several reasons the concept on a whole is a bad one.

1st My experience with collets is they, the tapers they fit in, and the nut all 
need to be properly cleaned and inspected before reassembly and a simple blast 
of air usually isn't good enough, especially if you are milling wood products.  

2nd Spindle bearing loading.  This kind of tightening load is going to directly 
load the spindle's bearings and will greatly reduce their service life (at 
least on high speed router spindles like we are talking about here).

3rd I think the potential of cross-threading or otherwise not starting the nut 
right are too high to rely on this for auto tool changes.

Todd Zuercher
P. Graham Dunn Inc.
630 Henry Street 
Dalton, Ohio 44618
Phone:  (330)828-2105ext. 2031

-Original Message-
From: Roland Jollivet  
Sent: Thursday, April 18, 2019 8:04 AM
To: Enhanced Machine Controller (EMC) 
Subject: [Emc-users] Something to think about re the hack-a-day tool changer

Sorry, guys, I really do understand how it works. I thought I was quite clear..

That's why I said the l steppers 'drive' the carousel around. Notice the '
'. Yes, to spell it out the motion of the X-Y steppers in a circular motion 
results in the carousel rotating. The carousel rotates as a result of the XY 
stepper motion...

The net result is that the carousel rotates while the nut is being loosened.
The motive force for this to occur is the stepper motors. The carousel itself 
is passive.

This is a bad idea because;
- stiction can result in a much higher force to loosen the nut than tighten it
- any lost steps in the stepper motors will result in a poor subsequent 
trajectory, it will be fighting all the way
- there is no simple way to monitor the torque necessary, or applied.

So... instead of using the steppers to effect the required action of loosening 
the nut, rather move... the motive force to the carousel.
Visually the result is the same, but since the carousel is now DRIVEN instead 
of just coasting around, the benifits are;
- geared DC motors of the required torque are readily available, or use a 
windscreen wiper motor with a further reduction
- the motor current is easy to monitor
- disable the stepper drives so the gantry becomes passive and is pulled around 
instead
- indexing of the carousel only needs a single switch to home it.

Regards
Roland



On Thu, 18 Apr 2019 at 13:50, andy pugh  wrote:

> On Thu, 18 Apr 2019 at 10:22, Roland Jollivet 
> 
> wrote:
>
> > To recap; small steppers 'drive' the carousel around to tighten the
> motor.
>
> No, the motor drives round the carousel to tighten the collet.
>
> It is really rather clever. Assuming that it can't work based purely 
> on guess-work seems silly.
>
> I suggested a very simple test using the same torque wrench as Gene 
> presumably uses now to tighten his collets to the correct torque to 
> test the hypothesis.
>
>
> But, assuming a 5mm pitch screw and 5:1 motor reduction:
> (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm 
> arm that is 235Nm.
> Correct torque for an ER20 is 75Nm.
>
> (Though I am not convinced that an ER17 can really lift a 300kg mass, 
> so suspect I messed something up. )
>
> --
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is 
> designed for the especial use of mechanical geniuses, daredevils and 
> lunatics."
> — George Fitch, Atlanta Constitution Newspaper, 1916
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

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

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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Andy Pugh]

> On 18 Apr 2019, at 14:04, Roland Jollivet  wrote:
> 
> - disable the stepper drives so the gantry becomes passive and is pulled
> around instead

That’s even worse than the risk of losing steps by driving the carousel with 
the axes. 

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

[Emc-users] Something to think about re the hack-a-day tool changer

[Roland Jollivet]

Sorry, guys, I really do understand how it works. I thought I was quite
clear..

That's why I said the l steppers 'drive' the carousel around. Notice the '
'. Yes, to spell it out the motion of the X-Y steppers in a circular
motion results in the carousel rotating. The carousel rotates as a result
of the XY stepper motion...

The net result is that the carousel rotates while the nut is being loosened.
The motive force for this to occur is the stepper motors. The carousel
itself is passive.

This is a bad idea because;
- stiction can result in a much higher force to loosen the nut than tighten
it
- any lost steps in the stepper motors will result in a poor subsequent
trajectory, it will be fighting all the way
- there is no simple way to monitor the torque necessary, or applied.

So... instead of using the steppers to effect the required action of
loosening the nut, rather move... the motive force to the carousel.
Visually the result is the same, but since the carousel is now DRIVEN
instead of just coasting around, the benifits are;
- geared DC motors of the required torque are readily available, or use a
windscreen wiper motor with a further reduction
- the motor current is easy to monitor
- disable the stepper drives so the gantry becomes passive and is pulled
around instead
- indexing of the carousel only needs a single switch to home it.

Regards
Roland



On Thu, 18 Apr 2019 at 13:50, andy pugh  wrote:

> On Thu, 18 Apr 2019 at 10:22, Roland Jollivet 
> wrote:
>
> > To recap; small steppers 'drive' the carousel around to tighten the
> motor.
>
> No, the motor drives round the carousel to tighten the collet.
>
> It is really rather clever. Assuming that it can't work based purely
> on guess-work seems silly.
>
> I suggested a very simple test using the same torque wrench as Gene
> presumably uses now to tighten his collets to the correct torque to
> test the hypothesis.
>
>
> But, assuming a 5mm pitch screw and 5:1 motor reduction:
> (0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
> arm that is 235Nm.
> Correct torque for an ER20 is 75Nm.
>
> (Though I am not convinced that an ER17 can really lift a 300kg mass,
> so suspect I messed something up. )
>
> --
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is
> designed for the especial use of mechanical geniuses, daredevils and
> lunatics."
> — George Fitch, Atlanta Constitution Newspaper, 1916
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[Gene Heskett]

On Thursday 18 April 2019 04:20:37 Roland Jollivet wrote:

> On Thu, 18 Apr 2019 at 01:22, Gene Heskett  
wrote:
> > On Wednesday 17 April 2019 15:06:02 Chris Albertson wrote:
> >
> > [...]
> >
> > > But if maybe you lock the spindle and turn the nut.
> >
> > This is the case, except you are turnbing the nut by using the xy
> > steppers to drive the carousel which is turning the nut.
> >
> > > Then your spindle
> > > lock needs to have a torque gauge fitted.  The gauge is either a
> > > spring and switch or a load cell.   The switch is much easier to
> > > interface with.
> > > 
>
> Hi Gene
>
> To recap; small steppers 'drive' the carousel around to tighten the
> motor. Bad idea.
I tend to agree, given the power limits available from the steppers.
>
> Here's a different way of looking at it;
> Keep the current stepper arrangement, but drive the carousel instead,
> with a windscreen wiper motor.
> You now have as much torque as you like in the carousel. Add a
> reduction drive if you like. If you mount this under the bed you can
> go to town on it.
>
> The process would be to move into tool position, Z down, then either
> track the motion of the carousel,

Difficult to synch, but probably doable.
> or disable the stepper drives and 
> let them coast as the carousel takes the spindle with it as it
> loosens/tightens. Once you've reached your target Amperage on the
> motor, stop, and Z up. The X,Y is now a 'don't care', so home the
> machine.

That would likely mean moving the tool changer at least to the middle of 
the Y range, or a long move to get y homed again. And the carousel 
becomes half a wheel that has to be in its own home position to allow 
the gantry risers to get past it.
The windshield wiper motor with a big geardown, say 6/1, could then use 
the wipers parking switch to position the carousel at one of 6 tool 
positions, with one more home switch. 

What I had in mind was to borrow the right rear 200mmx200mm so a gunstock 
blank could fit beside it. Then the gantry risers could clear it 
regardless.

And of coarse there's the y offset of the z mount, making the rear 80mm 
in-accessible. Currently the coolant tank occupies the space but theres 
nothing saying I can't make a riser for it to make 18" of working room 
for a swing in carousel under it. The limited Z range will be a problem 
for long tools in any event.

I'd like to steer clear of useing a transfer arm to reach into the 
working area, but powering a rotating socket on the end of it to do the 
tightening and loosening of the collet nut has a certain  cachette to 
it. That way the xy remains at the tool change location so a rehome 
won't be needed. And the transfer arm stays outside the path of the 
gantry when parked at the next tool.  And so do the arm driving motor(s, 
one to drive the arm and one to drive the socket). Driving the socket 
with a rack, with a bearing to back up the rack at the point of contact 
sounds good, and beats a whole chain of spur gears I saw in one design.  
So it sticks out but if the stickout is only when in position there 
shouldn't be anything but air for it to hit.  Design length enough in 
the rack to allow a full turn of the nut, and run it to one end or the 
the other depending on whether you are removing or inserting the tool 
before engaging the nut. Get the gear ratio needed by a bigger spur gear 
on the sockets base and a smaller gear on the wiper motor could do that. 
Thats 3 wiper motors to run.  Thinking... 

> You then further rotate the carousel to home with a single switch so
> it's always ready.

Or put the tool back in the pocket it came out of so the tool table can 
track it w/o that codeing hassle, then rotate to the new pocket before 
the arm picks up the next tool. And with all that, the tool changer is 
going to use up quite a bit of i/o. And a sheet of 1/4" ply cut up just 
to make try patterns. :)  All that bs makes buying an atc equipt spindle 
seem like the cheap option.

Frankly, neither R8 nor the various sizes of ER's were made with tool 
changers in mind. Either style demands a TLO calibration once the new 
tool is in place.  But I keep letting my imagination out to play. Code 
is a lot cheaper than hardware.

Thanks and take care, Roland.


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 



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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[andy pugh]

On Thu, 18 Apr 2019 at 10:22, Roland Jollivet  wrote:

> To recap; small steppers 'drive' the carousel around to tighten the motor.

No, the motor drives round the carousel to tighten the collet.

It is really rather clever. Assuming that it can't work based purely
on guess-work seems silly.

I suggested a very simple test using the same torque wrench as Gene
presumably uses now to tighten his collets to the correct torque to
test the hypothesis.


But, assuming a 5mm pitch screw and 5:1 motor reduction:
(0.5 Nm stepper * (2 * pi) / 5mm ) * 5:1 = 3kN. On the end of a 75mm
arm that is 235Nm.
Correct torque for an ER20 is 75Nm.

(Though I am not convinced that an ER17 can really lift a 300kg mass,
so suspect I messed something up. )

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916


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

Re: [Emc-users] Something to think about re the hack-a-day tool changer

[TJoseph Powderly]

Hi Roland

On Thu, Apr 18, 2019, 3:23 PM Roland Jollivet 
wrote:

> On Thu, 18 Apr 2019 at 01:22, Gene Heskett  wrote:
>
> > On Wednesday 17 April 2019 15:06:02 Chris Albertson wrote:
> >
> > [...]
> > >
> > > But if maybe you lock the spindle and turn the nut.
> >
> > This is the case, except you are turnbing the nut by using the xy
> > steppers to drive the carousel which is turning the nut.
> >
> > > Then your spindle
> > > lock needs to have a torque gauge fitted.  The gauge is either a
> > > spring and switch or a load cell.   The switch is much easier to
> > > interface with. <
> https://lists.sourceforge.net/lists/listinfo/emc-users>
> >
>
> Hi Gene
>
> To recap; small steppers 'drive' the carousel around to tighten the motor.
> Bad idea.
>
> Here's a different way of looking at it;
> Keep the current stepper arrangement, but drive the carousel instead, with
> a windscreen wiper motor.
> You now have as much torque as you like in the carousel. Add a reduction
> drive if you like. If you mount this under the bed you can go to town on
> it.
>
> Gene might include a visual
>


Gene could give a visual clue with each of these posts.

So search ' xatc ' in Google

Sorry I can't get this phone to copy the hackaday url ( stupid Newton!)

You will see he needs to move the spindle about the carousel center.
Not turn the carousel about it's center.
( The design as it is now)
That needs an xy move of the carousel center Or an arc motion of the center
or
some lever motion or ( insert favorite idea here).

The process would be to move into tool position, Z down, then either track
> the motion of the carousel, or disable the stepper drives and let them
> coast as the carousel takes the spindle with it as it loosens/tightens.
> Once you've reached your target Amperage on the motor, stop, and Z up. The
> X,Y is now a 'don't care', so home the machine.
> You then further rotate the carousel to home with a single switch so it's
> always ready.
>
> Regards
> Roland
>

Hth tomp

>
>
>
>

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

[Emc-users] Something to think about re the hack-a-day tool changer

[Roland Jollivet]

On Thu, 18 Apr 2019 at 01:22, Gene Heskett  wrote:

> On Wednesday 17 April 2019 15:06:02 Chris Albertson wrote:
>
> [...]
> >
> > But if maybe you lock the spindle and turn the nut.
>
> This is the case, except you are turnbing the nut by using the xy
> steppers to drive the carousel which is turning the nut.
>
> > Then your spindle
> > lock needs to have a torque gauge fitted.  The gauge is either a
> > spring and switch or a load cell.   The switch is much easier to
> > interface with. 
>

Hi Gene

To recap; small steppers 'drive' the carousel around to tighten the motor.
Bad idea.

Here's a different way of looking at it;
Keep the current stepper arrangement, but drive the carousel instead, with
a windscreen wiper motor.
You now have as much torque as you like in the carousel. Add a reduction
drive if you like. If you mount this under the bed you can go to town on it.

The process would be to move into tool position, Z down, then either track
the motion of the carousel, or disable the stepper drives and let them
coast as the carousel takes the spindle with it as it loosens/tightens.
Once you've reached your target Amperage on the motor, stop, and Z up. The
X,Y is now a 'don't care', so home the machine.
You then further rotate the carousel to home with a single switch so it's
always ready.

Regards
Roland

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