Re: [Emc-users] EDM forms?

[Gene Heskett]

On Thursday 01 August 2019 20:51:56 Gene Heskett wrote:

>
> If you can see why I'm only seeing a 90 degree arc in the backplot,
> teach me what is wrong.

Wrong K. Its running, but very very slowly. About .5 ipm  And from the 
looks of the carbide tool I'm using for electrode, its going away faster 
than the steel is.  So from the looks of it, I've gone up the hill to 
the shop building and picked my container of busted carbide to use for 
electrodes before I done.

And of course Murphy was watching carefully. My EDM psu died, transformer 
I think but its never been hot!  So I rounded up a 200 watt or so 
toroid, and the only other paper in a can capacitor I had, but cheap 
tests everything gismo says its ESR is somewhere in the 30 ohms range, 
downright sucky, and about 13 u-f.  So now I have a 70 volt open circuit 
supply and it seems to be making a decent fire. But the carbide is going 
away faster than that steel. I'm down .034" but the chord of the burn is 
just barely a quarter inch long.

At this point, I'm wondering if its worth it to stop, and see if I can 
use the THC module as a feed override, letting it run likity s--t until 
the average voltage is below 60, indicating the fire has started, in 
which case cut the feed override to about 5% while its burning, but run 
it at 25 ipm when its not.  That would cut the runtime quite a bit.

Ideas, as in has anyone done this?

Thanks Andy, all.

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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[Gene Heskett]

On Thursday 01 August 2019 13:01:25 andy pugh wrote:

> On Thu, 1 Aug 2019 at 16:47, Gene Heskett 
> wrote:
>
> Did that got g3 error, usual recipe, converted it back to g17, lcnc
>
> > bought it but back plot shows the arc as a straight line???
>
That turned out to be a missing F in the G2/3 lines.  So I have it 
working, sortof. arcgenm18.py is showing a full 180 degree path, but 
when that code is pasted into an ngc file, I am getting only the bottom 
90 degrees in the backplot.

> At this point I would give up and use R-format arcs.

Its getting tempting.

Here is the code:
%
( for a woodruff key, .750 long, .1885 wide, depth about 170 thou )
( from arcgenm18.py, z curve offset starting point for arc )
G20 (use inches)
#<_x_offset>= -0.3750
#<_y_center>=- 0.
#<_x_end>   = -#<_x_offset>
( Z motion from X#<_x_offset )
( G3 X0.3750 Y0. I0.3750 J-0. )
( translate to G18  y becomes z, j becomes k)
( G3 X0.3750 Z0. I0.3750 K-0. )
( comeing back )
( G2 X-0.3750 Y0. I-0.3750 J-0. )
(Translated to G18)
( G2 X-0.3750 Z0. I-0.3750 K-0. )
#<_z_tmp_start> = .3750
(intialize z-tmp)
#<_z_tmp>   = #<_z_tmp_start>
( tool is carbide, .125 diameter dull 2 flute mill )
(key is .1885 wide so)
#<_key_wobble>  = [[ .18855 -.125] / 2.]
#<_z_inc>   =   .001 (burn 1 thou per pass)
S300 m4 (run it backwards)

o50 WHILE [#<_z_tmp> gt 0.000]

G1 F10 x#<_x_offset> y[#<_y_center> + #<_key_wobble>]
G1 F10 z#<_z_tmp>
G18 G2 F1 X0.3750 I0.375000 K#<_z_tmp_start>
G17
G1 F5 y[-#<_key_wobble>]
G18 G3 F1 X-0.3750 I-0.375000 K#<_z_tmp_start>
G17
G1 F.2 y[#<_y_center> - #<_key_wobble>]
#<_z_tmp> = [#<_z_tmp> - #<_z_inc>]

o50 ENDWHILE
G17
m5
g0z5
M2
%

If you can see why I'm only seeing a 90 degree arc in the backplot, teach 
me what is wrong.

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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[andy pugh]

On Thu, 1 Aug 2019 at 16:47, Gene Heskett  wrote:

Did that got g3 error, usual recipe, converted it back to g17, lcnc
> bought it but back plot shows the arc as a straight line???
>

At this point I would give up and use R-format arcs.

-- 
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] EDM forms?

[Gene Heskett]

On Thursday 01 August 2019 05:52:06 Gene Heskett wrote:

> On Thursday 01 August 2019 02:59:37 andy pugh wrote:
> > On Thu, 1 Aug 2019 at 04:36, Gene Heskett 
> > wrote:
> >
> > output of arcgenm18.py is assumeing G17. not G18.  So I need to
> >
> > > translate the ijk into what it would be for G18 motion. Is there a
> > > rule of thumb for that?
> >
> > G17 XY IJ
> > G18 XZ IK
> > G19 YZ JK
>
> Thanks Andy. It would be very helpfull if that was added to the docs.
>
> > So to change a G17 arc to a G18 change Y to Z and J to K.
> >
Did that got g3 error, usual recipe, converted it back to g17, lcnc 
bought it but back plot shows the arc as a straight line???

Now clarify please. I have a #<_z_tmp>, which the gcode sets to=.1875
This equals the radius of the arc and also represents the depth as the 
while [#<_z_tmp> gt 0.], meaning that if no erosion on the carbide, 
the final pass, decrementing at -0.0005 per pass, the last pass will be 
a few thou over the depth the key sits in the slot. or it will be 
shallow if the carbide has seriously eroded.  The only y  motions are 
those that move to work the width of the key, ostensively to make the 
groove .1890" wide. those motions take place at the ends of the g2 or g3 
moves so that from above the tool is
  _
 ||
=$64K-Question=
Now in order to maintain the shape of the half circle, while effectively 
moving it to an ever increasing depth and running in G18 anything 
referencing z must have the z_tmp added so the center point decrements 
with z_tmp. Right?


So as I step thru the code, watch it get to setting the starting height, 
but instead of raising z to 0.1875, it stops at about .070" So x and y 
seem to be running  in inches, but z is maybe running in mm but showing 
inches. But thats just a theory at this  point. Different problem but 
this is 2.9 master too... I need to find that too

I noted the shaft was showing a teeny bit of rust from sitting in the 
distilled water over night, and that the blocks I had positioned and 
gooped to prop it into alignment hadn't stuck to the poly tray, so I 
dumped everything and cleaned up the shaft, and now have a 1x2 by 9 
piece of white ash gooped to the bottom of the water tray sitting out in 
the sun so maybe it will stick better. Thick enough the shaft can be 
laid lengthwise w/o the tray pushing on the ends.

So while all that glue is setting, I can work on duff code.
And maybe look around in my junk box for another useable capacitor. This 
sort of pulsing will kill an electrolytic in a short time, not to 
mention it can't deliver the amps to make a destructive spark like an 
oil/paper or mylar can.
 
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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[Gene Heskett]

On Thursday 01 August 2019 02:59:37 andy pugh wrote:

> On Thu, 1 Aug 2019 at 04:36, Gene Heskett 
> wrote:
>
> output of arcgenm18.py is assumeing G17. not G18.  So I need to
>
> > translate the ijk into what it would be for G18 motion. Is there a
> > rule of thumb for that?
>
> G17 XY IJ
> G18 XZ IK
> G19 YZ JK

Thanks Andy. It would be very helpfull if that was added to the docs.

> So to change a G17 arc to a G18 change Y to Z and J to K.
>
> You won't get a true segment of a cylinder with a cylindrical tool.
> Which was why I suggested a rotating disc that looks like an actual
> Woodruff cutter.


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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[Gene Heskett]

On Thursday 01 August 2019 02:59:37 andy pugh wrote:

> On Thu, 1 Aug 2019 at 04:36, Gene Heskett 
> wrote:
>
> output of arcgenm18.py is assumeing G17. not G18.  So I need to
>
> > translate the ijk into what it would be for G18 motion. Is there a
> > rule of thumb for that?
>
> G17 XY IJ
> G18 XZ IK
> G19 YZ JK
>
Thank you.

> So to change a G17 arc to a G18 change Y to Z and J to K.
>
> You won't get a true segment of a cylinder with a cylindrical tool.
> Which was why I suggested a rotating disc that looks like an actual
> Woodruff cutter.

Of no concern, the back of the key has a flat area, something like the 
rear .075"  is removed from a key that would be if a full 3/4" circle, 
such that the key is only .3" deep, not .375". So the flat spot in the 
middle that a cylindrical tool cutting on the end of it while doing a 
180 degree circle's run in G18 mode is actually fitting the key better 
than a regular key cutter. If in cuttable steel useing a key mill, one 
would have to cut straight in to needed depth, then advance the cutter 
lengthwise by at least 3/8" in order to duplicate the shape of the key.

With the lack of working room, and the need to figure out how to keep the 
key from tipping when its entering the groove in the pulley, its a right 
cast iron bitch to fit.  So IF I can locate another capacitor (the guy 
in Bulgaria that has all the soviet surplus is out of pocket till late 
August, caps like this have not been made in any great qty since there 
was a run on them  when it became sick bird to use pcb's over here, 
about 30 years back. It cost the station around $20K to get rid of most 
of what we had by having Westinghouse send a truck to pick it up and 
burn it at nearly 7000F, pcb's are dambed hard to destroy) This now 
failed one was spared the fire by not having a pyranol on its label at 
the time. Modern plastic caps to replace it are $100+ and probably only 
available as 30 yo NOS.

So, since that pulley now has a taperlock, I think I'll just make sure 
its pulled up as well as I can get it, relying on the taperlock to 
freeze it on the shaft. This new shaft, being unworn, fits the hub of 
that pulley with a lot less rattle room, and the twin grub screws can 
probably be adjusted to reduce the runout if I can get a dial to it to 
measure it.

But I need to find something ASAP or the shaft will start rusting sitting 
there in the water...

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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[andy pugh]

On Thu, 1 Aug 2019 at 04:36, Gene Heskett  wrote:

output of arcgenm18.py is assumeing G17. not G18.  So I need to
> translate the ijk into what it would be for G18 motion. Is there a rule
> of thumb for that?
>

G17 XY IJ
G18 XZ IK
G19 YZ JK

So to change a G17 arc to a G18 change Y to Z and J to K.

You won't get a true segment of a cylinder with a cylindrical tool. Which
was why I suggested a rotating disc that looks like an actual Woodruff
cutter.

-- 
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] EDM forms?

[Gene Heskett]

On Tuesday 30 July 2019 23:52:23 TJoseph Powderly wrote:

> you ask for long lasting
> that would be tungsten carbide
> but
> i suggest using easy to manufacture, easy to find, low cost
> because it will wear anyway
>
> so copper, with little lead ( often copper sold will be 'machineable'
> which means lead which means higher wear )
>
> for a key way you should be able to mill machine the electrodes
> and use the largest face ( dont cut with small end of rectangular
> prism , cut with side )
> ( do the hokey pokey - put your big foot outi used to teach the
> operators )
> use LOTS of pecking, just get a sizzle and jump away
>
> you dont have real edm equipment, just a hack so you gotta play safe
>
> and set it up vertical and submerged if possible so the jump opens up
> to a flood of the submerged fluid
> pee down the cut from above and to the side, you want to see the black
> puff out each jump
>
> Negative polarity ( tool neg, work pos ) If you use an r/c pulse
> generator then always negative tool.
> Short ontimes and hi freq ( say 10to50uS on and 200 or more off , 2000
> is good for lash-up )
>
> The exaggerated offtimes aid in letting the dielectric recuperate to
> non conductive state.
> The short cut times ( jump cycle ) aid in poor gap control, you
> actually cut while ' in the zome'
> but your system cant maintain the zone
> So you get efficiency by 'drive by edm', you go thru the zone each
> jump pass and get a little work done.
> You jump away right away and reduce the heat build up ( because you
> likely crashed into the work and shorted )
>
> you're a scope kinda guy, try this...
> reduce the gap using a resistor divider giving max 5v and watch the
> process when you get more 'chairs' then you are cutting more
> when you flat line, you're crashing
> when its open voltage all the time you're missing
> ( old elox guys look for 'grass' like on a lawn ,
>   there was a scope on all old elox generators,
>   the 'grass' was due to slow scope )
>
> hth
> tomp
>
Got it all hooked up, expected lots of noise because the capacitor was 10 
uf.  Note the /was/. From the symptons, taint no more, so while I am 
getting some fire, its dead silent! So I don't think that cap is a cap 
any more.. Fleabay, & see if I can find another. Big cap, around 64 
cubic inches, only 70 years old.

So I sat down to write a while loop, using Big John's arcgenm18.py. and 
haven't managed to get loadable code yet because I think I need the G18 
plane where y can be left out as the g2 nor g3 moves the y. But the 
output of arcgenm18.py is assumeing G17. not G18.  So I need to 
translate the ijk into what it would be for G18 motion. Is there a rule 
of thumb for that?

> On Wed, Jul 31, 2019 at 9:17 AM Gene Heskett  
wrote:
> > Greetings all;
> >
> > Like Jon said on the sheldon list. woodruff keys cut into a case
> > hardened shaft are a bear. Carbide makes a mark, for about 5
> > seconds.
> >
> > So it appears I will have to dig the keyways by EDM.
> >
> > Because the electrode will go away too, what makes the longest
> > lasting electrode material?
> >
> > And which polarity, if it makes a difference?
> >
> > Thanks all;
> >
> > 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)
> > If we desire respect for the law, we must first make the law
> > respectable. - Louis D. Brandeis
> > 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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[Gene Heskett]

On Wednesday 31 July 2019 09:39:22 andy pugh wrote:

> On Wed, 31 Jul 2019 at 15:29, Gene Heskett 
> wrote:
>
> Carbide was mentioned, I might even see how long this now dull tool
> would
>
> > work. Running backwards so the spiral is carrying fresh water down
> > to the fire on the bottom face?  Lots of things to try. :)
>
> Do you really mean "Woodruff" here?

Yes, these are not square keys. The rear face is a curve of about a 7/8" 
diameter washer, so the slots I'll be digging are only perhaps 3/4" 
long. One is for the 4 sheve stepped size pulley on the right end, 
driven by an identical pulley on the end of the motor countershaft 
faceing the other way, making about a 6/1 change in spindle speeds by 
loosening a toggle and moving the belt from groove to groove. And the 
2nd one is to lock the twin belt pulley that drives the spindle although 
I may leave it out now that it has a taperlocked hub.  They don't slip. 
Or at least I've never had one slip in 50 some years with up to 15 hp 
motors driving them.  And, because of working space once the taperlock 
insert is on the shaft, its a cast iron bitch to get in and out. So 
leaving it out and depending on the well tightened hub to hold it is not 
going to keep me up nights worrying about it. So the only key I'll cut 
is the one for the 4 step pulley.

I found a plastic pan but its a hair short, requiring the shaft to lay at 
an angle, so I painted the shaft with a narrow strip of goop, and its 
getting about an hours cure time gluing it to the bottom of the pan. In 
the meantime I've gently clamped the ends of the pan, with some wood 
blocks to discourage shaft movement, the dialed it so center drift is 
under a thou for a 6 inch lengthwise move.

That ought to be close enough axial alignment for the girls I go with.  
Then  next is hit the shop building up the hill and retrieve my EDM psu 
and stick it to whatever is handy and allows short leads.  Then sit down 
and write some code.  With luck. I might be making noise yet today.

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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[andy pugh]

On Wed, 31 Jul 2019 at 15:29, Gene Heskett  wrote:

Carbide was mentioned, I might even see how long this now dull tool would
> work. Running backwards so the spiral is carrying fresh water down to
> the fire on the bottom face?  Lots of things to try. :)
>

Do you really mean "Woodruff" here?

-- 
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] EDM forms?

[Gene Heskett]

On Wednesday 31 July 2019 05:10:08 Andy Pugh wrote:

> > On 31 Jul 2019, at 04:14, Gene Heskett  wrote:
> >
> > So it appears I will have to dig the keyways by EDM.
>
> A rotating disc electrode would seem to be a natural for EDMing a
> Woodruff slot.
>
Sourceing a piece of brass that thickness? Probably should have water 
carrying slots cut in the rim. Would probably have to be made. However, 
before I start, I am going shopping in the housewares department to see 
if I can find a bread pan or some such long enough this 10" shaft can be 
laid in it, and then covered with distilled water. Much easier than 
trying to seal both ends of a medicine bottle.  That should be enough 
water to do it in one pass.

That water I have several gallons of for the moisturizer bubbler my 
ladies oxygen generator runs the gas thru on its way to her. As long as 
its propped not to roll with the tables motions, it should just work and 
I could just step and repeat for both keys.

Thats 1566 steel, I wonder how far in I'd have to EDM, before I could 
switch to a regular carbide bit under power to finish the job precisely. 
IOW, how deep does the 60C case hardening go?

Carbide was mentioned, I might even see how long this now dull tool would 
work. Running backwards so the spiral is carrying fresh water down to 
the fire on the bottom face?  Lots of things to try. :)

But first find a pan. And have the pharmacy get refill auth for the 
metroprolol, the heart pill. Thats what they put me on after the 
pacemaker install.  That and rat poison for blood thinner. They said I 
could stop it, but I'm not bleeding worth a toot /with/ it when I nick a 
hand, so stopping it could easily lead to another pulmonary embolism. 
Thats a hell of a way to die, been there, done that, I am one of the 
less than 2% that survived the first one.

Thanks Andy
> ___
> 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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[Andy Pugh]

> On 31 Jul 2019, at 04:14, Gene Heskett  wrote:
> 
> So it appears I will have to dig the keyways by EDM.

A rotating disc electrode would seem to be a natural for EDMing a Woodruff 
slot. 



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

Re: [Emc-users] EDM forms?

[Gene Heskett]

On Tuesday 30 July 2019 23:52:23 TJoseph Powderly wrote:

> > So it appears I will have to dig the keyways by EDM.

> you ask for long lasting
> that would be tungsten carbide
> but
> i suggest using easy to manufacture, easy to find, low cost
> because it will wear anyway
>
> so copper, with little lead ( often copper sold will be 'machineable'
> which means lead which means higher wear )
>
> for a key way you should be able to mill machine the electrodes
> and use the largest face ( dont cut with small end of rectangular
> prism , cut with side )
> ( do the hokey pokey - put your big foot outi used to teach the
> operators )
> use LOTS of pecking, just get a sizzle and jump away
>
> you dont have real edm equipment, just a hack so you gotta play safe
>
Power supply is homemade, average voltage OC is around 40, 10 uf paper 
cap, hundred ohm 100 watt current limiter. Noisy is an understatement. 
probably does one bang per half cycle of line voltage when its cutting 
good.

> and set it up vertical and submerged if possible so the jump opens up
> to a flood of the submerged fluid pee down the cut from above and to 
the side, you want to see the black
> puff out each jump

With a wobbling, rotating 12 or 14 ga copper wire to stir the water, 
getting brushed on a file during the gcode motion to keep the end sharp, 
and moving lengthwise down one side of the slot. offsetting to the other 
side of the slot for the trip back.  Won't be very fast, but my back 
could use some rest before I tackle the re-assembly anyway. 
>
> Negative polarity ( tool neg, work pos ) If you use an r/c pulse
> generator then always negative tool.
> Short ontimes and hi freq ( say 10to50uS on and 200 or more off , 2000
> is good for lash-up )
>
> The exaggerated offtimes aid in letting the dielectric recuperate to
> non conductive state.
> The short cut times ( jump cycle ) aid in poor gap control, you
> actually cut while ' in the zome'
> but your system cant maintain the zone
> So you get efficiency by 'drive by edm', you go thru the zone each
> jump pass and get a little work done.
> You jump away right away and reduce the heat build up ( because you
> likely crashed into the work and shorted )
>
> you're a scope kinda guy, try this...
> reduce the gap using a resistor divider giving max 5v and watch the
> process when you get more 'chairs' then you are cutting more
> when you flat line, you're crashing
> when its open voltage all the time you're missing
> ( old elox guys look for 'grass' like on a lawn ,
>   there was a scope on all old elox generators,
>   the 'grass' was due to slow scope )

Chuckle, I haven't had a "slow" scope since '77.

That gives me an idea of how to use an offset module to jiggle the Z, but 
the response time of that big nema 34 motor raises an ugly limit as it 
can't follow an individual discharge. But with the electrode wobbling as 
its spinning its pretty self quenching anyway. I've used a 3" thin 
(0.03125") brass disk at 100 to 300 revs to cut a taperlock into leaves, 
usually 6, to give it room to get crushed by the taper, and I can plow 
right along, making a cut 7/8" deep in around 15 minutes with my jury 
rigged supply.  Thats where I found distilled water worked better than 
k2 for a dielectic fluid. K2 gets dirty faster and has to be changed 
more often.

I'm wondering if it would do any better than just doing the keys back 
profile at a decent feed rate. That, combined with the rotation and 
wobble ought to bring in fresh dielectric

> hth
> tomp

Makes me think about the "what if's", as usual, tomp and thats good, 
thanks.



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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[Gene Heskett]

On Tuesday 30 July 2019 22:52:06 Jon Elson wrote:

> On 07/30/2019 09:14 PM, Gene Heskett wrote:
> > Greetings all;
> >
> > Like Jon said on the sheldon list. woodruff keys cut into a case
> > hardened shaft are a bear. Carbide makes a mark, for about 5
> > seconds.
> >
> > So it appears I will have to dig the keyways by EDM.
> >
> > Because the electrode will go away too, what makes the longest
> > lasting electrode material?
> >
> > And which polarity, if it makes a difference?
>
> I think you want workpiece positive, and either graphite or
> copper makes a good electrode.
> Unless you do it with a second pass and new electrode, the
> slot is likely to get a taper as the electrode wears down.
>
> Jon
>
I'll make a container for distilled water, it seems to work better than 
k2, and write some gcode to sweep the bottom of the slot, going down 
about .0001 per pass around the loop. I've thought of using a motor 
brush like for a Dremel tool, but they're so short to start. Longer  
electrodes can be pulled and the ends squared up when the taper gets 
obvious.  Or a copper wire set to wobble, and bent to cover the needed 
width as its being rotated, could always be filed by dragging it on a 
file so as the keep the bottom of the slot square.  The shaft came with 
an expanded web sheathing which is padding it in the vice right now, so 
its probably isolated from the vice. That will allow the file to be 
clamped to the vice. For the reservoir, I'll machine a medicine bottle 
lid to be water tight on the shaft, and mill an electrode slot in the 
side of the pill bottle. Fill it to the slot in distilled water & write 
the gcode to pick the wire out and take it to the file, incrementing the 
depth there by that same .0001" per pass. Air hose the file occasionally 
to keep it clean. I haven't measured a key for height though, but can 
measure the key in the old shaft to get an idea of the depth needed.

Nobody said it had to be fast, and it won't be but by keeping the wire 
sharp, and wobbling about the key width (.189") it ought to work.

I damaged the inner edge of my taperlock hub with the trapped key trying 
to get it out, but I've now got that polished smooth a again, and have 
now machined a slot for key access, matching the slot remains in the 
pulley which should make re-inserting the key a heck of a lot easier.  
There's very very little room, and a relief cut in the small end of the 
taperlock will make it considerably easier to get everthing in time so 
the key just tips into place.  The flange of the taperlock robs me of 
the space to do that, so the key has yo be put in while the taperlock 
hub is backed away and loose but the shaft when the key is inserted is 
about 3/4" left of where it runs. Once its in, the shaft is moved to the 
right, carrying the key in past the taper, then the drawbolts are 
tightened to about 1/16th turn from broke, going around the hub until 
they don't move any more.  With the working room I have I'll have to 
make the wrench into a really short arm, and with the belts dragging, 
tightening will take at least an hour, it took 2 to get them all loose 
to take it apart. I've even considered leaving that key out and 
depending entirely on the taperlocks grip. There is no wear mark on the 
key now so it hasn't been slipping and slapping against the key even 
when the old belts are yelping at the rigid tapping turnaround while 
reversing an 8" 40+lb chuck.

And my back is still complaining about that 2 days it took to get it 
apart.

With the shaft re-installed, raise it as high as it will go,  to get 
enough room to join the powerbelts I'm going to try to put in, then see 
if it has enough tension when its been lowered to the max again. run for 
an hour, jack it back up and see if enough slack to take another link 
out can be found. The guy in the Fenner video, he doesn't have hands, 
he's got to have fingers with built in vicegrips disguised to look like 
fingers. I might wind up putting fresh but normal belts in it, in which 
case it won't be together for another 2 weeks because I'll have to take 
it all back apart again. Supposedly matched of coarse, these sure as 
hell aren't. But I'd guess they are 40 yo too.

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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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

Re: [Emc-users] EDM forms?

[TJoseph Powderly]

you ask for long lasting
that would be tungsten carbide
but
i suggest using easy to manufacture, easy to find, low cost
because it will wear anyway

so copper, with little lead ( often copper sold will be 'machineable' which
means lead which means higher wear )

for a key way you should be able to mill machine the electrodes
and use the largest face ( dont cut with small end of rectangular prism ,
cut with side )
( do the hokey pokey - put your big foot outi used to teach the
operators )
use LOTS of pecking, just get a sizzle and jump away

you dont have real edm equipment, just a hack so you gotta play safe

and set it up vertical and submerged if possible so the jump opens up to a
flood of the submerged fluid
pee down the cut from above and to the side, you want to see the black puff
out each jump

Negative polarity ( tool neg, work pos ) If you use an r/c pulse generator
then always negative tool.
Short ontimes and hi freq ( say 10to50uS on and 200 or more off , 2000 is
good for lash-up )

The exaggerated offtimes aid in letting the dielectric recuperate to non
conductive state.
The short cut times ( jump cycle ) aid in poor gap control, you actually
cut while ' in the zome'
but your system cant maintain the zone
So you get efficiency by 'drive by edm', you go thru the zone each jump
pass and get a little work done.
You jump away right away and reduce the heat build up ( because you likely
crashed into the work and shorted )

you're a scope kinda guy, try this...
reduce the gap using a resistor divider giving max 5v and watch the process
when you get more 'chairs' then you are cutting more
when you flat line, you're crashing
when its open voltage all the time you're missing
( old elox guys look for 'grass' like on a lawn ,
  there was a scope on all old elox generators,
  the 'grass' was due to slow scope )

hth
tomp



On Wed, Jul 31, 2019 at 9:17 AM Gene Heskett  wrote:

> Greetings all;
>
> Like Jon said on the sheldon list. woodruff keys cut into a case hardened
> shaft are a bear. Carbide makes a mark, for about 5 seconds.
>
> So it appears I will have to dig the keyways by EDM.
>
> Because the electrode will go away too, what makes the longest lasting
> electrode material?
>
> And which polarity, if it makes a difference?
>
> Thanks all;
>
> 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)
> If we desire respect for the law, we must first make the law respectable.
>  - Louis D. Brandeis
> 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] EDM forms?

[Jon Elson]

On 07/30/2019 09:14 PM, Gene Heskett wrote:

Greetings all;

Like Jon said on the sheldon list. woodruff keys cut into a case hardened
shaft are a bear. Carbide makes a mark, for about 5 seconds.

So it appears I will have to dig the keyways by EDM.

Because the electrode will go away too, what makes the longest lasting
electrode material?

And which polarity, if it makes a difference?


I think you want workpiece positive, and either graphite or 
copper makes a good electrode.
Unless you do it with a second pass and new electrode, the 
slot is likely to get a taper as the electrode wears down.


Jon


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

[Emc-users] EDM forms?

[Gene Heskett]

Greetings all;

Like Jon said on the sheldon list. woodruff keys cut into a case hardened 
shaft are a bear. Carbide makes a mark, for about 5 seconds.

So it appears I will have to dig the keyways by EDM.

Because the electrode will go away too, what makes the longest lasting 
electrode material?

And which polarity, if it makes a difference?

Thanks all;

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)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


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