|
John, could talk about/describe not using the compound to cut threads. What advantages have you found? What shortcomings? There is a discussion on the AtomicELS group that is touching on threading with a ELS system no one appears to understand threading without the compound. Of course Charlie might jump in also. Thanks, Ralph -- Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
Hi Ralph, ? I posted directly to the AtomicELS group with a description of how my ELS tracks the flank of the cut thread regardless of the use of the compound or not.? This approach is _not_ possible with a geared lathe that uses the threading indicator to engage the halfnut onto the lead screw since the Z always starts at the same point of the thread. ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? John Dammeyer ? "ELS! Nothing else works as well for your Lathe" Automation Artisans Inc. www dot autoartisans dot com ? ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Ralph Hulslander
Sent: October-17-20 2:56 PM
To: [email protected]
Subject: [digitalhobbyist] No compound? John, could talk about/describe not using the compound to cut threads.
What advantages have you found? What shortcomings?
There is a discussion on the AtomicELS group that is touching on threading with a ELS system
no one appears to understand threading without the compound.
Of course Charlie might jump in also.
Thanks,
Ralph
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
Thanks John, that helps to understand.
Ralph
On Sat, Oct 17, 2020 at 6:31 PM John Dammeyer < johnd@...> wrote: Hi Ralph, ? I posted directly to the AtomicELS group with a description of how my ELS tracks the flank of the cut thread regardless of the use of the compound or not.? This approach is _not_ possible with a geared lathe that uses the threading indicator to engage the halfnut onto the lead screw since the Z always starts at the same point of the thread. ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? John Dammeyer ? "ELS! Nothing else works as well for your Lathe" Automation Artisans Inc. www dot autoartisans dot com ? ? ? ? John, could talk about/describe not using the compound to cut threads.
What advantages have you found? What shortcomings?
There is a discussion on the AtomicELS group that is touching on threading with a ELS system
no one appears to understand threading without the compound.
Of course Charlie might jump in also.
Thanks,
Ralph
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
Hi Ralph, Much of the explanations are probably in the E-Leadscrew archives which unfortunately are pretty large. ? My ELS uses time to determine where the start of the thread is.? If the spindle speed is steady, which it is with a motor synchronized to the AC line and not underpowered, then the time between 1PPR is pretty well identical. ? The stepper motor (or Servo) has a fixed acceleration time to get up to a specific speed.? So it doesn't matter if it's today or a week from now.? It the target speed is 10 RPS and the acceleration is 100 rps/s then the motor will be up to speed in 0.1 second. ? Simple physics will tell you how far the motor has turned in that 0.1 second and after that how far it moves every second at a constant velocity. ?So you have several fixed values:? Distance motor moves during acceleration up to a set speed and the time it takes to get there. ? So if the spindle speed is reported as say 500mS per rev or 2 RPS (120RPM) and the Z axis is started on the index pulse then the spindle will have turned exactly 1/5th ?of a revolution by the time the Z axis is up to speed.? Since this is repeatable as long as the spindle speed is steady you can start Z at the same spot and again it doesn't matter if it's today or next week the tool bit will be at the same point along the Z axis and around the Spindle radially.? ? So what happens if you start the Z motion a bit closer to the headstock like 0.002"? ? Well the Z axis behaves as before.? It's up to speed in 0.1 second and the spindle has turned 1/5th of a turn.? But the tool bit is now over by 0.002" . ? Now for the math.? Say each compound depth of cut movement is 0.005".? That's the hypotenuse.? Our angle is 30 degrees so sin theta = opp/hyp and cos theta = adj/hyp. ? Sin 30 is 0.5.? Cos 30 is 0.866.? Divide each of those into the hypotenuse to get the Z (opposite) and X (adjacent) distance.? ? 0.005/0.866 = 0.00577 for X axis motion. 0.005/0.5 = 0.010 for Z axis motion.? ? So if the Z axis START position was say 2.000 from the headstock then the new START position is 1.990" and the X axis is moved 0.00577 towards the centerline. ?The tool ?bit will have moved 0.005" along what would be the 30 degree compound angle. ? Now for the fun part.? You could also do this by starting at the same Z position. ?After all you really just want the spindle to be at a different position when the tool enters the cut. First though an extreme example. ? Say you wanted to cut a 2 start 10 TPI thread.? Each thread is really a 5 TPI thread with a pitch of 0.2. ?The second thread just starts 0.1" over from the first.? So change the start position by 0.1 and run another 5 TPI thread and you get a nice neat 2 start. ? ? Instead of starting 0.1" over instead wait the spindle to turn far enough so the carriage will have moved 0.1" and then start the Z axis. ?On my ELS I'd measure spindle speed in ticks and then calculate how many ticks I have to wait before I pretend an index pulse happened. ? Or if you have the electronic gearing systems you count encoder pulses and start the lead screw earlier or later.? Now back to following the compound angle using X and Z.? Reference from the Index pulse the counts or time that have to occur to shift where the spindle is when the carriage starts moving. ? John Dammeyer ? "ELS! Nothing else works as well for your Lathe" Automation Artisans Inc. www dot autoartisans dot com ? ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Ralph Hulslander
Sent: October-17-20 4:11 PM
To: [email protected]
Subject: Re: [digitalhobbyist] No compound? Thanks John, that helps to understand. ? On Sat, Oct 17, 2020 at 6:31 PM John Dammeyer <johnd@...> wrote: Hi Ralph, ? I posted directly to the AtomicELS group with a description of how my ELS tracks the flank of the cut thread regardless of the use of the compound or not.? This approach is _not_ possible with a geared lathe that uses the threading indicator to engage the halfnut onto the lead screw since the Z always starts at the same point of the thread. ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? John Dammeyer ? "ELS! Nothing else works as well for your Lathe" Automation Artisans Inc. www dot autoartisans dot com ? ? ? ? John, could talk about/describe not using the compound to cut threads.
What advantages have you found? What shortcomings?
There is a discussion on the AtomicELS group that is touching on threading with a ELS system
no one appears to understand threading without the compound.
Of course Charlie might jump in also.
Thanks,
Ralph
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
Ralph and group.
I removed my compound slide about 2 yeras ago off my 7 x mini lathe.
Put on a riser block to accomodate the difference in height minus the compound slide required for the toolpost.
see photos.
photo 1 Riser block doubles up as my radius ball turner.
photo 2 New holes drilled and tapped for positioning the QC tool post, photo also shows the CSS Constant surcfae speed attachment.
This system controls increases or decreses the spndle speed when facing across large diamter bars, as a standard feature on CNC lathes.
photo 3 General arrangment.
photo 4 toolpost moved out to machine up to 7" diameter bar.and a support block added at the front of the cross slide.
With the ELS any manual and or auto screwcutting is a starightforward move in?
on the cross slide, and move out when returning to the start position.
With the RELS Richard UK added in a select option in the Eeprom a choice between?
a typical ELS 90 degree feed in (plunge) or using the new RELS a 30 deg (29.5) feed in.
photo 5 shows a modified original compond slide that allows angular adjustment without winding the top slide almost off the nut?
to get at the swivel clamping screws. It's always been Chinese?7 x lathe curse.
Anybody interested in this modified method please respond.
Both ELS RELS systems allows for taper turning and for example a 45 degree chamfer is already in the control box selection.
So again no need for a compound slide.
This original but modified compound slide is ready for the scrap heap, never used it for 2 years so why keep it.?
Hope all of interest
Stay safe and vigilant.
--
John
|
Actually John I think that Ralph was directing the question to John
Lindo!
Further to the comment you made that I repeat below, sorry John you
are incorrect. I am using the Russian ELS on on Arduino Mega and I
have "modified" the software. I have constant area cutting and the
operator can pick plunge or flank cutting. After reaching the
calculated depth for the Metric or Imperial thread further passes
are possible at increments chosen by the operator. I also have the
capability of clearing the tool away from the job? in X for testing
then subsequently moving it back to the original position if further
cuts are needed. As the system is encoder controlled the speed can
vary whilst cutting.
Richard
On 17/10/2020 23:31, John Dammeyer
wrote:
toggle quoted message
Show quoted text
?
Unless the math is
really clever on the Arduino based electronic gearing
systems they can't do it either.
?
?
_._,_._,_
|
John, John,? Jack or Jill I am so glad to get the responses.
I am one step closer to actually using some form of motorized Z with the Z motor mount.
Hopefully X will follow.
Ralph
On Sun, Oct 18, 2020 at 4:38 AM Richard < edelec@...> wrote:
Actually John I think that Ralph was directing the question to John
Lindo!
Further to the comment you made that I repeat below, sorry John you
are incorrect. I am using the Russian ELS on on Arduino Mega and I
have "modified" the software. I have constant area cutting and the
operator can pick plunge or flank cutting. After reaching the
calculated depth for the Metric or Imperial thread further passes
are possible at increments chosen by the operator. I also have the
capability of clearing the tool away from the job? in X for testing
then subsequently moving it back to the original position if further
cuts are needed. As the system is encoder controlled the speed can
vary whilst cutting.
Richard
On 17/10/2020 23:31, John Dammeyer
wrote:
?
Unless the math is
really clever on the Arduino based electronic gearing
systems they can't do it either.
?
?
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Richard
Sent: October-18-20 1:38 AM
To: [email protected]
Subject: Re: [digitalhobbyist] No compound? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
|
That was not me, believe me I am not even capable of thinking of what I might want to do.
That was Richard!
Ralph
On Sun, Oct 18, 2020 at 12:31 PM John Dammeyer < johnd@...> wrote: Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ? ? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
Sorry, I meant Richard. John ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of John Dammeyer
Sent: October-18-20 9:31 AM
To: [email protected]
Subject: Re: [digitalhobbyist] No compound? Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ? ? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
|
Wow John you have got me worried now! I did not think that it was
"clever math" so I had better go back and check what I did.
Richard
On 18/10/2020 17:30, John Dammeyer
wrote:
toggle quoted message
Show quoted text
Hi
Ralph,
So
you have written the 'clever math' to do this.? Super.? Looking forward to a
detailed explanation on how you've done that.
John
?
?
?
Actually
John I think that Ralph was directing the question to John
Lindo!
Further to the comment you made that I repeat below, sorry
John you are incorrect. I am using the Russian ELS on on
Arduino Mega and I have "modified" the software. I have
constant area cutting and the operator can pick plunge or
flank cutting. After reaching the calculated depth for the
Metric or Imperial thread further passes are possible at
increments chosen by the operator. I also have the
capability of clearing the tool away from the job? in X for
testing then subsequently moving it back to the original
position if further cuts are needed. As the system is
encoder controlled the speed can vary whilst cutting.
Richard
On 17/10/2020 23:31, John Dammeyer wrote:
?
Unless the math is
really clever on the Arduino based electronic gearing
systems they can't do it either.
?
?
?
|
Well if you are following the hypotenuse of the 28.5 degree triangle then for each pass X is a bit deeper and Z is closer to the headstock if the thread is cut towards the headstock.? Simple trig. ? But if you keep the Z starting position the same then to start the cut a bit further along the thread the spindle has to rotate a different amount while the Z axis motor is accelerating so the tool enters the work over by that Z amount. ? So I'm curious how your system tracks the flank of the thread. ? John ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Richard
Sent: October-19-20 12:52 AM
To: [email protected]
Subject: Re: [digitalhobbyist] No compound ? Wow John you have got me worried now! I did not think that it was "clever math" so I had better go back and check what I did.
Richard On 18/10/2020 17:30, John Dammeyer wrote: Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ? ? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
?
|
My X infeed per pass is based on a constant area calculation where
obviously the depth of the very first pass dictates the area of the
first cut and therefore subsequent cuts.
When commencing an Auto Thread sequence the operator either accepts
the default first pass depth for that pitch of thread or adjusts it
to a value appropriate to the material and tool. When the cycle runs
each subsequent pass depth increment reduces (by calculation) to
ensure the constant area. Limits have previously been set for Z. A
cutting pass will end at the Z end limit and after X is retracted
the carriage will then return to the Z start limit.
If the system is set to cut on the flank, having calculated the
required total depth for a pass a Z increment is calculated by
multiplying that pass depth by 0.566. At the start of a pass Z is
moved by this amount and X is moved into depth. Once these moves
finish the Z axis then starts its threading move on the zero count
of the encoder.
I did not think it was "clever" but maybe I am missing something?
Richard
On 19/10/2020 09:10, John Dammeyer
wrote:
toggle quoted message
Show quoted text
Well
if you are following the hypotenuse of the 28.5 degree
triangle then for each pass X is a bit deeper and Z is
closer to the headstock if the thread is cut towards the
headstock.? Simple
trig.
?
But
if you keep the Z starting position the same then to start
the cut a bit further along the thread the spindle has to
rotate a different amount while the Z axis motor is
accelerating so the tool enters the work over by that Z
amount.
?
So
I'm curious how your system tracks the flank of the thread.
?
John
?
?
Wow
John you have got me worried now! I did not think that it
was "clever math" so I had better go back and check what I
did.
Richard
On 18/10/2020 17:30, John Dammeyer wrote:
Hi Ralph,
So you have written
the 'clever math' to do this.? Super.? Looking forward to a
detailed explanation on how you've done that.
John
?
?
?
Actually John I think that Ralph was
directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry
John you are incorrect. I am using the Russian ELS on on
Arduino Mega and I have "modified" the software. I have
constant area cutting and the operator can pick plunge or
flank cutting. After reaching the calculated depth for the
Metric or Imperial thread further passes are possible at
increments chosen by the operator. I also have the
capability of clearing the tool away from the job? in X
for testing then subsequently moving it back to the
original position if further cuts are needed. As the
system is encoder controlled the speed can vary whilst
cutting.
Richard
On 17/10/2020 23:31, John Dammeyer wrote:
?
Unless the math
is really clever on the Arduino based electronic gearing
systems they can't do it either.
?
?
?
?
|
Hello, All, Like John Lindo, I have also used Richard's system to cut both plunge and 29.5° "compound simulation" threads, and can confirm that both work well. Unlike John, I have only had my compound off for about a year, and haven't needed it once since. -- Regards, Charlie New Jersey, USA
?
ANTHONY’S LAW OF THE WORKSHOP: Any tool, when dropped, will roll into the least accessible corner of the workshop.
THE LINDO COROLLARY: If you have cats, they will promptly hide the tool.
|
Ah.? So you do exactly the same thing mine does for following the flank.? The only difference is mine doesn't do the extra move in the Z threading direction before it waits for the index and does the pass.? ? Mine returns to that new calculated Z position and then waits for the index to start the next one.? Given how fast the spindle turns it's not a big deal to wait one extra revolution. ? I was under the impression you still closed the half nut to start each threading pass.? Maybe that's a different Arduino based system. ? And mine lets the operator decide how deep to make the first pas and then how deep for each of the subsequent main threading passes plus the number of spring passes.? Or, given the pitch can calculate how deep to make each pass and therefore the number of passes.? First pass and spring passes are still up to the operator. John ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Richard
Sent: October-19-20 4:55 AM
To: [email protected]
Subject: Re: [digitalhobbyist] No compound? My X infeed per pass is based on a constant area calculation where obviously the depth of the very first pass dictates the area of the first cut and therefore subsequent cuts.
When commencing an Auto Thread sequence the operator either accepts the default first pass depth for that pitch of thread or adjusts it to a value appropriate to the material and tool. When the cycle runs each subsequent pass depth increment reduces (by calculation) to ensure the constant area. Limits have previously been set for Z. A cutting pass will end at the Z end limit and after X is retracted the carriage will then return to the Z start limit.
If the system is set to cut on the flank, having calculated the required total depth for a pass a Z increment is calculated by multiplying that pass depth by 0.566. At the start of a pass Z is moved by this amount and X is moved into depth. Once these moves finish the Z axis then starts its threading move on the zero count of the encoder.
I did not think it was "clever" but maybe I am missing something?
Richard
On 19/10/2020 09:10, John Dammeyer wrote: Well if you are following the hypotenuse of the 28.5 degree triangle then for each pass X is a bit deeper and Z is closer to the headstock if the thread is cut towards the headstock.? Simple trig. ? But if you keep the Z starting position the same then to start the cut a bit further along the thread the spindle has to rotate a different amount while the Z axis motor is accelerating so the tool enters the work over by that Z amount. ? So I'm curious how your system tracks the flank of the thread. ? John ? ? Wow John you have got me worried now! I did not think that it was "clever math" so I had better go back and check what I did.
Richard On 18/10/2020 17:30, John Dammeyer wrote: Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ? ? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
?
?
|
Richard is that Oleg's method or did you eliminate his code and add yours?
Ralph
On Mon, Oct 19, 2020 at 12:08 PM John Dammeyer < johnd@...> wrote: Ah.? So you do exactly the same thing mine does for following the flank.? The only difference is mine doesn't do the extra move in the Z threading direction before it waits for the index and does the pass.? ? Mine returns to that new calculated Z position and then waits for the index to start the next one.? Given how fast the spindle turns it's not a big deal to wait one extra revolution. ? I was under the impression you still closed the half nut to start each threading pass.? Maybe that's a different Arduino based system. ? And mine lets the operator decide how deep to make the first pas and then how deep for each of the subsequent main threading passes plus the number of spring passes.? Or, given the pitch can calculate how deep to make each pass and therefore the number of passes.? First pass and spring passes are still up to the operator. John ? ? ? My X infeed per pass is based on a constant area calculation where obviously the depth of the very first pass dictates the area of the first cut and therefore subsequent cuts.
When commencing an Auto Thread sequence the operator either accepts the default first pass depth for that pitch of thread or adjusts it to a value appropriate to the material and tool. When the cycle runs each subsequent pass depth increment reduces (by calculation) to ensure the constant area. Limits have previously been set for Z. A cutting pass will end at the Z end limit and after X is retracted the carriage will then return to the Z start limit.
If the system is set to cut on the flank, having calculated the required total depth for a pass a Z increment is calculated by multiplying that pass depth by 0.566. At the start of a pass Z is moved by this amount and X is moved into depth. Once these moves finish the Z axis then starts its threading move on the zero count of the encoder.
I did not think it was "clever" but maybe I am missing something?
Richard
On 19/10/2020 09:10, John Dammeyer wrote: Well if you are following the hypotenuse of the 28.5 degree triangle then for each pass X is a bit deeper and Z is closer to the headstock if the thread is cut towards the headstock.? Simple trig. ? But if you keep the Z starting position the same then to start the cut a bit further along the thread the spindle has to rotate a different amount while the Z axis motor is accelerating so the tool enters the work over by that Z amount. ? So I'm curious how your system tracks the flank of the thread. ? John ? ? Wow John you have got me worried now! I did not think that it was "clever math" so I had better go back and check what I did.
Richard On 18/10/2020 17:30, John Dammeyer wrote: Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ? ? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
?
?
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta drill press, 40 watt laser, Consew brushless DC motors and a non working 3D printer
|
The half nut must stay closed through an automatic sequence. I felt
a constant cut area seemed more sensible and machine friendly. If
further passes are needed to achieve the required size or fit the
operator can set these. He (or she) also has the option of a rapid
move in X to clear the thread area for test. A further rapid move
takes the tool back to the Null position.
Richard
On 19/10/2020 17:08, John Dammeyer
wrote:
toggle quoted message
Show quoted text
Ah.? So you do exactly the
same thing mine does for following the flank.? The only difference is
mine doesn't do the extra move in the Z threading direction
before it waits for the index and does the pass.?
?
Mine
returns to that new calculated Z position and then waits for
the index to start the next one.? Given how fast the
spindle turns it's not a big deal to wait one extra
revolution.
?
I
was under the impression you still closed the half nut to
start each threading pass.? Maybe
that's a different Arduino based system.
?
And
mine lets the operator decide how deep to make the first pas
and then how deep for each of the subsequent main threading
passes plus the number of spring passes.? Or, given the pitch can
calculate how deep to make each pass and therefore the
number of passes.? First
pass and spring passes are still up to the operator.
John
?
?
?
My
X infeed per pass is based on a constant area calculation
where obviously the depth of the very first pass dictates
the area of the first cut and therefore subsequent cuts.
When commencing an Auto Thread sequence the operator either
accepts the default first pass depth for that pitch of
thread or adjusts it to a value appropriate to the material
and tool. When the cycle runs each subsequent pass depth
increment reduces (by calculation) to ensure the constant
area. Limits have previously been set for Z. A cutting pass
will end at the Z end limit and after X is retracted the
carriage will then return to the Z start limit.
If the system is set to cut on the flank, having calculated
the required total depth for a pass a Z increment is
calculated by multiplying that pass depth by 0.566. At the
start of a pass Z is moved by this amount and X is moved
into depth. Once these moves finish the Z axis then starts
its threading move on the zero count of the encoder.
I did not think it was "clever" but maybe I am missing
something?
Richard
On 19/10/2020 09:10, John Dammeyer wrote:
Well if you are
following the hypotenuse of the 28.5 degree triangle then
for each pass X is a bit deeper and Z is closer to the
headstock if the thread is cut towards the headstock.? Simple trig.
?
But if you keep the
Z starting position the same then to start the cut a bit
further along the thread the spindle has to rotate a
different amount while the Z axis motor is accelerating so
the tool enters the work over by that Z amount.
?
So I'm curious how
your system tracks the flank of the thread.
?
John
?
?
Wow John you have got me worried now! I did
not think that it was "clever math" so I had better go
back and check what I did.
Richard
On 18/10/2020 17:30, John Dammeyer wrote:
Hi Ralph,
So you have
written the 'clever math' to do this.? Super.? Looking forward to
a detailed explanation on how you've done that.
John
?
?
?
Actually John I think that Ralph was
directing the question to John Lindo!
Further to the comment you made that I repeat below,
sorry John you are incorrect. I am using the Russian ELS
on on Arduino Mega and I have "modified" the software. I
have constant area cutting and the operator can pick
plunge or flank cutting. After reaching the calculated
depth for the Metric or Imperial thread further passes
are possible at increments chosen by the operator. I
also have the capability of clearing the tool away from
the job? in X for testing then subsequently moving it
back to the original position if further cuts are
needed. As the system is encoder controlled the speed
can vary whilst cutting.
Richard
On 17/10/2020 23:31, John Dammeyer wrote:
?
Unless the math
is really clever on the Arduino based electronic
gearing systems they can't do it either.
?
?
?
?
?
|
My system is based on Oleg's with a number of changes. I certainly
would not have a system without that excellent base code. The
original software did not offer flank cutting or additional passes.
Richard
On 19/10/2020 17:30, Ralph Hulslander
wrote:
toggle quoted message
Show quoted text
Richard is that Oleg's method or did you eliminate
his code and add yours?
Ralph
On Mon, Oct 19, 2020 at 12:08
PM John Dammeyer < johnd@...> wrote:
Ah.? So
you do exactly the same thing mine does for following
the flank.? The only difference is mine
doesn't do the extra move in the Z threading direction
before it waits for the index and does the pass.?
?
Mine returns to
that new calculated Z position and then waits for the
index to start the next one.? Given how
fast the spindle turns it's not a big deal to wait one
extra revolution.
?
I was under the
impression you still closed the half nut to start each
threading pass.? Maybe that's a
different Arduino based system.
?
And mine lets the
operator decide how deep to make the first pas and
then how deep for each of the subsequent main
threading passes plus the number of spring passes.?
Or, given the pitch can calculate how deep to
make each pass and therefore the number of passes.?
First pass and spring passes are still up to
the operator.
John
?
?
?
My X
infeed per pass is based on a constant area
calculation where obviously the depth of the very
first pass dictates the area of the first cut and
therefore subsequent cuts.
When commencing an Auto Thread sequence the operator
either accepts the default first pass depth for that
pitch of thread or adjusts it to a value appropriate
to the material and tool. When the cycle runs each
subsequent pass depth increment reduces (by
calculation) to ensure the constant area. Limits have
previously been set for Z. A cutting pass will end at
the Z end limit and after X is retracted the carriage
will then return to the Z start limit.
If the system is set to cut on the flank, having
calculated the required total depth for a pass a Z
increment is calculated by multiplying that pass depth
by 0.566. At the start of a pass Z is moved by this
amount and X is moved into depth. Once these moves
finish the Z axis then starts its threading move on
the zero count of the encoder.
I did not think it was "clever" but maybe I am missing
something?
Richard
On 19/10/2020 09:10, John
Dammeyer wrote:
Well if you
are following the hypotenuse of the 28.5 degree
triangle then for each pass X is a bit deeper and Z
is closer to the headstock if the thread is cut
towards the headstock.? Simple trig.
?
But if you
keep the Z starting position the same then to start
the cut a bit further along the thread the spindle
has to rotate a different amount while the Z axis
motor is accelerating so the tool enters the work
over by that Z amount.
?
So I'm
curious how your system tracks the flank of the
thread.
?
John
?
?
Wow
John you have got me worried now! I did not think
that it was "clever math" so I had better go back
and check what I did.
Richard
On 18/10/2020 17:30, John
Dammeyer wrote:
Hi
Ralph,
So
you have written the 'clever math' to do this.?
Super.? Looking forward to a
detailed explanation on how you've done that.
John
?
?
?
Actually
John I think that Ralph was directing the question
to John Lindo!
Further to the comment you made that I repeat
below, sorry John you are incorrect. I am using
the Russian ELS on on Arduino Mega and I have
"modified" the software. I have constant area
cutting and the operator can pick plunge or flank
cutting. After reaching the calculated depth for
the Metric or Imperial thread further passes are
possible at increments chosen by the operator. I
also have the capability of clearing the tool away
from the job? in X for testing then subsequently
moving it back to the original position if further
cuts are needed. As the system is encoder
controlled the speed can vary whilst cutting.
Richard
On 17/10/2020 23:31, John
Dammeyer wrote:
?
Unless
the math is really clever on the Arduino based
electronic gearing systems they can't do it
either.
?
?
?
?
?
--
Clausing 8520, Craftsman 12x36 Lathe, 4x12 mini lathe, 14" Delta
drill press, 40 watt laser, Consew brushless DC motors and a non
working 3D printer
|
Mine also has the equivalent of a feed hold.? Press the stop button and it finishes what it was doing.? So if it's threading it stops at the end of the thread pass.? If it's retracting it stops at the end of the retraction etc. ? Once stopped you can move the tool anywhere you want to inspect, remove from between centers and replace if you are using a drive dog or camlock chuck.? Then press start and it moves back to the correct location and continues with the next threading pass. ?Really no different than if you were using the lathe with gears manually and watching the threading indicator for the next pass and using the compound to follow the flank at 29.5 degrees. ? BTW, if you are interested, the source code for my ELS has been posted first on Yahoo and now groups.io at /g/E-Leadscrew/files/ELS%201.00 as far back as 2008.? There have only been minor changes in the code over the 12 year period.? There is new stuff in the works.? The source code for that will also be posted. ? My web site has the schematics too.? Plus the first concept was an article published in Circuit Cellar Magazine back in 2006. ? John ? ? "ELS! Nothing else works as well for your Lathe" Automation Artisans Inc. www dot autoartisans dot com ? ? ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Richard
Sent: October-19-20 10:30 AM
To: [email protected]
Subject: Re: [digitalhobbyist] No compound? The half nut must stay closed through an automatic sequence. I felt a constant cut area seemed more sensible and machine friendly. If further passes are needed to achieve the required size or fit the operator can set these. He (or she) also has the option of a rapid move in X to clear the thread area for test. A further rapid move takes the tool back to the Null position.
Richard On 19/10/2020 17:08, John Dammeyer wrote: Ah.? So you do exactly the same thing mine does for following the flank.? The only difference is mine doesn't do the extra move in the Z threading direction before it waits for the index and does the pass. ? ? Mine returns to that new calculated Z position and then waits for the index to start the next one.? Given how fast the spindle turns it's not a big deal to wait one extra revolution. ? I was under the impression you still closed the half nut to start each threading pass.? Maybe that's a different Arduino based system. ? And mine lets the operator decide how deep to make the first pas and then how deep for each of the subsequent main threading passes plus the number of spring passes.? Or, given the pitch can calculate how deep to make each pass and therefore the number of passes.? First pass and spring passes are still up to the operator. John ? ? ? My X infeed per pass is based on a constant area calculation where obviously the depth of the very first pass dictates the area of the first cut and therefore subsequent cuts.
When commencing an Auto Thread sequence the operator either accepts the default first pass depth for that pitch of thread or adjusts it to a value appropriate to the material and tool. When the cycle runs each subsequent pass depth increment reduces (by calculation) to ensure the constant area. Limits have previously been set for Z. A cutting pass will end at the Z end limit and after X is retracted the carriage will then return to the Z start limit.
If the system is set to cut on the flank, having calculated the required total depth for a pass a Z increment is calculated by multiplying that pass depth by 0.566. At the start of a pass Z is moved by this amount and X is moved into depth. Once these moves finish the Z axis then starts its threading move on the zero count of the encoder.
I did not think it was "clever" but maybe I am missing something?
Richard
On 19/10/2020 09:10, John Dammeyer wrote: Well if you are following the hypotenuse of the 28.5 degree triangle then for each pass X is a bit deeper and Z is closer to the headstock if the thread is cut towards the headstock.? Simple trig. ? But if you keep the Z starting position the same then to start the cut a bit further along the thread the spindle has to rotate a different amount while the Z axis motor is accelerating so the tool enters the work over by that Z amount. ? So I'm curious how your system tracks the flank of the thread. ? John ? ? Wow John you have got me worried now! I did not think that it was "clever math" so I had better go back and check what I did.
Richard On 18/10/2020 17:30, John Dammeyer wrote: Hi Ralph, So you have written the 'clever math' to do this.? Super.? Looking forward to a detailed explanation on how you've done that. John ? ? ? Actually John I think that Ralph was directing the question to John Lindo!
Further to the comment you made that I repeat below, sorry John you are incorrect. I am using the Russian ELS on on Arduino Mega and I have "modified" the software. I have constant area cutting and the operator can pick plunge or flank cutting. After reaching the calculated depth for the Metric or Imperial thread further passes are possible at increments chosen by the operator. I also have the capability of clearing the tool away from the job? in X for testing then subsequently moving it back to the original position if further cuts are needed. As the system is encoder controlled the speed can vary whilst cutting.
Richard On 17/10/2020 23:31, John Dammeyer wrote: ? Unless the math is really clever on the Arduino based electronic gearing systems they can't do it either. ? ? ?
?
?
?
|
John,
I wonder why neither you nor Oleg, both creators of widely used and open-source ELS systems, don't use GitHub (or similar) service to publish source code.
I feel that publishing code on groups.io, internet forums or personal websites is more or less one-way affair (creator shares his content for other to copy / replicate).
Meanwhile a platform such as GitHub is very convenient for collaboration, it's very easy for anyone to fork the project, make some improvements, and then submit a pull request to the original author. They also have built-in issue tracker allowing users (or maintainers) to report bugs / ask for features. Consider this for your next version, please :)
Cheers,
--Gene
|
John Lindo
