开云体育

Art:
Thanks for info. I think the offsets are not being removed from X axis on M30. It resets to a value equal to original position + (or -) last coded position.
Tom

Hi Tom:

I don't know whats causing the slowdown, check to make sure the

pulse

width is not too high. If set to 0 or 1, the slowdown should not

occur, if

set too high, this may cause that type of problem at higher speeds.

>The incr. jog speed is at the current feedrate set.

I'll check out the X axis, you mean it doesn't reset in which way? to

zero?

or the offsets do not remove?

Thanks,
Art
www.artofcnc.ca

Hi Art,

I'm using spindle Step/Dir with V7.2, just changed from M1 where
spindle was OK, this is mill. Using 25Khz kernel freq, calibrated
FotV at half speed VFD op = 60hz,increase speed to 3/4 output = 83hz,
increase to max output = 87hz. (FtoV has been checked for linearity
using separate function generator).

Looking at the ouput of the D-FF pre-scalar at 1/2 speed frequent,
but random, change in output from 80uSec to 50uSec, at 3/4 speed
pulses change from 55uSec to 30uSec, not possible to measure pulse
train at full o/p - too random.

The output of the VFD dithers .2 to .3 in the Hz window at all ouput
frequencies.

Hope this helps

Bernard

Have you cut any threads with the latest Mach2?

Jerry

Art,
You can also load a file, jog 10", zero all or zero the axis you
jogged, hit the regenerate button and the extrema will be off by 10".
Todd
--- In mach1mach2cnc@..., "morgtod" <todmorg@h...> wrote:

Hi Art,
While you have the first file open move to any location other than
0,0,0 and do a zero all, close the file and load a new one and the
extrema will have wild min #'s instead of 0 ( my x & y min are

always

programed to 0 ) and the max will be the wild min + the correct

file

amount. confusing eh?
Todd

--- In mach1mach2cnc@..., Art <fenerty@a...> wrote:

Todd:

Can you explain this further, I cannot repeat it, when I load

another file,

the extrema changes to reflect the new file. Is this RC2 or 7?

Art
www.artofcnc.ca

Having Spindle correction at all times...Hmmm... Perhaps. The problem is I
use the actual feedback when a G32 is started as the start speed and
correction is done during the G32 to correct stepper movement to that rate.
I could I suppose have some way to set the "tune" speed. I don't want to use
commanded rate as many people when commanding a particular rate may be way
off that due to the conversion of values.
I'll look into it..
Thanks,
Art
www.artofcnc.ca

Hi again,
The z has to be at any # other than 0 ( say 2" before you start the
g28.1z1, just type the g28.1z1 into the mdi window twice and you
should see it happen.
Thanks
Todd
--- In mach1mach2cnc@..., Art <fenerty@a...> wrote:

Hi TorchHeads:

I need help with this one. On my system G28.1 Z-5 for example will

move the

Z to relative Z-5 and then do a homing routine. This is repeatable

time

after time. How can I make it do the move back to 1 or whatever.

Can you

send me a G-Code file which will cause this to happen repeatedly. I

do think

theres a bug here, but it seems line one of those that hides under

the

carpet.

Thanks,
Art
www.artofcnc.ca

Hi Art,
While you have the first file open move to any location other than
0,0,0 and do a zero all, close the file and load a new one and the
extrema will have wild min #'s instead of 0 ( my x & y min are always
programed to 0 ) and the max will be the wild min + the correct file
amount. confusing eh?
Todd

--- In mach1mach2cnc@..., Art <fenerty@a...> wrote:

Todd:

Can you explain this further, I cannot repeat it, when I load

another file,

the extrema changes to reflect the new file. Is this RC2 or 7?

Art
www.artofcnc.ca

On Wed, 06 Aug 2003 09:32:43 -0300, you wrote:

Thanks Steve:

Does the spindle speed work correctly in Mill on yours. (I was using mill
to troubleshoot this one, my mistake..)

Yes - works fine on mill, just been and tried it, it's within + or -
.05V on 20 attempts with lathe hardware. Reset toggle doesn't affect
spindle on either.

Speed correction is only turned on in G32 move. This is independent of
step/dir or pwm.

Ohh - anything stopping this being an option at all times?

The rest are on the current list...

Cheers!

Thanks for the bugs, clear and concise as always..

No problem :)
--
Steve Blackmore

Thanks Steve:

Does the spindle speed work correctly in Mill on yours. (I was using mill
to troubleshoot this one, my mistake..)
Speed correction is only turned on in G32 move. This is independent of
step/dir or pwm.

The rest are on the current list...

Thanks for the bugs, clear and concise as always..
Art
www.artofcnc.ca

On Wed, 06 Aug 2003 08:04:11 -0300, you wrote:

. To save much time I now wait for a second report before jumping
into the code. A single bug can take several hours of tracing to find and
I've spent hundreds of hours tracing non-existant ones.) This is not a
rebuke, I just wanted to make sure you guys know I'm not ignoring you on
bugs, just trying to prioritize the available time to be more efficient.

All Step/Direction spindle bugs V7.2 lathe

The problem I have, is although they are being generated the steps out
are reverting to some arbitrary value. Although Mach2 seems now to be
storing the value in XML, what it puts out isn't what I set. I can get
my 10V max out if I adjust motor settings, but on closing Mach2 and
restarting, although the values in steps per unit & velocity haven't
changed, max output voltage is down to 6V, re tune and it's back to
10V for that session only.

Can you also look at spindle & coolant toggles - sometimes
spindle fails to switch on - no LED and pin is not toggled. It seems
to happen more if you toggle coolant on first.

A long outstanding bug is it's impossible to set flood on extern2 and
mist extern3 - they always swap back.

Press Reset on and off then Spindle won't toggle.

Use spindle feedback for speeds has no effect using step/dir.

Requested speed and actual speed can be way out but no correction to
step rate is apparent.

Also V7.2 still reports itself as V5 in about screen.

--
Steve Blackmore

Hi TorchHeads:

I need help with this one. On my system G28.1 Z-5 for example will move the
Z to relative Z-5 and then do a homing routine. This is repeatable time
after time. How can I make it do the move back to 1 or whatever. Can you
send me a G-Code file which will cause this to happen repeatedly. I do think
theres a bug here, but it seems line one of those that hides under the
carpet.

Thanks,
Art
www.artofcnc.ca

Todd:

Can you explain this further, I cannot repeat it, when I load another file,
the extrema changes to reflect the new file. Is this RC2 or 7?

Art
www.artofcnc.ca

Hello All:

I have quite a list of small bugs and problems for next release. Please
hold off on any feature requests until after next version. I was only away
for a wekk, but the volume of mail is very high. Most are feature requests,
but many had some additonal bugs.

BTW: THC G28 bug will now be investigated further. I was waiting for
confirmation, and now we have 3 sepatate systems showing this bug. ( more
than 1 requires action.)

Thanks for your patience on these issues. I now wait for copnfirmation on
all bugs except the most serious. This is necessary as I recieve at least 5
bugs a day with 4 of them folling up with a "Never Mind..I Screwed up
report. ). To save much time I now wait for a second report before jumping
into the code. A single bug can take several hours of tracing to find and
I've spent hundreds of hours tracing non-existant ones.) This is not a
rebuke, I just wanted to make sure you guys know I'm not ignoring you on
bugs, just trying to prioritize the available time to be more efficient.

I have at leat 10-20 items for next weeks release and am working on them.

Thanks,
Art
www.artofcnc.ca

Hi Frank:

As a test lower the Z acceleration. This is the most common cause of this.
Rapid reversals of the Z is notorious for lost steps due to the huge inertia
of the Z axis. Usually it is the eaxis with the most torque required.
Thanks,
Art
www.artofcnc.ca

Hi Jeff:

Not yet possible, but hopefully will be as I get time. My list is very long
for the next version, but as soon as I can clear it all away, we'll look at
what can be done...

Thanks,
Art
www.artofcnc.ca

Hi Tom:

I don't know whats causing the slowdown, check to make sure the pulse
width is not too high. If set to 0 or 1, the slowdown should not occur, if
set too high, this may cause that type of problem at higher speeds.

The incr. jog speed is at the current feedrate set.

I'll check out the X axis, you mean it doesn't reset in which way? to zero?
or the offsets do not remove?

Thanks,
Art
www.artofcnc.ca

On Tue, 05 Aug 2003 22:11:49 -0000, you wrote:

i was wondering if mach1 / 2 is able to be interfaced to
potentiometers for controlling feed rates and bounceless
switches for controlling jog increments as well as axis
selection. this would resemble a bostomatic type controller
for anyone who knows those machines.

There are a couple of ways of doing this - one way is to drive the pot
via stepper using step/direction.

The other way is to replace the pot - generally the three connections
go +10, set speed, 0V. So supplying a 0-10V voltage to wiper and 0
will work, this can be done using PWM or Step/dir and there are two
circuits in the yahoo files section that work to put 0-10V out. BEWARE
though - the 0V on a lot of motor controllers is not at the same 0V as
the one on a PC and will blow your parallel port unless you isolate
it.

They reference 0V on non-isolated controllers to mains Neutral in most
cases, the effects of this is just the same as sticking 100V into the
parallel port. Most drive manufacturers make an additional isolation
board or sell isolated type controllers for this type of application.

There's a picture on my page at



that illustrates what could have happened!

Mach2 will run any type of switch and has it's own de-bounce routine
built in (setting under logic), 100 seems to work fine.
--
Steve Blackmore

Brian

Tried to send this direct so you could get the attachment but your e-mail bounced (twice)

I think that the attached macro (not with this post as Group will not allow them) coded for test as M90 (Save As it in your Mach2&#92;Macros folder) does the job you suggest. It would be useful to know if this is so or if I misunderstand the question. If it does work can you let me know what you tried that didn't so I can improve the documentation?

To pass an argument to a macro call you provide a P Q or L word in the call and pick up the values in the VB Script by a call like
xxxvalue = param1() where xxxvalue is a variable in your program and param1() is a function which returns the P word, (also param2() returns Q word value and param3() is supposed to returns L word value but I cannot get it to work)

Your example suggests that this is not what you want though. You can use variables in VB Script like in Basic so your example with "tool" as a variable is OK as far as it goes setting it.

There are two issues. (a) The variable "tool" will not be preserved between macros calls (so it will be unset in M6END) and (b) you need to pass a valid text string to Code.

You can save the tool in a G-code interpreter parameter say [10] by SetVar(10, tool) in the M6Start and get it back by tool = GetVar(10) in M6END

You should be able to format the string by something like:

code ("G43 H" & tool)

the variable "tool" is converted to a string and the operator & concatenates it with the fixed text "G43 H"

Hope this helps. I have not tested it all put together so Art may have to put us straight.

Leave out the ( and ) if you prefer - I think it reads easier with them but it's probably because of the languages I grew up with!

Best wishes

John Prentice

Well I thought I was home free and over the hump when I got the three axes
running off one port on the laptop (Compaq 1510 2.2Ghz P4) with Mach2 RC2.




This a router running simple rectangles, either multiple cuts to break
through or using tabbing and cutting the tabs after the first pass. In either
case I'm losing the zsteps in the last cut, so if I leave tabs 1/8" thick and
run I lose Z .125 with every tab. If I single step through the code I lose
nothing.


I'm posting out of BobCad and MasterCam (using VERY simple code, g90, g0,
g1,g03). All movements are in absolute except I and J in incremental. I'm
only feeding at 50 IPM at 2k steps/inch at Z and 120 IPM at 1.02K steps inch
in x and y, well within the capabilities of 25K pps. I can single step the
drives at anything up to 400 IPM x-y with an indicator and it doesn't drop a
step, put it an NC file to run and everything goes to hell unless I single
step that. Has anyone been through this?




Frank Carpenter

--- In mach1mach2cnc@..., "jacoby01" <BobJacoby@B...>
wrote:

Just jumping into this thread because it very topical - can anyone
recommend a msximum speed for my plasma cutting table - I'm in the
midst of assembling the electronics and designing the table.

By the "book" 16 gauge steel should be cut at 330 ipm (verse 3/8"

at

35 ipm) and 1/16" aluminum at 550 ipm - this sounds extremely fast -

I was initially going to design for a max speed of 200 ipm - now
I've moved it up to 300 - 350 ipm.

I'm anxious to hear others' experiences - any suggestions.

Bob Jacoby
Florida

There are a lot of factors that determine cutting speed. For long
straight lines perhaps the numbers from a table are appropriate but
most of my cuts are complex shapes and lots of short, quick cuts. I
doubt my table would hold together long with rapids of 300 to 500 IPM
and hundreds of abrupt stops and starts. While I can only relate to
14GA and thicker material (the thin stuff tends to curl when you do a
lot of close cuts) I find the 200IPM max my table will do quite
exciting enough. My table is a dual purpose router/torch design that
has the size, ridigity and strength to cut oak with a full sized
router so a 100lb gantry moving at 350 IPM is pretty scary

If your torch has an adjustable current setting you might consider
turning it down for thinner material and cutting slower.

I can tell you from experience that building a machine that will
sustain 350 IPM and not over torque something without spending big
$$$ will be a real challenge. If this is your first self-built CNC
project I recommend you scale back on the speed for this design.

There is not much room/time for error on a 48" wide machine when the
axis is moving at 5 in/sec. In the 2 secs it can take for you to
relize there is a problem and then hitting the e-stop button, your
machine has gone 10".