开云体育

--- In mach1mach2cnc@..., "jasonjet25" <jasonjet25@e...>
wrote:

Thank all of you for your input and opinions. Does anyone have
pictures of there chain drives and belt drives and rack and

pinions?

I think this will help alot. Also, about the 5 axis, I have seen

the

unit at rainnea and am very impressed. Is this something a

bigginer

should even dream of or should I walk away? Mostly refering to the
software of course.

If you haven't already, go to this site:

There is a plethora of various DIY machines and just as many
different drive systems.

Again thanks for all input

Jason
jasonjet25W@e...


--- In mach1mach2cnc@..., "rrc62_03782" <rrc62@a...>
wrote:

Chain and sprocket is OK for driving things in one direction

where

backlash isn't a problem, like ancient aero props ;) Not if

you

want

any sort of accuracy.

Not true....I'm driving my lead screws with chain and sprocket.

I

have a whisker of

backlash in the leadscrew nut but nothing...zero...between the

chain and sprocket.

I've been running it for a year and it's still as tight as the

day

I built it. In fact, even

when the chain is loose there is very little, if any,

backlash.

When I build my plasma

table I'll also use chain. It's cheap and it works very well.

I agree that belts are much quieter. I used #35 chain because I

couldn't find belts as

long as I needed.

Ross

Hi Robert - I made my own.
I use Amplified photo home switches, which in and of themselves are
optoisolated, my limits and estop go thru 4n25s, my step & dir line go thru
a filter/ amp/ inverter, and opto inverted gate at ultra high speed, and my
spindle & Vac controls are SSR'd..

Nobody offered this type of B.O.B. yet, you should look into it , isolate
those S&D lines. I have not had a single lost step since I did it.
JT

Thank all of you for your input and opinions. Does anyone have
pictures of there chain drives and belt drives and rack and pinions?
I think this will help alot. Also, about the 5 axis, I have seen the
unit at rainnea and am very impressed. Is this something a bigginer
should even dream of or should I walk away? Mostly refering to the
software of course.

Again thanks for all input

Jason
jasonjet25W@...


--- In mach1mach2cnc@..., "rrc62_03782" <rrc62@a...>
wrote:

Chain and sprocket is OK for driving things in one direction where
backlash isn't a problem, like ancient aero props ;) Not if you

want

any sort of accuracy.

Not true....I'm driving my lead screws with chain and sprocket. I

have a whisker of

backlash in the leadscrew nut but nothing...zero...between the

chain and sprocket.

I've been running it for a year and it's still as tight as the day

I built it. In fact, even

when the chain is loose there is very little, if any, backlash.

When I build my plasma

table I'll also use chain. It's cheap and it works very well.

I agree that belts are much quieter. I used #35 chain because I

couldn't find belts as

long as I needed.

Ross

One of the advantages of using the Sound Logic breakout board is that it
includes an optical isolator on the input. The optical isolator in effect
eliminated the very short false triggers. You can use either inductive
proximity sensors or micro switches.

Bob Campbell
Bob@...
www.campbelldesigns.com
Breakout board
THC board sets
CNC router plans
Stepper Motors

I use 10Ks with no problem, the reason is to supply a positive "Pull Up" to
the ttl signal, you only need a little bit of power, so 10K works fine,
Without them, you run the risk of false triggers. When the NC switch is
closed, it provides a path to ground, so the pin looks low, when the NC is
broken, the pin goes high, as ground is broken and the power thru the
resistor provides the +.
Without a resistor, with NC switches, the possible problems (baring a
massive short circuit) is false triggers, or wasted power, and possibly
overworking the par port, which according to spec can only source/sink MAX
<1A of current .
JT

I have no idea. I know that sound stupid but I have seen a few
diagrams that have that. Maybe domeone can answer that question.
Andre

--- In mach1mach2cnc@..., "george_barr" <george_barr@y...>
wrote:

I am using a proximity switch in NC mode. I do not use any resistor
(e.g. 4.7k) and it still works. Do I need a resistor for my limit
switches and what are the advantages/disadvantages with using a
resistor?
Thanks,

--- In mach1mach2cnc@..., "andrebarclay"
<andre.barclay@v...> wrote:

I'm trying to finish up my gantry router and I'm at the limit switch
part. Here's what I was thinking of doing. Take the 5v from the pc
and run it through a 4.7k resistor to a NC limitswitch and then to

the

11,12, or 13 pin on the port. Does anyone know if this will be a
problem and if there is a better setup? I thought it would be

better

to ask than to cry.

Andre

I am using a proximity switch in NC mode. I do not use any resistor
(e.g. 4.7k) and it still works. Do I need a resistor for my limit
switches and what are the advantages/disadvantages with using a
resistor?
Thanks,

--- In mach1mach2cnc@..., "andrebarclay"
<andre.barclay@v...> wrote:

I'm trying to finish up my gantry router and I'm at the limit switch
part. Here's what I was thinking of doing. Take the 5v from the pc
and run it through a 4.7k resistor to a NC limitswitch and then to

the

11,12, or 13 pin on the port. Does anyone know if this will be a
problem and if there is a better setup? I thought it would be

better

to ask than to cry.

Andre

I'm trying to finish up my gantry router and I'm at the limit switch
part. Here's what I was thinking of doing. Take the 5v from the pc
and run it through a 4.7k resistor to a NC limitswitch and then to the
11,12, or 13 pin on the port. Does anyone know if this will be a
problem and if there is a better setup? I thought it would be better
to ask than to cry.

Andre

I have a question about Mach2 limit switches:
When Mach2 senses that one of the limit switches has been triggered,
how do I jog off the end of one of the axis? All of my X axis limit
switches is connected to one contact closure or one parallel port
pin. I tried using the "auto limit override" button which turns
green, but the "override limits" is flashing red. Also, the
message "external EStop requested" is shown. I did not connect an
external EStop button yet. I assume I just press the EStop and
release this EStop so I can jog off the limits, correct? Will this
override allow me to jog off the limit AND prevent me from jogging
more into the limit?

Thanks,

Chain and sprocket is OK for driving things in one direction where
backlash isn't a problem, like ancient aero props ;) Not if you want
any sort of accuracy.

Not true....I'm driving my lead screws with chain and sprocket. I have a whisker of
backlash in the leadscrew nut but nothing...zero...between the chain and sprocket.
I've been running it for a year and it's still as tight as the day I built it. In fact, even
when the chain is loose there is very little, if any, backlash. When I build my plasma
table I'll also use chain. It's cheap and it works very well.

I agree that belts are much quieter. I used #35 chain because I couldn't find belts as
long as I needed.

Ross

Well If I had the money I'd do it with the Canadarm ;-)

But then what do you expect from a crazy Canuck (the correct spelling).

TTFN Vern.

A few newbie questions:

1. I set up the machine to be an x,y,z and c mill. Why on the main
screen (program1 display screen does A display rather than C. C
works ok, however.

2. I am somewhat familiar with the use of the g41 gcode, and am
implimenting it in the way I am used to using it. When it is on,
doesn't the tool path display show the tool's path? At this point, I
do not have a whole machine set up, only the motors, so I have no
easy way of knowing that it is even running on this downloaded demo
4.0. I am wondering if I have left some setting out that might be
disabling it.

That is all for tonight. Hope someone can help me a little more.
There is a lot in that manual to digest.

Thanks,

Tom

On Wed, 10 Mar 2004 15:11:13 -0600, you wrote:

It seems to me that, I've noticed here before, that for large size tables
chain and
sprockets work quite well. Remember the Wright Bros.

Chain and sprocket is OK for driving things in one direction where
backlash isn't a problem, like ancient aero props ;) Not if you want
any sort of accuracy.

Toothed belt drive is much better, and quieter!

--
Steve Blackmore

On Wednesday 10 March 2004 23:04, Jens Swales wrote:

hi all

if my machine coords are set and defined and i then jog the tool to a
start position at the workpiece with a 12 mm slip-gauge between the
tool tip and surface, i will set my prog dro to x=0 y=0 z=12, right?

now, if i move around a little, repos the tool, stop, rewind, jump a
few lines or whatever, is it still possible to find back to that
startpos again wit a single button click?

so far, ive only managed to point back to x=0 y=0 z=0.

the problem is the 12mm .. its just an arbitary value, there will not be a
standard procedure or OEM code for this, because you use 12mm, other people
might use somehting entirely different. .... so here are your options:

If you always use a 12mm slip, then create a button in the screen editor,
assign it some gcode like "G0 X0 Y0 Z12" call it 'return to ref 12mm; .. make
up some other for the other common sizes of slip you use.

better still:

Write a macro that cancels the G92 offsets (but leaves them set in the
interpreter, retracts the Z to the safe height, re-applies the offsets,
repositions X and Y to zero, moves Z to 12mm, this way you can be sure of
going to your reference position safely, and not ploughing into your part.

Assign the macro to a button using the screen editor, call it 'go to
reference' or somehting.

even better still:

write two macros, the first one is used to set X and Y to 0 and then it asks
you how thick the slip gauge is (offering you 12mm or whatvever you last used
by default) .. it then stores the slip thickness in a DRO, and sets Z to that
value. The second macro is used to return to the poasition. it reads the slip
thickness from the DRO and then does the retract, move, lower procedure,
except the distance it lowers to is read from the slip DRO.

assing the macro to a 'set reference' and 'goto reference' buttons
respectively.

ah well, there went your weekend ;)

--
RapidCut CNC Technology

CNC Plasma Cutter

hi all

if my machine coords are set and defined and i then jog the tool to a
start position at the workpiece with a 12 mm slip-gauge between the
tool tip and surface, i will set my prog dro to x=0 y=0 z=12, right?

now, if i move around a little, repos the tool, stop, rewind, jump a
few lines or whatever, is it still possible to find back to that
startpos again wit a single button click?

so far, ive only managed to point back to x=0 y=0 z=0.

is there any oem code for this?




jens

Robin,

Thank you for taking time to respond to my question. I will study
the movement more closely. I don't have the machine hooked up to the
pc yet, but I do have the motors turning on the bench, and I should
be able to tell whether the arcs are being produced by the x and y
motors, running real slowly. I just assumed that the display tracked
the actual tool path.

Tom Eldredge

My understanding is that the display may show a chord, (because the

display

point is only sampled 10 times a second or so) but the actual

machine will

follow a true arc.

One 'trick' you might like to try is to scale the part by 1000 and

scale the

motor tuning by 1000, effectivley your machine is now running in

microns

rather than mm, this can sometimes help with very small parts.

You could also try slowing down some. Typically the trajectory

planner plots

a linear move for the next 1/1000th of a second or so, if your arc

is only

going to take 0.003 seconds to cut at your chosen feed speed,

you'll get 3

short line segments. Slow it down and the situation improves. I'm

not sure

what Mach2 has set for the cycle time of the trajectory planner, so

I cant

really be sure if this is where your trouble lies.

My advice is ignore the display, slow it down , and then examine

the finished

parts on a shadowgraph to see what is actually coming out. You

might find

that the stick/slip in your bearings is too great to allow you to

acutally

cut a 0.08mm radius. Having a motor/leadscrew arangement with 1

micron steps

does not actually mean your router will move in 1 micron steps ...

--
RapidCut CNC Technology

CNC Plasma Cutter

Robin,

Thank you for taking time to respond to my question. I will study
the movement more closely. I don't have the machine hooked up to the
pc yet, but I do have the motors turning on the bench, and I should
be able to tell whether the arcs are being produced by the x and y
motors, running real slowly. I just assumed that the display tracked
the actual tool path.

Tom Eldredge

My understanding is that the display may show a chord, (because the

display

point is only sampled 10 times a second or so) but the actual

machine will

follow a true arc.

One 'trick' you might like to try is to scale the part by 1000 and

scale the

motor tuning by 1000, effectivley your machine is now running in

microns

rather than mm, this can sometimes help with very small parts.

You could also try slowing down some. Typically the trajectory

planner plots

a linear move for the next 1/1000th of a second or so, if your arc

is only

going to take 0.003 seconds to cut at your chosen feed speed,

you'll get 3

short line segments. Slow it down and the situation improves. I'm

not sure

what Mach2 has set for the cycle time of the trajectory planner, so

I cant

really be sure if this is where your trouble lies.

My advice is ignore the display, slow it down , and then examine

the finished

parts on a shadowgraph to see what is actually coming out. You

might find

that the stick/slip in your bearings is too great to allow you to

acutally

cut a 0.08mm radius. Having a motor/leadscrew arangement with 1

micron steps

does not actually mean your router will move in 1 micron steps ...

--
RapidCut CNC Technology

CNC Plasma Cutter

Robin,

Thank you for taking time to respond to my question. I will study
the movement more closely. I don't have the machine hooked up to the
pc yet, but I do have the motors turning on the bench, and I should
be able to tell whether the arcs are being produced by the x and y
motors, running real slowly. I just assumed that the display tracked
the actual tool path.

Tom Eldredge

My understanding is that the display may show a chord, (because the

display

point is only sampled 10 times a second or so) but the actual

machine will

follow a true arc.

One 'trick' you might like to try is to scale the part by 1000 and

scale the

motor tuning by 1000, effectivley your machine is now running in

microns

rather than mm, this can sometimes help with very small parts.

You could also try slowing down some. Typically the trajectory

planner plots

a linear move for the next 1/1000th of a second or so, if your arc

is only

going to take 0.003 seconds to cut at your chosen feed speed,

you'll get 3

short line segments. Slow it down and the situation improves. I'm

not sure

what Mach2 has set for the cycle time of the trajectory planner, so

I cant

really be sure if this is where your trouble lies.

My advice is ignore the display, slow it down , and then examine

the finished

parts on a shadowgraph to see what is actually coming out. You

might find

that the stick/slip in your bearings is too great to allow you to

acutally

cut a 0.08mm radius. Having a motor/leadscrew arangement with 1

micron steps

does not actually mean your router will move in 1 micron steps ...

--
RapidCut CNC Technology

CNC Plasma Cutter

It seems to me that, I've noticed here before, that for large size tables
chain and
sprockets work quite well. Remember the Wright Bros.

Cheers, Vern.

On Wednesday 10 March 2004 16:51, davesmachines wrote:

Gentlemen,

I am trying to set up Mach2 to mill some VERY tiny parts. Some of
the arcs have radii of .08 mm. In the tool path display, the
displayed contour shows the arcs to be broken into very large chords,
resulting in a small arc sometimes only having 3 chords in it. I
need to have these arcs cut with chords placed around every 5 degrees
of the arc.

My understanding is that the display may show a chord, (because the display
point is only sampled 10 times a second or so) but the actual machine will
follow a true arc.

One 'trick' you might like to try is to scale the part by 1000 and scale the
motor tuning by 1000, effectivley your machine is now running in microns
rather than mm, this can sometimes help with very small parts.

You could also try slowing down some. Typically the trajectory planner plots
a linear move for the next 1/1000th of a second or so, if your arc is only
going to take 0.003 seconds to cut at your chosen feed speed, you'll get 3
short line segments. Slow it down and the situation improves. I'm not sure
what Mach2 has set for the cycle time of the trajectory planner, so I cant
really be sure if this is where your trouble lies.

My advice is ignore the display, slow it down , and then examine the finished
parts on a shadowgraph to see what is actually coming out. You might find
that the stick/slip in your bearings is too great to allow you to acutally
cut a 0.08mm radius. Having a motor/leadscrew arangement with 1 micron steps
does not actually mean your router will move in 1 micron steps ...

--
RapidCut CNC Technology

CNC Plasma Cutter