¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
Date

Re: EMC API?

 

From ginger@... Tue Jun 15 22:45:25 1999
Date: Tue, 15 Jun 1999 22:48:37 -0400
From: Ron Ginger <ginger@...>
To: CAD_CAM_EDM_DRO@...
Mailing-List: list CAD_CAM_EDM_DRO@...; contact
CAD_CAM_EDM_DRO-owner@...
Delivered-To: mailing list CAD_CAM_EDM_DRO@...
List-Unsubscribe: <mailto:CAD_CAM_EDM_DRO-unsubscribe@...>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Subject: [CAD_CAM_EDM_DRO] EMC API?

From: Ron Ginger <ginger@...>

I am very interested in EMC. Id like to tinker with the GUI part of the
system- Ive done a Mill Power kind of clone alerady under Win95, Id like
to move that to EMC, maybe with Tk or Java. I assume there is a
programmer interface- API- soewhere in EMC that can be used this way,
but so far I havent found any docs about it.

can soeone give me a pointer to the right docs?

What seems to be really missing to me is a sort of overview doc that
describes all the pieces and how they work together- the WEB page junps
right to a description of an .ini file and lots of gory details, but
never explains the basic structure of things.

ron

The interface to EMC typically used by a gui is defined by the C++ header file
src/emcnml/emc.hh. and by the documentation for NML in
(C++ version) and
(Java Version).

NML provides the API for sending/recieving any message and the
emc.hh file defines the particular messages that EMC uses.

There is a CodeGenerator that automatically converts emc.hh from C++ to
Java which is typically run by the Makefile in src/java/emc/nml.

For example to send a jog message in Java:

import rcs.*;
import rcs.nml.*;
import emc.nml.*;

. . .

// Setup the NML channel
nml_emc nml_emc_object = new nml_emc();
NMLConnection NCCommandChannel = new NMLConnection(nml_emc_object,
"emcCommand","emcJavaGui", "emc.nml");

// Create and send a message to start jogging the X axis.
EMC_AXIS_JOG jog_msg = new EMC_AXIS_JOG();
jog_msg.axis = 0; // X
jog_msg.vel = 1.0; // units per second (units specified with
EMC_STAT.motion.axis[axis].units )
NCCommandChannel.write(jog_msg);

The classes EMC_AXIS_JOG and nml_emc are automatically generated from emc.hh,
and are are imported with "import emc.nml.*" the class NMLConnection is part
of the RCS libary and imported with "import rcs.nml.*"

As far as I know there is no Tk interface and I would have no idea how to
create one, since I am not a Tk programmer.

The classes in emc.hh at least have good names that are mostly
self-explanatory, but sometimes it can be unclear what the units a variable
uses are or which mode you need to be in to use a particular command. We need
better documentation and we have made several attempts in the past to do this
but I am not sure if any of those documents are current.



-- Will


Re: ball screws Backlash

 

With all this talk about Thomson, I thought I'd put in my $0.02. From my
limited experience, they make fine stuff - the Thomson tables, bearings
and supported shafts I have seem to be top quality. However, when I
e-mailed them about the new ballscrews, the sales-type insisted upon a
phone number before they would send specs and pricing. I hate
telemarketing, and told them that was why I didn't give out my number,
whereupon they got a bit snooty. After this experience, I'll be dealing
with _anyone but_ Thomson for my CNC needs.

YMMV

-Geoff


Re: EMC

Dan Falck
 

Unfortunately, I have been really busy the last couple of weeks. This last
weekend, I loaded the June 8 th version and got it running, but haven't cut
parts on it yet. I've been making a lot of parts on my othe mill for stuff
at work that's been keeping me busy.

Dan

At 09:18 PM 6/15/99 -0600, you wrote:
From: "Tim Goldstein" <timg@...>

Dan,

Haven't heard any more about your EMC adventures. Have you loaded up the
most recent version? Are you using it to make any parts yet? How is it
working for you?

Tim


(No subject)

 

Also a very special Thanx to Jon Elson!!!!!!
Tracey


Re: CAD program

 

Link to cadvance Free CAD 6.5/CAM program! Has
anyone used it? Figured it's free so I'm going to check it out.
Tracey


Re: starting from scratch

 

OK, let the adventures begin!
Apparently there are a lot of people just reading "lurkers" on this list with
as little background and knowledge of EMC, Linux, and CNC as myself. Maybe
all the help I have received, and hope to continue receiving, will get more
people involved!
I'm am going to be going through a complete conversion and set up of a
cnc mill and Linux on a new computer. So to start with the basics My first
purchase is going to be the Linux OS. The version they have at Barnes and
Noble is 5.2 is this the version I'm looking for?
Have seen talk on the list about 6.0 and haven't seen it anywhere, not to
mention it sounds like there are some unresolved problems for the EMC
application.
Thanx to Don, Ted, Matt, and everyone else that has been so patient and
helpful thus far.
Tracey DeChambeau
Tadguninc@...


Re: TCL/TK GUI and EMC API

Matt Shaver
 

From: David L Anderson <daveland@...>

Fred et al

I have a simple TCL/TK CNC control panel GUI that I was playing with for my
own use. I sent it out serveral months ago to Fred Proctor and some others
on the rt-linux NG. It is just a simple hack that at this point has no
interface to EMC. IF I could get some assistance in the details of the EMC
interface, I could try to get it to function. I am currently trying to get
EMC running after abandoning my own attemps to write stepper motor control
code (I learned a lot, Fred must work hard to get this stuff going!!!)
I'll dig up the TCL/TK code and place it on my web page this weekend.
I was wondering if you got around to uploading this as I'd love to see it!

From: Ron Ginger <ginger@...>

I am very interested in EMC. Id like to tinker with the GUI part of the
system- Ive done a Mill Power kind of clone alerady under Win95, Id like
to move that to EMC, maybe with Tk or Java. I assume there is a
programmer interface- API- soewhere in EMC that can be used this way,
but so far I havent found any docs about it.
Ron, meet Dave. Dave, Ron. Now that you two have been properly introduced
perhaps you'd like to work together on this ;)

While I'm asking for free software previews I though I'd ask if you could let
me look at your MillPower clone as well.

I sympathize with both of your concerns about the state of the EMC
documentation. I'll ask Fred Proctor next time I speak with him if there's a
good explanation of how to implement an EMC gui and communicate with the
motion control and I/O tasks. Also, in the very near future I'll try to make
a start on an EMC user's manual with this and other useful data in it. We'll
need better docs if the EMC is ever going to achieve its potential
popularity.

Matt Shaver
mshaver@...


Re: Problems with linux 2.2.10 + beta5 + emc-14-Jun-1999 - anyone ?

Matt Shaver
 

From: Max Heise <mahe@...>

Hi, I'm new to the list. Let me first introduce myself.
I'm student at the department of mechatronics/automation at the university
of applied sciences in Esslingen/Germany.
This is probably kind of offtopic, but is anyone using emc with linux-2.2.x
and the rtlinux patches for new kernels from ftp.rtlinux.org ?
The kernel and the rtl examples are running fine, but I get lots of errors
when building the emc-14-Jun-1999 package - as anyone tried this this yet
with success?
Actually, no. As several other folks have said, you'll need to use the 2.2.36
kernel with rtlinux-0.9J. Are you using RedHat 6.0 or perhaps S.U.S.E. ? If
Redhat 6.0, I don't know if you will be able to just drop in the 2.0.36
kernel, or whether you need to regress to RedHat 5.2. If S.U.S.E. (or any
other distribution other than RedHat) you are truly an EMC pioneer!

An other thing for the FAQ:
A month ago, i've spend two nights to figure out that emc and the shared
memory is quite sensitive to ipfilters like ipchains and ipfwadm. If you
get problems with these, flush input/output/firewall rules.
I just got IP Masquerading working on my Linux machine, and this was my first
exposure to ipfwadm (Since I'm running RedHat 5.2 I don't think I have
ipchains). Although I think that most machine tool controllers will be behind
the firewall, rather than being the firewall, it's probably important to
chase down this problem in order to at least understand what's going on. Fred
Proctor reads this list and he is also the rtlinux/shared memory guru. See:



If anyone needs the exact error message, send me an email.
Don't go to a lot of trouble, but I'd like to try mixing firewalling and the
EMC on my machine, just to see if I can duplicate your problem under RH5.2.
If possible, let me know what you did and what happened and I'll try it out
here.

Thanks,

Matt Shaver
mshaver@...


Re: EMC API?

mike grady
 

ron
I think you need to explain to most of these guys
what your millpower clone is/does
I wondered how
long it would be before it got ported to linux
mike

Ron Ginger wrote:

From: Ron Ginger <ginger@...>

I am very interested in EMC. Id like to tinker with the GUI part of the
system- Ive done a Mill Power kind of clone alerady under Win95, Id like
to move that to EMC, maybe with Tk or Java. I assume there is a
programmer interface- API- soewhere in EMC that can be used this way,
but so far I havent found any docs about it.

can soeone give me a pointer to the right docs?

What seems to be really missing to me is a sort of overview doc that
describes all the pieces and how they work together- the WEB page junps
right to a description of an .ini file and lots of gory details, but
never explains the basic structure of things.

ron

------------------------------------------------------------------------
Looking for a new hobby? Want to make a new friend?

Come join one of 170,000 e-mail communities at ONElist!
------------------------------------------------------------------------
welcome to CAD_CAM_EDM_DRO@..., an unmodulated list for the discussion of shop built systems in the above catagories.
To Unsubscribe:

bill, List Manager


Re: MaxNC modifications

Jon Elson
 

Jon Anderson wrote:

From: Jon Anderson <janders@...>

Some parts do not lose steps.
Each part will show a different pattern of lost steps. The pallet base
has a .500 dia hole that we indicate from to establish datum, all
pallets and all programming is referenced from this hole. It's real easy
to drop an indicator in and check for lost steps.
For example, one particular part run in multiples of 8 parts per pallet
might show .0013 lost in X and .0003 lost in Y every for every pallet
load. With the tolerances on this job, Dad can let this accumulate over
two pallets before having to rezero. This is so consistent that he
stopped checking after every two pallets and simply would jog X.0026
Y.0006 and rezero. After running several hundred parts, he'd check with
the indicator after the last two pallet loads and be off exactly
.0026/.0006.
Wow! this really DOES sound like a software problem. I really don't
believe this is the motor losing steps due to drag, electrical noise, etc.
It may be due to some characteristic of the control software, under just
the right sequence of G-code comands, produces step pulses too fast
for the motor to follow. But, that bit about fixing the axis offsets blind,
and being dead on after repeating the sequence 50 times, is just to
consistant to be anything other than software (one way or another).

Jon (increasingly happier that I went with servos)


Re: MaxNC modifications

Jon Elson
 

Tim Goldstein wrote:

From: "Tim Goldstein" <timg@...>

Jon,

I am sure you said, but what are you using for a controller? How many amps
are you running to the motors and at what voltage? What is the volt & amp
rating of your motors? What is the max feed rate you are trying to achieve?

I think you can safely assume your lost steps are not due to the software at
this point. From my playing with a number of software products I have pretty
well come to the conclusion that if you are having lost steps it is not the
software that is usually at fault. Do you have lost steps even at very slow
feeds or are the lost steps most common at fast rates?
Depending on the stepper drivers, the problem could be electrical noise. If
the stepper drivers take step pulses and a direction signal (and most work
that way), then stray electrical noise could cause extra step pulses to be
seen at the driver, even when the computer didn't generate them. You could
turn various motors on and off near the CNC setup while not commanding
any motion from the CNC control. If anything moves, even a couple of
steps, that's your problem. Grounding and shielding of the cables between
the parallel port and the stepper driver would be the first thing to try.

Jon


Re: Installing rotary encoders

Jon Elson
 

From: TADGUNINC@...

OK, at the risk of beating a dead horse, If you have access to .0001" DRO
scales, wouldn't they provide a more accurate method of locating the table,
than indirectly with encoders?
Yes! But, who can afford these? There are a number of innacuracies introduced
by using ballscrews and encoders. but, you should be aware that linear
scales are not a panacea! If the ways are worn such that the axes are not
orthogonal, or that an axis doesn't travel straight, then the linear scales won't
provide accuracy, either.

If not I'll be installing encoders on my mill, if so can the programs you
guys are running be set up to read from scales?
The computer should not be able to tell the difference.

Finally, does anybody have info about how well the BOBCAD program works? And
can it be used with the Linux programs for a source of the G codes?
It does not run on the Linux machine. they do have a DOS emulator, but I hear
news that the Windows emulator may be up and running. But, anyway, I
use Bobcad/CAM Ver 16.1 on a Windows 95 machine, and then send the
files over by network to the Linux CNC machine. I really haven't set Bobcad
up to have all the beginning and end functions I should have there, but it definitely
does work. I have cut several parts using Bobcad designs run through the
G-code generator of the CAM function. I find Bobcad cumbersome, and
the added things I have to do to specify roughing passes and finish offsets
to be pretty difficult. But, if the geometry is complicated, it definitely does
the job!

Jon


Re: EMC API?

Jon Elson
 

Ron Ginger wrote:

From: Ron Ginger <ginger@...>

I am very interested in EMC. Id like to tinker with the GUI part of the
system- Ive done a Mill Power kind of clone alerady under Win95, Id like
to move that to EMC, maybe with Tk or Java. I assume there is a
programmer interface- API- soewhere in EMC that can be used this way,
but so far I havent found any docs about it.
See : for a picture of the
screen. (This is from a month or two ago, they've been improving it still.)

can soeone give me a pointer to the right docs?
Umm, there is some documentation in the full installation package, but not
much on the UI itself. If you have specific questions, I'd be glad to answer
them, as I use it, and have done some serious poking around inside the UI
before Fred got it working so well.

What seems to be really missing to me is a sort of overview doc that
describes all the pieces and how they work together- the WEB page junps
right to a description of an .ini file and lots of gory details, but
never explains the basic structure of things.
Yup, and it is out of date, too (unless it's been updated recently, which may
be the case). There really is kind of an overview in there somewhere, that
shows the flow of messages between major components, and what, in
general, those components do. But, there could be more detail added.

Jon


Re: Problems with linux 2.2.10 + beta5 + emc-14-Jun-1999 - anyone ?

Jon Elson
 

Max Heise wrote:

From: Max Heise <mahe@...>

Hi, I'm new to the list. Let me first introduce myself.
I'm student at the department of mechatronics/automation at the university
of applied sciences in Esslingen/Germany.
This is probably kind of offtopic, but is anyone using emc with linux-2.2.x
and the rtlinux patches for new kernels from ftp.rtlinux.org ?
The kernel and the rtl examples are running fine, but I get lots of errors
when building the emc-14-Jun-1999 package - as anyone tried this this yet
with success?
I believe the kernel that Fred and everyone else is using is 2.0.36. I'm
pretty sure Fred said that he KNEW emc would not work with a 2.2.x
kernel, or specifically, Red Hat 6.0. I don't know the reason for this,
but it likely has to do with include files or something like that.

Jon


Re: re vacuum forming

Ted
 

rtr@...
Brian Robbins builds clear aircraft canopies for his Bright Star Millenium
and the Swift foot launched sailplanes by radiant and convection heating as
have been suggested on the list. He often uses a small sandwitch oven if
the part is small enough. He uses a bigger oven when the part blank needs
to be bigger. I hope to build him a radient heat source when time permits.
The little sandwich oven has a window so he can see when the blank is
sagging enough to conform to the mold properly.

When his shop vacuum didn't produce enough vacuum he put another
centrifical vacuum in series with it to get enough. This vacuum was
provided by Ron Wickersham, but almost any vacuum will work. His vacuum
forming machine for small parts consists of a five gallon plastic bucket
with the four inch shop vacuum line taped to its side. The form is
undrilled for air flow, but it is fastened to a piece of plywood which has
holes surrounding it, and in depressions in the mold if such are necessary.
These holes port to the 5 gallon bucket. He doesn't find it necessary to
provide a sealant along the lip of the plastic bucket. It seals adaquately
to the plywood board which holds the form.
Ted----------
From: MIADsgns@...
To: CAD_CAM_EDM_DRO@...
Subject: Re: [CAD_CAM_EDM_DRO] re vacuum forming
Date: Tuesday, June 15, 1999 2:44 PM

From: MIADsgns@...

In a message dated 6/15/99 5:39:57 PM SA Eastern Standard Time,
footman@... writes:

<<
I run a small company that manufactures foot orthotics (supports) -
these
are vacuum formed over plaster casts of peoples' feet - sounds like you
want
something that ends up with a flawlessly clear finish - with my equipment
(home made), my plastic (or carbon graphite & plastic) ends up with an
opaque
finish that is fine for orthotics, but would likely be lacking for your
purposes - let me know if you want more info . . .

Christopher Ellacott, D.Ch.
Chiropodist >>


Interested in your home equipment. I made a desktop vac former that works
well with thin plastic and uses a shop vac. but I need more succion to
form
with heavier plastic and finer detail. Any clues as to how to modify such
?

Mario

------------------------------------------------------------------------
Make your voice heard!

Sign up for the ONElist Weekly Survey. Go to homepage for details.
------------------------------------------------------------------------
welcome to CAD_CAM_EDM_DRO@..., an unmodulated list for the
discussion of shop built systems in the above catagories.
To Unsubscribe:

bill, List Manager


Re: programs to write G-code

Tim Goldstein
 

Pretty slick Jon.

I for one an pretty impressed with your talents. machinist, programmer,
electronic designer. You put us mere mortals to shame.

Thanks for sharing your hard work.


Tim
[Denver, CO]

-----Original Message-----
From: Jon Elson [mailto:jmelson@...]
Sent: Tuesday, June 15, 1999 11:27 PM
To: CAD_CAM_EDM_DRO@...
Subject: Re: [CAD_CAM_EDM_DRO] programs to write G-code


From: Jon Elson <jmelson@...>

I have put some programs that write G-Code for common
machining operations on my web page. Take a look at


The programs are in C, and I've made them as portable as
I can. They should run undel Linux with Gcc, or with
MS-DOS and the Microsoft C compiler, or Windows xx,
with Borland C. If you have trouble with them, let me know.

I will be adding some more programs as soon as I get them
converted.

Jon


------------------------------------------------------------------------
ONElist members are using Shared Files in great ways!

Are you? If not, see our homepage for details.
------------------------------------------------------------------------
welcome to CAD_CAM_EDM_DRO@..., an unmodulated list for
the discussion of shop built systems in the above catagories.
To Unsubscribe:

bill, List Manager


Re: re vac forming

Tim Goldstein
 

I have done some composite wing vacuum bagging and used one of the old
commercial freon tanks as a vacuum reservoir and it held 24" hg no problem.
I now use a section of 4" PVC pipe with endcaps for my vacuum tank and I
have tested it to 24" hg with no problems.


Tim
[Denver, CO]

-----Original Message-----
From: Jon Anderson [mailto:janders@...]
Sent: Tuesday, June 15, 1999 10:56 PM
To: CAD_CAM_EDM_DRO@...
Subject: Re: [CAD_CAM_EDM_DRO] re vac forming


From: Jon Anderson <janders@...>

MIADsgns@... wrote:

Any suggestions on buiding the pump with meter, sources for
the meters, hardware etc.?
I don't see any reason you can't use that tank as a vacuum plenum. Don't
know how many inches of vacuum you can pull with a shop vac, but the
tank will help with volume.
Valving to take the shop vac out of the loop could be a simple PVC ball
valve from the hardware store. Costs should be reasonable. Used ought to
work just fine if you can find it.
You can tolerate a slight leak here, you just don't want a high
vacuum/low cfm pump trying to suck air back through the shop vac while
it's running.

Some old friends have a business in town (Maier USA) making accessories
for dirt bikes and ATV's, vacuum forming plastic.
It's a trip watching a droopy sheet of plastic turn into an ATV
seatbase/fender in a matter of seconds! Their Thermwood 5 axis gantry
router is pretty neat too...

Jon

------------------------------------------------------------------------
How has ONElist changed your life?
Share your story with us at
------------------------------------------------------------------------
welcome to CAD_CAM_EDM_DRO@..., an unmodulated list for
the discussion of shop built systems in the above catagories.
To Unsubscribe:

bill, List Manager


Re: re vac forming

Tim Goldstein
 

Sort of funny you mentioned Lindsay Publication. I just got their catalog
yesterday and have been looking through it to see what books I want to get
on my first order.


Tim
[Denver, CO]

-----Original Message-----
From: MIADsgns@... [mailto:MIADsgns@...]
Sent: Tuesday, June 15, 1999 10:45 PM
To: CAD_CAM_EDM_DRO@...
Subject: Re: [CAD_CAM_EDM_DRO] re vac forming


From: MIADsgns@...

Thanks Tim,
My own search brought me to an excellent source of publications
for making
hard to find items such as Injection molding, vac forming
machines, etc, etc.
Lindsay Publications. The flyer I have its Called, Secrets.
(815) 935-5353.
It includes the book by Douglas E. Walsh. For anyone else interested in
making theuir own stuff, this is an ecellent source.

Mario

------------------------------------------------------------------------
Congratulations to "PAFind," our latest ONElist of the Week.

How is ONElist changing YOUR life? Visit our homepage and let us know!
------------------------------------------------------------------------
welcome to CAD_CAM_EDM_DRO@..., an unmodulated list for
the discussion of shop built systems in the above catagories.
To Unsubscribe:

bill, List Manager


Re: MaxNC modifications

Tim Goldstein
 

By the controller I mean the electronics that take the step and direction
signals and actually drive the motors.

OK, a few more questions. Don't know if I have an answers, but questions
seem in order!

Is the motor rated 5 amp as a unipolar or bipolar? Did I ask what the
voltage rating of the motor is?

I think your desire for 60 ipm may be a little much. 30 ipm may be a more
realistic number to shoot for. The Sherline I had with 125 in/oz motors
pooped out at about 35 ipm and 30 was pretty reliable. The Shoptask I have
now with 550 in/oz motors and ballscrews (rolled, not ground :( can achieve
about 33 ipm, but is rock solid reliable with 24 ipm as a max.

Tim
[Denver, CO]

-----Original Message-----
From: Jon Anderson [mailto:janders@...]
Sent: Tuesday, June 15, 1999 10:21 PM
To: CAD_CAM_EDM_DRO@...
Subject: Re: [CAD_CAM_EDM_DRO] MaxNC modifications


From: Jon Anderson <janders@...>

Tim,

Will have to sit down with my dad and gather up the details, but here's
what I can tell you right now.

Not sure what you mean by controller, Ah-ha communicates directly with
the stepper drivers via a special card.
Power supply is 40 volt more or less, unregulated, based upon the
schematic supplied by CyberPak for their drivers. Dad built the PS, I'll
get details from him.
Don't recall motor specs, will have to look them up again.
Think they are rated for 5 amps, we are running about 3 now, or so we
think. That is based upon the current limiting resistor.
Dad thinks we're not getting full power to the motors and tried to
measure amp draw INTO the drivers, came up with something like .5 amps
draw at rapid (40"/min). However, it's a chopper driver, not sure if
he's checking properly.
I've written little test routine programs. Single axis linear moves,
looped and run for 5-10 minutes show no lost steps. Interpolated linear
moves (XY) show no lost steps. Circular interpolation seems to be the
problem, but not always.

My dad machines little resistive elements for a customer of his. These
are generally under 1" dia and we run multiples of up to 8 at a time on
a custom pallet setup. We've tried using G92 to set offsets for each
nest, no subroutines; G92 with subroutines; G91 with and without
subroutines; and if a particular job is going to show lost steps, it
doesn't seem to matter how we approach it. Some parts do not lose steps.
Each part will show a different pattern of lost steps. The pallet base
has a .500 dia hole that we indicate from to establish datum, all
pallets and all programming is referenced from this hole. It's real easy
to drop an indicator in and check for lost steps.
For example, one particular part run in multiples of 8 parts per pallet
might show .0013 lost in X and .0003 lost in Y every for every pallet
load. With the tolerances on this job, Dad can let this accumulate over
two pallets before having to rezero. This is so consistent that he
stopped checking after every two pallets and simply would jog X.0026
Y.0006 and rezero. After running several hundred parts, he'd check with
the indicator after the last two pallet loads and be off exactly
.0026/.0006. (I don't recall exactly the sign (+/-) of the error or
corrections, but the numbers are real close to what he runs into)
This is with a program that does not use G92, it's done with geometry in
a subroutine, we move to a set entry point, enter the subroutine, enter
G91 and machine features, then back to G90 and exit the subroutine.
We lose NO steps in the Z axis.
Motors are wired with shielded cable, grounded at one end to the motor
PS ground. Logic lines are also shielded cable, grounded at one end to
the computer ground. Care has been taken to keep power and logic lines
well separated. If I didn't mention before, we've also run the MAX
software on two different computers, a 486-33 and 486-66 with identical
results. I'm close to getting my Cyrix P200 box done and will try
the Ah-ha from it, but don't think that's going to solve the problem.
Ah-ha warns about possible rounding errors and recommends programming to
4 decimal places. I edited a problem program to accomodate this and
there was no change in the lost steps, so that's not likely to be the
problem.

I printed your message, when this gets back to me I'll print it as well.
Dad will be by tomorrow and I'll give him both copies to go over and
fill in any details I missed or don't know.

We would both LOVE to solve this problem. Dad's good with basic
electricity, but we are both stumbling in the dark on this, and it's
getting expensive running around buying stuff trying to fix the problem.

As for desired speed, would like to see 60"/minute. This was part of the
reason I went with 16 pitch screws. I can accept the slight loss of
accuracy, the tradeoff is that for a given linear velocity we are
running the motors 20% slower. I may be all wet, but my seat of the
pants guess was that we were coming out ahead over a 20 pitch screw by
keeping the motor down in a broader part of the torque curve.

I have some top quality linear rail bearings and ballscrews, and want to
build a CNC conversion for my Hardinge. However, I have a hard time
justifying the time and expense in finishing it with the lost steps we
are getting. Even a single lost step per part would quickly cause major
problems in the X axis of a lathe.

Thanks,

Jon
(who btw, is highly impressed with the quality and quantity of posts to
this list!)

------------------------------------------------------------------------
Where do some of the Internet's largest email lists reside?

At ONElist - the most scalable and reliable service on the Internet.
------------------------------------------------------------------------
welcome to CAD_CAM_EDM_DRO@..., an unmodulated list for
the discussion of shop built systems in the above catagories.
To Unsubscribe:

bill, List Manager


Re: ball screws Backlash

Jon Elson
 

TADGUNINC@... wrote:

But about your lead screws, how do I contact Thompson?
Their phone # is (517) 776-4123
I'm sure they have a web site, but I don't have the URL.

As far as the
backlash error, if you preload two ballscrews against each other couldn't you
effectively eliminate the backlash?
That is how an anti-backlash ballnut works. It really preloads two nuts
against the same screw, that's easier.

Jon