Highest Resolution Linear Motion System?

Hello all-
I've been a long time lurker of this site and I've been thinking about/designing (and doing some building!) cnc's for a few years now. I'm trying to put together a small, high resolution cnc laser cutter and can't decide on the most appropriate linear motion system. I'm looking for ~4" of travel in the X and Y (no Z) with a resolution down to around 0.001mm (1 micrometre). So far, I've been mostly designing around steppers driving small, high tpi leadscrews. I've been having trouble sourcing high tpi scews that will give me enough travel. There are a few companies that make 200tpi screws mainly used in (I think) scientific apparatus, but the few that I found with enough travel were prohibitively expensive. McMaster sells a 41.7tpi miniature precision acme screw and, with microstepping, I think I can get the resolution I'm looking for.

I though I'd pose the question to the community here. Is there a better way to do it? I don't have much knowledge or experience with them, but would a servo system serve me better than steppers? Being a laser cutter, it won't be subject to nearly as much force as a mill would and, for my application, it can be run pretty slow. Could I get away with another drive system altogether? Maybe belts or a rack and pinion?

Thanks for the help

zaalexander

The only way to get that kind of accuracy ( 1 Micron ) is to use Ball-screws,which would be ground from C1 to C3 Grade, 3mm pitch, is a normal size that would work,then ac servos with 16 Bit to 20 Bit high count per rev encoders

You also could use a linear motor most can do ( 1 Micron ) positioning accuracy

Steppers belts, rack & pinion lead-screws, Etc it's not going to happen

You might get some ideas looking at <http://www.zaber.com>
Or you might get a heart attack from the prices.

There's a paper written by Cliff Mirman and Andrew Otieno on making a low-cost micromachining center, using ACME screw, steppers.. you should be able to find the white paper online,

Maybe find some used aerotec linear motor stages. You'll never get there with steppers.

Quote:

---

Originally Posted by vegipete

You might get some ideas looking at <http://www.zaber.com>
Or you might get a heart attack from the prices.

---

More likely to get a heart attack from the specs: A-LAR-E series: Backlash: <255 microns, or <0.255 mm. What we might call in the CNC world a rattling good fit, with an emphasis on the rattle.

You want accuracy over 4"? Linear motor with optical grating feedback, or if you can afford it, laser interferometer feedback. Both are $$$. You will also need an expensive air conditioner.

Cheers
Roger

First things first, how are you going to measure the resulting parts? Measuring to that resolution is extraordinarily difficult. You can't hold the part in your hand because in the time it takes to do the measurement, the size of the part will change due to thermal effects.

Machining to that level is a whole other world. Everything affects what you are doing. Dimensions are specified at a specifc temperature. When you machine a part, you might have to wait several hours before measuring it since the machining operation heated the part up enough to change the dimensions and you have to wait for it to re-stabilise at the design temperature.

OTOH you can simply tell everyone that you machined something down to micron accuracy. The chances that they can call you out on it and back it up are slim to none.

bob

For what its worth, I can watch my Taig tick by at .000025" per microstep, 40 microsteps per .001" with a Mitutoyo No. 2923-10 (which I have modified internally to get me almost 0.1 inches of travel)

Using a 40 tpi lead screw and off the shelf steppers a direct drive with a gekco stepper driver can do 12.5 micro inches per microstep, and you will be able to measure this with the appropriate equipment, but the problem is it will take you hundreds of hours of labor to lap the leadscrews to get any kind of real useful work from such a machine. you will need a spring to constantly provide tension to take up the backlash.
maping the screws is required.

1 micron is well within the functional tolerance of a cheap set of gauge blocks, but a 50-100$ tenthousanths dial or test indicator isn't good enough to measure below half a tenth.

btw, seems to me that this kind of accuracy would only be expected from an electron beam machine.. i don't know that you can even focus a laser good enough to achieve micron tolerances.

Hello

We provide the best lasers routes and plasma CNC machine. Need information. Please contact us

Regards
Casper
china famous of Pipe plasma/Fiber laser/Gantry plasma/Co2 cutting machine manufacturers and suppliers
Laser Solution - CO2 laser cutting machine, fiber laser cutting machine, waterjet cutting machine, CNC Router Machine, laser engraving machine

> i don't know that you can even focus a laser good enough to achieve micron tolerances.
Depends on the laser wavelength. 10.6 microns (industrial high-power IR CO2),I doubt it! UV laser - possibly.

Cheers
Roger

Sadly, I dont really agree with any of the above posts.

I actually build stuff to micron accuracies, industrially, and its not especially difficult, nor does it need most of the stuff mentioned.
I make high resolution screws, among other things, and have spent over 2000 hours to be able to accurately thread, on my cnc conversion.
I have studied the matter extensively for over 7 years, full time in cnc stuff.

My cnc conversion has 0.2 microns step size, and actual resolution is sub-micron.

Some refuttals to myths mentioned:
When working with industrial equipment in the 100.000$ range, of which I have sold dozens, only 2% of all machines (several thousands I had statistics on, over 20 years) have glass scales, and about 1% here in spain have air conditioning.

Standard lead screws, not ball screws, easily have resolutions, and can have accuracies, down to 0.5 microns.
Examples:
All old jig borers, from the 1960s forward, had 1 micron accuracies.
Mechanical screws, no ballscrews.
Moore (of Moore and Wright) were doing sub micron work maybe 30 years ago. Standard screws.
I learned a lot from their work Foundations of mechanical accuracy.

Manufacturers of accurate screws mention that the current high resolution screws have a 0.5 micron accuracy.
These are either direct 0.2 mm rise screws, or differential screws.
Thorlabs is one maker.
I have 3 Thorlabs screws, with 0.5 micron accuracies, differential.
They cost about 80€ each, here in spain, tax freight and duty paid.
Travel is about 3 mm, iirc.

A stepper driving anything is accurate to 0.5 step - not to microsteps.
They are springs, and lag by upto anythign less than 1 step, and do, under load.
Yet, this is not relevant to your non-contact application.

Measuring to 1 micron is easily achieved with basic, good quality, electronic micrometers, or electronic plunger indicators.
They are accurate to +/- 2 microns, and repeat to 1 micron.
Easily tested with gage pins (accurate to 1 micron).
Micrometers to 6", and indicators, cost 80€ each, or less.
Contact me off list if you want some.

Better micrometers are available.
Mahr and Federal both make sub micron resolution micrometers, from 800 to 2800$.

Small leadscrews are not, imo, a good solution to your motion control needs.
They will wear quite quickly, if the system moves fast, or a lot.

1 mm rise ground ballscrews, even used, will work fine.
Suggestion:
Buy c0 or c1 quality second hand ground ballscrews, around 100€ each. Ebay has them all the time. 10-20 cm usable lengths are available.

Movement:
Drive these with a servo, of 4000 counts, and you will have better than 1 micron resolution and repeatability.
Small ac brushless servos are 150€ each.
Contact me offlist.
A full brushless servo system is 290€ (taxes paid in the EU). Driver, motor, cables, encoder 4000 counts.
200 or 400 watts.
A servo positions, always, to max +/- 2 counts, and in practice to 1 count.
+/-1, of 4000, means you have 1/2000 count repeatability, and resolution.

With low speed, you dont need a high end controller like a CSMIO-IPS, which Is what I use and recommend.
A CSMIO for threading, like I use on my lathe, is about 1500 €, all in (CSMIO,2x24V DC PSU, ENC, MPG, extra IO, 500 ferrules, 100 pack din rail connectors, realys, etc).

You can use a cheap, very good, Pokeys-E (56€) with CNC-Addon (80€).
This gets you 4 axis, 125 kHz, miniature good quality ethernet system.
From Polabs in Poland.

Accuracy vs resolution:
The mentioned screws, 1 mm rise, C0 ballscrews.
Used screws, or unmapped screws, will get you repeatability and resolution down to 0.5 microns.
They are not particularly accurate, in that the screws have some (very) small error.

This means that you can make features that are accurate to 1 micron, but they may not always be positioned exactly at 1 micron accuracies true position.
You can map the screws with a cheap glass scale DRO, if you need.
DRO system with 1 micron scales is about 700€.
1 micron and 0.1 micron scales are readily available (down to 0.05 microns. Standard on Moore nanotech lathes, and better machining centers (Makino I think uses 0.05 microns).

Here above you have a shopping list and practical, workable, solutions.
I hope it was useful, not too long, and not too expensive.

Please let us know if you proceed with this.

Just because you can move something by 1 micron, does not mean its accurate to one micron. I do these accuracy's on a daily bases, on very expensive equipment. The last one the scales alone were 18k per axis. As I said before, steppers will never get there. I am lucky enough to have all the equipment to measure sub micron, and any mitutoyo will never get you there, you will need true LVDT gauging and that's 2k to get a basic set up. Then there is the autocolumators to check flatness, laser for position and squareness testing, an on and on. We do have one laser on aerotech stages that can hold a one micron tolerance on part profile. That was close to 1 million $.

Under .. its true that resolution is not accuracy.

Yet..
1 micron accuracy is std on manual, 40 year old equipment.

Today, advanced machines, like makino, deliver better than 1 micron machined, round, interpolation.
So does Moore Nanotech, tiny lathe. 0.3 micron machined accuracy.

And Mitutoyo is a well respected name that probably delivers what they promise, dont you think ?

Its quite easy to make small pieces to 1 micron accuracy. Say 3x3 cm.

Thermal effects become critical when you want to make larger ones.
Say 1 x 0.1 m.
Yet, glass scales are not expensive, any more.
They deliver sensing to much better than 1 micron.

The million-dollar FUD is a holdover from older times.

I found the article.... Original authors listed in the article.
\
[QUOTELow-Cost Micromachining Development and Application for Engineering and Technology Education
Cliff Mirman and Andrew Otieno
Department of Technology, Northern Illinois University, DeKalb, Illinois ([email protected])
Abstract
The concept of the micro-factory, conceived in the early 1990’s, led to the development of what is now deemed “the micro-machine-tool”. The increased need for miniaturization of parts has continued to play a major role in developing micro-manufacturing technologies. Micromachining has recently become the norm for producing very small complex parts with very tight tolerances and features. In order to move into this new micromachining area, companies must overcome several issues. The first is how to obtain low-cost, yet highly accurate micromachining equipment. The second is personnel that are skilled enough to operate these new generation micromachines. Because micromachining is a fairly recent technology, there is no machinability data to aid operators to perform the process efficiently. This paper reports on a machinability study carried out with the intent to develop optimum process parameters. The material used in this study was 316L stainless steel. It is demonstrated that the methodology employed can be used for any material. Detailed educational experiences which were developed utilizing micromachining techniques are also presented. Specifically, we propose the inclusion of micromachining courses within manufacturing, production or mechanical engineering curriculums. A sample of the proposed course is presented.
Keywords: micromachining, manufacturing curriculum, machinability,
1. INTRODUCTION
The increased need for miniaturization of parts has continued to play a major role in developing micro-manufacturing technologies. Micro-manufacturing focuses on technologies used to produce parts in the sub-millimeter range, and essentially bridges the gap between nano-scale and macro-manufacturing. According to the World Technology Evaluation Center’s (WTEC) commission report [1], micro-manufacturing has and will continue to have very significant impacts on national security, defense, energy, healthcare and domestic manufacturing base. Micro-parts are being utilized in the electronic and drive systems for small unmanned reconnaissance planes, for high precision parts used in missile guided systems, for medical devices to deliver medicines in tumors located in fragile internal organs, and many other significant applications.
There are many similarities and differences between macro- and micromachining. While the use of solid modelling and computer-aided manufacturing (CAM) software is similar, the operations and tooling are vastly different. Current research on micro-machining has focused mainly on developing the actual machine tools. In the U.S., micro-machining centers have been developed successfully at several institutions, including the University of Illinois Urbana-Champaign [1, 2], at the Florida International University [3, 4], and Georgia Institute of Technology [5, 6]. Although these machines, and those developed elsewhere nationally and globally, have been proven to work, complete commercialization has not been realized due to several factors. Firstly, the existing micro-machine tools cost in the region of US$80,000 to US0$200,000, which is very inhibitive for most companies intending to enter into this lucrative business. Secondly, micromachining applications dictate high precision, often in the sub-micron range. Inspection for tolerance and tool condition in this regime requires the use of expensive optical equipment that would not be feasible to incorporate into a micro-machine tool for practical applications. Moreover, tool wear monitoring 1
continues to be a challenging issue as there is no reported method that can be adapted to perform this task in a practical way. Most researchers [1, 3, 7] have found that in micro-machining, premature tool breakage is more is a common occurrence than tool wear. This has a significant influence on the resulting precision of the final product. In addition to tool condition, another factor that affects mach inability and precision is vibration. Vibration and chatter lead to poor surface finish. Increased vibration has been shown to increase chip load, thus significantly increasing cutting forces [2, 8]. In micro-machining, this may be a major cause for tool failure. Vibration modelling is essential to support chatter avoidance [8]. Research shows that with careful design of a rigid micro-machine, vibration can be predicted and isolation can be achieved [2, 8, 9]. However the influences of material properties at the micro-structure level, especially for steels, and cutting conditions are still uncertain and work continues to be done in these areas [2, 9, 10]. In addition to these factors, the mechanics of micro-cutting is still a relatively new area that is not well understood. In micro-cutting, there is no formal explanation of scaling effects (also referred to as “size effects”), for example, in the relationship between material removal rate and the specific cutting energy [5].
Through funding provided by the U.S. Army, Tank and Automotive Command (TACOM), Northern Illinois University (NIU) engineering and technology researchers developed a new generation low cost machine tool. In the same project, studies were conducted to examine material removal rates for different types of cutting tools at various speeds, feeds and depths of cuts for different materials. Lastly, this paper also presents an educational perspective for training engineers in micromachining.
2. DEVELOPMENT OF A LOW COST MICRO-MACHINING CENTER
2.1 Hardware
CNC machining at the macro level has been in existence for many years. Much has been learned about the process in relation to cutting parameters for different materials and different types of operations. However, compared to macro-machining, less is known about micromachining. As manufacturing needs have changed, and as parts grow smaller, there has been a major shift in how machining is accomplished on a micro scale. While macro-machining uses large tooling to cut material, in micromachining the tools are much smaller and the cutting process operates on different mechanics; thus there must be an understanding of this process. In the micromachining realm, cutting accuracies are on the micron level, and thus vibration and rate of travel have a large effect on the accuracy produced. Figure 1 shows examples of micromachined parts.
Micromachining is not a new technology, however the technology is still being researched and many micromachining companies are still in a transition to commercialize the process. Typical micro-machines are priced from US$80,000 to well in excess of US$200,000, depending on the usage and accessories that are required. This high cost places the machines outside of the budgets of many schools, small companies, and R & D laboratories. In this project, NIU engineering and technology researchers were given the task of developing a new generation of low-cost micro-machine (LCMM) which would be affordable and yet provide the required accuracies. It should be noted that the design engineers were given a time frame of approximately four months to outline, research, design, and construct the first generation LCMM.
Figure 1 - Example of two micro-machined parts
The following are the constraints that were placed upon the initial design of the LCMM:
• Material Costs - US$12,000
• Spindle speed – Between 10,000 - 100,000 rpm
• Accuracy – Between 0.001 inches – 0.0002 inches (25 microns - 5 microns)
2
• Work Area – 2 inch cube
• Open loop control – accurate without feedback
• Actuation – step motor/ lead screw drive
• Programming –G & M codes
• Tool Changer – 5 tool
• Number of Axes – 3 (x, y, and z)
The development of the first generation LCMM included the mechanical, electrical, and control aspects of the machine, as well as selecting the appropriate CAM software needed for solid modeling and the operator interface. There were two aspects of the machine that were at the forefront of the development – cost and accuracy. Since the overriding consideration in the development was cost, all components, software, and assembly needs were optimized, and off-the-shelf components were used whenever possible. The consideration of accuracy was more difficult since it is a compilation of many different tolerances. The accuracies are dependent on spindle run-out, actuator and motor resolution, and controller software, and the manufacturing tolerances. With all of these factors taken into account, the first generation, low-cost, 3-axis, micro-machine was designed, constructed and assembled and is shown in figure 2.
Figure 2 – Initial design of the low-cost micro machine
Feed in each of the three axes is achieved by a stepper motor and ACME lead screw combination. The lead screws are 0.25 inch (6.35 mm) diameter with a pitch of 0.05 inch (1.27 mm), utilizing an anti-backlash nut. The use of the anti-backlash nut is an important feature due to the vibrations that are encountered in the machine. The actuators have a rigid aluminum structure with roller bearing support, and thus provides the needed stability, as well as low friction motion. The stepper motors are NEMA 17 with micro-step ability, and Table 1 below shows the available accuracies for each step with and various lead screws. The stepper motor produces 6 oz-inches of torque which in hind site is more than enough to provide system motion.
3
Table 1 – Positional accuracy with motor/lead screw combination
The low-cost machine has been set-up to use 1/4 stepping for a positional accuracy of 0.0000625 inches (about 1.6 microns). It should be noted that in all cases, the accuracy achieved is dependent on other considerations such as manufacture and assembly processes employed. In addition, spindle run-out is responsible for much of the inaccuracy. Typical spindles have inherent run-out in the region of several microns. However, the accuracies are improving. The low-cost spindle which is being used in this application, shown in figure 3 below, has a low run-out; approximately 0.00024 inches (about 6 microns).
Figure 3 – Spindle in operation
2.2 CAM software
One of the initial design constraints developed for the LCMM was that of software and the human-machine interface (HMI). At the start, the machine was designed to utilize commercially available, low-cost CAM software that has the needed flexibility to produce three dimensional parts efficiently. Since the machine utilizes linear actuators on each of the three axes through stepping motors, the software needed the capability of providing control to each of these motors with the required accuracy. It also needed to have the capability to micro-step the motors and alter acceleration profiles. In addition, the CAM software to be used needed to have the ability to allow the transformation of 3-D part geometry into common G and M codes. It was desired that the software be able to interface with popular CAD and solid modeling packages like AutoCAD®, Solidworks®, and Pro/Engineer®. Initially, MACH 3™ was chosen as the control software, and the output was sent to a stepper motor driver controller. Since the computer was running the driver software and CAM software, the commands in MACH 3™ were not occurring in “real-time”, and thus, there was a slight time lag due to the multiple computer processing needs. Figure 4a shows a complex precision part cut using the MACH 3™ software and the resultant stair-stepping
4
on the cutting surface. This issue was fixed through the implementation of FlashCut™ control software. It is a “real-time” control software which does not run on the controlling computer. The FlashCut™ is run on a separate micro processor, thus, the software does not exhibit any time lags, resulting in better accuracy. The machined part using the new FlashCut™ software is shown in figure 4b. The software also integrates with the motor controller, and is relatively cost effective.
Figure 4a – Part cutting using Figure 4b – Part cutting using MACH 3™ software FlashCut™ software
The use of FlashCut™ also allows for auxiliary programming to incorporate control of motor spindle speed control through the user interface screen, in addition to managing some important parameters which greatly enhance the control and precision aspects of the machine developed. The new software and controller will allow an additional 4th axis to be added to the system, importation of common CAD and other solid models, and digital inputs and outputs for limit/home switches and sensors. Figure 5 shows the user interface for the FlashCut™ control software. In figure 5, the operator can see the tool path, coordinates of the cutter, G-codes, spindle speed, as well as other controlling interface buttons.
Figure 5 – FlashCut™ User Control Interface 5
Based on a 4-month fast development time for this first-generation NIU low-cost micro-machine, the authors developed a unique machine with many capabilities. With respect to the initial constraints which were outlined in section 2.1, the following were the outcomes and characteristics of this machine:
Final Results - Summary
• Component and material cost – US$7000 (at present)
• Work envelop – 4 in x 4 in x 4 in - (100mm x 100mm x 100mm)
• Portability – Weighs less then 30lbs (13.6 kg), with a 18in x 18 in x 18 in (450 mm x 450 mm x 450 mm) outer envelop.
• 3-axis machining – 4th axis will be added in the near future
• Accuracy – about 0.00012 inches (3 microns)
• Components – all off-the-shelf, and the reliability has been proven through 1 year of testing and part production.
• Actuation – Stepper motor (NEMA 17)/ lead screw drive with a 0.05 inch (1.27mm) lead using ¼ stepping (0.225 deg/step)
• Spindle speed – 40,000 rpm maximum.
• Low spindle run-out – about 8 micron.
• Software – ability to import many different CAM files. Relatively easy for operator to learn use of software. Allows the use of limit, home, and part height sensors. Flash cut software will support many import file types.
• Open loop control – accurate without feedback
• Actuation – step motor/ lead screw drive
• Programming –G & M codes
• Tool Changer – 5 tool
In general, the design team met or exceeded all of the initial design parameters.
2.3 Undergraduate student involvement
The authors and the Department of Technology at NIU have been involved in this project for over two years. During this period of time, several undergraduate students were involved in the design, construction, testing, and modification of the LCMM. At any given time, there was at least one manufacturing engineering technology and one electrical engineering technology student working with the authors on this project. Students were involved in all aspects of the projects. At the outset, students were involved in developing algorithms for controlling motors, both for the spindle and the three axes, as well as the user interface. Since the goal of this project was to develop a machining center which was capable of producing near micron parts, accuracy was an area which needed to be enhanced. Using metrology techniques, the students, working with the faculty members, developed methods of measurement and providing accuracy for the machine and parts that were cut.
3. PROPOSED MICROMACHINING CURRICULUM
This low-cost micro-machine was developed for use in industry but also in the classroom, where many institutions could not afford a high priced machine. In order to instruct students on the operation of this new realm or machining, a new micromachining course had to be developed. This course is a lab/lecture course which utilizes a hybrid laboratory component; both on-line and live. The following laboratory modules have been developed for this course.
Module 1 – Introductions to Mechanical Micromachining
Introduction to mechanical, physical, chemical properties, machinability, and typical applications of materials in micromachining, including metals, polymers, ceramics, and glasses. Students are made aware that materials used in micromachining are required to be free of inclusions typically found in castings and forgings used in conventional machining. It also reveals the similarities and differences between mechanical micromachining and conventional machining, including the effects of tool edge radius along the cutting edge in mechanical micromachining, in which the chip thickness becomes a comparable size to the cutting edge radius. Small chip load as compared to the cutting edge radius, the size effect, and the ploughing forces are also covered as significant characteristics in micromachining. [1]
6
Module 2 - Process Parameters in Mechanical Micromachining
The major object of a machining process is to maximize metal removal rate while maintaining good surface finish, long tool lives and lower power consumption by minimizing cutting forces. Whether it is turning or milling, or any other machining process, the proper choice of machining parameters is essential. This module introduces important parameters including cutting speed, feed, and depth of cut within the context of micromachining, and their critical impacts on the surface finish and integrity, tool lives, energy efficiency, and process stability. Tooling used in mechanical micromachining is much more costly than conventional machining, thus, preventing tool wear and breakage is critical in mechanical micromachining. Students are shown how to select correct process parameters for different tool-process-material combinations.
Module 3 - Tooling for Mechanical Micromachining
It is important in machining to ensure long tool life for good dimensional tolerance. Proper tool selection is therefore a key component of mechanical micromachining. This module emphasizes milling as the most common process in micromachining, and the most commonly available end mills. In addition, micro drills are also be covered in this module as a most-frequently used process in mechanical micromachining. Tools in micromachining that are commercially available are introduced, including tungsten-carbide micro end mills starting at 25 μm in diameter, and examples of their applications in fabrication of micro- molds and dies from tool steels for injection molding and micro-forming processes. [2,3,4]
Module 4 - Inspection in Mechanical Micromachining
Inspection of parts for dimensional integrity and quality control in micromachining is covered in this module, including optical and high-magnification microscopes, scanning electron microscopes, and tunneling electron microscopes. Students are introduced to basic metrology methods for inspection involving high magnification microscopes.
Module 5 - Interfacing with CAM Software and Understanding of NC codes
Students going into the micromachining professions need to have an in depth understanding of the GUI interface used by the CAM system. This module develops a knowledge base for using a low cost CAM interface, such as flash cut and micro mill, which will provide the students with a starting platform for use of the mechanical micro-machine-tools. Students learn to manually move the three axes, utilize the spindle controller as well as other inputs needed to control the system, and import part file format from any compatible software and start the cutting process. Students are instructed on spindle speed control, part programming and understanding of G codes, which is fundamental in developing CAM part files for the machines, and the modification of these codes to given parameters associated with the mechanical micromachining environment.
Module 6 – Maintenance Mechanical Micro-machine-Tools
Unlike conventional CNC machining centers, the new generation of mechanical micromachining centers will need different care and maintenance. This module introduces maintenance technologies specific to micromachining, as the process does not create conventional chips that a larger machine would. Techniques are introduced to handle the slurry of the removed material as a fine paste that gets into many areas of the machine. This section introduces students to general maintenance and protection procedures of micromachining centers by materials and by products of the cutting operation. Students learn about the fundamentals of electromechanical mechanisms, including those of the stepper motors and limit switches that are basic components of the mechanical micromachining centers.
4. PRELIMINARY RESULTS FOR MACHINABILITY STUDIES
As stated earlier, micromachining is a fairly new phenomenon in relation to macro-scale machining. For this reason, there is still no data on process parameters such as recommended feeds, speeds and depths of cut for various materials. One of the goals of this project is to complement studies in micromachining by developing methods to determine desirable machining conditions and reliable techniques for process control. This project also aims to enhance knowledge for micro-machining processes that will enable the development of next-generation micro-
7
machine tools. Some preliminary cutting tests have been performed to determine feasible cutting conditions for stainless steel 316L. Examples of cutting force measurements for micromilling at 30,000 rpm spindle speed with a 0.5 mm (0.02in) diameter cutting tool at various depths of cut (d) and feeds (f) are shown in figure 6 below. The depth of cut is in millimeters. Fx is the cutting force in feed direction while Fy is the force in direction perpendicular to feed. The sensitivity of cutting force with depth of cut is seen to be more significant than with feed. It should be noted that the feeds chosen were based on the maximum available for the micromachine. From these studies, it was concluded that the maximum achievable depths of cut for a 0.5 and 0.25 mm diameter tool 50 and 25 microns (2 and 1 microinches) respectively. Preliminary analyses also indicate high levels of burring at the selected cutting conditions, as shown in the high magnification image below (figure 7). It should be noted that more work is under progress the study tool lives and the machinability of other materials as well. Most important, it noted that the cutting conditions are likely to be affected by the design of the machine and these studies provide only a guidance for selecting the correct process parameters. Cutting force vs feed-2-1.5-1-0.500.511.5100200300400Feed (mm/min)Force (N)Fy - d=0.03Fy - d=0.04Fy - d=0.05 Cuttting force vs Feed-2-1.5-1-0.500.51100200300400Feed (mm/min)Force (N)Fx-d=0.03Fx - d=0.04Fx - d=0.05
Figure 6. Cutting forces on a 0.02 in diameter cutter at 30,000 rpm
Figure 7. Burring of the sample for a 0.02 in diameter endmill at 30,000 rpm. 8
4. CONCLUSIONS
This paper presents the development of a low cost micro-machine (LCMM). The first machine was built and has been undergoing tests for the last year. Two improved machines have just been built as part of this ongoing project and are due to be tested in summer 2009. Tests on this machine have indicated high levels of accuracy can still be maintained without closed loop control. The majority of the components used to build this machine are off-the-shelf components that are inexpensive and easy to install. The CAM software is easy to learn and use, and provides a very straight forward GUI. A proposed micromachining curriculum has been presented. This, together with the LCMM, offers a very affordable laboratory based curriculum to train engineers in this new area of manufacturing. Preliminary machinability studies have been carried out to determine optimum cutting conditions for stainless steel on this micro-machine. More testing is being done to generate machining data for different materials and cutting conditions.
Acknowledgement
The authors would like to thank the U.S. Army, Tank and Automotive Command (TACOM) for funding this project.
References
[1] WTEC “Assessment of International Research and Development in Micromanufacturing," World Technology and Evaluation Center, Inc., 2005, online http://www.wtec.org/micromfg/report/Micro-report.pdf
[2] X. Liu et al., “Cutting mechanisms and their influence on dynamic forces, vibrations and stability in micro-endmilling,” Proc. IMECE04 ASME Mechanical Engineering Congress, Anaheim, CA, Nov. 13 – 20, 2004. IMECE2004-62416,
[3] I. Tansel et al, “Micro-end-milling – I, Wear and breakage,” Int. J. Mach. Tool Manuf., 38, 1998, pp. 1419 – 1436.
[4] W.Y Bao and I.N. Tansel, “Modeling micro-end-milling operations. Part I: analytical cutting force model,” Int. J. Mach. Tool Manuf., 40, 2000, pp. 2155 – 2173.
[5] K. Liu and S. Melkote, “Material strengthening mechanisms and their contribution to size effect in micro-cutting,” Trans. ASME, 128, 2006, pp. 730 – 738.
[6] K. Liu, K. and S. Melkote, “Effect of plastic side flow on surface roughness in micro-turning process,” Int. J. Mach. Tool Manuf., 46, 2006, pp. 1778 – 1785.
[7] M. Rahman et al., “Micro milling of pure copper,” J. Mat. Proc. Tech., 116, 2001, pp. 39 – 43.
[8] T. Childs et al, “Metal machining – Theory and applications,” Arnold publishers, 1999.
[9] C. Mirman et al, “Development of a micromachining support platform,” Proc. 9th Annual IJME-INTERTECH Conf., Kean University, New York, NY, October 19-21, 2006.
[10] J. Tlusty and P. Macneil, “Dynamics of cutting forces in end milling,” Annals CIRP, 24, 1975, pp. 21 – 25.
[11] U. Engel and R. Eckstein, “Microforming – from basic research to its realization,” J. Mat. Proc. Tech., 125-126, 2002, pp. 35 – 44.
[12] M. Geiger et al, “Microforming,” Annals CIRP, 50(2), 2001, pp. 445 – 462.
[13] A. Otieno et al, “Image and wear analysis of micro-tools using machine vision,” Proc. 9th Annual IJME-INTERTECH Conf., Kean University, New York, NY, October 19-21, 2006.
9][/QUOTE]

Hi Louie

Really, really appreciated. A URL would also be appreciated if possible so we can see the pictures and graphs.

What I get from this is
* Micromachining does not have to cost $1M
* Stepper motors and screw-threads can work (especially plastic nuts)
* Spindle runout is crucial
* Cutting edges are also crucial
* Any backlash or chatter will kill the tiny tools involved.

But reading between the lines, I also get that 'micro-machining' does NOT equal 'ultra-precision machining'. That is, extrapolating from what dimensions you can machine to in a 100 mm work volume to what accuracies you might get over a 1 metre work volume is NOT realistic. Fair enough: they are two quite different subjects.

In more detail - it was not clear whether they were running Mach3 with an external motion control device such as the ESS. I got the impression they did use one for Flash-Cut. If not, that would make it an unequal comparison. Not complaining, just observing.

Also, it was not clear whether they had Mach3 set for continuous motion at the corners or absolute stop. The way Mach3 missed some fine details suggests that they had it set for continuous motion, which would allow mach3 to round off very fine detail. This may imply that the authors did not really have enough detailed knowledge of CNC machining, at least with Mach3.

Despite all that - a most interesting paper!

Cheers
Roger

Quote:

---

Originally Posted by RCaffin

Hi Louie

Really, really appreciated. A URL would also be appreciated if possible so we can see the pictures and graphs.

What I get from this is
* Micromachining does not have to cost $1M
* Stepper motors and screw-threads can work (especially plastic nuts)
* Spindle runout is crucial
* Cutting edges are also crucial
* Any backlash or chatter will kill the tiny tools involved.

But reading between the lines, I also get that 'micro-machining' does NOT equal 'ultra-precision machining'. That is, extrapolating from what dimensions you can machine to in a 100 mm work volume to what accuracies you might get over a 1 metre work volume is NOT realistic. Fair enough: they are two quite different subjects.

In more detail - it was not clear whether they were running Mach3 with an external motion control device such as the ESS. I got the impression they did use one for Flash-Cut. If not, that would make it an unequal comparison. Not complaining, just observing.

Also, it was not clear whether they had Mach3 set for continuous motion at the corners or absolute stop. The way Mach3 missed some fine details suggests that they had it set for continuous motion, which would allow mach3 to round off very fine detail. This may imply that the authors did not really have enough detailed knowledge of CNC machining, at least with Mach3.

Despite all that - a most interesting paper!

Cheers
Roger

---

I don't have it, but I do have the original PDF; if you PM me I can email it to you... But yes, accuracy, precision, and resolution are separate animals. The point of the article to me is the level of precision for the cost involved. Even then it's still well above what the average "hobbyist" spends on their machine...

Hi Louie

The offer is much appreciated. But I managed to find a source and get the full PDF after a bit of searching. Too many of the 'portals' want your grandmother's maiden name and SSN (or your money), so it took a bit of hunting.

Amusing: their micro-machine looks just like a scaled-down 'XYZ', where XYZ stands for any of the cheap 3-axis mills now available from around the world. It figures.

Cheers
Roger

Edit later:
Figure 7 shows 'Burring of the sample for a 0.02" end mill at 30k RPM'
I am not convinced. To me that looks like either vibration (less likely) or a chipped end mill. I mean, that's 0.5 mm full scale: no way is that a reasonable result.
On the other hand, I was getting something like that myself at one stage, and examination of the tip of the end mill showed it was chipped. Swarf was being dragged under the tip because of the chipping. So I resharpened the tip on a diamond wheel and repeated the cut. It came out perfect, like glass.
I am sure the authors are great on the theory, but whether they have the practical experience is another matter.

We have 7 Moore Nanotechs and half a dozen precitechs. Yes, they can hold .5 micron. Makino mori, whatever would not hold 1 micron accuracy, maybe in position, but no way in circular or contour. 1 Micron on a full size machine, your looking at Yasda, Mitsui Seiki, Sip, etc. And no, Mitutoyo is only shop floor accuracy. not bad products, but they have nothing to check sub micron. You will need Federal/Mahr, Lyon, Tessa etc. in either LVDT or CAP gauges for manual measuring. Even high end Leitz CMM's (1M+) have a hard time sub micron.

Can you share the link where you did finally find it?

Um ... does search again ... OK.
http://www.pdflibrary.org/pdf/low-co...ation-for.html

BUT: not the obvious big orange bars which pop up fast!

Wait until the page loading has finished - it takes a while, then go to the bottom left hand corner of the image of the PDF (don't scroll), to where you will find a small clickable orange icon labelled 'Download'. Click on that and save. It gave me the full PDF file.

Cheers
Roger

Hi all

I can understand that someone who has just spent $1M on a top of the line machining centre to get 1 micron accuracy might feel the need to justify their expenditure. But that should not blind anyone to recent advances in micro-positioning and micro-machining.

In the latest issue of Laser Focus World I see an ad from Aerotech for their FiberAlign series of motorised movements (p8). These have a step size of one nanometre. That is not a typo: not one micrometre, but one nanometer. There are 1000 nm to a um. XYZ configurations are available, along with motors and software. There are also tilt axes. Range - not given, but of the order of a few centimeters at least.

On the next page (p9) there is an article about laser machining. Some German researchers have been using a Trumpf YAG disk laser to machine an ablation depth of 55 um with a surface accuracy of <1 um. Mind you, they also reported that if you get the speed/feed combinations wrong, this can drift out to an ~10 um surface. So what's new about the need for tuning?

Let's face it, one micron is rather 'yesterday'.

Cheers
Roger
ducks for cover ...

we developed machines in the 70s that would position in nanometers. again positioning and accuracy are not the same thing. just by reading an article somewhere doesnt make it true. do you have any idea whats involved in just temperature control alone? just 1/100 of a degree will completely change size. the lathe im working on now is one micro inch resolution, during laser checking of positional accuracy the bay went from 68deg to 68.5 degrees and the machine grew 15 micro inch. this is a very heavy very high end machine. then you need to measure flatness squareness, pitch and yaw etc. it only takes one of these items to be out 1 micron, and that machine is no longer accurate to 1um.

Hi Under

I agree with everything you are saying about thermal expansion etc. Over 1 m it can be hugely significant.

The difference is the realisation that for many applications, a working volume of 100 mm is 'huge'. There are machines and companies which are making parts which are well under 10 mm in their biggest dimension. I have seen parts that are under 1 mm. That makes for a whole new ball game - and the market for this stuff is growing fast.

Cheers
Roger

I hear you RCaffin. Most of the parts we manufacture are in the 0.01mm to <5mm range. On the 350FG in the lab we have held tolerances of ±0.1um and underthetire is correct, in this regime temperature, not to mention tooling, are extremely critical. Who cares if your machine can hold 0.3um form if your tools aren't even round to a micron? That said, both the temperature in the machine enclosure and the measuring area must be known and controlled very precisely. Although it is nice being able to pull your part back in tolerance simply by breathing on it for a few seconds :D JK. All of our measurement systems are non-contact and most diameters are measured on laser micrometers from either Z-Mike or Keyence. As for the original post, I don't know what your budget is but the highest resolution linear scales of which I am aware are the Laser holographic linear scales used by Moore Nanotech. I am not sure but I believe they are made by Sony. *edit* also, we are disposing of two older precitech lathes in the near future. I am not sure what will happen to them. The laser scales on them are still good. I am not sure if they might be of use to you but I am willing to ask the uppers what the plan is for them. One machine is company owned the other is the property of the DOE. I am sure the DOE machine is off the table but I can probably ask about the other.

Hi MHoppe

OK, that's out of my class!

I have a machining volume of about 250x200x200 mm, which I do use, but I can't get down to micron accuracy. I can sometimes (with care) zero to about 5 microns, and certainly to 10 microns, and I can probe to significantly better than that. But the difference between a hot summer and a cold winter - no idea!

Cheers
Roger

Well, to be honest most of the tolerances requested are in the ~3-5um range. Which sounds tight but if you have ever used these machines you would quickly realize that you can hold that without breaking a sweat, especially on the small parts we manufacture. The yield on the 0.1um part was less than 100% lets say and that is not typical. Was an R&D part and the customer only required a few of them. Measuring to that accuracy was more of a pain than machining to it, and it wasn't on all dimensions just a small sinewave feature around a sphere. I am not sure what your equipment is but Id wager holding 5-10um on it is actually more of a feat than holding 2 or 3 on this machine ;) Just don't tell my boss that hehe

Quote:

---

I am not sure what your equipment is but Id wager holding 5-10um on it is actually more of a feat than holding 2 or 3 on this machine

---

Probably :-)
In some ways it is massively over-engineered. Instead of roller bearings on aluminium angle on MDF, or even 16 mm rails, the primary X axis uses 40 mm hardened and ground round steel rods with massive linear ball bearings. And that is for a 250 mm travel.
On the other hand, it's a bit hard to knock it out of alignment ...
Suits me.

Cheers
Roger

Quote:

---

Originally Posted by Mhoppe

I hear you RCaffin. Most of the parts we manufacture are in the 0.01mm to <5mm range. On the 350FG in the lab we have held tolerances of ±0.1um and underthetire is correct, in this regime temperature, not to mention tooling, are extremely critical. Who cares if your machine can hold 0.3um form if your tools aren't even round to a micron? That said, both the temperature in the machine enclosure and the measuring area must be known and controlled very precisely. Although it is nice being able to pull your part back in tolerance simply by breathing on it for a few seconds :D JK. All of our measurement systems are non-contact and most diameters are measured on laser micrometers from either Z-Mike or Keyence. As for the original post, I don't know what your budget is but the highest resolution linear scales of which I am aware are the Laser holographic linear scales used by Moore Nanotech. I am not sure but I believe they are made by Sony. *edit* also, we are disposing of two older precitech lathes in the near future. I am not sure what will happen to them. The laser scales on them are still good. I am not sure if they might be of use to you but I am willing to ask the uppers what the plan is for them. One machine is company owned the other is the property of the DOE. I am sure the DOE machine is off the table but I can probably ask about the other.

---

Trust me, the DOE machine will go back to the doe lab, then get sold at auction eventually after I steal parts to keep the other precitechs or Moore's running ;).

I guess if the DOE machine is missing too many parts, it may go for a fairly low price at auction then?
:)

Cheers
Roger

Usually yes. I've seen stuff go dirt cheap, and garbage go too high.

They are older Precitech Optimum 2000's. I am not sure what they might go for but Precitech no longer supports them. They still do their best to service them whenever they are out doing PM on our other machines. The lathes can still hold a few microns tolerance and are capable of sub 10nm RMS surface finish but they definitely show their age when sitting next to current offerings from either Moore or Precitech. Most in industry that use these type of machines are machining optics. We use them a bit differently. I am not sure what the market for such a machine would be. Ill hang it up here as I don't want to hijack the thread anymore than I have already :banana: <--- the song should embed with this graphic

This guy built a cnc lathe/grinder accurate to 1um...

https://www.youtube.com/watch?v=sFrVdoOhu1Q

That guy did a nice build. Nice thermally stable granite bed.