Frequently Asked Questions
Installation & Licensing
Remember to restart your computer after installing!
After undocking a laptop from the docking station, older versions of the Sentinel driver cannot find the hardware key that protects your CUTPRO/Shop-Pro license.
When your laptop is connected to your docking station, you can startup CUTPRO/Shop-Pro, using a license that is protected by a hardware key (dongle). However, when you use/start CUTPRO/Shop-Pro when your laptop is not connected to the docking station, CUTPRO/Shop-Pro can no longer find the hardware key. It is even possible that this issue still appears in the most recent version of the Sentinel driver, in which SafeNet has addressed this issue.
CUTPRO/Shop-Pro uses the Sentinel driver to locate the hardware key. This issue is known by SafeNet, the developers of the Sentinel driver and they have solved it in the latest version. However, if you have installed previous driver versions, these installations can still cause the described problem. Uninstalling older driver versions, does not always solve the problem.
To make sure all old driver versions are completely removed from your machine, you should run the SSD Cleanup program from SafeNet/Gemalto. This program removes all Sentinel entries from your machine.
Download the SSD Cleanup utility here.
After you have run the cleanup program, please download and install a Sentinel driver version with number 7.4.2 or higher.
Download the Sentinel Driver here
This error can be solved by updating the sentinel drivers and runtime here.
This can happen if the machine dynamics change at different speeds. When the machine is tap tested, it is assumed that the natural frequencies and stiffness of the machine remain the same at all speeds. However, the natural frequencies and stiffness of some spindles (but not all) change as the speed increases (especially after 10,000 rev/min). The variation may be due to:
– Automatic changes in the bearing preload mechanism;
– Thermal expansion of the spindle;
– Geared spindles (low speed – high torque, high speed – low torque stages).
It is impossible to predict the changes in the spindle dynamics with tap testing at zero speed. One has to measure the spindle at different speeds, which is not practical in the shop. The practical method is as follows:
– If this event occurs, measure the noise with MALDAQ. Obtain chatter frequency using with FFT or automated MALDAQ function.
– Shift the lobes manually to coincide with the speeds recommended by MALDAQ.
You can store the stability lobe as an ASCII output file using export function in CUTPRO. The text file can be used in any other program such as MATLAB and EXCEL.
The stability is calculated both using our unique frequency domain stability law and numerical simulation. The readers need to study the theory from our articles and textbook “Manufacturing Automation, by Y. Altintas, Cambridge University Press”
An analytical solution is done in the frequency domain, and it is fast, but also linear. When the cutting coefficient changes or you use exotic tools which deviate from cylindrical end mills; very high spindle speed with a very small radial depth of cut; serrated tools for roughing, we recommend the following steps. 1) Run the analytical stability solution which is fast (and most accurate linear theory around in frequency domain). 2) Select a speed region where you have an interest, and run the time domain stability lobes only for the limited region. You can push the limits of your cycle time reduction more accurately this way, if you need to. Time domain simulations are similar to trial cutting tests, and takes considerable time.
Note: You don’t need to use modal analysis module when you are interested only in FRF at the tool tip. CUTPRO Milling has an integrated modal parameter identification at the tool tip.
If you select modal parameters, and click new, a curve fitting – modal identification window will pop up within the machine Tool Structure and Workpiece Structure menu pages in milling option.
Multiple analytical stability lobes provide one curve for each radial depth of cut, and simulate several of cases and plot them together. It is identical to single analytical stability lobes except with multiple numbers of radial depths of cuts. It is also possible to run multiple cutting conditions by selecting “Batch” option in cutting conditions menu page.
Minimum spindle speed depends on the frequency content of your FRF. If you have high frequency modes, the minimum spindle speed in the analytical stability lobes will be higher. Please remove these modes in modal analysis, and try to run again.
You can also try time domain stability simulation. It is slower, but it will give accurate results at low rpms.
The FRF (tap test) of the spindles measured at zero speed change as the spindle speed increases. The sources of variations are as follows:
a) Spindle shaft and ball bearings thermally expand due to friction energy which increases as the speed increases. The thermally expand parts in the spindle system leads to increases in the bearing stiffness, which shifts the natural frequencies upwards. As a result, the stability lobes shift towards higher speeds. It is best to tap test the machine immediately after a 20-30 minute warm up to avoid any effects from cooling off.
b) On the other hand as the speed increases, the centrifugal forces and gyroscopic effects loosens the bearing stiffness. As a result, the natural frequency drops so as the stability lobes shift towards lower speeds. It is possible to calculate these effects using special bearing software such as SpindlePro, but simulations give some estimates which are not exactly the same as real measurement conditions. In addition to measuring FRF after a warm up period, it is better to measure the chatter frequency with a microphone, and enter the value as the natural frequency of the machine in CUTPRO simulations. This will improve the generation of chatter stabilities more accurately and effectively.
Frequency Response (FRF)
It is not trivial to measure long, slender end mills with an instrumented hammer. If you use hard steel tip, the hammer may bounce on the tool with several contacts, which leads to unacceptable Frequency Response Function (FRF). If you use soft tip, then you may not excite the natural frequency of the tool. Harder the tip is, wider the frequency range of excitation is but at the expense of bouncing. Softer the tool tip is, easier to create a single contact with the tool but at the expense of longer contact time hence lower frequency range. Therefore, you may not be measure the machine correctly.
First try some tricks:
*Put a layer of a scotch tape on the hammer tip, and see whether it reduces the bouncing.
*Try bronze or Aluminium tip.
*Use the smallest hammer you have.
If you are still unhappy with the bouncing or frequency spectrum range, do the following in MALTF:
1- Put the accelerometer at the tool tip, but hit the tool close to the chuck where it is more rigid. The bouncing will be reduced. Save the measurement as Cross FRF (i.e. xxx12.frf) using export command.
2- Move the accelerometer close to the chuck where you hit with the hammer. Hit the tool at the same location as in step 1. Save the measurement as direct FRF (xxxH22.frf) using export command.
3- Go to modal analysis module of CUTPRO. Select file -> select FRF files and click on Flexible Tool FRF option. Load both measurements collected in steps 1 and 2, and apply modal analysis, and curve fit the selected modes using any of the measurements.
4- Once you are happy about the curve fitting, click on H11 on the top right corner of Modal analysis. CUTPRO will estimate the FRF at the tool tip and display the results. You can “Save Parameters” under a file name which can be automatically used in milling/boring/turning simulations.
The program will estimate the correct FRF at the tool tip without having to hit the tool at its tip. It is not as accurate as measuring at the tool tip, but a lot better than not being able to measure it at all.
MAL receives significant number of questions about the tap testing and modal parameter identification of the machine. We prepared the following information for our users.
Accurate prediction of sweet spots or stability lobes of the machine requires
1- Cutting conditions (radial engagement of the cutter): This is trivial and entered by the user. CUTPRO allows changing cutting conditions along the cut, has many options for the user. Shop-Pro allows up, down or face milling configurations.
2-Geometric model of the cutter (cutting edge geometry, helix, rake, run-out, variation of tool geometry from one tooth to next, pitch angles, approach angle,..) CUTPRO has the most sophisticated but easy to use cutter geometry entry. You can pretty much enter any cutter geometry you want, ranging from simple helical end mills, to variable pitch-variable helix, serrated, indexed, arbitrarily distributed inserts or user defined complex cutters.
3- Work material properties during machining, cutting coefficients We provide more than 250 work materials in the database via material selection wizard. It checks the database and offers you the most accurate and up to date cutting coefficient for that material. Extensively used materials in Aerospace industry, such as AL6061, 7050, 7075 and titanium, Inconel, Wasp Alloy are prepared by our research laboratory and they are most accurate worldwide. Both CUTPRO and Shop-Pro share the same material database.
4- Frequency Response of the Machine and maybe workpiece as well if it is flexible. This is the most challenging task for most users, since it is time consuming and not easy to measure if the tool is long with small diameter. A flexible tool technique was developed by us as explained in the previous message.
We went further with our MALTF module. The tap testing is automated with human voice feedback. You either hear it from the speaker of the laptop, or headphones and get in the machine and tap test it without having to push buttons at each hit. The fuzzy logic based expert system automatically rejects-accepts the measurements, and tells you how it is progressing with a human sound. It also displays reliable measurement frequency range by checking the signatures of the data mathematically. MALTF has multiple measurement channels in CUTPRO, and single input/single output in Shop-Pro.
Also, we designed a three-level database design for our users. You can identify the spindles, and store them in a database. You can define the holder geometry, and cutter stick out and diameter by providing basic dimensions. MALTF will automatically predict FRF at the tool tip without having to do tap test. Alternatively, you can identify the spindle and tool holder assembly, and store it in the data base, and add only tool diameter and tool. MALTF will recommend you the proper tool length so that you can have a sweet spot (stability pocket) at a desired speed.
These will not be as accurate as direct tap testing on each tool, but our tests showed that the results are amazingly practical enough with shrink fit tools. All the algorithms are based on UBC’s five yearlong investigation on receptance coupling research which was supported by Mitsubishi Materials, Sandvik Coromant and Boeing. The tool optimization is also based on UBC’s chatter theory and optimization algorithm.
The ideal coherence value is unity (=1) for the frequency range of interest. However, it is always useful to look at the power spectrum of impact force and accelerometer. The impact force spectrum must not be close to zero if there is any mode (peak) in vibration spectrum. CUTPRO MALTF indicates the unreliable measurement range automatically.
If you cannot obtain good coherence and wide bad impact force spectrum, use flexible tool option of MALTF. The alternative is to buy a very light miniature hammer from PCB. However, such hammers are difficult to use.
It is not trivial to measure long, slender end mills with an instrumented hammer. For example, an end mill with 0.25 inch (6.35mm) diameter and overhang of 2 inch (50.8mm). The natural frequencies will range from 500 Hz to 5000Hz. Hammers with 500 N or higher force delivery with plastic tips will have a spectrum of 2000Hz, and will not be able to excite modes between 2000Hz and 5000 Hz. The following techniques can be tried in measuring the FRF of such a slender tool.
Let’s consider that you have a small Kistler, PCB, Dytran or similar hammer with 500N or 2000N force range. If you use hard steel tip, the hammer may bounce on the tool with several contacts, which leads to unacceptable Frequency Response Function (FRF) measurement. CUTPRO MALTF will inform you automatically that the measurement is not accurate. If you use a soft tip, then you may not excite the natural frequencies of the tool. Harder the tip is, wider the frequency range of excitation will be, but at the expense of bouncing (i.e. multiple impacts). Softer the tool tip is, easier to create a single contact with the tool, but at the expense of longer contact time which will lower the frequency range of the hammer. Therefore, you may not be able to measure the machine correctly.
First try some tricks:
*Put a layer of a scotch tape on the hammer tip, and see whether it reduces the bouncing.
*Try bronze or Aluminium tip.
*Use the smallest hammer you have, i.e. PCB’s or Dytran’s miniature hammer.
If you are still unhappy with the bouncing or frequency spectrum range, do the following
1- Put the accelerometer at the tool tip, but hit the tool close to the chuck where it is more rigid. The bouncing will be reduced. Save the measurement as Cross FRF (i.e. frf_12). Make sure you measure the distance between the impact and vibration measurement point.
2- Move the accelerometer close to the chuck where you hit with the hammer. Hit the tool at the same location as in step 1. Save the measurement as direct FRF (frf_11).
3- Go to modal analysis module of CUTPRO. Select file -> select f FRF files and click on Flexible Tool FRF. Load both measurements collected in steps 1 and 2, and apply modal analysis. The program will estimate the correct FRF at the tool tip without having to hit the tool at its tip. It is not as accurate as measuring at the tool tip, but a lot better not being able to measure it at all.
An alternative method is to use two hammers:
Use PCB’s or Dytran’s miniature hammer to measure the high frequency range (2000Hz and up), and use larger hammer for lower frequency range. Apply modal analysis to the two measurements separately, and enter them as modal parameters in CUTPRO. You need to train your hands not to cause multiple impacts with the miniature hammer.
Troubleshooting the source of vibrations during machining
The following procedure can be used to identify the source of the flexibility and weak joint on the machine/tool – fixture/workpiece set up:
1) Conduct a chatter test and measure the vibration using microphone or accelerometer. Use a new CutPro Data Acquisition Cutting Test workspace and enable the continuous data logging mode.
2) Apply FFT (Fast Fourier Transform) using Data Acquisition’s FFT button (located on the plot toolbar) where the chatter is evident with large amplitudes. Identify the chatter frequency and record it.
3) Conduct impact hammer tests on the cutting tool mounted on the spindle using CutPro Tap Testing. Check the natural frequencies from the Magnitude of the measurement. If the natural frequency of the tool measurement is very close to chatter frequency, the chatter is caused by the machine tool or cutting tool assembly.
4) Conduct hammer tests on the workpiece mounted on the fixture using CutPro Tap Testing. Check the natural frequencies from the Magnitude of the measurement. If the natural frequency of the workpiece measurement is very close to chatter frequency, the chatter is caused by the workpiece, fixture or clamping.
Identification of flexible joint that causes chatter
If the tool side is causing chatter:
(You can use 1D modal analysis and look at Lathe example in CUTPRO’s Example folder: C:\Program Files (x86)\CutPro\Examples\Modal Analysis\1D Modal Analysis)
1) Mark points along the tool – fixture – spindle (i.e. tool tip-1, middle of the stick out-2, close to the tool holder-3, tool holder but close to the tool clamping-4, tool holder close to the spindle flange–5, spindle nose-6). Stick the accelerometer at the tooltip (Point 1) and keep it at point 1, and apply impact at point 1 (FRF Measurement H11), point 2, …, point 6.
2) Load the measurements into CutPro Modal Analysis Module; select the high peaks (natural frequencies), especially the one which is close to the chatter frequency.
3) Display the mode shape which corresponds to the chatter frequency. Mode shape will show you how the tool-holder-assembly is behaving during chatter.
• If the mode shape shows that the deflection is a continuous from the spindle nose-holder-tool, the whole spindle is vibrating.
• If the tool holder is sharply deflecting away from the spindle nose, the holder-spindle interface is either worn or connection is weak due to dust or chip stuck in the spindle taper. If not, the holder is weak.
• If the tool is deflecting away from the holder-joint sharply, the tool-holder joint is not stiff enough. Tighten the holder or use shrink fit. If the tool is deflecting with the curved shape, the tool is too long and flexible.
If the workpiece side is chattering
(You can use 2D modal analysis example in the CUTPRO Example folder C:\Program Files (x86)\CutPro\Examples\Modal Analysis\2D Modal Analysis)
1) Mark points on the XY plane of the workpiece – fixture set up. Stick the accelerometer at one point (1) which seems to be most flexible.
2) Apply impact at each point while keeping the accelerometer at the same point (Point 1), and save the measurements.
3) Load the measurements in Modal Analysis (2D analysis). Check the mode shape which has the closest natural frequency with the chatter frequency. Mode shape will show you the clamping or workpiece locations which deflect most. If the deflections occur at the workpiece – fixture contact points, place the clamps at the points where you see the largest mode shape deflections.
Difficult to Cut Materials
The stability lobes in low speed milling require special attention to avoid using incorrect speeds and depths of cuts. The scenarios are divided into two categories: Low speed milling with small diameter end mills and low speed milling with large diameter cutters.
Low speed milling with small diameter end mills (d < 30mm): The natural frequency of such tools may change from 2000Hz to 4000Hz depending on the diameter and stick out length. Also, the tool holder may have frequencies ranging from 1500Hz to 2500Hz again depending on its diameter and length. Such high natural frequencies create usable stability lobes above 7500 rev/min for four fluted tools. If the speed is selected under these values, the contact between the flank face of the tool and wavy surface finish increases, and damps out the chatter. This is called “process damping” and can increase with the tool wear and smaller clearance angles ground on the tools. The users of machining software will obtain small depths of cut from simulations, but they will notice that they can cut 2-3 times deeper in reality due to this process damping mechanism. If chatter occurs, not the tool and tool holder but larger masses of the machine tool will cause the vibrations. It is recommended that the user sets the spindle speed at the desired level, and keep increasing the depth of cut from the predicted value until it chatters. The vibrations must be recorded by MALDAQ, and FFT needs to be applied to detect the structural modal frequency which is causing the chatter. This frequency must be compared against the natural frequencies measured using MALTF. Typically the spindle shaft, a column, or other large fixtures have lower natural frequencies (400Hz to 1000Hz) may be dominant at this speed. The user has to ignore the high natural frequencies of tool and holder (>1500) in simulating the stability lobes. MAL Inc. is developing new theories to account the process damping effects but this is rather a very problematic scientific problem to use in practice easily.
Low speed milling with large diameter cutters (>50mm): The same procedure needs to be used like in low speed milling with small diameter end mills, except that the user must be very careful in measuring the machine dynamics. If the cutter has up to 50-60mm diameter, one can use a hammer with 5000N force capacity and accelerometer which has above 50 Hz sensitivity. If the cutter is larger, one has to use a large sledge hammer which can deliver up to 20,000N force. The accelerometer needs to be sensitive from 10Hz upward (i.e. large accelerometer). Typically spindle shaft, housing, column table and parts of the set up with large masses, which have low natural frequencies (from 20Hz to 400hz) will cause the vibrations. For example, if the machine is chattering at 300 rev/min with 8 teeth, the chatter may occur between 10 Hz to 50hz, which will be due to spindle housing or whole column. In order to detect the mode causing the problem, the user needs to use MALDAQ to record the sound or accelerometer data, apply FFT and match the vibration frequency with MALTF (hammer) measurement. You can penetrate into the stability lobe by selecting only those low modes in the stability analysis.
MAL Inc. technical staff will be happy to assist our licensed customers by sending CUTPRO project files. The application will be kept confidential.
The spindle speed may be very low when you use cutters with large diameters. The useful stability lobe will depend on which natural frequency is causing the chatter vibrations.
We recommend the following:
- 1- Tap test the cutter on the machine. If the workpiece is flexible, tap test it as well.
- 2- Using MALDAQ, measure the vibrations during machining using an accelerometer attached to the spindle headstock or with a microphone. Take FFT, and check the frequency corresponding the highest peak. There may be some high peaks at the spindle speed (rev/sec) and tooth passing frequency (no. of teeth times spindle speed), and their integer multiples. These are forced vibrations in most cases, and you can neglect them. The chatter will occur close to one of the natural frequencies measured with the tap test.
- 3- If you use a large diameter cutter and low speed, the chatter will cause by low frequency modes of the machine (spindle, spindle housing). If the diameter is small and speed is also low, the chatter will be caused by the higher modes of the spindle but hardly by the high frequency tool modes except at excessive depth of cuts.
- 4- Apply curve fitting (Modal Analysis) to tap test but do not include modes which have values 1.5 times higher than the chatter frequency. Disregard the higher frequency range. Alternatively, using CUTPR milling module, you can use raw tap test data and indicate only the low frequency range to be read from the file.
- 5- If the chatter frequency and tooth passing frequency is close, or if the chatter frequency/tooth passing frequency is less than 5, you will see a useful chatter stability pocket by following the recommendation given in step 4.
- 6- If the chatter/tooth passing frequency is higher than 5, you can cut deeper than the lobes indicate due to process damping (ploughing between the flank and finish surface). If you want to increase the chatter free depth of cut a lot further, you need to design a variable pitch cutter using CUTPRO. If you want a slight increase in the depth of cut, you may wish to increase helix or approach angle of the cutter using CUTPRO. You can see how the change in cutter geometry affects the stability using CUTPRO simulations prior to manufacturing the cutter.
The limiting factor in milling titanium parts is tool temperature. Because of this limitation, spindle speeds are dramatically reduced. In this area of low spindle speeds, there’s no need for stability lobes because the distance between two peaks is so small and the difference between top and bottom of the peak as well. From time to time, the maximum depth of cut is 1mm when working on full diameter.
How can we increase radial and axial depths of cut dramatically for titanium parts in order to get high material removal rates?
While checking a range of low spindle speeds (500-800 rpm), the maximum depth of cut dropped from 1mm to 0.2mm (e.g. from 650rpm)at a certain speed in this range. How is this possible? I kept the depth of cut at 1mm (stable condition up to 650rpm) with a spindle speed of 800rpm (at this point 0.2mm was the maximum stable depth of cut) and it worked fine. Where’s the drop in depth of cut coming from and where we able to cut stable while the lobes indicated an unstable condition?
Milling of Titanium and Inconel at low speed is always a challenge, when the diameter of the cutter is small as well as the spindle speed. After the lobe number 7-8, the accuracy of chatter stability diminishes due to friction between the flanks and finish surface with vibration waves. This is called process damping and not possible to predict mathematically. It very much depends on the edge radius, clearance angle, speed, vibration frequency and amplitude, and work material. Researchers starting from Aachen and Leuven in 1960s have been unable to model this phenomenon mathematically.
The alternative is to use the knowledge in absolute stability (bottom border of lobes) and start testing the tool at that speed by increasing the depth of cut until chatter occurs. The chatter free depth of cut will increase as the speed is reduced after 7-8th lobe depending on the cutter geometry and vibration frequency. The productivity is improved only by controlling the temperature which depends on the radial depth of cut, feed and speed. First, you need to simulate temperature up to 90 degree immersion by dividing the radial depth of cut at 5 equal distances. Keep the cutting speed maximum same so as the feed, and record the tool life. Increase the cutting speed as you reduce the radial depth of cut by 20%. Decrease the radial depth by 20% and increase the speed and feed by the same amount). After simulations, try selected tests on the machine; make sure there is no chatter during cutting tests. Cut the material until the tool wear reaches to a limit, by measuring the wear at 4-5 minutes intervals. You will get a table of results which can be used in selecting proper cutting conditions in the factory.
Unfortunately, there is no magic way to estimate the tool life.
Note: The stability lobes at very low speed may be incorrectly calculated due to numerical problem caused by the resolution of FRF frequency. You need to ignore it.
Aluminium is a prime candidate to be machined at high surface speeds. Typically, bonding materials in carbide tools such as Cobalt, can withstand temperatures up to 850 to 1000 Celsius depending on the grain size and chemical compositions. However, Aluminium melts around 600 Celsius; hence tool wear is not a prime issue in machining Aluminium alloys. Aluminium can be machined as fast as the machine tool spindle can rotate as long as the spindle has enough power and dynamic stiffness to operate without chatter vibrations. Therefore, one needs to predict chatter stability lobes, or sweet spots as the operators call it, while avoiding the violation of power limit of the machine tool.
You need to know the Dynamic Characteristics of the machine tool at the tool tip which can be measured using any Fourier Analyzer or CUTPRO MALTF module. The second important parameter is the material data, and CUTPRO has extensive material library including Aluminium alloys. The rest of the data is trivial such as approximate dimensions and angles of the tool, radial immersion conditions. If the part is flexible, you can include its dynamic measurements as well to CUTPRO. Milling Module of the CUTPRO predicts chatter stability lobes most accurately, and you can select a certain depth of cut under the lobe which does not lead to chatter and feed rate. Milling module predicts everything about the process at this operating point, including forces, torque, power, vibrations at the tooltip, dimensional surface errors, chip thickness history, animation of the tool and surface motion.
When the tool and/or workpiece are flexible in peripheral milling, relative displacements occur between the two are imprinted on the surface finish, resulting in dimensional form or surface location errors. The errors are generated when the cutting edge is perpendicular to the finish surface. The displacements of the edge after leaving the normal surface are removed by the following tooth; hence they don’t affect the tolerance of the part. The helical edges leave periodic but saw teeth type of waves on the surface.
If the spindle rotates at low speeds, the form errors are generated mainly by the static deflection of the tool. If the spindle speed is high, the system either experiences forced (resonance) or chatter vibrations. Chatter leaves rough surface finish, while forced vibration leaves positive (down milling) or negative (up milling) offset on the surface finish. If such problems occur, the user needs to either avoid matching the tooth passing frequency (spindle speed times no of teeth) with the natural frequency of the system, or use zero helix and radial depth of cut which contains less than two teeth.
CUTPRO has been predicting the entire surface form since its inception in 1989. Unlike approximate solutions, CUTPRO milling module rotates/vibrates the tool while moving – vibrating the workpiece, which is the true kinematic model of milling. The chips are calculated by subtracting the digitized surface generated by one tooth by the previously generated surface, hence the surface errors, chips, forces, torque, power, vibration amplitudes are all calculated simultaneously most accurately.
The cutting force coefficients of the material affect the stability directly. The cutting force coefficients are related to the yield strength of the material, the friction coefficient between the tool and work material, rake face and cutting edge geometry, helix and approach angles. CUTPRO allows the creation of a fundamental orthogonal cutting database (shear stress, the friction coefficient of the chip-tool interface, and shear angle). CUTPRO automatically transforms these material properties to the oblique geometry of the tool using the laws of cutting mechanics. The theory can be studied from “Manufacturing Automation, by Y. Altintas, Cambridge University Press”.
CUTPRO allows you to define variable helix, serration, run out, variable pitch and arbitrary tool geometry, and you can see their effect on the process and stability.
CUTPRO Milling module allows you to enter both machine tool and workpiece flexibility exactly the same way. Similar to tap testing or defining the dynamic parameters of the tool attached to the spindle, you can measure – define the parameters of the workpiece in the “workpiece dynamics” menu. The flexibilities of both are included in the stability calculation.
The end mills need to be classified into three categories: Regular end mills with a diameter greater than 0.5 inc (12.5mm), small end mills with a diameter range of 0.2inc (5mm) to 0.5inch (12.5mm), and micro end mills with a diameter less than 0.2 inch (5mm). CUTPRO can handle any size end mill including micro mills provided that the user can enter Frequency Response Function (FRF) and material cutting coefficients correctly. CUTPRO uses exact trochoidal kinematics of end milling (tool rotates/vibrates while the part is linearly fed, hence any size tool with any feed rate can be modelled accurately. It can even predict the feed marks left on the micro-milled surface).
The machining issues with end mills can be classified also as follows:
– Chatter vibrations; Solution requires stability lobe prediction.
– Forced vibrations which lead to tolerance violations, undercut/overcut; Solution requires prediction of force and vibration magnitudes at given speeds and cutting conditions.
– Static deflections due to bending of the tool: Solution requires prediction of static forces and tool stiffness.
– Shank breakage if large chip loads and depth of cuts are selected. Avoidance requires maximum force and bending load prediction which should not exceed the ultimate tensile strength of the tool at the clamping point.
All vibration problems require measurement or estimation of Frequency Response Function at the tool tip either using impact tests or Finite Element (FE) models. FE models give approximate results and need to be tuned with experiments. Impact tests give accurate results, but they are not trivial to do when the tool diameters are small.
There are few ways to measure /estimate the FRF of small end mill:
1- End mills over 0.5 inch: You can use any small hammer with 100lbf-400lbf force delivery, and a small accelerometer. If the diameter is less than 0.5inch and the L/D > 6, you may have a difficulty to avoid double hit with a hammer. CUTPRO Modal analysis has a “Flexible tool FRF prediction” algorithm which helps you to estimate FRF by hitting on the shank close to the chuck only, while the accelerometer is mounted at the tool tip and shank.
2- End mills 0.2 inch to 0.3 inch range is better measured with a miniature hammer from Dytran or PCB with a small accelerometer. This is a challenge, but once you gain some experience (few hours of trials in the beginning) you can handle the miniature hammers.
3- End mills with an extremely small diameter – Micro end mills: Mounting an accelerometer on them will lead to incorrect FRF measurements because of added mass. You need to measure the natural frequency/FRF with a non-contact capacitive/inductive/laser sensors, which is time-consuming. Alternatively, you can measure the noise during machining with a microphone and FFT will give you the chatter frequency, you can estimate the stiffness using FE model or just use (3EI/L^3) and throw a very small damping (0.005). The biggest difficulty in micro end mills is the estimation of material and cutting edge geometry dependent cutting force coefficients. The coefficients given in CUTPRO are for regular end mills, and they are less than what micro end mills generate due to severe friction and size effect caused by micro chip loads.
In summary, as long as you can measure/estimate the FRF and cutting force coefficients, CUTPRO can handle any size/shape end mill. Shop-Pro can handle only cylindrical end mills, provide stability lobes and approximate torque/power. However, Shop-Pro can also recommend you the best speeds to avoid both chatter and forced vibrations.
The best way is to keep the webs as stiff as possible during milling. The procedure to machine them is explained as follows
1- Tap test the cutter, and obtain chatter free depth of cut, speed and width of cut.
2- Select the spindle speed, depth and width of cut which does not cause chatter.
3- Machine the part layer by layer without leaving any material at any layer. This way, the webs will always be stiff, and the dominant flexibility will come from the cutter.
4- If the top part of the machined web touches the cutter due to forced vibrations, reduce the cutter diameter by 0.15-0.20 mm after two times of the stable depth of cut. You will avoid cutter rubbing against the thin wall and exciting it at tooth passing frequency.
If you need to go far deep when the web is already thin and you are not able to implement the procedure above, follow the process outlined below
1- Tap test the cutter from the tip to the web height along the cutter axis. Select points with 20mm distance.
2- Repeat the process at the thin web which corresponds to the cutter. The web will be stiff at the bottom of the web.
3- Store all measurements from MALTF and export them as FRF files.
4- Use Modal analysis and obtain mode shapes of both cutter and flexible web separately. Save the modal parameters separately.
5- In CUTPRO Milling Module, select the modal parameters of the cutter and modal parameters of the web for the workpiece structure.
6- You need to use time domain simulation to get chatter stability lobes. Therefore select a speed range which is practical. You can use first the cutter at the tip and part at its thin tip and use analytical stability lobe first to get an idea about good speeds. Use the time domain stability later.
7- After you select a good depth and width of cut, and the speed, run time domain simulation of the process to see the vibration amplitudes and frequency.
This is a more complicated solution and one needs to be well trained and knowledgeable in milling dynamics of thin webs.
The frequency range of the machine tools can be listed as follows:
10Hz- 200Hz: Machine tool column, bed, guides, portal frames, spindle housings, tomb-stones have large masses with low natural frequencies. You need to use large hammers which can deliver more than 5000N force. The accelerometer must also be large which can cover this frequency range. CCS1 hammer package, which has large and small hammer sets, is recommended.
200-1500Hz: Machine tool spindles with tool holders or large boring heads typically have the first natural mode at around 500Hz, and higher modes at around 800 Hz, 1200 Hz to 1500Hz. A hammer with a force range of 2000N to 5000N can be used. Higher the force range is, the lower the measurement frequency range will be. If the work material is Titanium or Inconel, it is best to have a hammer which can deliver 5000N force. If the work material is Aluminum and machined at high speeds, 2000N force range with 4000Hz frequency range is sufficient. Miniature accelerometer with a small mass is recommended for the measurement. CCS1 hammer package, which has large and small hammer sets, is recommended.
Above 1500Hz – Typically, the tool holder and end mill modes will be dominant at this frequency range. The end mill diameters are typically less than 25 mm with 75-100mm stick outs. The hammer must be small with a frequency bandwidth up to 5000Hz. Steel tips can be used to widen the frequency range of excitation. The smallest possible hammer set should be used. Hammer set given in Shop-Pro set is sufficient.
Above 5000Hz: If the end mill has a diameter less than 12 mm, the frequency will be higher than 5000 Hz which is difficult to measure with standard hammers. Miniature hammers from DYTRAN and PCB are recommended. If possible, the displacement must be measured with laser vibrometers. Note that the measurement of such high frequencies is rather difficult in practice, and any equipment purchase must be consulted with experienced users or MAL Technical Personnel.
x=10^y and logx=ylog10=y
Example with the microphone at the —10Db setting:
20logx=—10db , logx=—10/20=—0.5 , x=10^(—0.5)=1/sqrt(10)=0.31
It means the microphone output is multiplied by 0.31 times.
Example with the microphone at the —20Db setting:
20logx=—20db , logx=—1, x=10^(—1)=1/10=0.1
It means the microphone output is reduced by 10 fold.
It is better to use —10dB (less attenuation) but you must make sure that the NI DAQ card must not saturate (do not exceed +-5V). If the noise is too high, microphone will create voltage beyond 5V which will saturate NI and you will always get 5V. In that case, use -20dB.
CUTPRO does not alter the sound dB measurement if you use a gain value of 1, what ever microphone is giving. This is recommended so that you can use microphones measurement directly without altering it.
This is a good question, and the preventive maintenance may save a lot of costs associated with spindle failure.
Whenever you receive a new machine or spindle, you need to measure its Frequency Response Function (FRF) by using a clean, master tool holder with a blank carbide cylinder. The stick-out must be kept constant and set. After the hammer test, you need to store the FRF data in a safe database. Periodically, depending on how much you abuse the spindle by running machines under chatter vibration, you need to check the magnitude of the FRF or magnitude of the imaginary parts of the FRF. Any increase in FRF magnitude means that the chatter vibration free depth of cut is reduced proportionally. If the magnitude of the FRF increased by 20% at any peak (natural frequency), that means you lost 20% of the dynamic stiffness and the depth of cut needs to be reduced accordingly if you were machining close to the stability lobe borders. The loss of stiffness may be due to wear of the spindle’s taper interface or bearings. It is up to you and spindle builder how far you can tolerate the wear of the bearings and spindle. I would not let it develop beyond 30%, but this number is totally dependent on each spindle builder. We wish that they know about the procedure and let their customers know. Every spindle has its own durability strength, and we cannot make assumptions about it.
Training & Demonstration Videos and help guides are now available on our website. Click here
You can also visit our YouTube Channel.