Effect of Temperature on CNC Machine Tools and Materials During Machining of Precision Metal Parts
We all know that temperature affects materials, but how does the final temperature affect our precision metal parts processing? On the surface, this will be a simple task to compensate. All we should do is use the coefficient of thermal expansion (CTE), look at the thermometer, and provide a solution. If it is so simple on the line. Before we can do this calculation, we have to consider some more variables. Here are the variables we really need to solve: Part of the CTE includes different materials in the same part, the machine's CTE and feedback gauge, and the machine geometry changes with temperature. Hot working process, by cutting coolants and lubricants add or remove heat.
After considering all of these variables, we have to determine the temperature we want to normalize the part.
What is the best way to control all of these variables? The best practice is usually to eliminate them. Set the ambient temperature of the machine and components to the desired normalized temperature; Let the parts and machine "soak" at this temperature for about 24 hours. Use cutting coolant that has been adjusted to the same temperature as the standardized environment, then monitor these conditions and control as needed. This is the perfect world, but it is not a norm. We usually see is designed to 20C or 68F size correcting part, and 70 ° F to 75 ° F in the processing environment. Part of the CTE is from 1 to 13 PPM.
Ambient temperature control
Ambient temperature control is the most common and how CMS typically refers to the specifications of our machines.
The standard is +/- 1 degree Celsius or about 3 degrees Fahrenheit. The steel has a CTE of approximately 7.3 ppm / F. For a 10 m machine with a 3 degree F temperature spread, = .0073 mm / m / 1F = 0.022 mm / m / 3F = 0.22 mm. As you can see, for a 30-foot machine, this is quite a lot of exercise. Aircraft trusses, helicopter blades, flight controllers and many other components are so large that they must be machines with tighter tolerances. The tighter you can control the better.
As an annotation, the Lawrence Livermore National Ignition Laboratory, which has many high-precision measurements, has a 10-metre cube laboratory and remains at 0.1 degrees F for testing.
There are also some standards, such as ASME 5.54, with thermal drift as one of the standard measurements in machine tool analysis.
In addition, the type of heat is also important. Radiant heat, convective heat and conductive heat make the material react differently. I won't go into details here, but the heat from the sun, the oven, the fan, and the air conditioning vents will affect the machine. The extent to which some manufacturers run liquid coolant even through racks and drive components, which we will discuss below.
Component temperature control
Another method is to monitor and control the temperature of the component. Gearboxes, motors, encoders and ball screws also change with temperature. For many years, even now, ball screws have been drilled through which the coolant controls their expansion. The spheroidal graphite is cast and still contained in a coolant jacket, also having a gearbox and motor. All modern spindles are liquid cooled or have more correct temperature control.
For large machines, some of these methods can be used; But for the long axis, 3 meters plus, the ball screw is not a good choice. Racks and pinions are often used instead of a one-piece construction. This allows the rack to move with the machine structure without having to fight two different alloys.
Structural temperature monitor
This is a method of using modern electronic devices to monitor the structure temperature and modify it to travel through the algorithms set up in the device.
The way to do this is to set up several temperature sensors along the axis of the machine. The data is returned to the control box, which modifies the encoder's pulse signal length to correct for thermal motion in the machine structure. This unit is completely separate from the machine controller. Both the temperature sensor and the machine encoder pass through this box. The encoder signal is then modified and sent back to the controller. This is a very good system, but it relies on two very important things to run correctly:
(1) Temperature changes must occur very slowly
(2) Must come from convection heat, ie air to material.
The added importance of this is that the system must be tested. The reason for the convection heat and slow change requirements is that the temperature sensor can only compensate for the entire length of the shaft, averaging all the sensors. We can place 100 sensors on the shaft, but if one area is hot, but the other area is cold, then the unit will only produce poor results in certain areas over the entire length of the average temperature. Currently, CMS is building a machine with this system, and we hope to provide a transformation before the end of the year.
Real-time positioning control
This is currently the best way to handle temperature changes in long machines. Not surprisingly, this is by far the most expensive. The system uses a laser interferometer to measure the distance traveled by the machine. This is not part of the machine controller. It is the same device used to calibrate the machine. The difference is that this laser is permanently mounted on the machine and is online, in real time. In the case of a gantry machine with a master-slave axis, two systems are required to correct the motion of the axis. It will also calibrate the temperature gradient and perform a self-calibration. The cost of the system is approximately $100,000 per axis. The CMS is also reviewing this system for possible availability in the future.
Parts itself as a variable
As I said earlier, we have to think about it. The machine's CTE will have a value, and some may vary greatly. How do we compensate? The ideal situation is to set the machine and parts at the desired temperature and stay very close. However, what if the nominal temperature is 68F and the machine and parts are 72F? There are several options at this time. If the machine is to be carefully maintained at this temperature, we can use the CTE of the material to calibrate the machine to the correct dimensions of 72F. This is a good solution. Now if the environment changes to 77F? The CTE of this machine will expand, that is to say, this 30 micrometers, and this part will expand by 10 micrometers. What are we doing now? We have several options. We can recalibrate the machine to this new environment by using the CTE of the part; We can also scale the part's program by subtracting the motion of the part from the motion of the machine and calculating the proportion of the overall size. Or we can take one of the active machine compensation devices we have been discussing.
Linear proportional control option
There is also a simple, effective and inexpensive device that is currently being used in the CMS to correct all these crazy moves. Before I describe this wonderful device, I must mention that it only works for the smaller long axis of 3 meters long. The device is a linear scale of the same CTE as the material being cut. As mentioned earlier, our choice of options is environment, component, machine or feedback device. In this case, the scale will be made of a material having the same portion of CTE, and the machine will be allowed to expand and contract at any rate, and the feedback device will only cause the machine to move the distance it reads. Another option here is to isolate the scale from the movement of the machine and to control the environment within or around the scale. This is the approach taken by CMS. We use a steel belt scale installed in a housing that is mechanically isolated from the machine, and we control the temperature inside the housing. This way, the machine is always calibrated correctly, regardless of external conditions. Now, part movement still needs to be considered, but this gives us a lesser variable to consider.
After considering all of these variables, we have to determine the temperature we want to normalize the part.
What is the best way to control all of these variables? The best practice is usually to eliminate them. Set the ambient temperature of the machine and components to the desired normalized temperature; Let the parts and machine "soak" at this temperature for about 24 hours. Use cutting coolant that has been adjusted to the same temperature as the standardized environment, then monitor these conditions and control as needed. This is the perfect world, but it is not a norm. We usually see is designed to 20C or 68F size correcting part, and 70 ° F to 75 ° F in the processing environment. Part of the CTE is from 1 to 13 PPM.
Ambient temperature control
Ambient temperature control is the most common and how CMS typically refers to the specifications of our machines.
The standard is +/- 1 degree Celsius or about 3 degrees Fahrenheit. The steel has a CTE of approximately 7.3 ppm / F. For a 10 m machine with a 3 degree F temperature spread, = .0073 mm / m / 1F = 0.022 mm / m / 3F = 0.22 mm. As you can see, for a 30-foot machine, this is quite a lot of exercise. Aircraft trusses, helicopter blades, flight controllers and many other components are so large that they must be machines with tighter tolerances. The tighter you can control the better.
As an annotation, the Lawrence Livermore National Ignition Laboratory, which has many high-precision measurements, has a 10-metre cube laboratory and remains at 0.1 degrees F for testing.
There are also some standards, such as ASME 5.54, with thermal drift as one of the standard measurements in machine tool analysis.
In addition, the type of heat is also important. Radiant heat, convective heat and conductive heat make the material react differently. I won't go into details here, but the heat from the sun, the oven, the fan, and the air conditioning vents will affect the machine. The extent to which some manufacturers run liquid coolant even through racks and drive components, which we will discuss below.
Component temperature control
Another method is to monitor and control the temperature of the component. Gearboxes, motors, encoders and ball screws also change with temperature. For many years, even now, ball screws have been drilled through which the coolant controls their expansion. The spheroidal graphite is cast and still contained in a coolant jacket, also having a gearbox and motor. All modern spindles are liquid cooled or have more correct temperature control.
For large machines, some of these methods can be used; But for the long axis, 3 meters plus, the ball screw is not a good choice. Racks and pinions are often used instead of a one-piece construction. This allows the rack to move with the machine structure without having to fight two different alloys.
Structural temperature monitor
This is a method of using modern electronic devices to monitor the structure temperature and modify it to travel through the algorithms set up in the device.
The way to do this is to set up several temperature sensors along the axis of the machine. The data is returned to the control box, which modifies the encoder's pulse signal length to correct for thermal motion in the machine structure. This unit is completely separate from the machine controller. Both the temperature sensor and the machine encoder pass through this box. The encoder signal is then modified and sent back to the controller. This is a very good system, but it relies on two very important things to run correctly:
(1) Temperature changes must occur very slowly
(2) Must come from convection heat, ie air to material.
The added importance of this is that the system must be tested. The reason for the convection heat and slow change requirements is that the temperature sensor can only compensate for the entire length of the shaft, averaging all the sensors. We can place 100 sensors on the shaft, but if one area is hot, but the other area is cold, then the unit will only produce poor results in certain areas over the entire length of the average temperature. Currently, CMS is building a machine with this system, and we hope to provide a transformation before the end of the year.
Real-time positioning control
This is currently the best way to handle temperature changes in long machines. Not surprisingly, this is by far the most expensive. The system uses a laser interferometer to measure the distance traveled by the machine. This is not part of the machine controller. It is the same device used to calibrate the machine. The difference is that this laser is permanently mounted on the machine and is online, in real time. In the case of a gantry machine with a master-slave axis, two systems are required to correct the motion of the axis. It will also calibrate the temperature gradient and perform a self-calibration. The cost of the system is approximately $100,000 per axis. The CMS is also reviewing this system for possible availability in the future.
Parts itself as a variable
As I said earlier, we have to think about it. The machine's CTE will have a value, and some may vary greatly. How do we compensate? The ideal situation is to set the machine and parts at the desired temperature and stay very close. However, what if the nominal temperature is 68F and the machine and parts are 72F? There are several options at this time. If the machine is to be carefully maintained at this temperature, we can use the CTE of the material to calibrate the machine to the correct dimensions of 72F. This is a good solution. Now if the environment changes to 77F? The CTE of this machine will expand, that is to say, this 30 micrometers, and this part will expand by 10 micrometers. What are we doing now? We have several options. We can recalibrate the machine to this new environment by using the CTE of the part; We can also scale the part's program by subtracting the motion of the part from the motion of the machine and calculating the proportion of the overall size. Or we can take one of the active machine compensation devices we have been discussing.
Linear proportional control option
There is also a simple, effective and inexpensive device that is currently being used in the CMS to correct all these crazy moves. Before I describe this wonderful device, I must mention that it only works for the smaller long axis of 3 meters long. The device is a linear scale of the same CTE as the material being cut. As mentioned earlier, our choice of options is environment, component, machine or feedback device. In this case, the scale will be made of a material having the same portion of CTE, and the machine will be allowed to expand and contract at any rate, and the feedback device will only cause the machine to move the distance it reads. Another option here is to isolate the scale from the movement of the machine and to control the environment within or around the scale. This is the approach taken by CMS. We use a steel belt scale installed in a housing that is mechanically isolated from the machine, and we control the temperature inside the housing. This way, the machine is always calibrated correctly, regardless of external conditions. Now, part movement still needs to be considered, but this gives us a lesser variable to consider.
