1. What is a grating ruler?
A grating ruler, simply understood, is a ruler. For example, have you ever used a vernier caliper, or a tape measure? There are scales on the ruler. These scales serve as a benchmark. You only need to compare the object to be measured with the scale on the caliper to measure the length. , size, depth and other information. Similarly, there are "scales" on the grating ruler. This scale is called a grating and is engraved on the ruler through photolithography. However, it does not read information through the human eye, but through a matching reading head. Get location information.
The actual principle is much more complicated: a grating is a series of stripes and slits made on a glass or steel tape ruler. The width of one stripe and one slit is called the grating pitch. A common grating pitch is 20um.
Every time the read head scans a grating pitch, it generates a sine wave signal cycle. This signal is then subdivided by an electronic circuit (either built-in in the read head or an external subdivision box), such as 5, 10, 50, 100 times subdivision. , so a very high resolution can be achieved.
For example, if a 20um grating pitch is subdivided 50 times, the period is 0.4um, which is the resolution mentioned by the manufacturer .
2. How the reading head scans the raster to read data
As shown in the figure below, the read head has an indicating grating (Scanning reticle) with the same grating pitch as the scale body, and the read head itself has an LED light source. When the read head moves relative to the grating scale (Scale), the LED light passes through the focusing lens. Afterwards (Condenser lens), the light is illuminated on the grating ruler, and then the light passes through the grating slit and is diffracted to the photodetector (Photocells) of the reading head, thus producing sinusoidal interference fringes of alternating light and dark on the detector plane.
Then, the detector converts these stripes into sinusoidal electrical signals, which are amplified and shaped by the circuit to obtain two sine waves or square wave signals A and B with a phase difference of 90 degrees.
The number of cycles of a sine or square wave is proportional to the distance traveled. When the ruler moves forward, the A signal leads the B signal by 90 degrees. When the ruler moves in the reverse direction, the A signal lags behind the B signal by 90 degrees.
If it is converted into a square wave signal and subdivided by 4 times, there will be 4 rising edges in one cycle. The resolution at this time only corresponds to one rising edge, which is 1/4 cycle (pulse).
It should be noted that in reality, the photodetector and LED are on the same side of the scale and are integrated in the readhead. When light irradiates the grating ruler, part of the light will be reflected back, and then pass through the focusing lens and irradiate back to the photodetector to form an electrical signal.
3. Why choose to use grating ruler?
Generally, calipers can only be read by the human eye. How do you transmit the data you read to the machine's motion axis, not to mention how you provide feedback and correction when the position is not correct. Even if you use a caliper that can output readings, the accuracy is not as high as that of a grating ruler.
How accurate can a general grating ruler be?
±15μm, ±5μm, ±3μm, ±1μm are available.
It should be noted that the accuracy mentioned here refers to the manufacturing error of the grating ruler, which refers to the actual possible error per 1m. If the distance is small, the error will be smaller, such as ±0.275μm/10mm, ±0.750μm/50mm, That is to say, the possible errors caused by movements of 10mm and 50mm are ±0.275μm and ±0.750μm respectively. This is the manufacturing accuracy of the grating itself and is also an important reference for selecting a grating ruler.
For example, we often hope that the positioning accuracy of an axis is ±3um/100mm or ±5um/100mm. Then when choosing a grating ruler, first of all, its accuracy must be higher than this requirement, such as ±0.5um/100mm or ±1um/100mm.
In addition, ±1um/100mm here cannot be deduced from ±10um/1000mm, because there is no linear relationship here, and often the grating ruler itself will have one or two short-stroke errors marked. For example, if the accuracy is ±10um/m, the error within 100mm will usually be less than ±1um.
Of course, when the read head subdivides the signal, it will also introduce errors, called differential errors, but this error is very small. For example, a grating ruler has a grating pitch of 20um, a resolution of 0.1um, and a period error (electronic subdivision error SDE) of ±0.15um. It refers to a grating with a grating pitch of 20um. After 200 times subdivision, the resolution is 20/200=0.1um. Within this 20um grating pitch, the error caused by system signal processing is ±0.15um.
What other factors are related to the measurement accuracy of the grating ruler?
It is usually related to the accuracy of the guide rail, structural rigidity, and the distance between the grating ruler and the end point. For temperature-sensitive systems, it also has a lot to do with the ambient temperature and the heat source near the grating ruler.
Usually the guide rail has straightness and parallelism errors, and the reading head can only read information about the position of the grating ruler. The position we care about is usually not the position of the grating ruler, but the position of the functional point on the structure, which means there is Abbe error.
Structural rigidity, the distance between the scale and the end point will all affect the Abbe error.
The rise and fall of temperature will cause thermal expansion of the grating ruler and will also introduce errors. Because the reference changes, the read head will interpret it as a slight movement of the grating ruler.
The grating pitch of the grating ruler is so small, how are the scales engraved? What materials are generally used for grating rulers?
As mentioned above, gratings are usually engraved on a glass or steel strip substrate. Glass is generally used for short-stroke rulers. Large-length rulers, such as 10 meters and 30 meters, use steel strips as the substrate.
As for how the grating pitch is engraved, as mentioned before, it is engraved using the principle of photolithography.
What is photolithography?
For example, if you look at an integrated circuit board, aren’t there metal lines on it?
The line is not drawn on, but a layer of copper is brushed onto the silicon board, and then a layer of wax is brushed on top. Then you use a knife to "carve" the wax off the part without wires, and then throw the board into the corrosive solution. Inside, the areas not covered by wax will be corroded, and then you take it out and the integrated circuit board is ready.
The same is true when making a grating ruler, but the grating pitch on the grating ruler is very, very small, and the spacing is on the micron level. There is no physical knife that can carve it out. At this time, we use "photolithography" .
Because light can be divided very finely, photolithography is to lay a layer of photosensitive film on the surface of the material to be engraved, and then shine it with light. Wherever the light hits, the photosensitive film will be "burned", and then this time You use a "piece of paper" with engraved lines on it to block the light, thus leaving the area where the grating needs to be retained, and then throw it into the corresponding liquid and soak it, and the grating ruler is ready.
What is the difference between an incremental encoder and an absolute encoder? What occasions are they used for?
Incremental gratings consist of periodic gratings. The reading of position information requires a reference point, and the position of the mobile platform is calculated by comparing it with the reference point.
Since an absolute reference point must be used to determine the position value, one or more reference points are also engraved on the incremental grating. The position value determined by the reference point can be accurate to one signal period, which is the resolution. In most cases, this type of scale is used because it is cheaper than absolute scales.
For absolute gratings, the absolute position information comes from the grating code disk, which consists of a series of absolute codes engraved on the ruler. Therefore, when the encoder is powered on, the position value can be obtained immediately and can be read by subsequent signal circuits at any time without moving the axis to perform the reference point zero return operation.
Because homing takes time, if the machine has multiple axes, the homing cycle can become complex and time-consuming. In this case, it is advantageous to use an absolute scale.
In addition, from the perspective of speed and accuracy, the maximum scanning speed of the incremental grating depends on the maximum input frequency (MHz) of the receiving electronic device and the required resolution. However, since the maximum frequency of the receiving electronics is fixed, increasing the resolution will result in a corresponding decrease in the maximum speed and vice versa.
Absolute encoders, on the other hand, are not affected by this situation and ensure high-speed and high-resolution operation. This is because the location is determined based on demand and using serial communication. The most typical application of absolute gratings is placement machines in the surface mount technology (SMT) industry. In this industry, improving positioning speed and accuracy at the same time is an eternal goal.
4. Selection of grating ruler
The first is accuracy.
Second, resolution.
Third, itinerary.
Fourth, the maximum detection speed.
Fifth, the electrical interface and wire length.
Sixth, installation method and installation space.
Seventh, anti-vibration performance.
Eighth, price . Absolute encoders are generally 20% more expensive.
Generally speaking, accuracy and resolution are the primary factors when we choose a grating ruler.
How to select a grating ruler based on positioning accuracy or repeat positioning accuracy requirements?
For example, to make a mobile platform with a stroke of 100mm and a system positioning accuracy of ±0.01mm (±10um), we can choose a grating ruler with a stroke of 120mm, an accuracy of ±0.5um/1m, and a resolution of 0.02/10=0.002mm. .
In this way, there is a 20mm margin for the stroke, which can be used for hardware protection, and the accuracy of the grating ruler itself is ±0.5um.
Regarding the resolution, the reason why the resolution is chosen as 1/10 of the positioning accuracy is because there is a control error, which is usually ±10cnt. The 1cnt mentioned here refers to the resolution based on the resolution of the grating ruler itself. The resolution after subdivision processing is usually 4x or 8x subdivision.
For example, with 4 times subdivision, the resolution of the above grating ruler with a resolution of 0.002mm will reach 0.002mm/4=0.5um. ±10cnt is also ±5um, leaving double the margin. The main consideration is that the system also has mechanical errors, such as the transmission system, structural rigidity, etc. These factors will also eat up part of the accuracy. Of course, these are the categories of geometric errors. The error curve can be obtained through measurement and partially compensated. However, there are still some dynamic errors that cannot be compensated.
In addition, many times, we do not care about absolute positioning accuracy, but only about repeated positioning accuracy. Generally speaking, the repeated positioning accuracy of the system is 1/2 to 1/3 of the positioning accuracy, and will not exceed an order of magnitude at most, that is, 1/10. For example, in this example, generally speaking, the repeatable positioning accuracy is between ±0.01mm/2～±0.01mm/10=±0.005mm～±0.001mm.
Of course, there is another point. The repeated positioning accuracy of the system is generally between resolution and positioning accuracy.
What is the most important thing to pay attention to when installing a grating ruler?
First, you need to reserve space for adjustment, because oftentimes there is insufficient space to adjust the reading head, causing debugging to take too much time and maintenance to be troublesome. Precise adjustments must be made after installation, and the quality of the adjustment can be judged according to the color of the indicator light.
In addition, the mounting surface of the general grating scale must maintain a certain parallelism with the guide rail surface, such as 0.1mm. This requirement is often twice as high as the swing parallelism requirement of the installation, such as the swing parallelism requirement of 0.2mm.