Step 10: The back-end process: Laser marking step by step
11/01/2000
BY MICHAEL POTTER
All laser vector marking systems work in essentially the same way. Typically, the vector marking system marks on a component while the component is stationary. The laser spot is moved over the surface to be marked in controlled X-Y coordinates, creating the character or image desired.
The first component of a laser vector marker is the source of laser energy. The energy source may be a neodymium: yttrium aluminum garnet (Nd:YAG) laser (often referred to as a YAG laser), a CO2 laser or any of a wide variety of other laser sources. Most likely, however, it will be a CO2 or Nd:YAG source. Depending on the marking application, the source will vary from a few watts of power to more than 100 watts. Energy from this source is directed through an optical path to a marking head.
 Figure 1. Vector marking system. |
It is important to know the location within the marking system where power is measured. Often, the power rating is that of the laser output before entering the optical path. Energy losses in the optical path can be significant with only a fraction of the original power available at the surface of the part. Complex optical paths and poor beam quality can reduce useable power and impact the quality of marking.
The optical path gets the beam to the scan head (which may require bending the beam around corners), manages the quality of the beam (beam conditioning), and properly achieves the desired spot size. Proper alignment of optics is critical for optimum laser-marking performance. Vibration will affect alignment over time and optics will require realignment.
 Figure 2. Flash-lamp pumped Nd. YAG laser. |
Energy reaches the marking or scanning head through an aperture and is then directed to the mirrors that determine where the energy will land on the marking surface. The first mirror the beam strikes is the Y mirror, whose position is determined by a galvanometer that is controlled by the computer generating the image coordinates. The position of the high-speed galvanometer and the mirror will determine where in the Y dimension the beam will strike the surface.
The beam will next strike the X mirror, whose position is also controlled by a high-speed galvanometer. The mirror determines where the beam will strike the surface to be marked in the X dimension. Control of the X and Y positions allows the placement of the beam anywhere within a predetermined marking field, thus allowing the creation of characters and images.
After the beam is reflected off the X mirror, it passes through a flat-field lens toward the surface to be marked. The flat-field lens is usually selected for a particular application based on a number of factors, including mark surface characteristics, part size and handling method, legend requirements, available energy, and required speed.
After the beam passes through the flat-field lens, it strikes the surface to be marked. When the energy hits the marking surface, the surface is rapidly melted or vaporized, leaving a trench or trough in the form of the image. Removal of material in this manner is called ablation (Figure 1).
Red vs. Green Lasers
Frequently, laser energy sources are referenced by color to indicate the wavelength of the output energy. Lasers, like the Nd:YAG and CO2, emit wavelengths in the infrared (IR) range and are therefore called "red" lasers. Nd:YAG is "near-IR" and CO2 is "mid-IR." These lasers emit the most common wavelength used in marking applications.
The use of "green" lasers for marking applications appears to be on the increase. Green laser energy is produced by passing a red laser output through a crystal device that reduces the wavelength by half, resulting in a doubling of frequency. The resultant 532-nanometer (nm) wavelength is particularly suitable for marking on silicon and metal packages. Red laser energy, on the other hand, is not well-absorbed by silicon. As a result, more energy must be used to mark the surface of the part and the unabsorbed energy may damage active circuitry beneath the surface of the part.
Popular Lasers for Marking Applications
One of the most common laser sources used in marking applications is the Nd:YAG. Nd is a rare earth element. The yttrium aluminum garnet material is doped with Nd ions, which are the source of the photons for the laser beam.
All YAG lasers consist of a pump energy source and a resonator comprising an Nd:YAG rod and reflective lenses. In marking applications, these lasers fall into two categories based on the pump source: flash-lamp pumped and diode pumped.
Today, the largest population of YAG lasers in the marking field is flash-lamp pumped. With a flash-lamp pumped YAG, the flash lamp acts as the source of energy, or the pump, and can produce high levels of laser energy (Figure 2). There are some basic disadvantages to this technology, however:
- Flash lamps have a limited life and must be changed frequently, causing downtime and incurring great expense. If the laser has a high duty cycle, it may need to be replaced as often as once per month.
- The flash-lamp YAG is inefficient and power is lost as heat. To manage and control this, heat chiller systems are used.
- Flash lamps consume large quantities of power. The combination of the power usage of the chiller and the laser can require as much as 15 times the energy of other types of lasers with similar power output.
More recent advances in YAG laser technology have resulted in the diode-pumped solid state (DPSS) YAG system (Figure 3). Here, the energy pump source is provided by laser diodes. These diodes pump out laser energy in the 808 to 815-nm wavelength that provides the same function as the flash-lamp energy. Initially, these lasers were limited to low-power output, but are currently available in high-power ratings.
 Figure 3. Diode-pumped Nd. YAG laser. |
DPSS lasers are much more efficient than flash-lamp systems. The laser diodes exhibit longer life than flash lamps and can, therefore, be changed less frequently. A typical lifecycle for a diode is about one year. However, diode life is significantly impacted by the level of power and the number of hours the laser is operated. Energy consumption can be reduced if a chiller is not required.
While laser diodes have a longer life than flash lamps, they are rather expensive. The number of diodes required is in proportion to the output power of the laser; therefore, the operating cost of high-power lasers can be prohibitive. However, the possibility exists to reduce operating cost as a result of the major reduction in power consumption when compared to a flash-lamp system.
Both types of YAG lasers are continuous wave (CW) and are also available as Q-switched models. With a Q-switched system, the energy builds to a high level before it is released in a short pulse. By building the energy to high pulse levels, extremely hard surfaces can be ablated. However, peak energy levels can vary from pulse to pulse, resulting in inconsistent marking depth. In applications where depth control is critical, Q-switched lasers may be impractical.
Addressing New Challenges
Manufacturers encounter new marking challenges because silicon is used as a final package material in many new semiconductor devices. The 1064-nm output of YAG lasers is not well-absorbed by silicon because most of the energy passes through the device; this can damage active components below the surface of the silicon. Also, because of the low absorption of this frequency, high power levels of red laser energy must be used to mark the surface. Green laser, at 532 nm, is well-absorbed by silicon, reducing the risk that energy passing through the device will damage internal circuitry. This results in a marking process with significantly less risk of yield loss (because of part damage) than is achievable with a red source. It also allows the use of a low-power laser for marking silicon surfaces.
 Figure 4. Diode-pumped green laser. |
Bearing in mind that green laser output is accomplished by taking the output of a standard red YAG laser and driving it through a crystal material, the YAG portion of the green laser is subject to the same operating costs and limitations as mentioned previously. The selection of the crystal material for doubling the frequency is critical to the long-term performance of 532-nm lasers. In cases where these lasers have been unreliable, the major problems have been in the output crystal modules. Progress has been made in improving the reliability of these modules and, in some cases, they are now field-replaceable (Figure 4).
The newest entry into the laser- marking field is a fiber laser. This diode pumps energy into a fiber cable that has a core that is doped with rare earth material (yttrium). The reflectance of the pump laser energy off the sides of the fiber cable and through the rare earth doped core causes the emission of photons and creates light amplification. The final output of the laser is at a wavelength of 1,100 nm.
Output power from fiber lasers ranges from about 9 watts to more than 20 watts, can be used for a variety of marking applications, and is particularly well-suited for marking plastic molded components.
Fiber lasers are small in size - 50 meters of fiber material coiled in the laser control box. There is no large rigid optical assembly, as with a flash-lamp or DPSS YAG system, so the connection to the marking head is by a flexible fiber cable. Fiber laser markers using a 915-nm diode have demonstrated lives that are four times those of the 810-nm diodes used in Nd:YAG systems. The fiber laser operates in CW mode, has a power monitor at the output and manages the output at a consistent level. This translates into tightly controlled etch depth of the mark.
 Figure 5. Fiber laser source. |
Fiber lasers use little power and do not require flash lamp replacement or annual laser diode replacement. The laser output from a fiber laser is coupled to the marking head through a collimator. This direct coupling eliminates the complex optical paths, thus reducing the need for expensive realignment and downtime. Optical realignment activity is the stage at which most injuries from laser energy occur (Figure 5).
How to Judge
One problem with using a method of speed measure of characters per second (CPS) is that the test is conducted using a manufacturer's choice of character font. Thus, a simple single-stroke font marks faster than a more complex or stylish font. The test is also done on black paper, removing ink to whatever level of legibility the manufacturer determines is acceptable.
To a large degree, CPS has become a relatively meaningless specification. Marking companies will suggest that the more powerful the laser is, the faster it will mark. This helps to sell more expensive lasers and also raises the replacement cost of laser pump diodes. The replacement cost of laser diodes for a 100-watt laser can be $50,000 over the course of 18 months.
While laser power impacts the theoretical speed of marking, particularly on harder surfaces like metal, the speed may be limited by the useable rate of the scanners in the marking head (regardless of the laser power available). What most often determines productive output will be the handling of the components in a system, or the complexity of the required legend on a part.
To determine marking rates and productive output of a system, sample prints determined by the customer to be of acceptable quality and are made using the required legend and logos should be evaluated. The time to make these prints can be offered by the laser-marking manufacturer so that marking time can be compared on an equal basis. The cycle time can then be compared to the ability of the handler to move parts, and to determine whether marking time gates productivity. With specifications that are vague and difficult to use as comparisons, this is a good way to determine if a particular marker will meet the requirements of the application.
MICHAEL POTTER, business segment manager, electronics, can be contacted at MARKEM Corp., 150 Congress Street, P.O. Box 2100, Keene, NH 03431; 603-357-4255, ext. 2431; Fax: 603-357-7413; E-mail: [email protected].