Test results show that optical trepanning can reduce system downtime and produce high-quality holes
Paul F. Jacobs
In 2002, Laser Fare conducted a Phase I SBIR contract for “Drilling 170 Micron Diameter Holes,” with the goal of finding the combination of laser parameters that would yield the highest quality holes, in the shortest time, at the lowest cost. The application involves drilling 24 million holes through 635-µm Inconel 600 stock, to build precision injectors for the USAF. Project specifications called for 99.9 percent of the 165-175-µm-diameter holes to have a standard deviation (SD) ≤1.65µm. The mean eccentricity (ε), had to be < 5 percent, with a mean taper (τ) <2 percent, and in production each hole had to be drilled in less than one second.
Laser pulse durations <10 ps can produce high quality holes, with little recast thickness, a small HAZ, and minimal eccentricity.1 However, short pulse duration lasers also require minutes to drill an excellent hole at high cost.
Laser Fare tested four laser drilling systems,2 in percussion and trepanning modes (see Figure 1): a 100fs to 10ps pulse duration Ti:sapphire laser, a copper vapor laser, a 10ns Nd:YAG laser at 1064 nm and 532 nm, and a 100-µs Nd:YAG laser at 1064 nm.
FIGURE 1. Laser Fare tested four laser drilling systems in percussion and trepanning modes.
Percussion drilling3 uses a laser spot comparable to the desired hole diameter and a series of laser pulses to drill through the plate thickness. Mechanical laser trepanning4 uses a smaller laser spot moved in a circular orbit, with the outer edge of the spot tangent to the intended hole diameter, or the workpiece is moved in a circle. Overlapping laser pulses trepan the larger hole diameter. Optical trepanning will be discussed below.
Laser Fare found four important results at the end of the Phase I contract. First, ultra-short pulses produce very high quality holes, but the time per hole was two orders of magnitude in excess of the program goal. Second, both Nd:YAG lasers percussion drilled holes in about two seconds, but the hole quality was poor (i.e. SD ≈10 µm, ε ≈ 15 percent, τ ≈ 10 percent). Third, laser trepanned holes, from 40ns and 100-µs Nd:YAG laser pulses also produced holes in about two seconds, but with significantly better hole quality (i.e. SD ≈ 5 µm, ε ≈ 7 percent, τ ≈ 5 percent). Fourth, lower laser-energy-per-pulse values improved hole quality. Prior tests used excessive peak irradiance values that generated a plasma. Self-absorption wasted laser energy with poor hole quality and large HAZ.
While the Phase I goals were not achieved, test results showed trepanning had clear benefits. Laser Fare’s Phase II proposal subsequently included two novel ideas.
Mechanical trepanning requires either the laser spot or the workpiece be moved many times per hole. For a 170-µm-diameter hole, 400 separate “actions” are needed per hole, and almost 10 billion individual actions are needed to trepan 24 million holes, leading to possible scanning system failures. We chose the “Optical Trepanning System,”5, 6 shown in Figure 1, where special optics transform a Gaussian laser beam into an annular beam, with an outer diameter, Do, set at the desired hole diameter. The annular optical trepanning spot remains stationary and the laser pulses generate the desired hole-diameter with no moving parts. System reliability is improved, and eliminating motion-related steps reduced trepanning time by a factor of three.
A key benefit of trepanning is that one needs to eliminate material only from a thin annular region, not the entire circle. For Ro = 85 µm, and Ri =65 µm, the annular material to be removed is only 42 percent of that for percussion drilling. Less than half as much laser energy is needed, less energy is transferred into the surrounding material, and recast thickness and HAZ are also reduced leading to a reduction in laser power. The second concept involves using a single laser source with an optical beamsplitter. The energy per beam is about 25 percent of that from the primary laser. The annular beam is optimized to minimize HAZ, improve hole quality, and effect an additional 3:1 time compression, since we can now trepan three holes essentially simultaneously.
Subcontractors on the Phase II program are: the University of Central Florida (optical trepanning and beamsplitter subsystem design), Computational Fluid Dynamics Research Corp. (numerical modeling), LASAG Industrial Lasers, PowerLase Ltd., and Spectra-Physics (evaluation of various laser systems and preparation of test coupons).
FIGURE 2. The OMIS II image shows laser drilled holes.
Percussion drilling and mechanical trepanning tests on 635-µm-thick Inconel 600 test coupons were completed with a decreased energy per pulse to avoid plasma self-absorption. Optical diagnostic tests, utilizing an ROI Inc. OMIS II showed that for this application trepanning produced much better quality holes than percussion drilling, and in comparable time. Figure 2 shows an OMIS II image of laser trepanned holes from Phase II.
The mean diameters are now close to 170 µm with the mean standard deviations for these 30 holes down to 1.67 µm, the mean eccentricities 4.5 and 6.1 percent, and the taper 1.8 percent. The time to trepan a hole was reduced to 2.2 seconds without optical beamsplitting. With optical beamsplitting, the time per hole will shrink to about 0.7 second. These results represent a significant advance in rapid, economical, precision laser trepanning.
FIGURE 3. In the axicon optical trepanning system the outer beam diameter can be varied by changing the distance between the focusing lens and the workpiece.
A practical benefit of the dynamic range of an axicon optical trepanning system7 (see Fig. 3) is that the outer beam diameter can be varied by changing the distance between the focusing lens and the workpiece.
From this work, we have been able to draw the following conclusions.
- Axicon optical systems can generate annular beams appropriate for precision optical trepanning.
- Optical trepanning can reduce system “down-time” relative to mechanical trepanning.
- The time and cost of precision trepanning can be reduced by eliminating the time required for numerous small movements of the laser spot.
- Optimum irradiance should stay below the threshold for plasma formation, at ≈ 108 W/cm2 for Inconel 600. Peak irradiance values near 5 x 107 W/cm2 are especially effective.
- An axicon-based optical trepanning system provides excellent hole diameter dynamic range capability.
- Narrow annular distributions produce the highest quality holes.
- Optical trepanning with 3:1 beamsplitting can reduce the time, and cost, of trepanning precision holes by nearly an order of magnitude
Dr. Paul Jacobs (pjacobs@laserf are.com) is vice president of R&D for Laser Fare, Inc., Smithfield, RI.
- Kamlage, G., et al., “Deep Drilling of Metals by Femtosecond Laser Pulses,” Applied Physics A: Material Science and Processing, Vol. 77, pp. 307-310, 2003.
- Jacobs, P.F., Phase I SBIR MDA / USAF Contract F29601-02-C- 0121 Final Technical Report, December 2002.
- Ng, G., and Li, L., “The Effect of Laser Peak Power and Pulse Width on hole Geometry Repeatability in Laser Percussion Drilling,” Optics and Laser Technology, Vol. 33, pp. 393-403, 2001.
- J. F. Ready and D. F. Farson, “Components for Laser Materials Processing Systems,” Chap. 4 in LIA Handbook of Laser Material Processing, pp. 134-139, Magnolia Publishing, Orlando (2001).
- M. Rioux, et al., “Linear, Annular, and Radial Focusing with Axicons: Applications to Laser Machining,” Applied Optics, Vol. 17, 1978, pp. 1532-1536.
- P.A. Belanger and M. Rioux, “Ring Pattern of a Lens-Axicon Doublet Illuminated by a Gaussian Beam,” Applied Optics, Vol. 17, pp. 1080-1086, 1978.
- D. Zeng, W. P. Latham, and A. Kar, “Temperature Distributions due to Annular Laser Beam Heating,” J. Laser Applications (in press).