-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueIPC APEX EXPO 2024 Pre-show
This month’s issue devotes its pages to a comprehensive preview of the IPC APEX EXPO 2024 event. Whether your role is technical or business, if you're new-to-the-industry or seasoned veteran, you'll find value throughout this program.
Boost Your Sales
Every part of your business can be evaluated as a process, including your sales funnel. Optimizing your selling process requires a coordinated effort between marketing and sales. In this issue, industry experts in marketing and sales offer their best advice on how to boost your sales efforts.
The Cost of Rework
In this issue, we investigate rework's current state of the art. What are the root causes and how are they resolved? What is the financial impact of rework, and is it possible to eliminate it entirely without sacrificing your yields?
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - smt007 Magazine
Drying Three Times Faster: Jet Manifolds vs. Air Knives
December 31, 1969 |Estimated reading time: 8 minutes
The physical removal of water on PCBs requires far less energy than evaporative removal. But unless air strikes with sufficient velocity to overcome the surface tension, physical removal does not occur.
By Jeff Jones
Because aqueous cleaning has become a widely accepted process, drying water trapped in tight spaces on boards has always presented a challenge. Water must be removed completely to prevent corrosion, shorts and circuit failure once the assembly is in the field. There are two primary modes of water removal: evaporation through convection or infrared heating, and physical removal via air impingement.
Physical removal is preferable for several reasons. First, it requires much less energy than evaporation. Second, evaporation tends to concentrate ionic contamination in the water that, in time, can lead to corrosion, current leakage and product failure. Finally, any heating used to evaporate the water droplets invariably warms the board and components, adding additional thermal cycles.
Figure 1. Residual water clinging to a component subjected to varying air velocities: a) before drying, b) low velocity, c) medium velocity, d) high velocity.
Removing water via impingement is typically thought to require a sufficient volume of air directed at an assembly to force it out of board topography. However, the primary rule is that the air striking a water droplet's surface must be sufficient in velocity to overcome the surface tension holding the water in place.
Figure 2. Theoretical minimum air speed vs. gap and hole diameter.
In a typical in-line process, board assemblies conveyored through the cleaning machine are sequentially washed, rinsed and dried. For drying, the board is impinged above and below the surface of the conveyor from by air and heat delivery systems. Figure 1 depicts the typical response of water clinging to a component to impinging air as velocity increases. The physical removal of water increases with increasing air velocity.
Whereas water configurations tend to establish themselves rather quickly, further impingement by air actually does little to physically remove the remaining water. For a completely wet board, air velocity tends to eclipse the effects of air volume applied (or of air or water temperature). Further, varying air impingement direction (rotation of air knives, etc.) does little to affect water removal in tight spaces if air velocity near the component is below that required to overcome surface tension. Accordingly, a study was conducted to evaluate air nozzle designs and their effect on the physical removal of water.
Theoretical Minimum Velocity Figure 3. CFD models of air-knife flow (in ips) shows a standard knife (a) with 0.040" slot onto a flat surface; a close-up of flow from a standard knife with 0.040" slot (b); air flow from an air knife with a 0.344" opening (c); a close-up of flow from a larger air knife traveling through and under the component (d); airflow from a knife with a 0.344" slot striking between two parts raised 0.050" above a planar surface (e); and a close-up (f) of flow under the right-hand component of 3e.
Theoretical calculations were performed to determine the minimum air velocity needed to overcome water surface tension in two simplified tight spots: a cylindrical hole and a gap between two planes. Figure 2 shows the minimum air speed required to push water through narrow planar gaps (e.g., between a chip carrier and a board and through holes). As the size of the gap or hole decreases, a significantly higher speed must be used to impel the water from the gap or hole. Note that for smaller dimensions, the required speed approaches nearly 250 fps. Using this as a bare minimum and understanding complex geometry further impedes flow, a speed of 275 fps was used in the computational fluid dynamics (CFD) models described below.
CFD Modeling Figure 4. Residual water vs. conveyor speed. The jet manifold system leaves 11.7 to 13.1X less water on the test vehicle than the air knife.
Figure 3 depicts, in cross-section, several computer models of air flowing from a typical air knife. Figure 3a is a 2-D, transient CFD model of air issuing from a 0.040" air knife slot at 110°F. Flow is simulated in 1.25 µs intervals and leaves the nozzle at a uniform speed of about 275 fps directed at a planar surface from a distance of 3.0". As shown, the high-speed air stream (in red) breaks up and slows considerably (approximately by a factor of 4) before it reaches the surface. The air no longer has the velocity to push water from the smaller spaces. It must be left to dry by evaporation.
In Figure 3b, a close-up of the same model, the flow breaks up at a distance of 0.750 to 1.000" from the nozzle, making the air knife less effective for drying boards that require higher clearances (greater than 1") for mounted components. Note also that the air surrounding the nozzle exit is not completely at rest. This is because the flow stream "entrains," or pulls, the surrounding air with it. Unfortunately, the same forces of friction that speed the surrounding air also slow the air-knife flow stream. Although more air is delivered, it moves more slowly and strikes the water with less force.
Figure 5. Residual water vs. dwell time.
The model in Figure 3c shows the effect of using a larger slot width. This is a 2-D, transient CFD model of air from a 0.344" air-knife slot with air at 110°F. Airflow, simulated in 10 and 20 µs intervals, leaves the nozzle at a uniform speed of about 275 fps. It is directed at a planar surface from a distance of 3.000" with a vertical connector placed so that the left edge of the upper opening is tangential with the right edge of the air-knife slot. Air leaves the nozzle at the same speed as in the earlier model and at the same distance from a simulated printed circuit board (PCB). Because the larger stream is more stable, it retains most of its velocity from the nozzle to the surface, forcing air through and beneath the connector.
Figure 6. Residual water vs. power usage.
Figure 3d is a close-up showing air flowing at high velocity through even the narrowest gap inside the connector (0.020" across). Flow continues down through the connector and underneath with sufficient velocity to force water out.
Figure 7. Residual water vs. total volume of air used.
Figure 3e depicts the same larger slot directing an air stream onto a surface between two simulated chip carriers. Each carrier is 1" wide, 0.250" high and at an elevation of 0.050" from the board surface. A close-up of the flow near the right-hand component is shown in Figure 3f. The high-speed flow was seen to penetrate beneath the component.
From the CFD models, it is apparent that to physically remove water from circuit assemblies at a distance of about 1" or greater, the slot width must be much larger than is typically used. A jet manifold has been developed that uses a series of large circular jets in place of the single narrow slot of the air knife. This device was tested in comparison to air knives.
Experimentation
Test vehicle and method. A portion of an old motherboard was used as the test vehicle to compare air delivery systems. The board was immersed in deionized water at ambient temperature, pre-dried on the bottom side only and placed on a horizontal conveyor set at a varying speed of up to 6 fpm. In most in-line cleaning, the rinse water is at an elevated temperature. For these tests, ambient temperature water was used to minimize evaporation. In addition, the bottom side of the assembly has no mounted components.
The air deliver system under test was mounted at a fixed height of 2.750" above the conveyor and is either an air knife (narrow slot) or a jet manifold (series of large circular jets). Test conditions and parameters are given in Tables 1 through 4. The operating pressure for both devices is 60" H2O (2.2 psig), which yields an estimated air velocity of 420 fps for the air knife and 357 fps for the jet manifold (air temperature is approximately 125°F).
Results are normalized by dividing the residual water weight by the surface area of the board. For all quantified performance comparisons, the lines giving the least-squares fit of the data (residual water vs. conveyor speed) are used for the conveyor speed range of 2 to 6 fpm.
Test results. Over the conveyor speed range, the jet manifold outperformed the air knife by more than an order of magnitude. The manifold left 11.7 to 13.1X less water (Figure 4) on the test vehicle at equivalent conveyor speeds, a performance ratio that will vary greatly from board to boarddepending on complexity and component population density.
Next, both devices were compared in terms of dwell time, defined as the time the test assembly is directly under the air delivery device. To simulate multiple air knives, the test vehicle was run beneath the same air knife for multiple passes. (For dwell-time estimation, the knives are assumed to be spaced 5" apart.) The jet manifold provided an equivalent performance to the air knife using 2.8 to 5.7X less dwell time. The results are plotted in Figure 5.
The next comparison was based on electrical energy use. For energy and air flow comparisons, those used were calculated as proportional to the ratio of the width of the test vehicle divided by the active conveyor width, which the air delivery systems covered (both were 10"). The jet manifold reaches equivalent results with 4.5 to 9.0X less energy (Figure 6).
Finally, flow measurements and dwell times are used in combination to quantify the total volume of air used. For these comparisons, equivalent results are obtained with the jet manifold using 3.3 to 6.6X less total air (Figure 7).
Conclusion
The emphasis on physical removal of water has required a new approach to drying for circuit assembly in-line cleaning. CFD modeling and experimentation have contributed to the design of a jet manifold system that is seen to be an improvement in performance over the standard air knife. Results were confirmed whether measured in terms of residual water vs. conveyor speed, required dwell time, electrical energy or total volume of air.
Acknowledgments
Special thanks are tendered to Austin American Technology for providing research resources; to Computer Consulting Associates for donating the board assemblies; and to Steve Stach (Process Sciences Inc.) and Bob Brady (Paxton Products Inc.) for helpful suggestions.
JEFF JONES may be contacted at Austin American Technology Corp., 12201 Technology Blvd., Suite 160, Austin, TX 78727; (512) 335-6400; Fax: (512) 335-5753; E-mail: jjones@jdaconsulting.com; Web site: www.aat-corp.com.
null
null
null