News

Using x-ray inspection, the efficacy of pin-in-hole reflow with different board finishes is studied.

Pin-in-hole reflow, intrusive reflow, pin-in-paste reflow and multi-spot soldering all describe the same reflow process, one that permits through-hole components to be soldered without hand soldering, wave soldering or selective soldering. The technique, an alternative to traditional production methods, reduces the number of process stages during manufacture by permitting surface-mount and insertion components to be processed simultaneously.

The pin-in-hole reflow (PIHR) technique with Pb-free materials has been featured at major trade shows in the past few years. The PIHR process used at a recent exhibition is described below,¹ as are the results obtained for the quality and quantity of the fill of the PTH joints manufactured during the event. During the study, the same Pb-free solder paste was used on boards with different surface finish and under two different reflow methods. The solder paste was a SAC alloy. Board finishes used were immersion gold, immersion silver, immersion tin and organic surface preservative (OSP or copper finish). The boards were reflowed under convection or vapor-phase reflow. The reliability of PIHR joints has been discussed; ² here we provide details from a U.K. National Physical Laboratory Pb-free joint testing project that shows that good through-hole joint cannot be broken, no matter how they are produced.

The PIHR Process

The basic PIHR process begins with stencil printing the surface of the board with solder paste. Stencil printing is used traditionally for depositing solder paste at surface-mount terminations. For PIHR, solder paste is also screen-printed over the through-hole pads and, therefore, into the holes. For the Nepcon U.K. "Pb-free Experience3" production, a SAC alloy was screen-printed through a 0.006" thick stencil.

To illustrate the efficiency and quality of the PIHR process, PTH connectors were used as an example of an in-process application. The pitch of the connector pins and the stand-off features of the connector body then determined the paste volume printed on the surface of the board. The thickness of the stencil used was ultimately a compromise between the needs of the connectors and the fine-pitch components on the test board. Proper design and selection of connectors permits the paste volume to be maximized on the surface of the pads and onto the surface of the soldermask. Guidelines on paste, stencils, paste volumes and appropriate inspection criteria are available from the SMTA.³

Successful screen-printing for PIHR has in the past used a rheometric, sealed print-head on the screen printer. This increases the penetration capability of paste into the hole, while minimizing waste associated with traditional blade-printing processes. The print penetration into the holes in Figure 1, for example, shows substantial fill, estimated at better than 80%. Standard metal blade printing can be used successfully for through-hole parts with a thicker stencil. A sealed head can improve hole filling capability.

Click here to see all tables and figures (696KB PDF file).

After printing, surface-mount components are placed on the surface of the paste prior to insertion of connectors into the pasted through-holes. However, with appropriate connector selection and design, connectors may be inserted first. Historically, it has often been difficult to automate odd-form assembly because of the need for additional odd-form equipment. With certain placement machines having the capability to insert all popular surface-mount shapes and odd-form parts, it widens the opportunity for use of PIHR technology. The board design on the test production line featured two connectors, 0201 and 0402 chip components, and BGA, flip chip and QFP devices.

The two connectors were a 96-way right-angled connector and two six-pin IDC sockets. Vision checks can be made on each part to align leads with plated through-holes. It should be noted that accurate insertion requires vision systems to have the capability to check through-hole position accuracy, as well as the traditional fiducial marks associated with SMT (Figure 2). Drilling and photoimaging of printed boards can have large tolerances; the base PCB materials can change their dimensions by 0.001" per inch. Ideally, all components would be provided in tape-and-reel formats for production, but long parts like the 96-way connector can take up feeder space, so versatility is also important on packaging selection for odd-form parts. Ideal packaging options are tape-and-reel for IDC connectors and machined plates for 96-way edge connectors. As the plates are used in an automatic tray feeder, part positions were also provided for the other connector design to give added flexibility.

Following automatic assembly, the board is transported to reflow. Here, the surface mount and PTH joints can be simultaneously reflowed. For this experiment, both convection reflow and vapor phase were available, as both methodologies have been successfully used for PIHR.

The size of the reflow oven for PIHR in terms of length and number of zones depends on the throughput required. In production, a reflow oven is selected for its throughput and its compatibility to the product type and peak temperatures required. Alternatively, the most popular Pb-free solder pastes can be processed at 240�C, or 230�C with vapor phase reflow. If convection reflow is used, then it is important to correctly profile the temperature across the entire board. The connector undergoing PIHR may be the largest component on the surface of the board, in terms of mass, and can act as a local heatsink. Therefore, monitoring the PTH parts in the profile will ensure that any differential temperature is noted and the profile adjusted accordingly.

With a double-sided surface-mount assembly, the connector-side of the board would be processed second. However, it is also possible to process connectors and other through-hole components on both sides of the board, if required. Figures 3 and 4 show typical solder joints produced with the PIHR assembly process on a har-busHM+ and 96-way right-angled connector.

Solder joint strength and long-term reliability are no different than with conventional wave or manual soldering operations.⁴ PIHR joints also meet the visual requirements for international standards such as IPC-A-610D. This has been shown by the U.K. National Physical Laboratory, which conducted a Pb-free soldering project that permitted customers to compare the results from Pb-free production to that from long-term reliability testing. NPL has also been conducting a U.K. government-funded (DTI) project in collaboration with industry to evaluate the solder joint reliability of various Pb-free solders for many years. Reports can be obtained from npl.co.uk/ei.

The NPL test board featured the same 96-way connector used at the "Pb-free Experience3" processed using PIHR. The NPL reliability project consisted of 145 boards with mixtures of the following alloys: SnAgCu, SnAgBiCu, SnCu, SnPbAg and a SnPb control. That test board consisted of a selection of surface-mount components, as well as the connector. To date, the samples have been through 2200 cycles of thermal testing (-55° to +125°C) with dwell times of 40 min. None of the test boards has exhibited electrical failure. A further feature of the NPL pin-in-hole reliability project involved the use of different stencil apertures in the printing process to obtain a range of solder joint volumes.

By producing solder joints with varying volumes, the NPL project permits a fair representation of joints to be produced that may be likely in production. It also covers the basic range of criteria of IPC-A-610D. As a result, this makes the final reliability results of the NPL report more meaningful when compared to the visual joint quality used in the industry.

Quantifying fill percentage by x-ray. IPC-A-610D section 7.5.5.1 indicates that the target for solder vertical fill within supported holes is 100%. The acceptable level for classes 1, 2 and 3 is at least 75% fill; anything less is deemed unacceptable. Confirming actual fill level optically of anything less than 100% fill is almost impossible, however, as one cannot see down the joint to determine a level (Figure 1). The alternative technique to qualify the joint strength, and by extension have a measure of the fill in the joint, is to pull the pin in a controlled manner and detect the force required until it is removed from the hole.

This latter method requires destruction of the sample. X-ray inspection offers a way of nondestructively investigating the PIHR joints and permits the fill percentage of these joints to be calculated. Achieving all this at an angular view is essential for PIHR investigation, as looking solely from above will not show detail through the hole (Figures 5 and 6). In addition, recent x-ray developments enable other aspects of the quality of the Pb-free process to be further investigated.^5,6 A digital x-ray inspection system was used for analysis in the "Pb-Free Experience3" tests. This system had an open x-ray tube with submicron resolution that provided 16-bit grey-scale sensitivity with an x-ray image size of 1.3 megapixels. The system was able to provide oblique angles of up to 70� at any point 360� around any position on the test board without compromising available magnification. This is achieved through tilting the x-ray detector instead of tilting the board. Software on this x-ray system measured the fill percentage of the PIHR joints.

Figures 7 and 8 show examples of how this is achieved. Cursors are placed on-screen that highlight the PTH joint and additional lines can be added at different percentage levels through the hole. In this example 0, 50, 75 and 100% levels are shown but as many extra levels as required can be added. In this example, the partially filled joint is substantially less than 50% filled and therefore would be unacceptable against the IPC610-D criteria.

Figure 9 shows an x-ray navigation map of the test board. This shows the layout of the connectors and other components; it is also used to locate the positions of any failures if found through a location rectangle superimposed on this map (Figure 10).

Results

Test boards with all four finish types and reflowed under either convection or vapor phase were examined by x-ray. Images of the connectors were taken at oblique angle views (see Figures 10, 11, 12, and 13). The plan was then to calculate the variation in fill percentage for each joint in the smaller connectors and in a representative sample of joints in the 96-way connector. It was then hoped that there might be some correlation between any variation of fill percentage in the PIHR joints and the board finish or reflow method. However, the results of all boards tested showed 100% fill for all PIHR joints of all the connectors regardless of finish or reflow method. Therefore, no measurements of any variation could be taken. Some joints showed a little more voiding than others, but this was not consistent across all joints in the same device (see the top left joint in Figure 13, for example) and specific conclusions cannot be reliably drawn.

Although the sample set was very small, all PIHR joints made were shown to have 100% fill percentage, regardless of board finish or reflow method. Therefore, perhaps the best conclusion that can be reached from this work is that if you consider using the PIHR technique as part of your production then it would appear to be a robust process and provide good results. We feel this is a reasonable comment in view of the fact that the "production line" at the exhibition was set up in a matter of hours, in non-ideal conditions, but was still able to achieve consistent results even as the board finish and reflow method changed. This consistency of result may well have been caused by our considering and implementing the following checklist for PIHR as part of the process used on the line.

PIHR Checklist

1. Are your through-hole components compatible with reflow soldering temperatures? Do they meet minimum requirements of the IPC/IEC component compatibility standards of 250°C for Pb-free?

2. Can through-hole components in packaging suitable for automatic assembly be procured? Manual assembly is always possible but automation will improve consistency.

3. Has the through-hole and lead-to-hole ratio for automatic and manual insertion been calculated? This may be required if holes are typically grouped to reduce the number of drill sizes during bare board manufacture. Lead size plus 0.010" is normal.

4. What is the standoff height of the components, where are the component standoff feet located and will they contact the paste deposit? A minimum standoff height should be 0.010".

5. Has the solder resist and paste been tested during reflow? Does it cause solder balling? It is often necessary to print paste onto the resist to obtain the correct volume of solder to fill the hole after reflow.

6. Has the stencil thickness required to fill the PTHs with solder after reflow been calculated? The following calculation is one of the basic references available: Volume of paste = (volume of PTH - volume of pin) x 2.

7. Has the stencil manufacturer been informed that through-hole apertures are required on the new stencil? Is the stencil supplier typically directed to take them out? Has the supplier been shown a connector?

8. Have changes to internal soldering standards for PHIR assembly been discussed with the Quality department and customers? While 100% fill can be achieved, positive fillets are more difficult. Joints will also look different but it is easy to meet the requirements of IPC-A-610D.

9. Have component lead lengths been specified, and can they be controlled? Lead length control is crucial and should ideally give a protrusion of 1 to 1.5 mm below the board.

10. Do you know how strong solder wave and hand soldered joints are? In fact, they are no different than through-hole reflowed joints.

References

1. Results from Hands-on Pb-Free Experience, Nepcon UK, 2003, 2004 and 2005. See smartgroup.org for more details.

2. M. Wickham, A. Brewin, L. Zou and C.P. Hunt, "Report MATC (A) 141: Code of Practice for the Use of Electronic Components and PCBs in Pb-Free Processing," npl.co.uk, April 2003.

3. SMTA (smta.org), and Smart Group (smartgroup.org and leadfreesoldering.com).

4. M. Dusek, M. Wickham and C. Hunt, "Report MATC (A) 156: The Impact of Thermal Cycle Regime on the Shear Strength of Pb-Free Solder Joints," npl.co.uk, November 2003.

5. D. Bernard, N. Hoo and D. Lodge, "Use of Digital X-Ray Imaging as a Process Control Tool for Pb-Free PWB Assembly," SMTA International Proceedings, September 2004.

6. D. Bernard and K. Bryant, "Does PCB Pad Finish Affect Voiding Levels in Pb-Free Assembly?" SMTA International Proceedings, September 2004.

This article was originally published at SMTA International in September 2005 and is used with permission of the authors.

Dr. David Bernard is product manager, x-ray systems at Dage Precision Industries (dage-group.com); d.bernard@dage-group.com. Bob Willis is process engineering consultant at Electronic Presentation Services (bobwillis.co.uk); bob@bobwillis.co.uk.

• Prev
• Next