Because of the low stand-off of QFN components (around ½-1 mil) common on printed circuit boards (PCB), it is very difficult to get enough contact time with a cleaner to break through the flux dam and break down all the residues. For proper cleaning, the following impacts effectiveness of the process: solvent match, contact time, and agitation.
As a standard service we provide customers, Techspray’s lab (Techlab) evaluated this challenge for a mobile device manufacturer. They were having trouble cleaning RMA tacky flux from under QFN using a solvent aerosol. There was no question Techspray product, G3 Flux Remover (part #1631-16S), was able to solvate the flux, but we wanted to evaluate different techniques to find the best for cleaning under such a low stand-off.
G3 Flux Remover cleaning effectiveness was already established with a static cleaning test. The soil, in this case an activated RMA tacky flux, is soaked in room-temperature for 30-seconds without agitation. If the flux residue is fully broken down and running off the substrate, solvency of the soil is established.
G3 Flux Remover has low surface tension to allow it to flow under tight stand-offs. The high pressure spray of the aerosol can provide contact and agitation, so it is a matter of finding the right impingement angle, spray position, and spray duration. These are the three variables that were tested.
Testing Manual Cleaning Protocols for QFN Components
Multiple process variations were tested to verify the most critical parts of the process:
- Board angle (A in fig 2) was constant at about 30°
- Final flush was constant, but tested at 3 and 5 sec.
- Spraying 4 sides of component, 2 adjacent sides for longer period of time, and diagonal
- Different spray times: 3, 5 and 10 sec.
- Straw angle (B) at 10° and 30°
- Oscillating straw or centered
- With and without a pre-rinse
The standard TechLab test vehicle (fig 1) was used with only the 2 QFN slots populated (as marked on fig 1). Because the customer’s QFN did not have central ground pad, that aperture was masked off the stencil. Kester FL250D paste was used for stenciling, and Alpha R100 liquid flux added under component. Reflow was achieved using a hot air work station set at 450°C. Techspray’s G3 Flux Remover (1631-16S) was used to clean the boards with straw attachment. For evaluation, of cleanliness under the QFN, they were mechanically (and forcefully!) removed.
Fig 1 – TechLab PCB
The best results were met with the following cleaning process (ref fig 2):
- Board angle (A) - @30°
- Straw angle (B) - @10°
- 10 sec spray on 1 side of component, then immediately spray the adjacent side for 10 sec
- Slightly oscillating straw along the side of the component as it is being sprayed seems to improve performance.
- Follow up quickly with a 5 second rinse – spraying over and around component while holding board at an angle. This helps flush out loose residues still suspended in solvent.
The above process cleaned most of residue (fig 4-6), and with additional spray time, should reliably clean all residues.
Fig 2 – Aerosol positioning
Fig 3 – control / uncleaned QFN & pad
Fig 4 – Cleaned pad U2
Fig 5 – Cleaned pad U1 (note flux residues)
Fig 6 – Cleaned underside of QFN from U1 (note flux residues)
Electronic assembly, rework, and repair of high reliability PCBs is particularly challenging on the benchtop, where it is more difficult to establish repeatable processes. There are steps you can take to develop standard cleaning procedures that will better ensure effectiveness. Techspray’s lab is available to help you qualify products, establish cleaning procedures, or diagnosing cleaning issues. For more information, go to www.techspray.com/techlab
or call 800-858-4043.