Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

Glue Dispense Process Control



Glue Dispense Process Control | 10 August, 2001

We are an OEM automotive electronics supplier and began using SMT in our products a couple of years ago. We use a glue and wave process, and one of our main issues has been our glue dispensing process. We have a Camelot 3800 machine, and we have been controlling the process by controlling the ambient temperature in the machine cabinet. This seems to work well, but we still seem to have issues with the viscosity of the adhesive changing from lot to lot. We're using Loctite 3615. Has anyone else experienced this? Can you recommend a similar product from another supplier? Should we measure and track dot size statisticly? How do you measure this parameter?

Thanks in advance, John

reply »


Glue Dispense Process Control | 10 August, 2001

John, Instead of dispensing the adhesive, why not screen print the adhesive on the PCB? The deposition is very repeatable and accurate.

reply »



Glue Dispense Process Control | 10 August, 2001

Hey John,

You're not alone, a lot of people have the same problem with different machines and material. What is your specific problem, just keeping SPC control? Or are your defects at the wave from the glue too?

If it's a process problem, I found that by placing about 5 dots on an empty spot on the board or break away, gets the machine going pretty good before it has to place a dot that a component is being placed into. Another trick to avoid the lot to lot problem is to get as big of container of material as possible. Sometime changing over to the 5 gallon can of material that gets pumed to the head via high pressure hose is the way to go. Depends on your thru-put.

If it's an SPC problem, I've used the video micro-scopes and test boards before. This works well, and you can actually mease the dot size from a 2-dimensional view. You can measure right on your production boards or have special placement boards made and you can wipe the glue off and use em' again. Also, you can keep stats after wave to see if your dots are being placed properly center. Measuring these two things should keep you on track.

reply »


Glue Dispense Process Control | 10 August, 2001

When you say that glue viscosity changes for lot to lot, is this LOT TO LOT of your product OR Loctite�s product? And when you say that glue viscosity changes, what do you mean [eg, increase by � , decrease by � , er watt?]. What specifically is the problem? Viscosity of the adhesive is usually blamed for stringing problems, but in most cases, improper hardware setup is the culprit. Viscosity impacts dot size far more than it affects stringing.

PROCESS CONTROL: Several points to consider are: * Put an adhesive witness mark on every board. [Check the fine SMTnet Archives for more.] They fun, cheap, and take no time to do. * Glue likes to be shelf-life and environmentally controlled. Check with your supplier for recommendations. * SMT Adhesives as a collective group are hydroscopic. An increase in humidity in your facility will lower the viscosity of certain materials. * Recognize that even though you�re controlling cabinet temperature, changes in the room temperature reflect into the cabinet. * Adhesive viscosity varies within a single cartridge. How can this happen? The shear frictional forces that are applied to the adhesive as the piston moves down the cylinder will cause the material to heat up. Many dispensers offer heated boots around the nozzle to maintain a temperature above ambient and balance the viscosity as the piston travels. But adhesives are cured by heat - the cross-linking process is initiated in the nozzle. [C Shea] * Shot sizes must be changed as the piston travels down the cylinder. The medium applying pressure to the piston is air - a compressible gas. The shot size of air required to move the piston and extrude a set volume of adhesive is smaller in a full cartridge than it is in a half-empty or nearly empty cartridge. As the cartridge empties, shot sizes should be increased. Smaller cartridges (5 or 10 cc) are more forgiving than larger (30 or 50cc) cartridges. [C Shea] * Purge the nozzles regularly. As the adhesive in the nozzle cures, it reduces the size of the nozzle, eventually violating the 50% diameter rule. A quick purge, or continuous flow of adhesive through the nozzle, will remove most of the buildup inside the nozzles. Purge nozzles after as little as fifteen minutes of downtime or after two hours of production. [C Shea] * Keep the nozzles clean. A fresh nozzle every day helps to guarantee the proper diameter. Nozzles can be cleaned in simple bench top ultrasonic cleaners with approved solvents. The best practice is to have two sets of nozzles, so one can be cleaned while the other is in production. This method limits downtime to a few minutes for nozzle changeover. Removal and cleaning at the start of the shift also provides an opportunity to inspect the nozzles. Any dents or dings in the nozzle change the diameter and affect the glue dots. [C Shea]

MATERIAL: Loctite 3615 is a fine material. Look to solve your in-house problems before stepping down on material.

MEASURE & TRACK DOT SIZE WITH STATISTICS: Sure, the witness mark gives you that option. Using an optical measuring device like a microscope reticule, the process engineer or technician can measure dot sizes and implement a program that calls for standardized hardware, shot size increases at certain fill levels of the syringe, and set intervals for cleaning.

MORE: Check the fine SMTnet Archives [both the Forum & papers (whatever they call that)]. I want to say that Chris Shea wrote a couple papers on dispensing.

reply »

Thermal Interface Material Dispensing

Cost-effective Conformal Coating Machine