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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Glue "decouple"

Views: 3647

#59918

Glue "decouple" | 21 September, 2009

Hi,

Our SMT line has placement capacity of 80K cph, but only 21K cph gluing capacity.

The rate between reflow and glue of our products is around 5 to 1. The line works great for the reflow boards, but when we produce glue boards the glue dispenser is a bottleneck. This is so bad that we end up using the line for reflow only 45% of the time, and 55% of the time is needed for the glue boards (representing only 20% of total volume).

In the future we'll buy a new line exclusively for glue boards, with a placement capacity balanced with the glue dispenser (around 20K cph). Meanwhile, we're considering to "decouple" the glue dispensing process (i.e. removing the glue dispenser from the rest of the line and make it run all day long). This would allow us to always run the line at its full speed, no matter the attachment process.

The problem is that we would have delays of up to 24 hours between glue dispensing and chip placement, and from what I heard this could mean trouble.

Has anyone here ever seen/tried something like this? What problems do you think we could expect?

Thanks in advance for any help on this.

Best regards, Santiago

reply »

#59921

Glue | 21 September, 2009

Why not: * Stencil print glue? * Stencil print the majority of the glued components and then use your dispense head for the tough spots?

reply »

#59923

Glue | 21 September, 2009

We don't have any experience printing glue. Our equipment supplier told us it's an ugly process, mainly because of screen problems.

We asked some other companies in the area and they all avandoned this technology because they were not able to achieve good results.

Do you think we should give this a try? Most of the components are 0603 (nothing smaller than that also)

Thanks

reply »

#59924

Glue | 21 September, 2009

What babies!!!! Sure, give it a try. * Stencil material: metal or plastic, no difference in print, metal is easier to clean, no static problem, available * Stencil thickness: Chips 0.006", SOIC 0.010" [print height equals 1/3 thickness] * For 0603, aperture size was an oblong shape , 0.030" long x 0.015" wide. A minimum of 0.010" gap between the aperture and pad, both sides. * Snapoff: to increase dot height, upto 0.040" can be used * Squeegee material: metal on metal, polycarbonate on plastic * Print speed: 1 to 2" per sec * Squeegee pressure: 0.2 to 0.3 kg/cm ... enough to clean wipe the stencil * Separation speed: Slow ... 0.1 to 0.5 mm/sec over 3mm separation distance * Print sequence: print print

reply »

#59926

Glue | 21 September, 2009

Anytime we use a glue/wave solder process we print the adhesive using a stencil and our screen printer and we have excelent results.

It is in no way a difficult process.

reply »

#59933

Glue | 22 September, 2009

The only time I ever saw problems with stenciling glue was when one know it all operator was using IPA to clean the stencil. He didn't seem to be able to understand that alcohol was a curing agent for that particular glue.

But other than that, as long as you were only talking about chips it was great.

reply »

#59944

Glue | 22 September, 2009

I think I forgot to mention a little detail here: our boards go through automatic insertion before entering the smt train. We use old machines (universal radial 3 and radial 5), so clinching is not perfect.

Do you still think glue printing is a good idea here?

Thanks

reply »

#59953

Glue | 23 September, 2009

Yes, after you move SMT to before PTH like everyone else.

I'm guessing that you only have SMT on one side of those boards? And you must be an OEM. A privately owned one?

How did you ever start doing PTH before SMT?

reply »

#59959

Glue | 23 September, 2009

Santiago can still maintain his current PTH, SMT, Reflow, Glue, Wave solder process flow and print glue with a stencil. In this, the bottom-side [contact-side] of the glue stencil requires cutting pockets that correspond with the clinched PTH leads.

Here are links to articles on this type of stencil: * http://www.circuitnet.com/experts/images/stencil_design_through_hole_smt_mixed_technology.pdf

* http://www.trafalgar2.com/documents/Technical_Articles/6.09-coleman.pdf

reply »

#59960

Glue | 23 September, 2009

Thanks to all for the great replies. I guess we'll buy a stencil for a few hundred bucks and try this.

Anyway, nobody expressed an opinion about the "decoupled" glue dispensing. What do you think about this idea?

Thanks again for your help.

Best regards, Santiago

reply »

#59961

Glue | 23 September, 2009

We apply temporary solder mask off line, using a programmable dispense head. Works fine. Purchasing a placement machine with a dispense head usually only works-out well for the sales-type. Not in every case, but usually.

reply »

#59969

Glue | 24 September, 2009

Those articles are talking about screening before PTH placement. They are talking about pin in paste. He would still have to stencil the glue before the PTH parts. How would the squeege go over the clinched leads?

And if he does get a new line I would strongly recommend visiting a CM before he does.

reply »

#59978

Glue | 24 September, 2009

Both of the articles talk about pockets cut in the contact side of the stencil that correspond to components [or PTH leads] that would allow PTH first, glue second, and finally second-side SMT placement.

reply »

Plasma Prior to Conformal Coating

ii-feed SMD Intelligent Feeder