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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

uBGA (.5mm pitch) printing woes

Views: 6513


uBGA (.5mm pitch) printing woes | 6 September, 2017

Hi All,

We are being intermittently challenged by a .5mm pitch uBGA paste release issue. We don't build with many of these so don't have a lot of history here.

We've gone through 3 different stencil designs and the most consistent performer was 4 mil thick with round 12 mil apertures (pads are 11 mil). At one point during testing we ran 31 straight panels (10 uBGA/panel) without a paste release issue. This is stainless steel stencil of standard grain size material with no coating or other secondary treatment.

We started running it again this week and had more problems. Moved the stencil to a different line and performance improved but is not what I would call acceptable. The defects are randomly distributed on the panel.

My question is, how many of you bit the bullet and went with more expensive, finer grained materials to solve this problem, how many are having a coating applied by your stencil house, and how many of you have great results without either of these "upgrades"?

Thanks in advance!


reply »


uBGA (.5mm pitch) printing woes | 7 September, 2017

we use 0.12 mm stencil with 0.3 mm apretures for 0.5mm step FBGA153 memory without problems

the solder paste is SAC305 Class4

reply »



uBGA (.5mm pitch) printing woes | 7 September, 2017

In my experience with similar sized parts, the nano coatings (or what ever your supplier might call them) do help with repeatable paste release.

How are you cleaning your stencils? A big red flag to me is that your stencil worked fine for the first run, but had issues on the next. If the apertures aren't clean, it might impair paste release.

reply »


uBGA (.5mm pitch) printing woes | 7 September, 2017

Chrys Shea has published extensively on stencil materials, coatings and paste release mechanisms. Many of the papers are available to download on her website. Just search for Shea Engineering Services. I've worked with Chrys in the past, she is definitely an expert worth consulting.

reply »


uBGA (.5mm pitch) printing woes | 7 September, 2017

HI Tsvetan,

Can you tell me what material your stencil was fabbed from?

I have seen (from different suppliers) references to at least 3 different grain sizes of stainless for laser cut, plus electroformed nickel for laser cut, electroformed nickel, nano coatings, etc..

I'm trying to determine if the advanced materials and coatings are the right solution if all other design parameters are exhausted, or if we have another issue. Thanks!

reply »


uBGA (.5mm pitch) printing woes | 7 September, 2017

That is a good point, and was something that we discussed yesterday in our process eng. meeting.

I don't have an answer right now as to the current cleanliness but I know they were washing (U/S) the stencil repeatedly during the previous run and inspecting under 40X when it came out of the washer.

reply »


uBGA (.5mm pitch) printing woes | 7 September, 2017

Two magic bullets help us greatly with fine pitch printing: 1) nano coatings (we use a DEK product) YES it really does work 2) temperature control in the printer (AC unit).

Of course good quality stencil and proper cleaning of the stencil are important, too. You might consider having a set time-frame where you pull the stencil and clean it.

reply »


uBGA (.5mm pitch) printing woes | 8 September, 2017

The material is 0.12mm stainless steel, electropolished it's thick for fine pitch but we have lot of other big components on the board which do not solder good if we use 0.1mm thickness stencil.

The 0.3 mm apretures are big enough and have good release even if there are paste remains on the apreture walls after some use, the amount of paste which is released is good enough for reliable soldering. If you look at the paste after printing with camera it's with trapezoidal shape instead of cylinder shape.

reply »


uBGA (.5mm pitch) printing woes | 19 September, 2017


The 4 things I would do, and in this order:

1. Clean your stencil with stencil cleaner wipes after it comes out of the stencil cleaner. I like MicroCare ProClean (MCC-PROWR). I will even saturate the wipe with a bit of isopropal just till the wipe is damp, at least on the 1st pass. I wipe the bottom vigorously for 20-30 swipes forcing the solvent up through the apertures, then wipe the top once from left to right. If the wipe has any "dark" material at all on it i repeat the process till it does not. The stencil must be absolutely clean for these types of parts. Trusting in your wash blindly will not cut it.

2. Make sure your paste is fresh. It is not usually a problem this time of the year, but I don't know where you are at, so it bears mentioning. I have seen good paste added to dried out paste in an effort to "save" money. Now we are throwing out 2 jars of paste instead of one. If paste is dried out, toss it. Reliability of your circuit cards is greater than the cost of a jar of paste.

3. Environmental conditions. Temperature and humidity, at least inside of the screen printer, can affect print quality. Read the TDS for whichever paste you use. Meet or exceed the recommendations.

4. Nano coatings. If you need them, then you need them. You should at a minimum be using quality stencil squeegee blades that are coated. This is such an easy thing to do, and it makes a small difference. Sometimes a small difference is all you need to make an iffy print into a good print. I like Permalex blades from Transition Automation. There may be others that are better, i wouldn't know. These guys product is so good I haven't bothered to shop around. Replace them as suggested. I think they recommend 6 months under heavy use, and 1 year under normal use. They will call and make sure you are happy, and remind you when it is time to replace them.


reply »


uBGA (.5mm pitch) printing woes | 19 September, 2017

Thanks, all, for your responses. It's always helpful to know what everyone else is doing, whether to validate your own experience or to force you think in more detail.

At this point we've determined that no amount of aperture design, material control, environmental monitoring, printer setting adjustments (all of the basics) were going to have any measurable affect.

The most effective thing we did with what we had to begin with was optimize the setup with a little better support placement, but ultimately it was a nano-coating that got us over the hump. The only negative effect in using that was that we got a little too much paste in another location (if we had auto SPI I suspect we'd be seeing better than 100% TE) but we think we've solved that problem with a reduction in blade pressure.

The next step will be to try a finer grained stencil w/o the coating to see if we can get around adding a day to stencil lead time and concerning ourselves with stencil washer detergent concentration more than we'd like to.

reply »


uBGA (.5mm pitch) printing woes | 22 September, 2017

Just for the sake of passing along the information, the fine grained stencil provided zero improvement over the standard PhD stainless and was not nearly as helpful as the baked-on nano-coating. We probably won't be trying that again unless we go under .5mm pitch and also plan to coat it.

reply »

pcb components vacuum pick up

IPA Alternatives