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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Tombstone caused by flux residue

Lloyd

#35351

Tombstone caused by flux residue | 7 July, 2005

This problem exists at a very low level but happens on all our discrete components, all our products and in two plants. Between the PCB land and the component termination there is a barrier of flux residue, this actually lifts the component slightly at one end making it look like a very minor tombstone and difficult to catch with AOI. Oven profile (RSS peak 215C) well with in the normal range and placement etc all fine. Paste manufacturer suggesting component terminations but if I touch it with an iron it wets fine and it happens on several different component manufacturers! Any suggestions as to what is causing this? It is now becoming a pain. Paste RP11 PCB Finish HASL REHM ovens with cooling zone

reply »

#35355

Tombstone caused by flux residue | 7 July, 2005

Sounds like you need to turn-up the dial of your oven. Solderabilty protection finishes on components are changing, whether it makes sense or not.

reply »

sarag

#35366

Tombstone caused by flux residue | 7 July, 2005

Hi Lloyd, Have you done any ionic analysis to find out what the flux residue is composed of? If you can eliminate the source residue, then your problem obviously should go away. Though tombstoning isn't known to be much of a functional problem it is sure still frustrating and visually unappealing!

Regards, Sara Gorcos www.residues.com

ps - if you have a picture of the tombstone, you can send it to me and I can post it on my website so that others can see what you are talking about.

reply »

Inds

#35370

Tombstone caused by flux residue | 7 July, 2005

how fast are you ramping your profile..and how long are you soaking it.. are you giving enough time for the flux to get activated...

This message was posted via the Electronics Forum @

reply »


RDR

#35372

Tombstone caused by flux residue | 7 July, 2005

Sarag, FYI tombstoning is a functional failure. Tombstoning is where one end of the component is not soldered and is in the air above the board/solderjoint surface. You may be thinking of billboarding (part on its side with both ends soldered down).

reply »

sarag

#35373

Tombstone caused by flux residue | 7 July, 2005

Ahh. sorry for the foul up - and thanks for stopping the spread of misinformation!! Sarag

reply »


RDR

#35375

Tombstone caused by flux residue | 7 July, 2005

What size discrete components and have you evaluated your pad geometries? If pads are too far apart this will happen. The flux resudue is simply there because you didn't solder the component. It will stay on top of the solder bump. You can see this by looking at unloaded but printed locations if you have any on your boards. I have never heard of RP11 paste, is it used at your other plants? I am looking for the commonality between the sites be it either the paste or design.

reply »

#35503

Tombstone caused by flux residue | 14 July, 2005

Folks,

Thanks for all the info and suggestions but let me add a bit more to the pot.

Our products work in a harsh environment and are classed as safety critical. Customer returns are our biggest concern, we don't get very many and feeling here seems is that if we don't change anything then it will stay that way. My question is how can I solve problems when my hands are tied so much? The fault mode is so small that running sample runs is not feasible. Any way: Some bright spark has included the oven profile in our customer specification (peak temp, heating & cooling rates and overall times), this means I can't increase the settings as I would like to without having to get the whole product re-approved by our customers, nice eh!

Oven profile is as follows: Heating rate of around 3 Deg/Sec, and soak time of about 2 minutes and cooling rate about 3.5 Deg/Sec.

The components in question are 0603.

I haven't looked in detail at the PAD layout as I again come up against the customer specification, not to mention the fact that our product is RF based and layouts are very precise. What�s the norm for 0603 components?

RP11 is an old Multicore paste (15 years), and again is in our customer specifications. I'm trying to get a new paste qualified for new products as they come in but I'm still left with the problem on our existing products.

Is there any way I can post a photo in the Forum? I've sent one to Sarag, but I'm sure if you guys could see the problem it would become much clearer.

Thanks for the interest in my problem

reply »

saragorcos

#35509

Tombstone caused by flux residue | 14 July, 2005

Hello Lloyd, I have posted your picture in my photo library for all to see - you can find the page at http://www.residues.com/circuitree_pictures.html

Hopefully that helps!

Sara

reply »

Bman

#35522

Tombstone caused by flux residue | 14 July, 2005

Considering all the constraints you are faced with, I would try changing the apatures for that part on the stencil. It seems as though a no-clean "homeplate" appature would put less paste on the board. Less paste means less flux, and it might be enough to prevent the tombstoning. I'm sure some of the SMT masters will chime in with better ideas, but that's what I would try.

reply »


RDR

#35538

Tombstone caused by flux residue | 14 July, 2005

Lloyd, could you get us a pic from the side? It looks to me as if the pads may be too small or far apart, but I can't tell. We normally use a 35 x 35 pad size on .07 centers when we have the luxury of designing them that is.

reply »


RDR

#35539

Tombstone caused by flux residue | 14 July, 2005

Can flux really prevent wetting? This seems odd to me. I am actually thinking the opposite of Bman, I am wondering if using a homeplate with an overprint condition might help?

Russ

reply »

skandaae@vsnl.net

#35549

PROBLEM IN ONE COMPONENT (0603) | 15 July, 2005

PLEASE SUGGST ME FOR A PROBLEM CAUSED DURING MY PROCESS

In one of the product I am faceing one component tilt after tht reflow oven

1. check the thkickness of the paste found the paste was more reduse the thickness by adujesting the screenprintg snap of reslut was 2 boards was fine again the same problem repted.

2. The measureing the components we came to know the thicknes was wearing do this any of the above problem cause the components to tilt this is abosered for only one component.

please comment to slove the problem.

reply »

Brian

#35565

Tombstone caused by flux residue | 15 July, 2005

We have seen this problem in the past and had to deal with similar customer constraints. It's not the flux causing the wetting, but the placement and pad geometries as some have mentioned. To reduce or eliminate the problem, we have changed the apertures on the stencil to reduec the spacing undeer the component and underprint a little more paste under the part. we have also used the homeplate design with this method to keep solder balls to a minimum.

Good Luck, Brian

reply »

#35589

PROBLEM IN ONE COMPONENT (0603) | 16 July, 2005

skandaae: By "tilt" do you mean that one end of the 0603 has lifted from the board surface? If so, we call this "tombstone".

In general, tombstoning is caused by an imbalance in temperature or forces on the component. * Convince yourself about the possibility of a temperature imbalance by running the board in the opposite direction in oven and observing any change in the direction of tilt. * Determine if this is a component problem by placing a different 0603 component [using the same program] and reflowing the board. * Investigate the possibility of imbalanced forces by: -- Comparing the pad layout of this component with the pads of other 0603 on the same board [sometimes a wide spacing between pads will cause this] -- Inspecting the paste pring process to assure that equal amounts of paste are applied to both pads. -- Reviewing the accuracy of the component placement to assure that the component is positioned between the pads.

Consider searching the SMTnet Aechives for previous discussions on trombstoning.

reply »

lloyd

#35723

Tombstone caused by flux residue | 25 July, 2005

Folks thanks for all the info, I'm gonna check out our pad geometries and see how they compare with industry standards.

Brian, could you elaborate a bit more on the stencil design you have mentioned. Do you mean that you have offset the stencil apertures so the print is slightly off the pads (therefore closer together) and then added the homeplate design?

reply »

Brian

#35732

Tombstone caused by flux residue | 25 July, 2005

Exactly. We bring the paste apertures closer together, so that the part sits more in the center of the paste. As a CM, our customers do not always accept our recommended design changes, so we use this method to reduce our tombstoning if the pad geometries are wrong. It does help.

Brian

reply »

IPC Training & Certification - Blackfox

Fluid Dispensing, Staking, TIM, Solder Paste