Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Tombstoning issues!!!

Tombstoning issues!!!

Subscribe to Thread

Hi. I know that this is one of the mos common threads her... - Apr 09, 2008 by Manuel_R

what is your conveyor speed? stencil thickness? ... - Apr 09, 2008 by kircchoffs

conveyor speed: 30 in/min. 8 zones stencil thickness 6 mil ... - Apr 09, 2008 by Manuel_R

Is the stencil opening rectangular? Home Plate? What is the... - Apr 09, 2008 by elias

openings, rectangular no home plate. pad spacing don't have ... - Apr 09, 2008 by Manuel_R

We wonder about: * Solder mask encroaching onto the compone... - Apr 09, 2008 by davef

Stencil design and pad layout are key. Too far apart - bad.... - Apr 10, 2008 by Dirk Nuendyke

If I read you right the issue is mostly with components from... - Apr 10, 2008 by ScottE

Umm.... he states he is using the Induium paste for 2 weeks ... - Apr 10, 2008 by Real Chunks

Update: Recently, we ran a (old) product on which they ma... - Apr 14, 2008 by Manuel_R

Many factors will make tombstone happen. from design/stencil... - Apr 22, 2008 by fowlerchang

Here is my email as below: ... - Apr 22, 2008 by fowlerchang

It is CSP reliability issue, which is the mobile CPU from Sp... - Apr 22, 2008 by fowlerchang

Manuel_R

#54323

Tombstoning issues!!! | 9 April, 2008

Hi.

I know that this is one of the mos common threads here, I have been doing some research, but i didn't find any similar.

The main issue is that I have seen tombstoning since two weeks ago, it is more characteristic on 0603 resistors (Stackpole electronics) also another board has 0402 with tombstoning also, but is more at 0603; i'm observing this issue on my three SMT lines. The first questions that I made when see tombstone are:

Is the printing aligned? - Yes it is centered, ok.

Is the component centered? - Yes, not significant coorinate displacement.

Is the right profile? - For solder paste Indium NC-SMQ92J. 0.9�C/sec ramp up, 20 sec hang between 175-180, peak 220�C. this is the "optimized profile" recommended from indium for tombstoning.

Might the component/board contaminated? - is on several resistor Part numbers, several references random. I tried changing date codes on components, not positive result.

Is the solder paste ok? - during the last two weeks, we have been using the SAME SOLDER PASTE, with the difference of the packaging... it came with "proflow cassette" presentation, instead of regular 700gr cartridge. I bring indium here to see what can be wrong, he brought me few solder paste from other lot. checked the same variables that I mentionet before. Everything OK, changed solder paste and tombstone reduced, but not eliminated... that was last friday. today (wednesday 09) I observe the same issue again on the three lines again.

We're not using the proflow cassete packaging anymore since we have "regular" cartridges, lot# is different, process still the same.

My thought is that the solder paste is the one that is causing us the issues. Can you please tell me what else can I verify, or any other experiment to do in order to see the variable (or variables) causing the issue? I cannot (or maybe can, only for troubleshooting purposes) test with other solder paste due to customer requirements (automotive).

Issue observed? Tombstoning

Where? mostly 0603 resistors (stackpole) randomly. on three different production lines.

solder paste? Indium NC-SMQ92J (63-37 90% metal load)

Reflow profile? Indium "recommended" 0.9�C/sec ramp up, 20 sec hang between 175-180, peak 220�C.

Equipment? printing-MPM UP3000/Hi-E ... Placement-CP6/CP4/IP3 ... Reflow - Vitronics Magnatherm 820N/Heller 1800.

Thank you in advance!!!!

reply »

kircchoffs

#54324

Tombstoning issues!!! | 9 April, 2008

what is your conveyor speed? stencil thickness?

reply »

Manuel_R

#54325

Tombstoning issues!!! | 9 April, 2008

conveyor speed: 30 in/min. 8 zones stencil thickness 6 mil for most, for the one with 0402 5 mil (due to a 20mil pitch QFP) laser cut.

thanks.

reply »

elias

#54326

Tombstoning issues!!! | 9 April, 2008

Is the stencil opening rectangular? Home Plate? What is the spacing between the pads on the stencil for the component?

reply »

Manuel_R

#54328

Tombstoning issues!!! | 9 April, 2008

openings, rectangular no home plate. pad spacing don't have data right now. the problem is those products have from 1 year to 5, running constantly whitout problems and observe the issue around 3 lines no matter of the product.

reply »

davef

#54329

Tombstoning issues!!! | 9 April, 2008

We wonder about: * Solder mask encroaching onto the component pads. Please describe the solder coverage on pads where there is tombstoning. * History of the issue. How long has this been going on? Have you ever produced non-tombstone board consisently? * Breadth of the issue. You appear to say that you have this issue on multiple products. Is that accurate? * Heat uniformity. What happens when you reflow boards rotated 90* from the normal dirction?

reply »

Dirk Nuendyke

#54334

Tombstoning issues!!! | 10 April, 2008

Stencil design and pad layout are key. Too far apart - bad. If your pad geometry is correct, try the inverted home plate. That has worked well for me in the past. That, coupled with a 5-mil stencil.

With regard to solder paste, some manufacturers like Henkel-Multicore and AIM, make an "anti-tombstoning" formula consisting of Sn62Ag2Pb36. The 2% silver will have the effect of a "plastic range" (liquidus of 179�C) to mitigate the tombstoning.

See below chart from Alpha's website for aperture design ideas: PS-good job SMTnet on giving us the ability to post pictures!

Attachments:

reply »

ScottE

#54336

Tombstoning issues!!! | 10 April, 2008

If I read you right the issue is mostly with components from one supplier. I'd suggest looking at the components under a microscope. Also, if you have some old pre-tombstoning parts from this same vendor take close-ups of both and compare them side-by-side. We once had a chip supplier increase the plating thickness and we got a spike in tombstones. The thick plating made the terminations rounder than the older parts, thus easier to rock up and tombstone. I seem to remember we also had a problem once when a supplier changed manufacturing locations and simultaneously changed the plating. The change in plating made the parts less solderable, leading to inconsistent wetting and an increase in tombstones.

Good luck.

reply »

Real Chunks

#54339

Tombstoning issues!!! | 10 April, 2008

Umm.... he states he is using the Induium paste for 2 weeks and experiencing problems. The problem percentage drops when he changes lot # of paste. Problem occurs on 3 different lines (assuming 3 different product). What's the common variable on all of this? The paste!

Try a different paste! Take only minutes to perform and vendors will give you samples for free!

Jeez, asking him to change pad designs of product that have been running for years?

reply »

Manuel_R

#54381

Tombstoning issues!!! | 14 April, 2008

Update:

Recently, we ran a (old) product on which they made a new stencil, appertures 1-1 on 0603. As soon as they began to run, tombstones appear. I went to trashcan to rescue old stencil, and noticed that pads came with a "c" shape reduction, I asked to put the stencil just to run a few boards and voil�! tombstoning desapear!!!

Now the dilema is the followining, I know that excesive solder on pad (and non optimum pad design) are a contribuitor for this issue, but there's other factor that trigger the problem because I have a lot of products withou apperture reduction on my stencils and ran without this issue before. by now I'm gonna begin to order stencil with apperture mods and review pad design on the coming products.

As all the people mentioned on their posts (thanks a lot) "material" is the main contributor, might be solder paste, component plating or HASL on board. one of this (or more than one) togheter with excesive paste is creating this issue, but the good new is that i can contain with stencil mods.

I will keep you posted to see what the "X-factor" is.

regards!!!

reply »

fowlerchang

#54456

Tombstoning issues!!! | 22 April, 2008

Many factors will make tombstone happen. from design/stencil aperture/soler paste/placement/reflow and so on. here is one article about tombstone which I prepared long ago. send me your email if you need it.

This message was posted via the Electronics Forum @

reply »

fowlerchang

#54457

Tombstoning issues!!! | 22 April, 2008

Here is my email as below:

fowlerchang@rayval.com

This message was posted via the Electronics Forum @

reply »

fowlerchang

#54458

0.5mm CSP reliability issue(Sorry that I can't use new thread) | 22 April, 2008

It is CSP reliability issue, which is the mobile CPU from Spreadtrum.

All the tests passed and shipped to customer. They assembly them and tested but failed. After they reflow this CSP with hot air gun, some of the boards passed.

We got some boards to do cross section and dye penetration. No evidence of solder joint crack were found after failure analysis.

This CSP is 0.5mm pitch without underfill process. It is a lead free component, and we try them with leaded and lead free solder paste. But same problem happened.

We doubt whether it is component or design problems. But we don��t know what��s the next step.

Thanks! fowlerchang@rayval.com

This message was posted via the Electronics Forum @

reply »