Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Tombstoning

Tombstoning

Subscribe to Thread

What are the parameters to be adjusted to prevent this defe... - Dec 16, 1999 by Henry C. Yao

Hey Henry, Tombstoning can result from many different condi... - Dec 16, 1999 by TNT

Depending on the part say a cc1206 capacitor try a 7 degree... - Dec 16, 1999 by

Henry, TNT makes some good points. I have also found that t... - Dec 17, 1999 by Christopher Lampron

Hi Henry, amoung all things that cause this funny effect th... - Dec 17, 1999 by Wolfgang Busko

Hello Henry, Tombstoning is caused (In most cases) by a for... - Dec 17, 1999 by Mike Naddra

If i have problem with tombstone, first i would look at the... - Dec 18, 1999 by

All of the information so far has been great, I would like... - Dec 21, 1999 by RDR

Interesting thing but what actually is the "homeplate" desi... - Dec 22, 1999 by Wolfgang Busko

Wolfgang: From US baseball parlance: Home Plate. A five ... - Dec 22, 1999 by davef

Here's the picture!!! Not to say that people aren't correc... - Dec 22, 1999 by davef

I thought it might be something from baseball but I must ad... - Dec 23, 1999 by Wolfgang Busko

Hi Niel Manson of Alpha Metals U.K. did extensive research ... - Jan 07, 2000 by Mark Alder

Henry C. Yao

#8138

Tombstoning | 16 December, 1999

What are the parameters to be adjusted to prevent this defect. What are the things that I have to look at. I believe that when the components get smaller, occurrence of this defect increases. Also, there is the problem of solder paste condition.

reply »

TNT

#8139

Re: Tombstoning | 16 December, 1999

Hey Henry, Tombstoning can result from many different conditions: 1. Look at the part placement under a microscope before the board has been reflowed. Is it centered perfectly on the pads? If not, the part might migrate towards one pad during reflow. 2. If the parts are only tombstoning in certain areas there might be a heavily grounded area of the board that is causing unequal heat dispersion. 3. Paste placement could also be an issue. Pad dimensions could be different, or the paste height could be fluctuating. Some other possible solutions: 1. Increase placement pressure 2. Use a small glue dot between the pads that will cure before the solder flows. 3. Try using nitrogen in your reflow process. It aids in a more equal heat transfer.

I hope this helps.

reply »

Lloyd Nelson

#8140

Re: Tombstoning | 16 December, 1999

Depending on the part say a cc1206 capacitor try a 7 degree theta rotation.

reply »

Christopher Lampron

#8141

Re: Tombstoning | 17 December, 1999

Henry, TNT makes some good points. I have also found that the use of a 2% silver solder paste (provided that it's allowable) can help reduce the occurance. You may want to run a thermal profile on the PCB is question. Is the board thermaly equalized just prior to reflow? If not, try adjusting your belt speed down to allow longer soak times. The idea is to make all of the solder on the board liquidous simutaniously.

Hope this Helps

Chris

reply »

Wolfgang Busko

#8142

Re: Tombstoning | 17 December, 1999

Hi Henry, amoung all things that cause this funny effect the most effective measures in my experience were good PCB design (pads and equal thermal conditions for both sides of the components), placement accuracy and profiling, profiling, profiling. Look at your ramps (or slopes), try to get 2�C/S. Look at your temp. after preheat and the ramp you get when going into liquidous there the things really happen. Consider also - paste - printing - contamination (parts and PCB) - reflow system (is there any air stream involved ?) - handling - ...

M2C

Wolfgang

reply »

Mike Naddra

#8143

Re: Tombstoning | 17 December, 1999

Hello Henry, Tombstoning is caused (In most cases) by a force imbalance on either side of the component. As the solder becomes liquidous the surface tension of the liquid will pull the component to the center of the liquidous area. To solve the problem of tombstoning, you need to asalyze the root causes of this force imbalance. Are the land geometries correct in relation to the component(If the pads are too far apart and the component is not placed dead center one side of the component will see greater surface tension force and will raise up (Tombstoning).

Is the profile correct are you heating the board evenly , ramp rate, time at preheat , time above liquidous etc.- If one pad of the component goes liquidous before the other, as that second pad goes liquidous there will be greater surface tension force on the first side and depending upon amount of solderable component mass presented to the second side, the surface retention force may or may not be enough to prevent toumbstoning.

Is the solder paste uniform - how well are your solder bricks centered on the land pads.

Component placment how well are the components centered on the centroid.

Hope this helps

reply »

sin

#8144

Re: Tombstoning | 18 December, 1999

If i have problem with tombstone, first i would look at the placement of my components. if placement is 50% greater on the pad, it would be ok to reflow. even you have bad profile, you would have some other problem not tombstone.

then i look at the contamination part and pads, hardly this is problem.

then i look at my stencil spec and printer stability.

sorry can not help much.

reply »

RDR

#8145

Re: Tombstoning | 21 December, 1999

All of the information so far has been great, I would like to add one more thing to the puzzle. We have utilized a "homeplate" aperture on the stencil when the pad geometry or PCB design prohibits all of the above mentioned remedies. The hompeplate design allows for the largest mass of paste to be located directly underneath the wettable area of the component. This allows the solder to pull the part down instead of to the center of the pad or wherever it wants to go. We reduced our defect level from 16000 DPMO to 12 by doing this.

Russ

reply »

Wolfgang Busko

#8146

Re: Tombstoning | 22 December, 1999

Interesting thing but what actually is the "homeplate" design ? Is there somewhere a source where someone else could have a look at it ?

Just to satisfy my curiousity

Wolfgang

reply »

davef

#8147

Re: Tombstoning | 22 December, 1999

Wolfgang: From US baseball parlance:

Home Plate. A five sided, two dimensional, closed shape where three equal length sides form a cube with one side removed and two additional equal length sides that form interior angles of 135�, 90�, and 135�.

As you say, a picture is worth a thousand words.

Essentially, it is an aperture area reduction approach to reducing the volume of material under the component terminations.

Dave F

ps. I'll try to find a picture

reply »

davef

#8148

Re: Tombstoning | 22 December, 1999

Here's the picture!!!

Not to say that people aren't correct that a home plate didn't correct their tombstoning problem (Is that enough negatives in one sentance to total obscure the meaning er what?), but it's not the first thing I'd change. Some of the earlier comments seem more inline with the traditional approach to fixing the problem.

Dave F

reply »

Wolfgang Busko

#8149

Re: Tombstoning | 23 December, 1999

I thought it might be something from baseball but I must admit that my concentration watching this game was more on the outfield action and on trying to understand the rules. And besides, couple of years passed since my stay in US.

Yeah at first sight this might be helpful in avoiding solderballs but looking at tombstoning the other mentioned approaches should be considered more closely first.

IMHO

Wolfgang

reply »

Mark Alder

#8150

Re: Tombstoning | 7 January, 2000

Hi Niel Manson of Alpha Metals U.K. did extensive research on this subject. He can be contacted at neilmanson@alphametals.cookson.com.

Regards Mark Alder

reply »