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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Solid Solder Deposit

AJ

#29813

Solid Solder Deposit | 3 August, 2004

Hi All,

I have a question for everyone regarding SSD - Solid Solder Deposit. I have read many good articles on this "new" process and was wondering why it has not caught on. It is better known in the US as SIPAD or Precision Pad Technology. Here is a link that explains the technology: http://www.sipad.net/Circuitnet%20Exclusive.htm Most of the articles I have come across are by Matt Kehoe. Why has this technology not caught on? Or am I just a nieve intern? Your (unbiased) input would be greatly valued. Thanks for your time.

AJ

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »


RDR

#29815

Solid Solder Deposit | 3 August, 2004

It is great for single sided boards, but you will have to use a glue/wave process or something else for second side assemblies.

Russ

reply »

celldude

#29823

Solid Solder Deposit | 3 August, 2004

SSD can be used on both sides of the pcb. Although , it is not very logical to use it on the bottom side of a pcb that has numrerous Through-hole components, that are to be wave soldered. There are many advantages to SSD. The amount of solder is much more consistent as opposed to paste. Also there is less room for operator error due to accidentally wiping paste off of pads. Also, The sticky flux that is applied to the ssd can sit out much longer than solder paste and not dry out.

reply »

celldude

#29824

Solid Solder Deposit | 3 August, 2004

I have heard the reason it has not caught on is because the SSD machine used i nthe process is very expensive, and there is only a few in the U.S. The company i work for has our own SSD Machine. We SSD 75% of our boards we build, and found it is much better to work with than paste.

reply »

Haran

#29870

Solid Solder Deposit | 5 August, 2004

What is the major difference between the SSD and the normal Solder Paste Printing?Appreciate if could provide more information.

reply »


RDR

#29894

Solid Solder Deposit | 5 August, 2004

What happens to the flux and pads on the second side of a double sided board during the first reflow? It seems to me like the solder would dome and the flux would be used up and migrated to other areas.

Russ

reply »

celldude

#29899

Solid Solder Deposit | 5 August, 2004

During the reflow of the first side, the solder on the other side does not reflow. With this being true, when you reflow the second side, the components already reflowed on the first side do not fall off. The sticky flux is applied one side at a time. (Example: ssdbtm side->wash->ssd top->wash->sticky flux bottom side->place smt components->reflow->wash->sticky flux top side->place topside smt components->reflow->wash) ssd is basically the process of applying solder paste to the pcb with a stencil and the solder reflowed and flattened out. Our company does this all in-house so we do not have to apply our sticky flux (which is water soluable) until right before we run our board through the pick and place machine.

reply »

AJ

#29921

Solid Solder Deposit | 6 August, 2004

Celldude,

What kind of parts do you use for SSD? More specifically does your company use 0402, 0201, BGA's, etc? Also are the results of using SSD better than the normal process of applying soder paste? Also if anyone else uses SSD, what are your experiences with it? Thanks a lot.

AJ

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »


RDR

#29925

Solid Solder Deposit | 6 August, 2004

I see how that works in your shop, I was thinking of when you have someone else do it for you and then both sides would have to be done at the same time and then double reflow isn't very viable with out glue.

How do you flatten the solder after reflow? How do you apply the flux? I like the SSD process and would like to investigate bringing it into our shop so we could do it one side at a time and enable us to do the double sided boards

Russ

reply »

celldude

#29942

Solid Solder Deposit | 7 August, 2004

We use ssd on all components, from 0201 to very large fine pitch and bga's. With some very small micro bga's , if there is a problem with bridging, we will tape off the stencil during the ssd process and just use bare pads...along with the solder from the BGA. During the SSD process, the solder paste is applied with a 4 to 6 mil stencil, and flattened with an air knife that is part of the machine itself. we apply the sticky flux with a 4 mil stencil.

reply »

AJ

#29950

Solid Solder Deposit | 9 August, 2004

celldude,

I have a couple more questions for you. Where can one look into getting these machines necessary for SSD? Also, knowing that the machines are expensive..Does your company find they are saving money (in the long run) by using SSD? More specifically, do you find that there is less rework involved? Lastly, if there are any articles that you have come across related to SSD that you found helpful could you shoot them over my way? My email is: Aalap.Jha@EmersonProcess.com My co-op rotation is over Aug. 27th, hopefully by then I can convince my supervisor that SSD is the way to go! Thanks for all your help!

AJ

reply »

celldude

#29956

Solid Solder Deposit | 9 August, 2004

We have definitely seen money saved. There is a lot less rework after reflow. Money is also saved in overtime. for example...when we are building a first article board with many placements, it is not always possible to get the first board through the machine in an 8 hour shift, therefore the operator needs to work OT to finish it before the paste dries out. With SSD, it can be finished the next day without worries. The sticky flux is good for days. Also there is less money wasted in unused paste that dries out. Those are just a few i can think of off the top of my head. I do not have any access to figures or statistics to support these statements. I also have no idea where we obtained the machine. Our company is very tight lipped about the machine. I do get to work with the boards that are SS'D, but i do not get to see the process, or the machine itself. Im at the bottom of the corporate ladder. Sorry i could not be of any further help.

reply »

AJ

#29958

Solid Solder Deposit | 9 August, 2004

Thank you all for replying and helping me out. Obviously there is much more information to be found on this technology. Before I leave for school I will post all the info (links to articles and other bits of info) I have gathered on SSD, for you guys. Hopefully people keep replying to this thread to give their experiences with SSD.

Regards, AJ

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

#29970

Solid Solder Deposit | 10 August, 2004

Hello all,

Wow, I cannot believe that I had missed this forum thread. Maybe in reality is was for the best as all of the excellent questions and answers about SIPAD and solid solder in general have been very accurate and positive, reinforcing our commitment to this exciting technology.

Now that I have been made award of this thread however, I feel compelled to add my 2 cents to the discussion so, please read on. A formatted version of this will be posted at http://www.sipad.com/ssd_forum.htm

Is anyone using/tried this product/process in production?

There are over 100 current customers using SIPAD ssd today including several of the largest CEM in the world, NASA, and many other large and small companies.

If so what advantages/disadvantages have you discovered?

Perfect soldering results, less cleaning, less voids, hand place BGA's and LLP's, via in pad filling, more at http://www.sipad.com

What's the pricing like compared to standard PCBs?

Pricing is set and based on UPM's, Units Per Minute, in the shop. Most boards are priced between $4.00 to $40.00. 1-5 pcs cost $195.00 lot charge (approximately $40.00 ea) and once the 300 pc quantity is reached, unit prices drop as low as $3.95, ($4.00) ea.

We run very high mix of products on the same line and am curious as to whether it is worth investigating as part of a changeover time reduction project.

Absolutely yes!

(http://www.smtnet.com/Forums/Index.cfm?CFApp=1&Message

I have a question for everyone regarding SSD - Solid Solder Deposit. I have read many good articles on this "new" process and was wondering why it has not caught on.

One of the things that has created the impression that SIPAD ssd has not caught on is the reality that it is a niche. Some assemblies will not realize the cost savings to offset the cost of adding ssd. It depends on the assemblers capabilities and equipment. If an easy board is being assembled with older antiquated equipment it can benefit from ssd however, if the same easy board is being assembled using sophisticated equipment then the yields will be higher and the cost for ssd would be hard to justify. Additionally, many sophisticated assemblers realize that the cost for ssd is justified because it can do things that traditional paste and place technology cannot including helping with vi in pad technology, eliminate voids in BGA's, allow for hand placement of BGA's and fine pitch components etc. This means that ssd will not be a revolution completely replacing traditional assembly methods but, has found its way into many seembly environments where it has shown tyo be without equal.

It is better known in the US as SIPAD or Precision Pad Technology.

SIPAD ssd is the original ssd developed at Siemens in Germany. PPT is a second ssd technology that was developed after SIPAD but came to the US first.

It is great for single sided boards, but you will have to use a glue/wave process or something else for second side assemblies.

This is true but great advances have been made in the 2 sides ssd technology. Its not the 2 side that is the challenge, it is the 2 pass eflow that has been tricky. A lot of progress is being made that involves the assembler adding the flux to the second side for the secoind reflow cycle.

Also, The sticky flux that is applied to the ssd can sit out much longer than solder paste and not dry out.

The SIPAD flux ahesice is guarentted for 6 months and we have documented results of the flux lasting for 2 years with the protective paper in place. Also, we have had recent examples of the flux lasting for more than 5 days without the protective paper waiting for customer first article approval. Once the approval was granted the board assembly was comleted and reflowed with perfect results.

I have heard the reason it has not caught on is because the SSD machine used i nthe process is very expensive, and there is only a few in the U.S.

SIPAD ssd is offered as a service.Machines are available and cost less that 100K.

What is the major difference between the SSD and the normal Solder Paste Printing?Appreciate if could provide more information.

The major difference is that ssd will not smear or smudge when assembled. The tacky flux is stronger than any solder paste and holds components more securly. Blind leaded parts, (like BGA's), cannot be reliably placed in wet paste by hand and are subject to shorts, opens and voids.

Below are some emails beginning in November 2003 that were sent to me from the Marshal Space Flight Center in Huntsville Alabama. According to these test results, SIPAD BGA's are really proving to be more reliable and longer lasting than traditional paste and place technology when thermal cycled.

What happens to the flux and pads on the second side of a double sided board during the first reflow? It seems to me like the solder would dome and the flux would be used up and migrated to other areas.

You are correct. That is whay the flux is left off of the second side and applied after the first side is reflowed.

During the reflow of the first side, the solder on the other side does not reflow. With this being true, when you reflow the second side, the components already reflowed on the first side do not fall off. The sticky flux is applied one side at a time. (Example: ssdbtm side->wash->ssd top->wash->sticky flux bottom side->place smt components->reflow->wash->sticky flux top side->place topside smt components->reflow->wash) ssd is basically the process of applying solder paste to the pcb with a stencil and the solder reflowed and flattened out. Our company does this all in-house so we do not have to apply our sticky flux (which is water soluable) until right before we run our board through the pick and place machine.

Escellent job celldude!! I agree 100%

What kind of parts do you use for SSD? More specifically does your company use 0402, 0201, BGA's, etc? Also are the results of using SSD better than the normal process of applying soder paste? Also if anyone else uses SSD, what are your experiences with it? Thanks a lot.

Check out the emails below from NASA concerning SIPAD BGA thermal cycleing as it relates to reliabilty of BGA's. A formal report is due out soon and will be available at http://www.sipad.com The emails start at the bottom and progress to the top by dates beginning last year.

--------------------------------------------------------------------------------

----- Original Message ----- From: Strickland, Mark To: Matt Kehoe (mkehoe@sipad.net) Sent: Tuesday, March 16, 2004 9:07 AM Subject: SIPAD Testing The thermal cycle testing of the SIPAD boards vs standard paste boards is going well. We are at 1860 thermal cycles and have observed that the BGA's on the SIPAD boards are failing at a slower rate than those on the standard paste boards. The graph below shows the distribution of failures, however, the chronological failure log shows even better results for the SIPAD boards. When testing is complete, the Weibull plots will give a better picture of the failure distribution. We started with sixteen assemblies (six BGA's and four QFP's per assembly.) We currently have 15 boards in test; one solder paste board that has completely failed is in destructive physical analysis. After DPA is complete we can hopefully definitively state what the failure modes are. In light of these early results, our plan is to start incorporating SIPAD into our designs for engineering units and research test vehicles that we will fabricate in the lab. This approach is supported by the ease of SIPAD use in a small volume operation and the apparent quality/reliability that it affords. When the testing is complete we will formally publish and present the results of the study. Thanks for assisting us in this effort. Mark Strickland NASA/MSFC Avionics Dept Electronic Packaging/ED16 256-544-7432

--------------------------------------------------------------------------------

----- Original Message ----- From: Strickland, Mark To: 'SIPAD Systems' Sent: Tuesday, February 17, 2004 10:23 PM Subject: RE: Whats new

We are at 1600+ cycles and SIPAD is doing well. We've had a few failures of components on the SIPAD PWB's but not as many as with the standard HASL/paste boards. We've started sectioning one HASL/paste board that has totally failed, but don't know the failure mode yet. None of the SIPAD boards have totally failed.

Do you have any data on the relative flatness of SIPAD versus a typical HASL PWB? --------------------------------------------------------------------------------

----- Original Message ----- From: Strickland, Mark To: 'SIPAD Systems' Sent: Wednesday, November 26, 2003 12:23 PM Subject: RE: Updates

Sorry it has taken me so long to get back with you. We are still testing and our now at 891 thermal cycles (-55 to 100C). Thus far it appears that the SIPAD is doing quite well. Any final conclusions will be made after we microsection to verify failure mode. The enclosed Powerpoint chart depicts the failures by type after 891 cycles. The test will continue for a few more months. However, we'll probably start microsectioning some of the boards in a few weeks once all the devices on a particular board have failed. Call if you have questions. Mark Strickland My email address has changed to: mark.strickland@nasa.gov or Stephen.M.Strickland@nasa.gov Electronic Packaging NASA/MSFC/ED16 256-544-7432 http://workmanship-pbma-kms.intranets.com/login.asp?link=

I see how that works in your shop, I was thinking of when you have someone else do it for you and then both sides would have to be done at the same time and then double reflow isn't very viable with out glue.

How do you flatten the solder after reflow? How do you apply the flux? I like the SSD process and would like to investigate bringing it into our shop so we could do it one side at a time and enable us to do the double sided boards

The flux is applied with a stencil

We use ssd on all components, from 0201 to very large fine pitch and bga's. With some very small micro bga's , if there is a problem with bridging, we will tape off the stencil during the ssd process and just use bare pads...along with the solder from the BGA. During the SSD process, the solder paste is applied with a 4 to 6 mil stencil, and flattened with an air knife that is part of the machine itself. we apply the sticky flux with a 4 mil stencil.

This applies only to PPT. SIPAD paste is also applied using a stencil but the paste is reflowed, washed, fglattened and fluxed on finished boards reday to go to assembly.

I have a couple more questions for you. Where can one look into getting these machines necessary for SSD?

Call Matt Kehoe at 770-475-4576

Also, knowing that the machines are expensive..Does your company find they are saving money (in the long run) by using SSD? More specifically, do you find that there is less rework involved?

Rework is normally eliminated.

Lastly, if there are any articles that you have come across related to SSD that you found helpful could you shoot them over my way? My email is: Aalap.Jha@EmersonProcess.com My co-op rotation is over Aug. 27th, hopefully by then I can convince my supervisor that SSD is the way to go! Thanks for all your help!

We have definitely seen money saved. There is a lot less rework after reflow. Money is also saved in overtime. for example...when we are building a first article board with many placements, it is not always possible to get the first board through the machine in an 8 hour shift, therefore the operator needs to work OT to finish it before the paste dries out. With SSD, it can be finished the next day without worries. The sticky flux is good for days. Also there is less money wasted in unused paste that dries out. Those are just a few i can think of off the top of my head. I do not have any access to figures or statistics to support these statements. I also have no idea where we obtained the machine. Our company is very tight lipped about the machine. I do get to work with the boards that are SS'D, but i do not get to see the process, or the machine itself. Im at the bottom of the corporate ladder. Sorry i could not be of any further help.

Ahmen Celldude. I agree.

reply »

Fluid Dispense Pump Integration

Boundary Scan