Simultaneous Double Sided Reflow

I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing.

Is there any one who has been running this process in production ?

Yes I do mean reflowing both sides at the same time it does work.

reply »

| I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | Is there any one who has been running this process in production ? | | Yes I do mean reflowing both sides at the same time it does work. | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types).

Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's.

At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications.

Earl Moon

reply »

| | I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | | | Is there any one who has been running this process in production ? | | | | Yes I do mean reflowing both sides at the same time it does work. | | | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types). | | Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's. | | At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications. | | Earl Moon | Earl- Could you use your infallible knowledge base to further explain? How would one keep the components on the side which was printed first from being disturbed as the second side is printed? This seems to be the biggest roadblock in my mind, assuming the parts' joints have sufficient holding power.

Ryan Jennens Phoenix Engineering dEsign, Inc.

reply »

| | | I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | | | | | Is there any one who has been running this process in production ? | | | | | | Yes I do mean reflowing both sides at the same time it does work. | | | | | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types). | | | | Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's. | | | | At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications. | | | | Earl Moon | | | Earl- | Could you use your infallible knowledge base to further explain? How would one keep the components on the side which was printed first from being disturbed as the second side is printed? This seems to be the biggest roadblock in my mind, assuming the parts' joints have sufficient holding power. | | Ryan Jennens | Phoenix Engineering dEsign, Inc. | Ryan,

Please understand I would freely give you all everything I know. Just now is not the time and I'm not being evasive. It's just that I am a consultant, though cheap and not very good, and the time is right for at least three things for me to continue making a living. One is the issues being discussed with increasing interest. Another is inert atmosphere soldering. Yet another is using GDT as a concurrent engineering and continuous process improvement tool.

I will say this: There are, as there has been, several companies out there who initiated the simultaneous double sided practice. I worked with some. Right now, it is not being discussed freely.

I must almost apologize for this reply as you know I usually offer all I can remember (not much) free to you all on this great forum. Also, I know someone such as you will ponder this question and figure it out on your own sooner than later. This is because you care and have the talen to do so.

Hope I haven't offended,

Earl Moon

reply »

| | | | I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | | | | | | | Is there any one who has been running this process in production ? | | | | | | | | Yes I do mean reflowing both sides at the same time it does work. | | | | | | | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types). | | | | | | Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's. | | | | | | At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications. | | | | | | Earl Moon | | | | | Earl- | | Could you use your infallible knowledge base to further explain? How would one keep the components on the side which was printed first from being disturbed as the second side is printed? This seems to be the biggest roadblock in my mind, assuming the parts' joints have sufficient holding power. | | | | Ryan Jennens | | Phoenix Engineering dEsign, Inc. | | | Ryan, | | Please understand I would freely give you all everything I know. Just now is not the time and I'm not being evasive. It's just that I am a consultant, though cheap and not very good, and the time is right for at least three things for me to continue making a living. One is the issues being discussed with increasing interest. Another is inert atmosphere soldering. Yet another is using GDT as a concurrent engineering and continuous process improvement tool. | | I will say this: There are, as there has been, several companies out there who initiated the simultaneous double sided practice. I worked with some. Right now, it is not being discussed freely. | | I must almost apologize for this reply as you know I usually offer all I can remember (not much) free to you all on this great forum. Also, I know someone such as you will ponder this question and figure it out on your own sooner than later. This is because you care and have the talen to do so. | | Hope I haven't offended, | | Earl Moon | Earl- Like I said, I understand.

Ryan Jennens

reply »

| | | | | I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | | | | | | | | | Is there any one who has been running this process in production ? | | | | | | | | | | Yes I do mean reflowing both sides at the same time it does work. | | | | | | | | | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types). | | | | | | | | Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's. | | | | | | | | At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications. | | | | | | | | Earl Moon | | | | | | | Earl- | | | Could you use your infallible knowledge base to further explain? How would one keep the components on the side which was printed first from being disturbed as the second side is printed? This seems to be the biggest roadblock in my mind, assuming the parts' joints have sufficient holding power. | | | | | | Ryan Jennens | | | Phoenix Engineering dEsign, Inc. | | | | | Ryan, | | | | Please understand I would freely give you all everything I know. Just now is not the time and I'm not being evasive. It's just that I am a consultant, though cheap and not very good, and the time is right for at least three things for me to continue making a living. One is the issues being discussed with increasing interest. Another is inert atmosphere soldering. Yet another is using GDT as a concurrent engineering and continuous process improvement tool. | | | | I will say this: There are, as there has been, several companies out there who initiated the simultaneous double sided practice. I worked with some. Right now, it is not being discussed freely. | | | | I must almost apologize for this reply as you know I usually offer all I can remember (not much) free to you all on this great forum. Also, I know someone such as you will ponder this question and figure it out on your own sooner than later. This is because you care and have the talen to do so. | | | | Hope I haven't offended, | | | | Earl Moon | | | Earl- | Like I said, I understand. | | Ryan Jennens | | Ryan, you are a class act. Again, you'll figure it out. It's just that it takes some resources maybe beyond the average person's grasp. Besides, it might take some regressive thinking, within everyone's grasp, as using focused IR with glass plate baffles or?

Go get 'em and With great respect,

Earl Moon

reply »

| | | | | I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | | | | | | | | | Is there any one who has been running this process in production ? | | | | | | | | | | Yes I do mean reflowing both sides at the same time it does work. | | | | | | | | | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types). | | | | | | | | Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's. | | | | | | | | At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications. | | | | | | | | Earl Moon | | | | | | | Earl- | | | Could you use your infallible knowledge base to further explain? How would one keep the components on the side which was printed first from being disturbed as the second side is printed? This seems to be the biggest roadblock in my mind, assuming the parts' joints have sufficient holding power. | | | | | | Ryan Jennens | | | Phoenix Engineering dEsign, Inc. | | | | | Ryan, | | | | Please understand I would freely give you all everything I know. Just now is not the time and I'm not being evasive. It's just that I am a consultant, though cheap and not very good, and the time is right for at least three things for me to continue making a living. One is the issues being discussed with increasing interest. Another is inert atmosphere soldering. Yet another is using GDT as a concurrent engineering and continuous process improvement tool. | | | | I will say this: There are, as there has been, several companies out there who initiated the simultaneous double sided practice. I worked with some. Right now, it is not being discussed freely. | | | | I must almost apologize for this reply as you know I usually offer all I can remember (not much) free to you all on this great forum. Also, I know someone such as you will ponder this question and figure it out on your own sooner than later. This is because you care and have the talen to do so. | | | | Hope I haven't offended, | | | | Earl Moon | | | Earl- | Like I said, I understand. | | Ryan Jennens | The process I have been using is a standard convection reflow oven using standard paste, process etc. The only different thing is using a glue after the first print pass.

The glue is not cured but it does provide about 200g force in its green tack state which will allow chip, SOIC16, SOT23, SOT89 and TSOP parts to be held in place.

There should be no cpntact with any pad or compenent in a normal production process for DSRS so contact is not the problem. The movement and shock of machine stops, load unload is the real problem with the process. The flexture of the board can also be an issue. I have been also evaluating two new adhesives with fast UV light cure to increase the component retension force to meet my requirement of minimum of 500g shear force.

Hope this helps.

PS. Some people get upset but I will just tell you that there are two video clips showing this process on a new CD ROM from ECD to be seen at Nepcon West !! |

reply »

| | | | | | I have been working on this process for a couple of weeks and I know that there are a few other companies in Europe and Japan who are also playing. | | | | | | | | | | | | Is there any one who has been running this process in production ? | | | | | | | | | | | | Yes I do mean reflowing both sides at the same time it does work. | | | | | | | | | | | Some of us began single pass, double sided reflow in 1994 in limited volume. None the less, it was production for a particular application. Others of us began earlier with very specific applications for chip carrier device types only (no other device types). | | | | | | | | | | Though it doesn't count in today's "modern" reflow world, we used vapor phase to accomplish these requirements for LCCC's. This process was used to assemble military/aerospace requirements in the late 70's through the 80's. | | | | | | | | | | At another company (1995), we began running double sided BGA's using single pass process management. Very soon, I will be working with a company to take the process into higher production with dedicated designs more focused on this type process while limiting device types and taking more advantage of ASICS and other "single chip" applications. | | | | | | | | | | Earl Moon | | | | | | | | | Earl- | | | | Could you use your infallible knowledge base to further explain? How would one keep the components on the side which was printed first from being disturbed as the second side is printed? This seems to be the biggest roadblock in my mind, assuming the parts' joints have sufficient holding power. | | | | | | | | Ryan Jennens | | | | Phoenix Engineering dEsign, Inc. | | | | | | | Ryan, | | | | | | Please understand I would freely give you all everything I know. Just now is not the time and I'm not being evasive. It's just that I am a consultant, though cheap and not very good, and the time is right for at least three things for me to continue making a living. One is the issues being discussed with increasing interest. Another is inert atmosphere soldering. Yet another is using GDT as a concurrent engineering and continuous process improvement tool. | | | | | | I will say this: There are, as there has been, several companies out there who initiated the simultaneous double sided practice. I worked with some. Right now, it is not being discussed freely. | | | | | | I must almost apologize for this reply as you know I usually offer all I can remember (not much) free to you all on this great forum. Also, I know someone such as you will ponder this question and figure it out on your own sooner than later. This is because you care and have the talen to do so. | | | | | | Hope I haven't offended, | | | | | | Earl Moon | | | | | Earl- | | Like I said, I understand. | | | | Ryan Jennens | | | The process I have been using is a standard convection reflow oven using standard paste, process etc. The only different thing is using a glue after the first print pass. | | The glue is not cured but it does provide about 200g force in its green tack state which will allow chip, SOIC16, SOT23, SOT89 and TSOP parts to be held in place. | | There should be no cpntact with any pad or compenent in a normal production process for DSRS so contact is not the problem. The movement and shock of machine stops, load unload is the real problem with the process. The flexture of the board can also be an issue. I have been also evaluating two new adhesives with fast UV light cure to increase the component retension force to meet my requirement of minimum of 500g shear force. | | Hope this helps. | | PS. Some people get upset but I will just tell you that there are two video clips showing this process on a new CD ROM from ECD to be seen at Nepcon West !! | | | Hell Bob, who would get upset with such a shy, conservative, fashionable (where has your portrait gone - the Louvre of course) gentleman as yourself. Besides, you could have just come out and declared your glued up process success. Of course, you're leaving out some of the best parts of the process, but I'm sure they're for sale on a CD. Can't wait to get mine.

No sour grapes. Just wish I could afford to sell your way. But that's the price I pay for being a cheap consultant - though not very good.

You're the best,

Earl Moon |

reply »