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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Ryan

#11206

Process Change | 3 June, 1999

In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated.

-Ryan Jennens TelGen Corporation

reply »

Earl Moon

#11207

Re: Process Change | 3 June, 1999

| In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | -Ryan Jennens | TelGen Corporation | Ryan,

Good to see from you here again. Pretty busy these days as are we all?

We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F.

We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes.

I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already.

Regards,

Earl Moon

reply »

Ryan Jennens

#11208

Re: Process Change | 3 June, 1999

| | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | -Ryan Jennens | | TelGen Corporation | | | Ryan, | | Good to see from you here again. Pretty busy these days as are we all? | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | Regards, | | Earl Moon

Thanks, Earl, for the condenscending reply. Sounds like you have a very robust process running.

Let me rephrase: Does anybody have experience with low-temp solder alloys and their resultant solder joints? Is it worth sticking with 63/37 and finding a way to adhere heavy parts? What are the disadvantages of using low-temp solder alloys, besides having to purchase two sets of solder products?

-Ryan |

reply »

#11209

Re: Process Change | 3 June, 1999

| | | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | | | -Ryan Jennens | | | TelGen Corporation | | | | | Ryan, | | | | Good to see from you here again. Pretty busy these days as are we all? | | | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | | | Regards, | | | | Earl Moon | | Thanks, Earl, for the condenscending reply. Sounds like you have a very robust process running. | | Let me rephrase: Does anybody have experience with low-temp solder alloys and their resultant solder joints? Is it worth sticking with 63/37 and finding a way to adhere heavy parts? What are the disadvantages of using low-temp solder alloys, besides having to purchase two sets of solder products? | | -Ryan | | | | Ryan: Too bad you have to do this, but people did it all the time in making hybrids

Anyhow, 43Pb43Sn14Bi is a nice low temperature solder with good fatigue resistance. Its matt finish will be easier to inspect than your 63/37. I would be uneasy using 43Pb43Sn14Bi with 20 pitch and below because (I've heard of, but never seen) dendrites forming.

Good luck

Dave F

reply »

Earl Moon

#11210

Re: Process Change | 3 June, 1999

| | | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | | | -Ryan Jennens | | | TelGen Corporation | | | | | Ryan, | | | | Good to see from you here again. Pretty busy these days as are we all? | | | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | | | Regards, | | | | Earl Moon | | Thanks, Earl, for the condenscending reply. Sounds like you have a very robust process running. | | Let me rephrase: Does anybody have experience with low-temp solder alloys and their resultant solder joints? Is it worth sticking with 63/37 and finding a way to adhere heavy parts? What are the disadvantages of using low-temp solder alloys, besides having to purchase two sets of solder products? | | -Ryan | | | | Sorry about the condescending remarks - my ass! I was simply trying to say there are easier ways to do hard stuff.

Luck be with you,

Moonman

reply »

#11211

Re: Process Change - Mascots???? | 3 June, 1999

| | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | -Ryan Jennens | | TelGen Corporation | | | Ryan, | | Good to see from you here again. Pretty busy these days as are we all? | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | Regards, | | Earl Moon | Wattz a "Mascots?" Are you employing the San Diego Chicken, Purdue Boilermaker, and Toledo Mudhen to assemble your boards? Don't those goofy costumes with all the feathers cause a bijundous ESD problem?

reply »

JohnW

#11212

Re: Process Change | 3 June, 1999

| | | | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | | | | | -Ryan Jennens | | | | TelGen Corporation | | | | | | | Ryan, | | | | | | Good to see from you here again. Pretty busy these days as are we all? | | | | | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | | | | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | | | | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | | | | | Regards, | | | | | | Earl Moon | | | | Thanks, Earl, for the condenscending reply. Sounds like you have a very robust process running. | | | | Let me rephrase: Does anybody have experience with low-temp solder alloys and their resultant solder joints? Is it worth sticking with 63/37 and finding a way to adhere heavy parts? What are the disadvantages of using low-temp solder alloys, besides having to purchase two sets of solder products? | | | | -Ryan | | | | | | | | Ryan: Too bad you have to do this, but people did it all the time in making hybrids | | Anyhow, 43Pb43Sn14Bi is a nice low temperature solder with good fatigue resistance. Its matt finish will be easier to inspect than your 63/37. I would be uneasy using 43Pb43Sn14Bi with 20 pitch and below because (I've heard of, but never seen) dendrites forming. | | Good luck | | Dave F | Ryan,

The only advive I got is just don't do it!. Dave's right the grain growth in it is pretty bad. the other thing you got to look at is how much heat is the board gonna produce in it's working life..??? and where is it gonna be used, is the low melting point solder gonna get the job done, answer is ..probably, but only for a short time. If grain growth doesn't kill you the operating temp of the board will. Go for the std solder 63/37 or maybe a 2% silver. If you really want to do a double sided reflow. Run the topside 1st like you need to then flip it after your's hit it with the inline xray and do the bottom's but head for the shallow end of the paste profile so if it's window is 30-45secs aim for about 35.I've got 1 board with 240 pin QFP's on both sides that I got going thru' double sided reflow..man what a buzz. Or if you got thru hole stuff why not just glue the bottom's?? then wave it mauch easier. Anyway's basically it can be done real well with normal solder if you got a good profile, good board design and good paste (note the word good!)

Enjoy!

JohnW

reply »

JohnW

#11213

Re: Process Change | 3 June, 1999

| | | | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | | | | | -Ryan Jennens | | | | TelGen Corporation | | | | | | | Ryan, | | | | | | Good to see from you here again. Pretty busy these days as are we all? | | | | | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | | | | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | | | | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | | | | | Regards, | | | | | | Earl Moon | | | | Thanks, Earl, for the condenscending reply. Sounds like you have a very robust process running. | | | | Let me rephrase: Does anybody have experience with low-temp solder alloys and their resultant solder joints? Is it worth sticking with 63/37 and finding a way to adhere heavy parts? What are the disadvantages of using low-temp solder alloys, besides having to purchase two sets of solder products? | | | | -Ryan | | | | | | | | Ryan: Too bad you have to do this, but people did it all the time in making hybrids | | Anyhow, 43Pb43Sn14Bi is a nice low temperature solder with good fatigue resistance. Its matt finish will be easier to inspect than your 63/37. I would be uneasy using 43Pb43Sn14Bi with 20 pitch and below because (I've heard of, but never seen) dendrites forming. | | Good luck | | Dave F | Ryan,

The only advive I got is just don't do it!. Dave's right the grain growth in it is pretty bad. the other thing you got to look at is how much heat is the board gonna produce in it's working life..??? and where is it gonna be used, is the low melting point solder gonna get the job done, answer is ..probably, but only for a short time. If grain growth doesn't kill you the operating temp of the board will. Go for the std solder 63/37 or maybe a 2% silver. If you really want to do a double sided reflow. Run the topside 1st like you need to then flip it after your's hit it with the inline xray and do the bottom's but head for the shallow end of the paste profile so if it's window is 30-45secs aim for about 35.I've got 1 board with 240 pin QFP's on both sides that I got going thru' double sided reflow..man what a buzz. Or if you got thru hole stuff why not just glue the bottom's?? then wave it mauch easier. Anyway's basically it can be done real well with normal solder if you got a good profile, good board design and good paste (note the word good!)

Enjoy!

JohnW

reply »

Earl Moon

#11214

Re: Process Change - Mascots???? | 3 June, 1999

| | | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | | | -Ryan Jennens | | | TelGen Corporation | | | | | Ryan, | | | | Good to see from you here again. Pretty busy these days as are we all? | | | | We have nine different board assembly types with very complicated, low volume/high mix stuff. This includes hand loaded through hole (using Mascots mostly), non-fine pitch SMT, fine pitch stuff, extra fine pitch, BGA, uBGA plus some really ugly via in pad nonsense. To do all this, up to 16 million placements per month on six lines, we use only two paste types (both 63/37 eutectics) and two reflow profiles with wave solder temperature set at 470 F. | | | | We use little selective wave palleting. Mostly we use goop mask applied with Camelots. Also, we use some pin in paste for certain applications. We're moving more to one additional solder paste and flux type as no-clean. We do use one reflow profile in air and another in nitrogen - and on it goes. | | | | I guess what I'm saying is a lot of thought went into these complicated processes. It has paid off for us though very few outside CM's are even interested in our business or processes. Maybe you could do this more complicated work with a little more thought, though I know you have put a lot into this already. | | | | Regards, | | | | Earl Moon | | | Wattz a "Mascots?" Are you employing the San Diego Chicken, Purdue Boilermaker, and Toledo Mudhen to assemble your boards? Don't those goofy costumes with all the feathers cause a bijundous ESD problem? | Hell, create ESD problems? Quite the contrary my friend. Ours are built very close to the ground (actually rubbing it at all times during movement) having no artifical fiber composition. This ensures, along with other grounding (drool, urination, etc.), our Mascots do the job effectively and offend very little save a little excessive gas passing when fed improperly. Also, minimal pooperscooper processing is required though advisable once a week.

Moonman

reply »

Steve Gregory

#11215

Re: Process Change | 6 June, 1999

Hi Ryan!

First off, who's telling you you need to do things bass-ackwards from what a normal process dictates? The x-ray vendor, or someone within your company? The only reason that I can think of to do something like that is because that decoupling caps sometimes fall right beneath QFP leads and whatnot and might obscure the x-ray image from a few leads if it were a transmission xray. So what? Look at at the 90% of the rest of the leads and you can comfortably say that your soldering process is sound.

Laminography xray does solve that problem, as does something like a FeinFocus xray. But you're stuck with what you have. What suprises me, is that it sounds like you're about to use xray on every board as a normal part of the process.

I know that's a "whiz-bang" thing to be able to tell your customers, that you xray every board as a normal part of the process, but that's totally un-necessary! Kinda' goes back to a statement an Ol' Salty QA manager once told me; "People get hung up on lookin' at things to reject em', instead of lookin' at em' to accept em'..." Never a truer statement was spoken...

Xray should be used only as a process verification tool, nothing else! Anybody that tries to tell you different has no idea about processes! In fact, there's talk out that xray'ing an assembly could reduce reliability, but there's no concrete stats to back that up (yet anyway). I picked that bit of information up at the IPC National Chip Scale and BGA Symposium in March.

My advice is to NOT go to two different alloys, and try as hard AS HELL to keep from doing things other than in the normal way! Why complicate things? Our jobs are tough enough as it is!

Good Luck Buddy!

-Steve Gregory-

reply »

#11216

Re: Process Change | 6 June, 1999

snip

| Ryan, | | The only advive I got is just don't do it!. Dave's right the grain growth in it is pretty bad. the other thing you got to look at is how much heat is the board gonna produce in it's working life..??? and where is it gonna be used, is the low melting point solder gonna get the job done, answer is ..probably, but only for a short time. If grain growth doesn't kill you the operating temp of the board will. | Go for the std solder 63/37 or maybe a 2% silver. If you really want to do a double sided reflow. Run the topside 1st like you need to then flip it after your's hit it with the inline xray and do the bottom's but head for the shallow end of the paste profile so if it's window is 30-45secs aim for about 35.I've got 1 board with 240 pin QFP's on both sides that I got going thru' double sided reflow..man what a buzz. | Or if you got thru hole stuff why not just glue the bottom's?? then wave it mauch easier. | Anyway's basically it can be done real well with normal solder if you got a good profile, good board design and good paste (note the word good!) | | Enjoy! | | JohnW | Ryan: Two points:

1 Following on John's point about not wanting to have solder liquidous temperature too close to equipment service temperature:

A rule of thumb: Liquidous should be at least 2X service.

2 If you're going to use low temp, why not use it on both sides?

TTYL

Dave F

reply »

#11217

Re: Process Change | 23 June, 1999

| In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | -Ryan Jennens | TelGen Corporation | Ryan: Look at this:

http://jefry.ipc.org/SCRIPTS/WA.EXE?A2=ind9702&L=technet&D=&H=&T=&O=&F=&P=1524

Dave F

Dave:

Thanks for the reply. I saw this one when searched the archives. I am a little confused by it. The charts I have from Indium show that 43Sn43Pb14Bi has a liquidus at 163 and solidus of 144. This is why I am confused why Dave Hillman refers not only to a eutectic melting point, but also at 98C. Any thoughts?

-Ryan Jennens

Ryan: You're correct about the temps for Sn43/Pb43/Bi14.

I think Hillman is saying that Sn43/Pb43/Bi14, at least in part, transmorglifies to Sn26/Pb20/Bi54, er maybe Sn25/Pb25/Bi50, er maybe both. He wasn't real specific about the details of the alloy in his reply, as you know.

Anyway, these new alloys are eutechic at 102 and 93C respectively. 98C is pretty much in the middle between 102 and 93C. Take a look at Manko "Solders And Soldering" page 139 for more details on these and many other Sn/Pb/Bi "low temperature" alloys. Or maybe they're on the chart from Indium.

Nothing says that something that starts as Sn43/Pb43/Bi14 can't become something(s) else. I mean, who knows what metals you're adding to the mix from the component terminations. There isn't a huge volume of material on some SMT pads.

So if Hillman's correct and if you go with my rule of thumb that operating temperature can't be more than 1/2 of solidus, you gets to ~50C for an op temp. Hell I fart at that temperature.

Scary eh? ... the op temp, not the farting.

Dave F

reply »

#11218

Curious on how this thread developed | 23 June, 1999

Dave,

Looks like you just replied to Ryan's original question, but the message itself shows that the two of you have been going back and forth. Did you copy from your email discussion? Just curious whether something is going wrong with the Forum.

Thanks,

Cunli

| | In a month or so, we will be getting an in-line x-ray inspection machine. It will require that we solder our most comlicated side (normally top) first. This is opposite to our process now where we solder our passive (bottom) side first. We cannot reflow the complicated side down, because not all of the boards have enough pad area to hold the heavy parts on above liquidus. My plan was to solder the complicated side with eutectic solder, x-ray inspect, solder passives with low-temp solder (43Pb43Sn14Bi), and automated-optical inspect those. I have heard from the vendor that Bismuth alloys will give dull solder joints. I have also heard that Indium alloys will be more expensive and still be dull or grainy looking. How dull will these low-temp joints be? I am weary of using high-temp solder on the complicated side because the melting temp will be pushing the limits of some of the components and the needed ramp-rate may be too high. Any input will be greatly appreciated. | | | | -Ryan Jennens | | TelGen Corporation | | | Ryan: Look at this: | | http://jefry.ipc.org/SCRIPTS/WA.EXE?A2=ind9702&L=technet&D=&H=&T=&O=&F=&P=1524 | | Dave F | | Dave: | | Thanks for the reply. I saw this one when searched the archives. I am a little confused by it. The charts I have from Indium show that 43Sn43Pb14Bi has a liquidus at 163 and solidus of 144. This is why I am confused why Dave Hillman refers not only to a eutectic melting point, but also at 98C. | Any thoughts? | | -Ryan Jennens | | | Ryan: You're correct about the temps for Sn43/Pb43/Bi14. | | I think Hillman is saying that Sn43/Pb43/Bi14, at least in part, transmorglifies to Sn26/Pb20/Bi54, er maybe Sn25/Pb25/Bi50, er maybe both. He wasn't real specific about the details of the alloy in his reply, as you know. | | Anyway, these new alloys are eutechic at 102 and 93C respectively. 98C is pretty much in the middle between 102 and 93C. Take a look at Manko "Solders And Soldering" page 139 for more details on these and many other Sn/Pb/Bi "low temperature" alloys. Or maybe they're on the chart from Indium. | | Nothing says that something that starts as Sn43/Pb43/Bi14 can't become something(s) else. I mean, who knows what metals you're adding to the mix from the component terminations. There isn't a huge volume of material on some SMT pads. | | So if Hillman's correct and if you go with my rule of thumb that operating temperature can't be more than 1/2 of solidus, you gets to ~50C for an op temp. Hell I fart at that temperature. | | Scary eh? ... the op temp, not the farting. | | Dave F | |

reply »

#11219

Re: Curious on how this thread developed | 23 June, 1999

| Dave, | | Looks like you just replied to Ryan's original question, but the message itself shows that the two of you have been going back and forth. Did you copy from your email discussion? Just curious whether something is going wrong with the Forum. | | Thanks, | | Cunli | Cunli: No everything is fine. I found that our email conversation was getting interesting, so I pasted several of our emails to get the information on and involve the forum. Ta. Dave F

reply »

#11220

Re: Curious on how this thread developed | 23 June, 1999

Thank you for sharing your discussions. Cunli.

| | | | Cunli | | | Cunli: No everything is fine. I found that our email conversation was getting interesting, so I pasted several of our emails to get the information on and involve the forum. Ta. Dave F |

reply »

SMT Services - STI Electronics

Solder Paste Inspection