|Ask the Experts|
May 1, 2018
Lead-Free Profile Using Leaded Paste
We assemble a PCB that includes an 8 converter modules mounted close together. The converter modules are installed after the reflow process. We tried to run these parts through the normal reflow but found some did not solder well because of insufficient heat in the center of the module group. I am concerned that increasing the heat profile will cause soldering problems with lead free BGAs on the same PCB. The board is processed as lead free, but this board is exempt from RoHS requirements. We are considering using leaded solder paste while using the existing lead free reflow profile. The intent is to lower the solder reflow temperature at the converter modules while maintaining the correct profile for the BGAs. Do you know if this has been successfully done and is it a reliable soldering method?
|Expert Panel Responses|
Mixing solder alloys on the same board is never a good idea, but it is a cleaver thought. It won't be long before we are all required to "get the lead out" so let's consider the real problem: getting enough heat into the Modules to let them solder safely along with your BGAs. First, make sure you are profiling the temperatures of the critical parts, at least the BGAs (those parts at risk) and your "Mdules" (the parts that are not getting hot enough). This data will be critical in finding the best reflow recipe. With convection reflow ovens, you have three options: higher temperature, higher convection rate (air flow), or more time, and combination of the three. Since over temperature of the BGAs is the main concern, I don't recommend higher temperatures in the zones. This just opens you up to the BGAs going well above their kill temp. You can increase the convection (air flow rate) IF you have this option on your reflow oven. Some do, some don't. This takes the form of fan RPMs, fan frequency (Hz), PSI control, a Low, Medium, High setting and other recipe settings depending on the oven. This increased setting will allow you to heat thermally massive parts without increasing the set point temperature in the zones. This lowers the risk of getting the BGAs too hot, because the zone temp is still the same, while pushing more heat into your Modules. It will force the BGAs to get up to temp faster, thus increasing their time above temp, so you should make sure this time limit is not violated. If you can't increase the convection rate, either because you do not have the control, or it is already at maximum, then the final option is to increase the time in each zone. This is of course done by lowering the conveyor speed. More time in each zone, again without increasing the zone temperature set points, will allow more time for the thermally massive Modules to get to the needed temp without risking the BGAs maximum temperature. Again, this will increase the time the BGAs spend above temp, so this should be watch to make sure you are not violating this time limit. Another approach is to "heat shield" the sensitive BGA with the same composite material used to make wave solder pallets. We make wave solder pallets to protect parts on the board from the wave, and we can do the same thing for reflow. Make a reflow pallet that covers (top side) the BGAs and other sensitive components and protect them from the heat convection because it acts as an insulator to slow the heat rise for these parts. Yes, the "pallets" would have to be applied to every board being reflow soldered, however, they are reused 1000s of times, just like in wave soldering. If you have already attempted these ideas, then it may be time to try a different reflow process, like Vapor Phase. This process can force heat into the heaviest of thermal masses without risking temperature sensitive components, because the maximum the parts will see temperature is controlled by the boiling point of the liquid used in the system, typically 230C or 240C. It worth a renewed look at this older, but returning reflow process, now that new "ozone safe" liquids have been developed.
Senior Project Engineer
Electronic Controls Design Inc
The increase in the needed energy for good lead free solder joints to be formed is significant. If the target PCB has a wide variety of parts and the Reflow system lacking in thermal capability it become an issue of protecting smaller parts from reaching temperatures that exceed their operating specifications while providing adequate heat [energy] to reflow the heavier parts like BGA's and connectors. The only true and safe method of guaranteeing a "not to exceed" peak temperature while ensuring the PCB and ALL the components reflow in an Inert environment is to reflow using Vapor Phase technology. A System that utilizes an inert vapor as the mechanism for thermal transfer.
When using a Pb-free profile for Pb-free BGAs & Sn/Pb paste, the good thing is that the Pb-free BGA balls will melt completely and a homogeneous joint is achieved. Since a Sn/Pb paste is now exposed to higher temperatures, some of the things that need to be considered are:
Technical Manager - Europe
PCBs with a mix of tin-lead and lead free components or high mass and low mass components has a similar solution if redesigning the PCB is not in the cards. You can define each component with its own unique process specification visa-a-vie the thermocouple(s) attached, from which you let the prediction software find you a solution that balances your PCB. What do I mean? Let's say you have a BGA with a peak spec of 235-250 C. 235 for the solder joint and 250 as the maximum temperature of the component. Meanwhile on the same PCB you have a few heat intolerant components that cannot go above 245C. It doesn't help that one of these components is in close proximity to your higher mass BGA. So what do you do? Using KIC2000 software, you can choose to create individual specs for each thermo couple. In this example, I would define the peak range for my BGA as 235-250C, while for my smaller heat intolerant component I would set a limit of 245C. Go ahead and run a profile and with KIC's Navigator software you will find out two pieces of information. First, are both components in spec? Secondly, if not, what do you need to change with your oven recipe to bring these components into specification. In practice, I have found sometime there is no solution, especially if you are using an older reflow oven designed for tin-lead processes. Nevertheless, if there is a possible solution the software will find it for you with the gear you have to work with. Sure beats having to redesign a board or send PCBs to rework.
Global Account Manager
I would highly suggest not to use you lead-free profile to run you leaded paste. There is melting temp difference between to two products. Profiling your boards correctly is one of the most important processes to perform proper reflow. You should profile each different board that you reflow because there is a differences depending on board population, thickness and ground plans which effect your reflow process.Joe Mekus, M.C. Miller Co.
I like out of the box thinking and, in theory,your solution may work. The questionthat you have to ask yourself is how confident are you that you can reliablyapply leaded and lead-free paste to the two different areas. Also what method to you have in mind to dothis.Robb Spoerri,
It is possible, but has somechallenges. Short of board redesign, andassuming that a profile cannot be created that will meet the thermalrequirements of all parts with the equipment that you have, as is, I would gowith the suggestion made by another respondent.
Design reflow pallets that will effectively insulated your lower massparts and provide greater thermal transfer resistance while leaving the highmass parts clear/open. This may bedifficult to dial in on first effort so you will likely want to order one andTC and run it to see where you're at and make adjustments as necessary.
You're not alone, considering soldering SAC305BGA's with Sn63 paste at a lead free profile temperature. Many have foundthemselves cornered into this situation by the supply chain only being able tosupply mixed BGA alloys, at least for a period of time. One BGA may only beavailable in Sn63, while another is only available in SAC305, and you have to decidehow to make shipments.Jack Lucas, Ametek Programmable Power, USA
It may be obvious that soldering Sn63 BGA's with leadfree solder paste, melting at 217 C or above, is a bad idea. But there is alot of anecdotal information to suggest that many are soldering lead free BGA'swith Sn63, and I have heard from multiple sources that the reliability of thesejoints, with respect to extended temperature cycling, is adequate.
So I haveallowed this practice on a temporary basis. I think the resulting leadpercentage in a relatively homogeneous solder joint is roughly 3%. You mustachieve close to 235C peak and be above liquidus of the lead free alloy forabout 60 seconds.
Finally, I actually saw a Circuit Insight publishedarticle on a scientific study that supports the above mentioned anecdotalinformation about the reliability of this scenario.(http://www.circuitinsight.com/programs/53246.html)
Long term you should solder lead free BGA's with leadfree compatible paste, but temporarily, depending on your customers' acceptanceand your comfort level, soldering lead free BGA's with Sn63 paste may beacceptable.
|Submit A Comment|
Comments are reviewed prior to posting. You must include your full name to have your comments posted. We will not post your email address.
Free Newsletter Subscription
Circuitnet is built for professionals who bear the responsibility of looking ahead, imagining the future, and preparing for it.
Insert Your Email Address