|Ask the Experts|
November 4, 2020
Competing Reflow Oven Zones
We have an 8 zone reflow oven in our SMT production line. We recently purchased an oven profiler and I'm having some issues with our tin lead process. There is a fairly large list of paste options within the oven profiler software, however the part number of our paste doesn't match any of the options.
The issue we have is some of the zones in the oven are competing with each other to get to the correct temperatures (within 2C of setpoint).
How do I find the best way to prevent the oven from having competing zones while also getting an accurate process?
|Expert Panel Responses|
It is always wise to contact your paste manufacturer's Technical Support for targeted process and profiling assistance. While the profiler software is a great tool, Your profile will depend upon many factors including customer requirements,component limitations, oven characteristics, board layout, etc.
Ultimately, quality requirements should drive the process, not adherence to these guidelines. Ideally, profile measurements are to be collected on a populated assembly with the reflow profile recorded for each product being processed.
It is common for the same profile settings to be used for multiple assemblies. Reflow profile data should be collected, analyzed and recorded for each assembly part number at the beginning of individual production runs.
There are two basic profile types: Ramp-Soak-Spike (RSS) and Ramp-To-Spike (RTS). RTS profiles are suitable for use in most applications for enhanced solder performance. RSS profiles are appropriate when the assembly has a large thermal mass or large ∆T.
Senior Applications Manager
Ovens do their best to maintain isolation between zones. But there are practical limits to what they can achieve because there is open space between zones. Most can do 20 to 30 C between adjacent zones. These limits may be stressed by the assumed demands of the solder paste.
However, most solder paste spec's are very broad for one simple reason, to accommodate the capabilities of the oven and the limits of you components being soldered (reflow). I would suggest you can "straight ramp" to liquidus + 10 to 15 degrees C using your 8 zone oven without difficulty and the paste would be OK with it. If you MUST soak, use the first three to ramp to your soak temp, soaking for three more zones, then use the last two zones to achieve the desired peak temperature.
If you need further assistance achieving the needs of your select paste, I'd consult the paste manufacture for help determining the best "profile" shape requirements.
Senior Project Engineer
Electronic Controls Design Inc
Each oven manufacturer should specify the maximum delta between zone temperature set points (usually 40°C-65°C). As for your solder paste, you can ask your solder paste supplier to send their part number and spec sheet to the profiler company for addition to their solder library, or contact your profiler company directly to get it added (we do this all the time).
After getting your process parameters set, you should be able to use the prediction features of your profiling software to obtain an optimal compromise between zone set points and conveyor speed (smaller zone deltas usually require slower conveyor speeds).
M.O.L.E. Line Product Manager
Electronic Controls Design, Inc. (ECD)
You will want to start by getting the recommended thermal profile for the solder paste you have selected. You should be able to get that from the paste manufacturer - typically easy to find in material data sheets on the manufacturer's web site. If you are relying on guidance provided through the profiler software, you may want to contact the technical support people at the profiler supplier.
Failing that, you could start by selecting one of the eutectic tin/lead that are already included in the software. I wouldn't expect much difference between different part numbers of solder pastes with similar alloys and flux chemistry. If the profile this yields is not close enough to what you need for your solder paste, you should be able make minor adjustments to your zone settings.
Principal Product Engineer
Benchmark Electronics, Inc.
It sounds like you are using an auto-prediction feature that came with your profiling package. Although typically very accurate, there are situations where the oven can't handle the difference in zone set temperatures suggested by the profiling software.
On some profiling software packages you can override the suggested profile and manually make adjustments to get your optimal results. I would suggest that you contact the manufacturer of the profiler and discuss this with them. If by chance it is a Fluke Process Instruments (Datapaq) profiler, please feel free to reach out to me directly.
National Sales Manager
This issue can occur when the different zones' settings are far apart from a temperature variation point of view. Sometimes, close enough is good enough so run a test board (if possible) and evaluate the results. With the tin lead and 8 zones, you can pretty much set a good profile without issues. Based on the board population though, some challenges might occur.
You can contact me with more documentation (picture and description of the board, type of paste used, oven settings from your profiling equipment) and I will try to help.
Engineering and Operations Management
It is best to keep the oven temperature settings less than 30 degrees C from one zone to the next. Some ovens may not be able to maintain zone-to-zone differences higher than this. Generally speaking smaller zone-to-zone differences are easier for ovens to maintain.
I suggest talking to your solder paste supplier to get the recommended process windows for your reflow profiler and the solder paste that you are using. Your solder paste supplier can give specific settings to program into the profiler rather than trying a pre-programed option that doesn't match the solder paste.
You express two problems, and they are interlinked. The first is to define the process window for your tin/lead paste. If the manufacturer of your profiler does not include the paste process specs in its library, you can ask them to include it, and they will likely do so. Alternatively, you can go directly to your paste manufacturer and ask them to supply the specs.
If your new profiler has an automatic oven setup software, it should be able to identify the correct oven recipe for your application. (Please note that you may want to edit/shrink the paste process window if you have components in your assembly with lower specs).
Should the profiler not include the optimization software you should be able to purchase that option. It is well worth the money as you will achieve a more robust process, and you will be able to set up your oven quickly.
PS: For the sake of full disclosure, I work for KIC which manufacturers thermal profilers.
inspīre solutions LLC
The 'recipes' that are included in profiling software are provided as a user convenience. They should not be relied on for establishing the best profile for your process. The best profile should be determined by your solder joint quality, not adherence to a profiling recipe. The profiler is simply a tool to measure the process and the software should not dictate the process.
Often times QA auditors will wrongly insist the profile has to match the TDS or profiler recipe when in fact, product quality should drive profile development. Contact your solder paste vendor for profiling assistance as they have experts available to provide both telephone/on-site support. If yours doesn't, promptly find one that does.
Director of Product Management
Since most thermal profiler manufacturers have an extensive library of solder paste profiles, my first suggestion would be to contact the profiler manufacturer and request a profile for the paste you would like to use. My second suggestion would be to contact your solder paste supplier since they would most likely have a recommended profile or suggested process parameters you can use as a baseline to create the appropriate profile.
When you contact your over supplier about the profile, you should also ask them how to prevent cross talk between the oven zones. Cross talk between zones can be greatly affected by oven loading and the thermal mass of your board assemblies and your oven supplier should be able to coach you on custom profiles and preventing cross talk.
The first part concerns the paste specifications. Most profiling software has the ability to change the profile target on an existing paste or add new pastes. It is best to contact your paste manufacturer to obtain his current recommendation, but remember it is only a recommended starting point and may not be the best for your application.
The second part refers to what is called zone separation. To fully understand the issue we need to think about how a board is heated in a convection reflow oven. Gas is heated in a plenum and then forced to the board surface with blowers. The gas gives up some of its heat (AKA Calories or BTUs) when it contacts the board and then returns to the plenum to be reheated.
The cycle is repeated over and over with the temperature being controlled by thermocouples and the gas flow (impingement velocity) by the blower speed. The space between zones is an open window, therefore some of the hot gas and heat can move to an adjacent zone. When the lower temperature zone can’t give up the added heat fast enough, reheat of the gas is not required and it can sometimes runs above its set point.
There are a few things that can be done to increase the zone separation. One method is to allow the problem zone to give up some of its heat. This can be done by lowering the set point of the proceeding zone. For example if zone 6 is the problem and zones 5 & 6 are both at 180 C dropping the temperature of zone 5 slightly will give zone 6 a place to loose heat.
The worst case is when automated profiles ask for the set point of a zone to be lower than those on each side. This allows both the previous and subsequent zones to overpower the middle zone and it does not control. IF this happens you will need to manually insert set points in the prediction area of the software to see if you can come up with a set points that work.
Following are examples of good and poor set points combinations
Z1 Z2 Z3 Z4 Z5 Z6 Z7 Z8
Good 120 140 150 170 180 185 230 255
Good 120 140 150 170 175 180 230 255
Poor 120 140 150 170 180 175 230 255
Note: I rely on predictive profiling software to help me do my job, but it attempts to find the best set points for all profile parameters and sometimes comes up with results that don’t work.
(Caution the following may not apply to all brands of reflow ovens) If you have an air oven you can open the dampers on the blowers to allow more cool air into the lower temperature zone. On a nitrogen oven you can increase or turn on the nitrogen flow into the lower temperature zone. In some cases the nitrogen inlet may need to be moved to the zone requiring cold gas.
Additionally, heat barriers can be placed between zones to help block the gas and heat flow between zones.
Manager, Process Technology
Kind of funny none of the experts remembered the most important tip of all, the effect of loading the oven in a production scenario. Consider; after you have used a scrap production board or something really close to it to set up your profile, somewhat close to matching the suggested profile on the TDS as a guideline, think about adding several bare PWBs both in front and behind the profiling board, so you can see what effect that has on the overall tunnel or zone temperatures. If you can put at least one or two boards ahead of and behind your profile board with the datalogger still attached, you will see that some ovens, or some zones in a given oven cannot recover between CCAs as well as other ovens can. This can cause a huge difference between what your profiling board and what your production boards see. You may have obtained the "perfect" temperature profile using your scrap populated CCA, but in real life (production) that program simply may not be good enough to recover between CCAs, and the results can be very poor, indeed.
|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