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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Convection Oven and thermo profiling of Circuit Boards

Russ Cutler

#4403

Convection Oven and thermo profiling of Circuit Boards | 17 April, 2000

We just bought a new Heller 1800 EXL reflow oven for reflowing SMT boards. We us a low temp RMA solder paste with 2% silver. Everything we build is to IPC Class 3 specifications. We have approximately 600 different types of circuit boards we build, and will reflow in the Heller oven. The question is, should we have 5 or so profiles that are "good enough" to fit the entire 600 very different boards into, or should we have a "unique" thermoprofile and recipe for each and every board. Our boards range the entire spectrum from double sided boards to 14 layer with 2 oz. copper internal gound planes. What is proper, and makes the most sense to ensure that we get the proper match of the thermoprofile to the exact board, and have optimum solder joint quality and reliability? Any pointers would be great. Thanks!

Russ Cutler

reply »

#4404

Re: Convection Oven and thermo profiling of Circuit Boards | 17 April, 2000

Sounds like a true IR guy to me!!!

Your profile should reflow solder on the board properly to produce a reliable product for your customer. If that requires more than one proifile, so be it!!!.

That being said, we went from hundreds of IR profiles to one profile, when we made the change you've just made. Still have one profile, years and hundreds of board P/Ns later.

Have fun with your new oven Dave F

reply »

Russ Cutler

#4405

Re: Convection Oven and thermo profiling of Circuit Boards | 18 April, 2000

Let me be more specific about my dilema. We have used a vapor phase reflow process for many years. We just bought a Heller 1800 this year (2000). We may very well be able to run all 600 very different board types we build in 10 recipies on the Heller oven. In our efforts to acheive this, it is necessary in my opinion to thermo-profile each of the 600 boards individually, so we have an accurate idea of which of the 10 recipes will fit each board best. This involves a considerable amount of my time. Am I out in left field? Is there any better ideas? Since we have no baseline comparisons for convection reflow, we have to start somewhere, and that is by getting a profile on each board so we know how it reacts to the oven. All comments are welcome, please.

Russ Cutler

reply »

Arturo

#4406

Re: Convection Oven and thermo profiling of Circuit Boards | 19 April, 2000

Russ:

you could divide your 600 in let's say 10 or more categories,category #1 could be the thinnest thickness and smaller board size and smaller components;category #2 thinnest thickness smaller board size and mixed small and big components; and so on with these three parameters combinations(thickness,board size and type of components) and make one profile for one PCB per category. Of course you have to check the quality of your solder each first time you run a different model and if it does not convince you then you have to profile that model.

It worked for me.

Arturo.

reply »

Rudolf Niebling

#4407

Re: Convection Oven and thermo profiling of Circuit Boards | 19 April, 2000

You need a KIC profiler whitch is able to find optimal profiles for your boards. The KIC works very well with the Heller ovens. With the KIC is also possible to find one good profile for different boards. Get in contact with KIC +1-858-673-6050

reply »

Russ Cutler

#4408

Re: Convection Oven and thermo profiling of Circuit Boards | 19 April, 2000

Let me see if I've got this straight...rather than hook up a MOLE or a KIC profiler to each and every board (600 P/N's in my case), most people are making judgement calls based solely on how similar a board looks to another board, for which they have an established and solidly reliable profile? And then as a check on this assumption based on looks, they are verifying solder joint quality? Seems like this approach would save a lot of time, but how does one know in reality how hot a board is getting, if the time is not taken to profile each board for specifics? Still a question I'm not real comfortable with yet...

Russ Cutler

reply »

Mike Flori

#4409

Re: Convection Oven and thermo profiling of Circuit Boards | 19 April, 2000

If you can divide your boards into different categories it allows you to determine a different starting profile for each category, based on size, # of layers, etc. When you get a new board to profile you'll have an idea what settings will get you close to what you want and then just have to tweak it. I think you will find that you will end up with profiles that tend to group around common values. The fewer categories you have the less time you will spend changing settings and waiting for the oven to stabilize before you can run the first board.

reply »

#4410

Re: Convection Oven and thermo profiling of Circuit Boards | 24 April, 2000

My agreement with all this started by Arturo's message...

Just as a reminder, I guess he suggested to have still "one profile per board" which I completely agree... So, Board #(1,600) could be using a Recipe #(1,10) but it should still be saved as board_type_XXX so that if, for any reason, you want to modify the profile in future, the change will only affect board type XXX rather than Recipe Y which was initially used by many others...

reply »

armin

#4411

Re: Convection Oven and thermo profiling of Circuit Boards | 25 April, 2000

If I were you, I would divide my boards into categories...and I'll categorize them by their components e.g. Bds with BGA, with QFP, Bds. with plain chip components...and so on...this is to create a recipe for each components with QFP's, BGA's etc...to avoid future profile problems. Just think about this, if u got a 10 profiles for those 600 bds. and encountered solderability problem in the future on 1 of your product, once u adjust the temp to suit that problematic bd. u need to readjust back for the other bds. on that same recipe.

Happy profiling...

reply »

Reflow Oven

reflow oven profiler