Reflow For Rigid Flex

Phil
And welcome to Board Talk with Phil Zarrow and Jim Hall, the Assembly Brothers, here to solve, or add more confusion, to whatever you are pondering in the area of surface mount technology.

Jim
Today we have a question from S.J.  We have a new PCB assembly that includes a section of flex cable and FR4 circuit board.  What are the reflow rules I need to consider while processing a mix, FR4 and flex cable assembly through the reflow process.

Phil
Well, first of all I believe that's known in the vernacular of our industry as rigid flex, clever name don't you think? That's and interesting question with regard to reflow. Reflow is reflow. You follow all the other rules that you need driven by the specs of solder paste and of course your most vulnerable component on the board.

Jim
Your flex cable might become your most vulnerable component.

Phil
You may be fixed ring so remember when you do profile to use the fixture.

Jim
I want to go over my brother's head and say you should be fixturing. You really should very carefully look at fixturing, keeping that thing level, particular if there are surface mount parts on the flex section of your assembly.

You want to make sure that's not moving around and flexing during reflow and moving your components.

Phil
Generally most people build the entire assembly, either reflex or rigid flex in a fixture, right through printing, pick and place, and reflow. So they avoid disturbing stuff through handling.

But with regard to the reflow question I think that's straightforward, but the thought I am wondering about is with regard to more important issue particularly with rigid flex is storage and handling.

Jim
And why would you worry about storage and handling?

Phil
Because rigid flex boards tend to have an interesting moisture absorption aspect and sometimes we get involved with bake out.

Jim
Flex materials tend to absorb moisture much more quickly and so proper storage and perhaps baking before reflows is a significant consideration.

Phil
So that would probably be the main concern. There are some pieces of wisdom coming out of IPC on handling flex and rigid flex, I believe it's 1601 on the circuit board handling and storage specification. 

So anyway beyond whatever you do, don't solder like my brother.

Jim
Don't solder like my brother. 

IPC 1601 Printed Board Handling and Storage

Depending on the overall thickness of the flex, bake times vary. I would recommend that you contact your board manufacturer for guidelines, as Polyimide is very hydroscopic.

I just felt that it should be mentioned that IPC-1601 has been superseded by IPC-1602, Standard for Printed Board Handling and Storage.

Thanks for a great write up on rigid flex. I am curious what other people are using for moisture bake out techniques of PI materials prior to assembly. In the past we have used recipes such as 150 @ 4 hours. Interested to know what others use.

Rigid-flex, actually all flex can heat up more than twice the PWB temperature seen during normal profiling, but even much more than that if there is a significant portion of infrared energy. It absorbs the IR spectrum and gets hotter than the FR-4 or Polyimide PWB materials. Be sure to shield it well, as Mr. Tulman has noted. This is especially true when working around IR rework equipment, but remember that ALL convection ovens have around a 5 to 6% IR energy percentage. So your flex can be getting hotter than the PWB.

And if you have more than one flex connecting the rigid PWBs (actually quite common), it is also a good idea to place a thin layer of Durostone or some other insulating material between the flex circuits so they do not stick to each other after reflow.

Baking flex-rigids including Polyimide (normal stuff) is necessary, but remember that PI picks up moisture quickly once again, so solder immediately.

Flex-rigids should be throughly dried before reflow. With very hot and long assemblies it should be a common practice to build a "cap" over the flexes to avoid heating them too much. The adhesive used for the coverlay is acrylic based and its Tg is too low.