Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
two-sided design guidelines

two-sided design guidelines

Subscribe to Thread

Is anyone out there aware of any standard or even a section ... - Oct 09, 2002 by Steve Thomas

Steve, I am not aware of any standards so to speak. But IP... - Oct 18, 2002 by RDR

Check out ... - Oct 21, 2002 by Stephen

There is some pretty useful info in there ... - Oct 22, 2002 by RDR

I want to know how many PTH or DIP components is the critica... - Oct 23, 2002 by pizz

It sounds like you already know the answer! How much does i... - Oct 23, 2002 by RDR

Thanks, Russ, and Stephen. Valuble comments all. I also dow... - Oct 23, 2002 by Steve Thomas

thank ur kind help. the above two processes,which is more ... - Oct 24, 2002 by pizz

Steve Thomas

#21878

two-sided design guidelines | 9 October, 2002

Is anyone out there aware of any standard or even a section of a standard that specifically addresses design guidelines for two-sided assemblies, particularly one that focuses on different processes?

I'm being asked to produce information from said standards, and I have no recollection of the existance of anything of the kind. Specifically, we're looking for guidelines for design to optimize a glue/wave bottom side process AND the same for a reflow/selective wave (with pallet) process. Currently the boards are designed with all pth on the top and all smt on the bottom, but that may change depending on what we find and can convince engineering to do.

Again, I doubt the existance of such but pointers to anything close would be appreciated. And yes, I tried the obligatory archive search but combinations of dfm, bottom-side, two-sided, circuit side, produced nothing of any value in this context.

Secondly, I'm looking for any information re: any relationships between overall pcb size and smt reliability. We're trying to decide if we should reduce the size of some boards to avoid potential failures derived both from rough handling during assembly and mechanical stresses in the use environment.

If it's not already apparent, we don't do testing here to find this stuff out for ourselves. We just hope someone else has and has written a paper publishing the results so we don't have to.

reply »

RDR

#22011

two-sided design guidelines | 18 October, 2002

Steve, I am not aware of any standards so to speak. But IPC 782 and 770 have some useful info regarding mounting and pad geometries for wave etc...

Since I do not know what type of components you will be wave soldering I will just give you some input on things I have learned.

Orient all chip parts perpendicular to the wave, extend length of pad .2mm or so and reduction in width is sometimes beneficial SOIC packages should be mounted with the leads perpendicular to wave with solder thieving pads at the trailing edge. Make sure that the height of components is not so much that they interfere with nozzles etc... Component layout also needs to address potential shadowing by placing taller/longer components in front of smaller parts. Does your wave have a good "chip wave" such as the rotary (electrovert) or Gemini (Sensby)? A vibrating laminar flow (omega option on electrovert, oops I mean speedline) option also helps quite a bit in the elimination of skips especially with Tant Ds and similar components (tall height relatively short pads). We use the chip wave to solder the parts and the laminar wave to cleanup shorts, excessives, etc... I should note that caps and resistors can be soldered with a laminar flow alone with great results

For selective wave pallets you want to ensure that you have a minimu of .150" from PTH to the closest components to be masked if they are caps and resistors and such(possibly greater depending on component height, the taller the fixture the more clearance is needed to ensure that the solder can get up to the board). Also the height of components is critical so your fixture is not to deep and prevent solder from reaching board for a long enough time or you heve to turn the wave up so high that when the fixtures hits it you squirt solder out from behind it so far that the operator has to run and duck. Call pentagon EMS (503-924-2747) and talk to Chad for some more specifics on pallet design and requirements.

I know nothing about size and reliability except speculation that the bigger it is the harder it falls?

Hope this was helpful

Russ

reply »

Stephen

#22061

two-sided design guidelines | 21 October, 2002

Check out http://www.aimtronics.com/DFMDecember2000.pdf I"m not sure if there is anything usefull to you in it, and if you are going to check it out check it out soon, the company is in recievership.

reply »

RDR

#22075

two-sided design guidelines | 22 October, 2002

There is some pretty useful info in there

reply »

pizz

#22101

two-sided design guidelines | 23 October, 2002

I want to know how many PTH or DIP components is the critical point,when we choose the process,two side reflow or one reflow other wave soldering? it is not easy to make decision ,the cost of the hand soldering is cheap in our country. any comments will be appreciated.

reply »

RDR

#22108

two-sided design guidelines | 23 October, 2002

It sounds like you already know the answer! How much does it cost you to solder P.T.H. by hand (number of pins/per hour, / cost of assembler) vs. wave preparation and time of loading/masking etc... Cost of rework from each process?

Only you can make that determination.

Russ

reply »

Steve Thomas

#22109

two-sided design guidelines | 23 October, 2002

Thanks, Russ, and Stephen. Valuble comments all. I also downloaded the Aimtronics doc. Now, to convince someone to read something like that.....

reply »

pizz

#22124

two-sided design guidelines | 24 October, 2002

thank ur kind help. the above two processes,which is more reliably,other words,robust? I am in a dilemma. some suggest that 50 pins is the critical point,once exceed the quantity,the hand soldering is not practical. I prefer to two sides reflow.but it results the heavy labour.

reply »