Soldering on a Ground Plane

Subscribe to Thread

We've got this hard-to-Wave-solder board where all of the S... - Feb 10, 2000 by C.K.

CK: If you�ve done everything under the sun process-wise a... - Feb 10, 2000 by davef

Dave, to answer your questions, point-by-point: � Just o... - Feb 11, 2000 by C.K.

CK: I hear you about pads. Most secondary side SMT pads a... - Feb 16, 2000 by davef

C.K.

#1977

Soldering on a Ground Plane | 10 February, 2000

We've got this hard-to-Wave-solder board where all of the SMD's are on a ground plane - in fact the entire bottom-side is a ground plane. Our designers have incorporated thermal relief on some of the pads, but for reasons unknown to us, they don't have them on ALL of the pads. The result is at least 2 or 3 skips per board (out of 50 or so opportunities). The skips usually occur on those parts without thermal relief. I've done everything under the sun process-wise at the wave: different fluxes, more heat, less heat, more contact time, less contact time, and I'm running out of ideas. Even our touch-up operators have to hold the iron to the parts longer than usual when theey touch up the skips.

Any suggestions (design and/or process) would help.

reply »

davef

#1978

Re: Soldering on a Ground Plane | 10 February, 2000

CK: If you�ve done everything under the sun process-wise at the wave: different fluxes, more heat, less heat, more contact time, and less contact time; and are running out of ideas; let me table some thoughts.

� "Don�t mess with Mother Nature." Natural processes are defined, structured, and repeatable. You have a situation where the heat sinking ability of your board is greater than capability of your heat source. Your option: Relay-out the ground plane. Ya know, solid fill ground planes are (1) bad practice, because they create unbalanced construction, in addition to the process problems you state and (2) are usually not required, but designers don�t take the time to determine their radiation and design a grid based on actual radiation. And then, they get all huffy when questioned on why the solid fill is necessary. Geesh � And touch-up operators are probably doing something bad to the pooch when they hold their irons to the parts longer than usual, when touching up the skips. Mother Nature says "pay me now or pay me later." � Just out of curiosity, how are second side, skippy components oriented on the board with respect to the wave? � Just out of curiosity, how are second side, skippy components separated from other components? � What criteria are you using to judge these solder connections? Like is the fillet smooth, showing good wetting, and flowed greater than 25% up the component termination? Er something like that? � Finally, in your "everything under the sun" dance, did you try waving the board at an angle to the wave other than 90�, say like 45�?

My2�

Dave F

reply »

C.K.

#1979

Re: Soldering on a Ground Plane | 11 February, 2000

Dave, to answer your questions, point-by-point:

� Just out of curiosity, how are second side, skippy components oriented on the board with respect to the wave?

The SMD's are all oriented in the "correct" direction with respect to wave direction(i.e. solder connections on a part touch the waves at the same time).

� Just out of curiosity, how are second side, skippy components separated from other components?

There's lots of real-estate on the bottom side; hence, all components are nice distances apart from each other, and there's no higher-profile parts in front of lower ones. The skippy components are the ones where there's no thermal relief provided.

� What criteria are you using to judge these solder connections? Like is the fillet smooth, showing good wetting, and flowed greater than 25% up the component termination? Er something like that?

For the most part, we use IPC-class ii workmanship standards. When i say "skips", I mean there's no solder there what-so-ever.

...and lastly, I haven't tried solder @45deg. yet. You do bring up good points on your first bullet, though, where you say,

"solid fill ground planes are (1) bad practice, because they create unbalanced construction, in addition to the process problems you state and (2) are usually not required, but designers don�t take the time to determine their radiation and design a grid based on actual radiation. "

I forgot to mention also that these skippy parts are ALL 0603 componentry. We DON't get skips on the 1206's. My original theory with this was that the pads on the 0603's are a problem because 1.) there's not enough darn pad to solder to, and 2.) the flux burns off at a faster rate. On a positive note, the next rev., there'll be ALL 1206's.

reply »

davef

#1980

Re: Soldering on a Ground Plane | 16 February, 2000

CK: I hear you about pads. Most secondary side SMT pads are primary side SMT pads that are on the secondary side. This only matters when boards are wave soldered. Rather than changing to larger components, consider a bigger fight than that ... getting designers to use different pads on the secondary side.

WRT the component, secondary side pads are: * Narrower than the component * Wider, may be twice as wide, as a primary side pad

Reasons for doing this are: * Good fillet formation * Larger chip components are getting tougher to find

One source of the layout for these pads is Wassink "Manufacturing Techniques For SM Assemblies" ISBN 0901150304

Good luck

Dave F

reply »