LQFP vs BGA hand soldering?

[FJB2069]

Anyone have any experience with BGA hand soldering?   I have always tried to stay away from BGA parts as far as prototyping and being able to rework a pcb.  But now with choices between small pitch LQFP parts and BGA I am wondering if a 176 - 206 ball bga may be easier to work with than the same pin count LQFP?

 

Any feedback is appreciated.

 

 

[Rodo]

I think LQFP is better than BGAs.  I've hand soldered both but there are advantages to LQFP. A BGA board may require more layers on the PCB to route all "balls" out to other components and you won't be able to test the pins (balls) unless you fan them out.  Production cost may increase for fine pitch BGAs where the vias to route out traces are very small (0.50mm pitch on BGA).

[FJB2069]

Thanks for the advice!  I didn't think about the via and and layer requirements.

[Tesla DeLorean]

Via in-pad, shrinkage of solder mask and proper fill / planarization can also be issues depending on your design rules, and PCBA limits 

But as @Rodo suggests, the size trade-off tends to increase board costs, and rigors of board manufacture and assembly. High ball counts make pin escape more challenging, and you need to think about how the AF muxing can help or hinder pin choices.

FMC and LTDC tending to have large counts and limited choices/options

Even for small pin-count, my PCBA guys are much more comfortable with the LQFP or LQFN type devices over BGA or WS

[gregstm]

I've always used LQFP parts up (to 100 pins) because I can hand solder them and I can get away with a two layer board (using the bottom layer mostly as a ground plane). I have dreamed of using a BGA for its small size but I think it is a game for the big players. But I might still give it a go one day.  I have thought of trying the UFBGA 100 package because with certain applications you could possibly get away with a 2 layer design - but it seems that package is not as common as others (and I'm not sure it would be worth the effort)