Effects of moisture on CPLD

Dave,

The packages may have absorbed moisture, and if you put them through a normal solder wave machine, you will have popcorn.

The devices are prone to crack, and shatter due to the moisture turning to steam, and having no where to go.

Parts may be baked before assembly to remove the water.

Aust> I have twenty Xilinx XC9572XL VQFP64. Their sealed package has been open two

two

Now I

people's

Thanks Austin - It sounds like we we would definitely notice if there was a problem then?! From the white paper it suggests a minimum of a 12 hour cycle at 125degC is needed.

Dave

two

people's

Reflow soldering without baking them (probably 120 C for 24 hours) might give problems. Hand soldering should be OK: I never have problems, anyway.

Leon

Now

Thanks Leon. That's interesting, as the first few are going to be hand soldered (and I don't have a proper oven at the moment).

Dave

Dave,

Yes, you would notice a difference! Not all parts will pop, and those that do, may not even be noticed initially. Later they will fail as they will now absorb water like crazy from all of the cracks.

Aust>

open

anyway.

If you get a PCB assembly company to do the others, they should be able to do the baking for you. They probably do it all the time.

Leon

I'm not so sure about that. I bought a handful of microcontrollers about a year ago and stuffed boards, and all was fine. I never resealed the bag (I don't have the facility) so they sat until I used them a month ago. I handsoldered one prototype and it failed in a particular way. After much gnashing of teeth (I hate shotgunning parts), I removed the micro and replaced it with another from the same bag, which worked. I have since hand-built three more prototypes and I've had a 50% fallout. I guess I should've baked the parts...

-a

anyway.

Thanks Andy. That's interesting - One of the assembly houses I've come across on the web says the parts would need baking before they will hand solder them . This seems to concur with your experiences.

Dave

Perhaps one of those BlackNdecker toaster ovens will do the job since they were mentioned awhile back as being useable for home based reflow ovens in another thread on BGA pcb work.

regards

johnjakson_usa_com

Yes, I was thinking about one of those too (just as long as they don't mind running constantly for 24 or 48 hours.)

Dave

I also have been hand-soldering various Xilinx and other chips for years. I've done Xilinx Spartan FPGAs in the PQ208 package without any problems, and 9500 and Spartan parts in the PC84 also.

The chip never gets very hot in this process, maybe some local areas around the edge reach 100 C when I'm having trouble and clearing out some solder bridges. This is nowhere near the temperatures maintained when IR reflow soldering boards full of these components. Now, the lead-free stuff will push these processes to even higher temperatures.

Jon

I am sure any oven that will maintain the required temps will do. But I am not convinced that hand soldering without oven drying will cause failures. I have had many boards built by hand without any problems that I can relate to moisture.

I have real life experience with moisture effects with BGA parts, Bake them at the recommended temp for the time specified on the bag, or yoiu will get disconnected wires to the pins. I found that my assembler used parts stored in an 'office environment' for several days, and *ALL* parts turned out to have bad connections when the boards were done. The humidity issue is very real, follow the directions on the bag, really.

A burned designer

two

people's