TECHNET Archives

February 2007

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Stadem, Richard D." <[log in to unmask]>
Reply To:
TechNet E-Mail Forum <[log in to unmask]>, Stadem, Richard D.
Date:
Mon, 12 Feb 2007 14:55:33 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (1 lines)
Werner, thank you for your note. I totally agree. 

I misread Inge's original email. I thought he said putting a thermocouple on top of the SOLDER PAD was misleading. What he actually said was putting a thermocouple on top of the component body was misleading. That is true. 

Another note. I totally agree the best way to thermocouple a profile test board is to drill up into the solder joint, insert the TC from the bottom so the weld is inside the solder ball. However, I then backfill the hole with thermally conductive epoxy with a minimum thermal conductivity of 14 BTU/in. hr. F. This helps reduce the insulative effect of the PWB just below the weld inside the hole. After the thermally conductive epoxy is cured, I then measure the electrical resistance of each of the thermocouple wires to the solderball via. I should get less than 1 ohm. If I do, then I know that the weld is making full mechanical contact with the inside of the solderball and/or the pad. When the mole or data recorder runs through the reflow oven, and I see a very solid line on the graph, I know I am measuring the actual temperature of the solderball/pad mechanical junction. If the profile line is not a solid line throughout, but is "noisy" with little spikes in both directions, then I know I am only measuring air temperature under the BGA, and I disregard that particular reading. The attenuating effect of full metal contact is easily seen.

 

Incidentially, I recently wanted to verify that a small bake oven was getting a particular subassembly up to a certain temperature in a pre-baking operation. When I suspended  thermocouples around the component within the chamber (measuring the air temperature), the 4 thermocouples measuring the air temperature consistently measured 6 degrees C. less than the 5 thermocouples that I had embedded into the subassembly using the method I just described. This was long after all of the TC readings had flatlined. 

The calibration lab questioned as to how an object being warmed by the air currents inside the oven could be hotter than the air currents themselves. The only explanation I could offer is that the thermal transfer efficiency of air directly to a thermocouple weld is less efficient than a piece of metal heated by the air making direct contact with the weld. I proved this by taking the thermocouples measuring the air and switching them with the thermocouples measuring the subassembly. The measurements were exchanged, one for one. Again, the TCs measuring air were all exactly 6 or 7 degrees cooler than those embedded in the assembly. I then took one of the TCs measuring air and clamped it to the metal rack in the oven. It went up 6 degrees.

 

The same phenomena holds true for a thermocouple weld sitting just above a BGA pad, measuring air temperature alone. I notice it always measures a few degrees cooler than a TC with a good mechanical attachment to the solder or the pad, even when the solder is in liquidus.



________________________________



From: [log in to unmask] [mailto:[log in to unmask]] 

Sent: Monday, February 12, 2007 1:30 PM

To: Stadem, Richard D.; [log in to unmask]

Subject: Re: [TN] FW: Report from a BGA





Hi Richard,

Ingemar is absolutely correctthe only way you know what the t's are at the solder balls is to put thermocouples in them. It is indeed standar5d procedure see IPC-7530, 7.9.5.

BGA solder balls get heated by conduction through the PCBmany people do not realize thisand BGA component body is in fact a heat sink and a shield preventing both convection and radiation heat transfer.



Regards,

Werner Engelmaier

可靠性先生

Engelmaier Associates, L.C.

Electronic Packaging, Interconnection and Reliability Consulting

7 Jasmine Run

Ormond Beach, FL 32174 USA

Phone: 386-437-8747, Cell: 386-316-5904

Werner Engelmaier 


ATOM RSS1 RSS2