Reflow profile negative temp ramp rates

Subscribe to Thread

Question. When profiling and looking at the profiles cooling... - Apr 09, 2002 by Craig

Craig, it depend's n what you want to look at. In term's ... - Apr 09, 2002 by JohnW

Read IPC-7530 "Guidelines for Temperature Profiling for Mass... - Apr 09, 2002 by davef

Hey Bud. Allow me to 'turn' your point about 'grain structu... - Apr 09, 2002 by davef

Dave, I sand corrected since I know you wouldn't steer me wr... - Apr 11, 2002 by JohnW

John, I would not INTENTIONALLY steer you wrong [most of ... - Apr 11, 2002 by davef

Craig

#19428

Reflow profile negative temp ramp rates | 9 April, 2002

Question. When profiling and looking at the profiles cooling rate. Over what temperature range should the slope be monitored and what sort of cooling rates should be looked for. Paste supplier says no more than 6degC per sec, some components maybe lower, I try to aim for no more than 4degC per sec. But have noted some steeper drops on some parts in the cooling zone. I'm not sure how much of a temperature drop is bad eg is a drop of 5.5degC/sec for 30degC as bad as 5.5degC/sec for 100degC.

Thanks

reply »

JohnW

#19432

Reflow profile negative temp ramp rates | 9 April, 2002

Craig,

it depend's n what you want to look at. In term's of te gran structure of the solder jont that's formed when the joint is cooling so you'd want to monitor the temperature drop from the peak or say 215dg C for ref down to probably about 150 and you want to keep the ramp as slow as possible. A 6 deg Ramp at cooling will give you a fairly course grain structure I would ave thought, generally I'm aiming for 2-3deg C max. Things like voiding and so forth in BGA joint's should be settled by that point as well. The important thing is how many samples your software uses to calculate the slope, if you use more you'll get a more accurate indication of what the ramp up or own is and obviously the inverse would be true. John

reply »

davef

#19435

Reflow profile negative temp ramp rates | 9 April, 2002

Read IPC-7530 "Guidelines for Temperature Profiling for Mass Soldering Processes (Reflow & Wave)".

There is no advantage in dropping off much faster than 4�C/sec. In fact, you risk fracturing your newly formed solder connections due to board warping because of the thermal shock temperature gradients of high ramp cooling.

That being said, I know nothing compared to your paste supplier. It's tough to figure that their recommendations would not be well researched to produce excellent solder connections.

reply »

davef

#19436

Reflow profile negative temp ramp rates | 9 April, 2002

Hey Bud. Allow me to 'turn' your point about 'grain structure' just a bit.

Fast cooling rates / coarse grain structures was a bit of red herring in the early 80's. Faster cooling DOES produce finer grained solder connections. Accelerated testing of these connections show a ~2X longer life. Unfortunately, the grain structure of solder coarsens with time and temperature. After one year on the shelf, you cannot tell the difference between solder connections with coarse or fine grain structures. As you'd expect, the rate of 'grain coarsening' is faster in most in-use products than those in storage.

That is why both IPC-SM-785 and IPC-9701 strongly recommend artificial aging to coarsen the solder grain structure prior to temperature cycling to prevent unrealistically high test results.

reply »

JohnW

#19467

Reflow profile negative temp ramp rates | 11 April, 2002

Dave, I sand corrected since I know you wouldn't steer me wrong...although the lst time I looked at the IPC spec for a solder profiel it was still designed for older generations of paste and were specifically water wash biased although that doesn't change the metarlurgy side of life. John

reply »

davef

#19473

Reflow profile negative temp ramp rates | 11 April, 2002

John,

I would not INTENTIONALLY steer you wrong [most of the time].

But this is just a GREAT example of how we learn something and take it to mean something, only later to learn that while what we learned is true, it means something else.

reply »