Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Reflow Profiling - Cooling Rate?

Reflow Profiling - Cooling Rate?

Subscribe to Thread

Who's controlling the cooling rate when profiling? If so, w... - Jan 31, 2020 by dontfeedphils

We never really "controlled" it, per-se...more a f... - Feb 03, 2020 by robgd3

A few lifetimes ago I went to a seminar on reflow profiles. ... - Feb 03, 2020 by Stephen

Reason I ask is because I've never really paid much attentio... - Feb 03, 2020 by dontfeedphils

There are two areas that are important when discussing cooli... - Feb 04, 2020 by Fred Dimock

Hello Phil, As the other respondents have mentioned, the co... - Feb 10, 2020 by SMTA-Tony SMTA

on 10 zone ovens with 3 cooling zones, my negative slope is ... - Feb 10, 2020 by Evtimov

The ideal cooling zone curve should be a mirror image of the... - Apr 22, 2020 by SMTA-Samy SMTA

This is the first time I heard that the heating and cooling ... - Apr 23, 2020 by Fred Dimock

dontfeedphils

#84345

Reflow Profiling - Cooling Rate? | 31 January, 2020

Who's controlling the cooling rate when profiling? If so, what's the max cooling rate you're trying to hold?

reply »

robgd3

#84356

Reflow Profiling - Cooling Rate? | 3 February, 2020

We never really "controlled" it, per-se...more a function of the speed coming out of liquidous.

But, the accepted best practice is 4°C-6°C/second.

Cheers, ..rob

reply »

Stephen

#84359

Reflow Profiling - Cooling Rate? | 3 February, 2020

A few lifetimes ago I went to a seminar on reflow profiles. The speaker emphasized more than once that the cool down rate is important. Someone asked him about controlling the cool down rate and he acknowledged that you can't. And IIRC it was an oven manufacturer hosting the seminar.

One lifetime ago, we had a board with a QFN. Someone had tweaked the profile to where there was no voiding. The customer wanted the profile to match the data sheet. I had to peak on zone 7 of a a 9 zone oven to get the cool down rate to what the data sheet said. We had some voiding after that.

One thing to watch out for is to make sure that your profiling software does not have too short a time to base the cool down rate on. If it is too short it can give you a higher cool down rate than is accurate. If there is a little fluctuation that is longer than the time in the program then it can give a high number. I'm not sure if I worded that well.

reply »

dontfeedphils

#84360

Reflow Profiling - Cooling Rate? | 3 February, 2020

Reason I ask is because I've never really paid much attention to the cooling rate, but I'm currently running a 10 zone oven and using all 10 zones for heating whereas I've almost always used a 10 zone as 9 heating and the last as a cooling.

I'm currently hitting about 4.5c/sec in my initial profiles, so I'm happy.

reply »

Fred Dimock

#84362

Reflow Profiling - Cooling Rate? | 4 February, 2020

There are two areas that are important when discussing cooling rate.

One is the rate thru the liquidus, where a fast rate produces small grains in the metal and a slow rate results in larger grains. Although smaller grains are stronger normal SMT cooling rates of 1 to 6 C/sec don't seem to make a big difference. (note a max of 6 C/sec is specified for many boards)

The second area is high stress during cooling on large BGAs etc. due to large delta Ts on the parts. Sometimes large components are run with cooling rates of less than 0.5 C/sec.

Some reflow ovens have the capability to control the cooling rate by varying the cooling fan For very slow cooling some people use the last zones in the oven for cooling and some ovens have heat added to the cooler.

reply »

SMTA-Tony

#84389

Reflow Profiling - Cooling Rate? | 10 February, 2020

Hello Phil,

As the other respondents have mentioned, the cooling rate is not easily controlled and varies based on the oven and board design, as well as the settings of the last few zones and cooling zones. In general, the board is going to cool in air at ambient temperature at a certain rate which is determined by the board design and construction. Boards with higher thermal mass will cool more slowly than boards with low thermal mass. The cooling rate can be accelerated by increasing the air movement around the board.

Faster cooling rates are recommended to minimize inter-metallic growth and optimize the solder joint grain structure. In some cases I have seen fans added at the exit conveyor to help increase the cooling rate with high thermal mass boards. This is not normally needed for most surface mount assemblies. Please contact us directly if you would like to discuss this in more detail.

Thank you, Tony Lentz

This message was posted via the Electronics Forum @ SMTA SMTA

reply »

Evtimov

#84391

Reflow Profiling - Cooling Rate? | 10 February, 2020

on 10 zone ovens with 3 cooling zones, my negative slope is between 2.5 to 4 deg C/second based on board size and thermal mass.

reply »

SMTA-Samy

#84787

Reflow Profiling - Cooling Rate? | 22 April, 2020

The ideal cooling zone curve should be a mirror image of the reflow zone curve. The more closely this curve mimics the reverse of the reflow curve, the tighter the grain structure of the solder joint will be upon reaching its solid state, yielding a solder joint of higher quality and bonding integrity.

Source: https://smtnet.com/library/files/upload/how-to-profile-PCB.pdf

This message was posted via the Electronics Forum @ SMTA SMTA

reply »

Fred Dimock

#84789

Reflow Profiling - Cooling Rate? | 23 April, 2020

This is the first time I heard that the heating and cooling rates should be identical so I downloaded a copy of the paper to see what it said. There is one short paragraph claiming that the rates should be equal but no references or data. This doesn't mean it is wrong but it would be best to have more information before modifying our processes.

Two comments - There are times that a slow cooling rate is needed i.e. large BGAs, and the Feb 10th post by Tony Lentz clearly states how the cooling rate can be slightly modified in a reflow oven.

reply »