Author Topic: My first baked boards have had a couple of issues... (Read 6445 times)

alank2 · « **on:** July 01, 2014, 01:44:33 am »

A couple weeks ago I did my first bake of 10 boards. 2 of them had issues related to a weak connection. 1 I tried to rebake with the display on it melted the display so I tossed it (I didn't know if the display could tolerate the heat, I found out it couldn't).

They have a quad 7 segment display, and the other problem board would dim the top segment of the display if you pushed slightly int he middle of the board from the backside (flexing it ever so lightly). Taking the soldering iron to that signal path with a bit more solder fixed the issue.

I used a 5 mil stencil and the nominal 7351 footprints, but to be honest being a hand solderer I'm use to seeing quite a bit more solder at joints than what this resulted in. Does anyone use a stencil thicker than 5 mil? I got the stencil at oshstencils and they had 3mil and 5mil. Should I move to the larger footprints "7351 most"? Should I try to find a thicker stencil?

My issue might have also been that I didn't get my reflow oven hot enough so perhaps I should have pushed it a bit hotter.

Thanks,

Alan

Psi · « **Reply #1 on:** July 01, 2014, 04:15:30 am »

You also may have hot/cold spots in the oven

tautech · « **Reply #2 on:** July 01, 2014, 05:02:23 am »

You might find some clues in this thread:
https://www.eevblog.com/forum/altium/altering-the-paste-mask-for-correct-paste-coverage-on-large-pads/

David_AVD · « **Reply #3 on:** July 01, 2014, 06:31:14 am »

Just curious - leaded or lead-free solder?

alank2 · « **Reply #4 on:** July 01, 2014, 12:05:47 pm »

Leaded.

Psi · « **Reply #5 on:** July 01, 2014, 12:06:40 pm »

Was your solder paste stored in the fridge?
and was it brought up to room temp before reflow?

alank2 · « **Reply #6 on:** July 01, 2014, 12:09:10 pm »

Yes and mostly.

Psi · « **Reply #7 on:** July 01, 2014, 12:30:32 pm »

A picture of the bad solder joints may help us tell what the issue is.

alank2 · « **Reply #8 on:** July 01, 2014, 01:42:07 pm »

Here are pics:

Pin 9 you can see where I took the soldering iron to it to fix the flexing the board dims the top segment issue:

http://home.earthlink.net/~alank2/reflow/bake1.jpg
http://home.earthlink.net/~alank2/reflow/bake2.jpg
http://home.earthlink.net/~alank2/reflow/bake3.jpg

Thanks,

Alan

philpem · « **Reply #9 on:** July 01, 2014, 11:32:08 pm »

Quote from: alank2 on July 01, 2014, 01:42:07 pm

Here are pics:

Pin 9 you can see where I took the soldering iron to it to fix the flexing the board dims the top segment issue:

http://home.earthlink.net/~alank2/reflow/bake1.jpg
http://home.earthlink.net/~alank2/reflow/bake2.jpg
http://home.earthlink.net/~alank2/reflow/bake3.jpg

Thanks,

Alan

The chip looks like it's floating on top of the solder -- like it never sank down into the solder when it reflowed, or the pins were bent. Is that an optical illusion, or is the chip really like that?

I'd be looking at adding an oven fan to the reflow oven; I did that to mine and it massively improved the reflow consistency.

IanJ · « **Reply #10 on:** July 01, 2014, 11:46:55 pm »

There's some dull pads.......looks like lack of heat. Not much self centering going on either.

Ian

Psi · « **Reply #11 on:** July 02, 2014, 12:03:38 am »

Yeah, im going to say not enough heat as well.

Same reasons as others above..

Unless the solder melts on all the chip pads the chip wont sink down, which it looks like it didnt from the pic.
(Pressing down on the chips when you place them gets rid of this prob but shouldnt be needed)

Also a lack of heat stops the passives from centering on the pads (they can only do this when the solder is in fully liquid state) and i can see a passive at an angle which implies you had that issue.

Also the solder isnt bright and shinky which is another sign it didnt get up to temp.

alank2 · « **Reply #12 on:** July 02, 2014, 12:25:25 am »

Thanks guys, I'll push it hotter next time and see how it goes.

Do you guys think there is enough solder? Should I move to the larger sized "most" pads too?

Psi · « **Reply #13 on:** July 02, 2014, 12:28:06 am »

Solder amount looks ok to me.

marshallh · « **Reply #14 on:** July 02, 2014, 05:12:48 am »

If it failed you only have 2 possibilities:

1. Poor paste quality
2. Wrong reflow profile/inconsistent heating

In this case it look like the connections didn't flow properly, probably because the flux burned off prematurely. You had too much dwell/soak time or took too long to get up to reflow temperature.
Even leadfree shouldn't pose that much of a problem.

alank2 · « **Reply #15 on:** September 09, 2014, 09:31:10 pm »

I'm still at trying this guys. I've moved to connecting the temp probe to the pcb by putting it through one of the through holes at an angle for physical contact.

How does this profile look (for leaded)? Any changes or places I should extend the time. The blue is SP, red is PV, and thin green is the CO. I drove the SP to 230 at the end knowing I wouldn't get there but wanted to push the CO to 100%.

Here is the profile I'm using (custom software):

<Profile>
<Name>Amtech 4300 63Sn/37Pb Profile A</Name>
<Step>
<Action>set_auto</Action>
<Name>Stage 1 - Preheat</Name>
<SP>140</SP>
<MaxChange>0</MaxChange>
<Fan>1</Fan>
</Step>
<Step>
<Action>delay_threshold</Action>
<Name></Name>
<Test>Greater Than Or Equal</Test>
<PV>140</PV>
<Seconds>0</Seconds>
</Step>
<Step>
<Action>set_auto</Action>
<Name>Stage 2 - Soak</Name>
<SP>170</SP>
<MaxChange>0</MaxChange>
<Fan>1</Fan>
</Step>
<Step>
<Action>delay_threshold</Action>
<Name></Name>
<Test>Greater Than Or Equal</Test>
<PV>140</PV>
<Seconds>90</Seconds>
</Step>
<Step>
<Action>set_auto</Action>
<Name>Stage 3 - Reflow</Name>
<SP>230</SP>
<MaxChange>0</MaxChange>
<Fan>1</Fan>
</Step>
<Step>
<Action>delay_threshold</Action>
<Name></Name>
<Test>Greater Than Or Equal</Test>
<PV>210</PV>
<Seconds>10</Seconds>
</Step>
<Step>
<Action>set_auto</Action>
<Name>Stage 4 - Cool Down (Open Door)</Name>
<SP>0</SP>
<MaxChange>0</MaxChange>
<Fan>0</Fan>
</Step>
<Step>
<Action>delay_threshold</Action>
<Name></Name>
<Test>Less Than Or Equal</Test>
<PV>70</PV>
<Seconds>0</Seconds>
</Step>
<Step>
<Action>end</Action>
<Name>Completed</Name>
</Step>
</Profile>

tggzzz · « **Reply #16 on:** September 09, 2014, 09:49:49 pm »

Quote from: philpem on July 01, 2014, 11:32:08 pm

I'd be looking at adding an oven fan to the reflow oven; I did that to mine and it massively improved the reflow consistency.

What fan did you use? (I'm concerned about fans @ 200C)

hamster_nz · « **Reply #17 on:** September 10, 2014, 01:54:23 am »

Quote from: alank2 on September 09, 2014, 09:31:10 pm

I'm still at trying this guys. I've moved to connecting the temp probe to the pcb by putting it through one of the through holes at an angle for physical contact.

The profile looks good - what were the results like? In the earlier pictures it looked like quite a few things didn't "wet" - e.g. R2 & R3.

I found this paper in google : http://www.nordson.com/en-us/divisions/efd/Literature/White-Papers/Solder/Nordson-EFD-First-Principles-Solder-Reflow.pdf

Quote

Beyond the 130°C point, where water and most low boiling point materials will have finished evaporating, you have the temperature range up to alloy solidus. At these temperatures, the activators in the flux fulfill their function, cleaning the solder and substrate of oxides. Spend too long here and the flux activity can be mostly or completely consumed, resulting in poor wetting and what appears to be unreflowed solder despite achieving a peak well above alloy liquidus. Longer heating processes can be accommodated with higher activity flux formulations. As a goal, do not spend any more time between 130°C and the alloy solidus than is required for acceptable point-to-point temperature differential on your product.

Your problem looks like this.

Humm - Is the past that you are using OK to use for reflow? do you keep it in a fridge? maybe it has gone stale?

alank2 · « **Reply #18 on:** September 10, 2014, 02:00:07 am »

I just did a very tiny board 0.4" x 0.4" and it has a couple of 0603x4 resistor networks where the paste did not look shiny/pretty but more of a gray/peppered. On the smaller SC-70-6 it was actually bright silver looking, but I placed one unstraight and it was hard to get paste on the pads nicely with the stencil I was using. I didn't have the temp probe on the tiny pcb, but a much larger pcb that I had it mounted to so maybe that and not stirring the paste was the problem. I am going to retry it.

alank2 · « **Reply #19 on:** September 12, 2014, 01:48:51 am »

Did my second "run" of boards tonight, 14 of them. I did them all at one time and I noticed that the solder didn't melt as quickly on the front ones near the glass so I ended up leaving the door closed for an extra 10 or 15 seconds. My graph showed it made it up to 235 even though the power was cut off at 210 and when I opened the door it was smoking a bit. Not sure if the flux was burning off but a small bit of smoke was visible and smelled. I was a bit concerned, but all 14 boards program and seem to work perfectly. Should there be a bit of smoke? I somewhat remember something a bit similar the first time, but not as strong of obvious.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: My first baked boards have had a couple of issues... (Read 6445 times)

Share me