The chips are not always adequately soldered mystery

[michael]

Loose

I just had the very unwelcome experience of a big RF module go sliding off the PCB after a couple weeks of light use. It is a Microchip RN2483 with castellated edges.

---

---

See all the edge pads? I would have thought it took a gorilla to pry off the chip, but (without dropping the PCB and) through normal use it came off after two weeks.

Suspicion

This is almost surely due to me not using enough solder paste before reflowing, and I use too little paste probably when looking at the small pads and thinking:

I better not put too much paste on there otherwise it will short to the next pad.

Theory

I guess half the solder moves up the leg, pin, or IC contact when reflowing. Only the other half stays on the pad and serves to hold the parts together.

My new idea is to apply more paste, even if it's overflowing and covering two pads. I assume that once reflowing begins, half the paste will disappear into corners rather than creating a bridged short.

To be clear, this would happen with any reflow and solder application procedure, like with stencils.

Anyone with more experience care to confirm or reject these theories?

[cborn]

In your photo there appears to be adequate paste, but it doesn't appear to have "wet" at the connection to the module. Were the pads on the module pre-tinned, soldered, or bare?

My son just finished mid-year exams in his elec. engineering course, and when he came home last night I went to show him the V-One and a circuit I had made on it. When I picked it up the chip fell off the board, while all the passives appear to be bonded quite strongly. It had been working, but must have had minimal bonding. I wonder if it is due to different metal types, perhaps pre-treated contacts, lack of flux, too high/low temperature or hold time, or some other variable?

[michael]

@cborn Good idea, but I'm not sure what the condition a RN2483 has when it leaves the factory. I didn't manually apply solder, flux, or anything to the dozens of tiny pads so I guess we consider them bare?

Burnishing

On top of your questions about metal types, flux, and temperature, I wonder if the age of the ink or solder matters, and how hard we need to burnish the ink. I burnished the RF module pads quite a lot according to memory.

Reproduction

If this is a reproducible problem, it would be great to know under what conditions parts are not soldered adequately.

Which type of chip is the one that fell of your PCB after you soldered it? SOIC, DFN, QFN, or some other one? Maybe the mass of the part matters?

[cborn]

In my case it was the standard N555 supplied with the Hello World kit, but I think the age of the solder paste had a lot to do with it.
I've done a lot of large modules (the EMX module from GHI), and the thermal mass of the modules does mean that they will take longer to get up to temperature than passives. In the reflow oven it isn't such a big thing, but on the heated plate of the V-One with cooler ambient temperatures (winter here now) the larger parts may now make temperature, or hold there long enough. For reference, the IR camera showed the plate temperature at 180-190C, while the top of the PCB was 140C, when the GUI was showing the temperature as 210C, how does that compare?

[michael]

@cborn So you observed a SOIC-8 N555 chip improperly soldered and you think your paste is too old, is that it? The advice about thermal mass is great in my case, and I'll use a manual reflow profile to try to correct for the large RF module.

But large chip absorbing energy can't explain why a small N555 didn't reflow properly, so that's still a mystery unless you have a new idea.

I'm going to rename this topic from 'I didnt apply enough paste' since we are no longer considering that a cause.

In the meantime, I've remounted the RN2483 that broke off of my PCB and hand soldered the individual pads without applying new solder. Then I applied new solder to the four corner pads to try to hold the chip down without trying to solder all 48 pads and probably create short circuits.

Keep your fingers crossed while I test this...

[michael]

@cborn Do you have the impression that the N555 was solidly fixed to the PCB for a number of days or weeks, and that the binding quality wore off with time? Was it 10-15 days like my case with the RN2483?

[cborn]

I can't say it was solidly soldered, although I didn't expect it to fall off when I picked up the PCB!

However in another case I had used the V-One to apply the paste for an industrial board, it all seemed quite solid (pressing on the IDC connector with my thumb didn't shift it), but it came back after a week or so with the connector broken off, the solder had let go.
Talking to Voltera guru @mewertow, his thought was that as the PCB was a commercial one with 60/40 lead based HASL finish, this could interact with the standard V-One paste to end up with a weak connection. I purchased some standard T4 solder paste to use for these applications, but haven't had chance to fully test it.
However @mewertow did some testing and has a suitable paste up onthe Voltera store now.

[rkagerer]

I know this is an old topic but did you guys ever isolate a likely cause for the chips that fell off the board?

I’ve done two projects recently and had IC’s fall off on both. One was the Hello World. It soldered well and seemed good. I had it in my knapsack for a couple weeks and I guess all the banging around was just too much - the chip eventually fell off. I’m not too upset or surprised about that one.

More recently I had an MSOP-8 (with a thermal pad underneath) come off when the amplifier chip attached to it got too hot. i.e. My theory is the chip got too hot and desoldered itself.

My paste was old, especially in the latter case. Under the microscope I could see the individual solder beads (not sure if that’s normal).

[eric.bumbalough]

If you can see individual solder beads, the solder never reflowed properly. We see this regularly on large components and occasionally on smaller things. Either do a manual bake or try to touch it up with an iron or hot air.

You can tell with the naked eye whether or not the solder reflowed. The color changes from dull grey to shiny metallic when it has melted properly.

[rkagerer]

I meant I could see them in the paste BEFORE the reflow / soldering.