3
\$\begingroup\$

I have been doing hardware design and PCBA prototyping for a while and I have issues with quality of reflow soldering. I have a small lab including a pick'n'place machine, stencil printer (eC-stencil-mate) and reflow machines (Zb5040hl and ec-reflow-mate). I thought the issues that I am facing is caused by the Chinese reflow machine so recently I purchased a second hand reflow machine, manufactured by EuroCircuits. But still I have same issues:

  • solder balls around the SMT component
  • stains on the pcb due to fluid of the paste
  • solder balls on the solder joint

It looks like the paste explodes during preheat/soak or it just floats away from the pad of the components...

The soldering procedure is done like this:

  1. Get out the paste from fridge (4°C). Paste was opened ~4 month ago and it is stored in fridge for that time.
  2. Mix the paste with hand tool for ~2 min
  3. apply paste to a clean stencil
  4. print the paste with 1 swipe on the stencil
  5. remove PCB from stencil printer and place components to the board (manually or with machine)
  6. put it into the reflow machine and reflow

I selected the SN42/BI58 paste because of its low temperature specifications. Sometimes I work with temperature sensitive elements and it is nice if we don't apply more than 200°C. The reflow temperature is ~139-141°C with peak soldering temperatures of 175°C. I tried to follow the reflow profile as much as possible with the EC machine but still I see the same issue.

Datasheet states a 2.5 °C / sec rate for heating and it is not violated. The highest rate is ~1.2 - 1.5°C / sec at preheating stage.

Soldering with chinese machine (peak 200°C is reached in 280 sec): zb5040hl

Soldering with EC-Reflow-Mate (peak 180°C reached in 210 sec): ec_reflow_mate_1

Soldering with EC-Reflow-Mate (Peak 180°C reached in 330 sec): enter image description here

Soldering profile for the last (330 sec): enter image description here

What can be the issue? Can the paste be scrap? Am I introducing bubbles into the paste during mixing?

Update: Testing the suggestions, I made major improvement. Using brand new paste, not mixing, same profile as 3. test (330 sec) I do not see much solder balls, just small flux. I don't know if the flux is normal or still I need to tune the profile.

Stencil printing: Stencil printing

Result: Soldering with new paste New old

Pastes (1. new not opened, 2. Freshly opened, 3. old paste. I do not see too much differences between old and new): pastes

\$\endgroup\$
6
  • \$\begingroup\$ Paste was opened ~4 month ago and it is stored in fridge for that time. you didn't mean "Opened and left in the fridge with its lid's taken off", did you? Don't get offended please, it's my English... \$\endgroup\$
    – Rohat Kılıç
    Nov 21, 2022 at 15:56
  • \$\begingroup\$ hey, I opened a fresh new paste like 4month ago. I put the lid back but the original packaging was removed (plastic cover). I meant, I used the same paste for this experiment that I used for 4 month of period already. When the paste was not used it was stored in fridge. But tomorrow I will take some pictures from this "old" paste and a new one without mixing as @voltage-spike recommended. Sorry I am not a native speaker. \$\endgroup\$
    – D_Dog
    Nov 21, 2022 at 16:16
  • 1
    \$\begingroup\$ This has helped me with reflow issues... smtnet.com/library/files/upload/SMT-troubleshooting-guide.pdf \$\endgroup\$
    – user324996
    Nov 21, 2022 at 17:16
  • 1
    \$\begingroup\$ Some crisp photos of the paste before population and before soldering might help too. Solder balls could be from smeared paste. \$\endgroup\$
    – Spehro Pefhany
    Nov 21, 2022 at 17:18
  • \$\begingroup\$ I added some photos. Looks like the paste that I used was in bad condition. Still I think the reflow profile needs some adjustments, because I think too much flux remained around the pads. \$\endgroup\$
    – D_Dog
    Nov 22, 2022 at 14:51

1 Answer 1

Reset to default
1
\$\begingroup\$

I've never heard of mixing the paste before stenciling and in 15 years of prototyping I've never had to mix paste. If the paste looks like it's separated from the solder then you might have a storage issue or you might want to contact the manufacturer. The flux and paste should be homogeneous.

So it could be the flux or the mixing of air into the paste which could create issues.

It does look like it's some kind of flux issue because the solder has melted.

You could do some experiments and put some rework flux on it and heat it up with a soldering iron and see what happens.

You might also just want to kick the oven profile up 10 or 20c and see if you get better results on a test board

\$\endgroup\$
7
  • \$\begingroup\$ Hey, thanks for the comment. Tomorrow I will give it a try to this. I have multiple batch of paste (fresh new, not opened) so I can open a new one and try it without mixing. The thing is I used this same procedure for ~1 years of period and the quality was changing between manufacture. Sometimes it looks perfect, no balls, sometimes there are some solder balls and sometimes almost the whole assembly is ruined by the balls and flux stains... \$\endgroup\$
    – D_Dog
    Nov 21, 2022 at 15:56
  • 1
    \$\begingroup\$ FWIW, I always mix, but also allow the paste to warm to close to room temperature (30 min+) . Jacking up the temperature to a lead-free profile with PbSn eutectic covers a multitude of sins. \$\endgroup\$
    – Spehro Pefhany
    Nov 21, 2022 at 17:26
  • 1
    \$\begingroup\$ @Spehro Thats interesting, 2 for mixing one not. I figure that mixture is homogeneous, and if you keep it stored properly it should stay that way (which is why you want to keep it cold). My guess is if you do decide to mix, don't mix air into the mix. \$\endgroup\$
    – Voltage Spike ♦
    Nov 21, 2022 at 18:14
  • 2
    \$\begingroup\$ @VoltageSpike Air doesn't seem to be a problem. In any case there's rougher mixing in distributing it on the stainless stencil and mushing it with the squeegee. \$\endgroup\$
    – Spehro Pefhany
    Nov 21, 2022 at 18:19
  • 2
    \$\begingroup\$ @SpehroPefhany Good enough is good enough, eh? \$\endgroup\$
    – Voltage Spike ♦
    Nov 21, 2022 at 18:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.