9
\$\begingroup\$

I have an issue with solder paste, I would like to know its origin so I can fix the problem and solder by components properly.

I use a lead free Sn42/Bi57.6/Ag0.4 solder paste, manufactured by ChipQuick Here is the datasheet The syringe I'm using has been opened three weeks ago and stored at ambient temperature until now (I close it with the protective cap between each use, of course).

I ran some tests before actually using to solder components : I simply deposited several bits of it on a copper board, which I previously wiped with alcohol.

I have to my disposal a soldering oven (not some salvaged toaster, a real oven designed for this application). However, it works like regular timer-ovens : set a time with one button, and set a temperature with another.

So this is the process I used so far :

  1. I put the board in the oven, at ambient temperature
  2. I start the oven at 90°C and I wait one minute
  3. I set it to 140°C and wait for two minutes
  4. I set it for 180°C and wait for the solder paste to "melt" and get transformed to actual solder
  5. Finally, just after the activation, I turn off the oven and open the door to allow a quick return to ambient temperature.

Problem is : I always end up with a nice sphere instead of observing spread solder across the copper face. Exactly like this : this

I want to know if I am doing something wrong during the process, or if this is surely linked to the storage conditions of the solder. Note that the manufacturer indicates a good "shelf life" but I don't know if it implies that the container should not be opened.

\$\endgroup\$
  • 1
    \$\begingroup\$ Can you post a picture with your culinary achievements? \$\endgroup\$ – Dmitry Grigoryev Aug 25 '16 at 15:34
  • 2
    \$\begingroup\$ I wanted to, but I can't take pictures on my workplace. Just imagine a perfect metal sphere on top of flat copper with flux around it. edit : I found an equivalent picture on google, I'm adding it \$\endgroup\$ – MaximGi Aug 25 '16 at 15:37
  • 1
    \$\begingroup\$ @MaximGi Did you check reflow profile in the datasheet? You didn't provide amount of time which it takes for the solder to melt, but it might be too slow. \$\endgroup\$ – AndrejaKo Aug 25 '16 at 16:26
  • 6
    \$\begingroup\$ The Google picture looks like a "cold solder joint". A cold solder joint is when there is enough heat to melt the solder but not enough heat/time to overcome the heat sink offered by the parts being soldered. Happens a lot where larger masses are being soldered such as relays etc. \$\endgroup\$ – Harvard Aug 25 '16 at 16:29
  • 1
    \$\begingroup\$ I would also recommend a thermocouple attached to the board to put a bit of truth into the temperature readings. The solder "melting" means it just reached 138C, not necessarily the peak 165C specified. Try cooking until it wets the pad. \$\endgroup\$ – W5VO Aug 26 '16 at 4:17
19
\$\begingroup\$

My guess would be that the copper board is not being given enough time to heat up. Due to its thermal mass the copper heats up much more slowly than the solder, and the solder melts before the board reaches the correct temperature. If you choose a smaller piece of copper, or an etched PCB with less copper on it, or leave the copper board in the reflow oven longer, the solder will eventually flow as expected. It's probably just that the large thermal mass can't heat up enough before the solder melts.

\$\endgroup\$
  • \$\begingroup\$ Thank you for you answer, after some more tests, one of which consisted of baking the entire board at 240°C and hope the solder wets (cf question comments). It was actually successful, so I think you are right, and I will use an higher melting point solder paste \$\endgroup\$ – MaximGi Aug 26 '16 at 13:29
10
\$\begingroup\$

I've heard bad things about desktop ovens like this. They don't necessarily have the oomph to get the job done correctly. Saying "I turned knob X to temperature Y and waited Z minutes" does not mean that you have any idea what's happening to your board. The only reliable way of knowing would be to measure, maybe with a thermocouple in contact with the board (not perfect, but likely close enough).

You're obviously reaching an adequate temperature, because the solder is melting. It's certainly possible that your oven doesn't provide enough oomph to actually heat the board, and the solder is melting on top of cold componets. You may also be having problems with the flux. Either the flux in the paste is past it's prime, or the heating profile you're actually getting isn't giving the flux enough time to do its job, or the flux is activating too long before you bring your solder past the melting point, and a new oxidation layer is forming.

My advice is actually to forego no-lead solder unless there is some regulatory reason why you need to work with it. It's just harder to use- requires higher temperatures, which makes it tougher to come up with the right temp profile short of using real equipment. You may still have problems with lead, but probably less so.

Just as an aside, regardless of the nature of your oven, if it doesn't have heat-ramp-soak control with feedback, it isn't "meant for this purpose".

Update -- given the low temp nature of the Chipquik, the comments on no-lead solder don't apply. I think it might highlight the issue of premature and prolonged activation of flux though, if the oven is a very powerful one. No real way of telling whether its that or a cold board, though, without measuring. Temp crayons might shine some light on this.

Lead solder might actually help. Flux activation temps are better documented, so the soak profiles can be tweaked to slow things down before activation to avoid oxidation.

\$\endgroup\$
  • 1
    \$\begingroup\$ QuickChip is a low-temperature soldering paste, working, melting around 140°C, so I don't think that the general lead-free warning applies here. I do agree about the amount of heat in the board. \$\endgroup\$ – AndrejaKo Aug 25 '16 at 16:27
  • 1
    \$\begingroup\$ @AndrejaKo -- wow -- just looked at the profiles, and that certainly is low-temp! Is it a eutectic melt? \$\endgroup\$ – Scott Seidman Aug 25 '16 at 16:40
  • 2
    \$\begingroup\$ I'm not sure now, but I think it is. If I remember correctly (this was from few years ago), it was designed to allow very easy desoldering of large components, with the idea that you'd use it to basically replace the existing solder with this one, which takes forever to cool down, so you have lots of time to remove the component. Oh, yeah, and I missed the name in my original comment. \$\endgroup\$ – AndrejaKo Aug 25 '16 at 16:48
  • 1
    \$\begingroup\$ What the hell is a "oomph" ? Also you are right about the oven, I was told it was designed for soldering smd and I didn't question that. But then I found what a real soldering oven looks like, and I think my predecessors who bought this crap have been scamed \$\endgroup\$ – MaximGi Aug 26 '16 at 13:31
0
\$\begingroup\$

I suspect a problem with the copper board. What about the surface, is it bare copper only, or is it covered with tin? I would try a conventional soldering iron and lead rosin-core solder to do some test joints. If the solder does not flow well, there is something wrong with the board. The surface may be oxidized or the copper areas are to large to get heated. Wipping the board with pure alcohol does not remove copper oxide from the surface, very fine abrasive paper does.

\$\endgroup\$
  • 1
    \$\begingroup\$ The solder flows well using an iron with leaded solder. However, i'm not supposed to use leaded solder \$\endgroup\$ – MaximGi Aug 26 '16 at 13:23

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.