I used to do this for a living. I'd strongly recommend a hot air station, decent tweezers and liberal use of flux. TBH, I never found solder paste to have any advantage over solid.
I tried a bunch of different vacuum pens when I worked in the industry and didn't like any of them.
Another tip, have a good think about order of placement so you don't impede access with a previously placed part.
Also, don't rely on stensils, you often won't have them, or won't be able to use them because the board is already populated.
Another tip, have a good think about order of placement so you don't impede access with a previously placed part.
Also, don't rely on stensils, you often won't have them, or won't be able to use them because the board is already populated.
Do you lay down solder (solid) with an iron, then place the parts, and then use a hot air station?
And what temp would you set a hot air station for solid, the same or different, I went with 300 deg F for the hot air station w low air speed/pressure. It was slow maybe, but careful.
I have several tweezers from a gifted Ifixit kit, they're labeled ESD Safe, whatever that's worth, they are well aligned, very pointy (finger stab capable), and have a good springy feel, so I'm happy enough with those.
Yes, prepare the pads, place part hold in place with tweezers if necessary, then apply heat. as for temperature, it really depends what you're working on, but paste and wire generally have similar melting points and characteristics. You need to have the temperature a bit higher than the melting point of the solder you're using.
I'm not saying that you will get a better result from solid over paste, rather that there's no advantage with paste (unless you're using a stencil). But then if you're using a stencil you're not really gaining anything unless you're reflowing in an oven.
In fairness, I was diagnosing manufacturing faults and replacing a single, or at most a few components, not populating entire boards.
Not an issue if you use stencils and an oven, or hot-air come to that. SMT and soldering iron is a compromise - general advice is more flux, less solder (to avoid solder bridges between IC pins).
I find I always get way too much solder onto the board when I use solid solder. That's in part because I use 0.5 mm solder. 0.2 mm diameter would be better.
I have as well and I wasn't super impressed either. But my current solution is to pinch the part between the points on the tweezers and then force the points apart to drop the part. All without changing the location of the part. It's an issue on 5x5 mm QFN and larger which are larger than the distance between the points of my tweezers.
With QFP packages I usually just grab two pins and move the chip over. That seems to work.
Tom
Good job. I started doing SMT about 18 months ago. Back then I could hardly do a 1206. Now I routinely do 603 on all my board. I can do 402 but it’s not worth the effort to me as it easily takes twice as long as a 603.
I actually have just been using a syringe of 183C leaded paste with a needle outlet for most of my work. I’ve gotten pretty good at dabbing a good amount on the pads, then I just align the component and use the hot air gun. I even do QFP this way, although sometimes I’ll drag solder it. I often end up with a bridge or two but I just clean it up with the soldering iron and some flux. I bought a toaster oven at one point to do boards, but I find the cost of the stencil prohibitive for one off boards so I just raw dog em using my syringe of solder paste.
I actually have just been using a syringe of 183C leaded paste with a needle outlet for most of my work. I’ve gotten pretty good at dabbing a good amount on the pads, then I just align the component and use the hot air gun. I even do QFP this way, although sometimes I’ll drag solder it. I often end up with a bridge or two but I just clean it up with the soldering iron and some flux. I bought a toaster oven at one point to do boards, but I find the cost of the stencil prohibitive for one off boards so I just raw dog em using my syringe of solder paste.
Tom, try using a way bigger tip than you would normally and use plenty of flux. It works surprisingly well. When soldering an SOIC for example, my tip would span at least three pins. Surface tension would leave just the right amount of solder on the joint. We soldered at 600°F, with 63/37 solder, that may have impacted results.
If you see a bridged pin, flux the spot, clean your iron and swipe straight out. Half the solder will stay on the tip. I use 0.5mm solder only.
My results with fine tips and solder were always worse than using the method above.
@cowanaudio, what do you use for 0603 and such? That's where I typically end up with too much solder. I think the tip on my iron is 1.6 mm wide could be 1.0.
Tom
Tom
Hakko makes a micro soldering handle and tips that can work with some of its solder iron controllers. Using very fine gauge solder and the smallest chisel tip, it can solder fine pitch SMD one pin at a time perfectly. That very tiny tip is kind of delicate though. The next size chisel tip up from that is pretty rugged and can be used for most small work. Its still much smaller than any of the tips that work with the standard size soldering handles.
Think about the chisel tips versus the knife edge and other tips is that IME at least heat transfer is much better with chisel tips
Think about the chisel tips versus the knife edge and other tips is that IME at least heat transfer is much better with chisel tips
I use a 1.6mm 30 degree chisel tip for most of my small soldering, including 0603. If the second joint has too much solder, just one small drop of flux and a quick pass with the clean iron tip and it should be perfect. You loaded the first pad without the component there so it's easier getting just the right amount of solder on that one. I don't use conical tips for any soldering tasks.
The Pace Iron I use also has a range of tips called MiniWave, which have a small concave area on the flat of the tip. They work really well too and allow you to remove a little more solder with a swipe. They are not as universal as the 1.6mm chisel, which usually stays on the iron unless I am doing a bunch of QFP packages or similar. When soldering a QFP, I fill the MiniWave tip to a little convex and run the tip allong the fluxed pins. After a bit of practice it's easy to get very good consistency. I did a Pace training course that taught me this technique many years back, but Youtube will show you exactly the same today.
Drag soldering works with a chisel tip, too. You just need to load the tip a little more often.
Another really usefull tool are hot tweezers. I can pluck most surface mount packages straight off the board in one movement, with the correct tips. Without the hot tweezers, a piece of .001" or .002" stainless steel shim works really well. Heat the pin, slip the shim under, repeat. In a minute the shim is under the whole component. A big tip speeds this up a lot, too. Hot air allows you to do this under a BGA, too.
The Pace Iron I use also has a range of tips called MiniWave, which have a small concave area on the flat of the tip. They work really well too and allow you to remove a little more solder with a swipe. They are not as universal as the 1.6mm chisel, which usually stays on the iron unless I am doing a bunch of QFP packages or similar. When soldering a QFP, I fill the MiniWave tip to a little convex and run the tip allong the fluxed pins. After a bit of practice it's easy to get very good consistency. I did a Pace training course that taught me this technique many years back, but Youtube will show you exactly the same today.
Drag soldering works with a chisel tip, too. You just need to load the tip a little more often.
Another really usefull tool are hot tweezers. I can pluck most surface mount packages straight off the board in one movement, with the correct tips. Without the hot tweezers, a piece of .001" or .002" stainless steel shim works really well. Heat the pin, slip the shim under, repeat. In a minute the shim is under the whole component. A big tip speeds this up a lot, too. Hot air allows you to do this under a BGA, too.
Last edited:
Yep! I wouldn't attempt SMD rework without them. Although, if you have the room, you can often just push the component off by heating up both ends with a wider chisel tip.
Tom
Hot tweezers, would have to practice a lot with those! I'm going to buy a few more of those practice kits, try to get more confident / competent before doing something that matters. I tried the sweep method on a QFP (didn't know that's what they are called until Tomchr used the term) and I made a wreck of it.
No. But I do still build prototypes and test circuits by hand. Once I have the circuit dialled in, I'll commit to a production run. So the rework is on the prototypes and test circuits.
Since I don't have X-ray vision, I have to rely on inspecting it from the outside. All QFNs I've used have had non-pullback leads, i.e., you can see part of the lead on the side of the part. That makes it easier to line the part up correctly before reflow. It also makes it easier to check for solder bridges and such. At least you know where the pads are...
Of course, none of that can check if there's a solder bridge or - worse - a void in one of the solder joints. So I test the board thoroughly after soldering.
I've had relatively few issues with the QFNs, actually (knock on wood). I think I've only had one case where I had to remove the part, tin the pads manually, and reflow with a hot air tool. I think it's mostly a function of me getting better at spreading the solder paste. If you get a clean 'print' of paste the parts will solder without issues. But if the paste starts to smear you can end up with solder bridges.
The pros have this down to a science. It's all about getting the stencil thickness right, getting the correct apertures defined in the stencil, printing or otherwise depositing the solder paste cleanly, and having the right reflow temperature profile. The pros have X-ray vision - or at least the outfit I use does. You can pay them to inspect each solder joint with X-ray, but I don't see a need unless you're planning to send your circuits into space.
Tom