Hi,
Over the past few months I gave at a local ham club a series of lectures about working with SMD components. I started from quick introduction to SMD technology, which was followed by a lecture discussing techniques useful for amateurs working with surface mount parts at home, then by live demo of building a small board, and in the end, by discussion of rework of botched SMD job, and demo of working board, interleaved with short explanation how to measure frequency response of a circuit, and how frequency response analyzers work. The lectures have been recorded and uploaded to youtube.
Even though the class is very slightly RF-flavored, to make it more attractive for the audience in the room, the topic is generic enough to be useful for beginners wanting to start building circuits using SMD components. I based the class on my own experiences, so I thought that it might be useful to share the videos with members of the EEVblog forum.
Part 1: SMD basics
During this lecture I gave very quick, 15 minutes introduction to SMD components. I talked briefly about basics and advantages of SMD technology, types of components, gradual abandonment of imperial measure units by device manufacturers and transition to metric system, and confusion that may result from it.
Part 2: How I Learned to Stop Worrying and Love SMD Components
During this lecture I discussed workflow and techniques useful for working with SMD components at home, required tools, and how to assemble boards with surface mount components without a need for specialized and expensive tools.
Part 3: Live board building demo
During this lecture I will show how to use some of the techniques mentioned in second lecture to build a small power RF amplifier using MMIC.
I will explain how to prepare board for placing the components, then show how to place them and, finally, how to solder the board using hot air reflow process, and how to use board preheater. I will also explain why reflowing electrolytic capacitors or other sensitive components can be dangerous for them.
Part 4: Board rework, frequency response, and measurement of working circuit
During this lecture I talk about rework done to the MMIC amplifier board which I built during lecture 3, explain how to measure frequency response of a circuit, talk about inner workings of a frequency response analyzer, and in the end I demonstrate that the board really works and amplifies the input signal.
I published on my blog slide decks I used during the lectures. I added some information that I did not have time to squeeze during the class, screenshots with graphs from frequency response analyzer which were shown live during the last lecture, and (by request from the audience) a section about how to read SMD markings. Enjoy!
http://rroeng.blogspot.com/2015/08/smd-soldering-at-home-slides.html
No time to look at the videos now, but I bookmarked the page. We will see if you can move me from through hole. Thank you for putting this together.
a mixed pot of components ...
... feel free to re-use
i started this chart to give myself an idea of what the solder pad sizes are like (esp diodes). unlike resistors you can "read" the size straight.
it is very useful especially for the hundreds of types which share a common landing pad, and in many cases, different names of the same foot print. esp when shopping using size as selector ...
Thank you for putting this together. I dabble on SMD here and there, but just like GreyWoolfe, perhaps it will convince me to move to SMD once and for all.
No time to look at the videos now, but I bookmarked the page. We will see if you can move me from through hole. Thank you for putting this together.
Thank you for putting this together. I dabble on SMD here and there, but just like GreyWoolfe, perhaps it will convince me to move to SMD once and for all.
It's a common misconception to think about SM components as miniature TH components, and many people are traumatized by the though of soldering all those tiny leads one by one. They are not. SMT is a different technology, and requires different tools and methodology. To see how easy it is to build a board with SM components, check the third video. It's a recording of a class in which I built a board live, while explaining in details every step to 30+ people.
What I'd recommend to someone who just starts to work with surface mount parts, is to get a good, temperature controlled soldering iron, with good tip (don't fall tor the very thin tips. They are useless) and a hot air station. You will quickly see that soldering boards with SMD is actually easier than soldering TH parts. What takes time is placement of the components; Soldering time does not depend much on the number of parts or pins. For some parts it is much faster to solder SM component, than its TH counterpart. If you don't believe me, see how quickly it is to solder fine-pitch (0.65mm) IC with 28 tiny pins:
People were amazed when I showed the video during presentation
What takes time is placement of the components;
This. To me it is similar to the task of painting a room: the preparation takes a lot longer than the job itself.
In addition to the tools you mentioned (which I was fortunate enough to be able to purchase over the years) is a good optical solution - either good magnifying glasses or a stereo loupe/microscope. That is one of my last requirements for having a better SMD experience but, due to space constraints, it is postponed for the foreseeable future...
What takes time is placement of the components;
This. To me it is similar to the task of painting a room: the preparation takes a lot longer than the job itself.
When you work with TH components you also have to insert them in the holes, so the workload is similar in both cases. What you don't have to deal with when working with SMD is bending and trimming the leads. Clear win for SMD.
BTW. I consider placement of components as a part of the soldering process.
In addition to the tools you mentioned (which I was fortunate enough to be able to purchase over the years) is a good optical solution - either good magnifying glasses or a stereo loupe/microscope. That is one of my last requirements for having a better SMD experience but, due to space constraints, it is postponed for the foreseeable future...
I also thought in the beginning that I will need at least a good loupe, but turned out to be not necessary. Down to 0604 I can handle the parts just fine having good light in the workspace. I think 0402 should also be manageable, but I haven't used them much. Anything what does not provide stereo vision is useless. I tried to work with USB microscope, but quickly tossed it away.
Home is the key word that got my attention.
I've used SMD for ~5 years, BUT with a mixture of TH thrown in for 2 main reasons.
1. Routing. I do homemade SS PCBs boards only at this time. TH resistors in particular, provide essential pathways for routing traces when no other possible paths exist. You can use links sure, but if 2 holes need be drilled why not use a TH resistor when I have thousands. It also removes a few components from the SMD side and helps get density higher and PCBs smaller.
2. TH parts on hand. Big power devices, power resistors etc.
0605, SOIC, even SOT23-6 is easy and even with my ageing eyes and a cheap magnifing headset it is not difficult at all. Any project that I have to make a PCB for I will ultilise SMD as much as possible.
But it does take some practice, there was a good thread about SMD practice recently, i'll see if I can find it to link here.
For those starting in SMD, I'd recommend you don't try to push the boundaries of what is achieveable, stick with those sizes I've suggested and save yourself some grief untill some experience is gained.
Thanks for the videos. I will watch them when I'm home.
Thank you for this - I am just getting started in SMD and it was really helpful.
I have had some shocking boards so far, but am slowly getting to grips with it. My worst results have been using solder paste - I am going to stick to fine solder wire from now on.
The other thing I found really useful as a noob was to buy some flux pens - they make putting flux in the right place so much easier than with a brush or such.
The other thing I found really useful as a noob was to buy some flux pens - they make putting flux in the right place so much easier than with a brush or such.
Interesting how we all use differing techniques, I've never had a need for additional flux or braid.
But boy am I anal about a clean tip and PCB.
And you can stick Pb free solder where the sun don't shine. It's muck.
$1 shipped for a board + components. That's an amazing deal.
How is this even possible, is the postage paid by the Chinese government or are they working for free?
Postage is heavily subsidized by the government.
Home is the key word that got my attention.
I've used SMD for ~5 years, BUT with a mixture of TH thrown in for 2 main reasons.
1. Routing. I do homemade SS PCBs boards only at this time. TH resistors in particular, provide essential pathways for routing traces when no other possible paths exist. You can use links sure, but if 2 holes need be drilled why not use a TH resistor when I have thousands. It also removes a few components from the SMD side and helps get density higher and PCBs smaller.
2. TH parts on hand. Big power devices, power resistors etc.
0605, SOIC, even SOT23-6 is easy and even with my ageing eyes and a cheap magnifing headset it is not difficult at all. Any project that I have to make a PCB for I will ultilise SMD as much as possible.
But it does take some practice, there was a good thread about SMD practice recently, i'll see if I can find it to link here.
For those starting in SMD, I'd recommend you don't try to push the boundaries of what is achieveable, stick with those sizes I've suggested and save yourself some grief untill some experience is gained.
That's a very safe, conservative approach, but I found the opposite to work best for me. For the first project I made learning to work manually with SM parts I used only SM components.
This happened also to be one of my fist boards I made learning to use toner transfer method, so it was a double challenge. As much as I learned with this single project challenging myself and pushing the boundaries, I wouldn't have learned doing 10 projects going the "easy path".
The second picture shows two earlier iterations of the project. You can see how much my technique improved just by working on these boards.
One thing that a friend pointed out to me, which I've found valuable, is to consider avoiding SMD for large components (large inductors, large capacitors etc). So you might have a board that is 100% SMD, except for a few large electrolytics and inductors. The through hole connection is much more sturdy, and soldering SMD inductors which have really thick leads and huge thermal mass is an exercise in frustration unless you have a full on reflow oven. I've melted the plastic on a surface mount electrolytic cap before the paste started melting, because the 2oz board I was working with was wicking away heat faster than I could add it -- and the cap contacts were hidden under the body of the component, practically unreachable!
So yeah, stick to through hole for the really big, chunky components.
One thing that a friend pointed out to me, which I've found valuable, is to consider avoiding SMD for large components (large inductors, large capacitors etc). So you might have a board that is 100% SMD, except for a few large electrolytics and inductors. The through hole connection is much more sturdy, and soldering SMD inductors which have really thick leads and huge thermal mass is an exercise in frustration unless you have a full on reflow oven. I've melted the plastic on a surface mount electrolytic cap before the paste started melting, because the 2oz board I was working with was wicking away heat faster than I could add it -- and the cap contacts were hidden under the body of the component, practically unreachable!
So yeah, stick to through hole for the really big, chunky components.
Use common sense, but don't try to constrain yourself just to avoid challenge, because that's when you really learn new things.
So yeah, stick to through hole for the really big, chunky components.
Use common sense, but don't try to constrain yourself just to avoid challenge, because that's when you really learn new things.
Of course -- I only mention it because sometimes when you're designing a circuit, you can just reflexively set the search parameters to "surface mount only" -- and sometimes, design tools like Webench might give you parts lists with huge SMD components. It's time like these when you want to consider
explicitly making a point of finding corresponding through hole parts, especially in a hobbyist setting. Unless you want a challenge! But there's a fine line between "challenge" and "poor engineering"
So yeah, stick to through hole for the really big, chunky components.
Use common sense, but don't try to constrain yourself just to avoid challenge, because that's when you really learn new things.
Of course -- I only mention it because sometimes when you're designing a circuit, you can just reflexively set the search parameters to "surface mount only" -- and sometimes, design tools like Webench might give you parts lists with huge SMD components. It's time like these when you want to consider explicitly making a point of finding corresponding through hole parts, especially in a hobbyist setting. Unless you want a challenge! But there's a fine line between "challenge" and "poor engineering"
Failing is also an excellent way to learn
I learned how to solder big smd electrolytic capacitors by frying several of them
But you made a very good point, of course.
Nice!! i was look something for this, thank you.