So, You’re Scared Of Surface Mount | Hackaday

If you are lucky enough to come across some home-brewed electronic products from the 1950s, it is very likely that these components are assembled on solder labels under the lid, each with long leads, and installed on the rack for The socket of the tube. Easy to assemble, using the most agricultural soldering iron.

Open home buildings in the 1960s or early 1970s, and you may find the same passive components and germanium transistors installed through holes in the bizarre, larger interposer boards, or even homemade PCBs with thick and wide tracks.

By the late 1970s and early 1980s, you will find a more familiar sight. The dual-in-line IC through holes on the ribbon board with a pitch of 0.1 inches, and self-made PCBs began to appear on the glass fiber board. Easy to use and easy to solder. Familiar with safety. Almost forty years later, what you will see on the breadboard is still the same, but it can still be seen from many kit manufacturers.

But we all know that progress in the field of electronic components has not stalled. The history of surface mount components can be traced back to the 1960s and has appeared in consumer devices since the late 1980s. More components per square inch, smaller, cheaper equipment. Today, they are everywhere, and more and more through-hole versions no longer offer these new components. If your experiment is limited to 741 and 555, this is not a problem, but if your taste extends to new types of sensors for microcontrollers or RF work, then this will constrain your style.

This development triggered a series of reactions. Many people happily embraced the new media,

The result is full of really smart SMD projects. But many people still cannot upgrade to SMD, maybe because of smaller SMD components, special tools they may need, or even new skills they must learn. When you sell a kit with SMD components, you will receive responses from people who like the kit but want to get it through the through hole, so this article applies to them. In order to expose the mystery of using SMD, and prove that the work of SMD should be controlled by almost anyone who can use a soldering iron.

This may be the first reaction of a lifetime through-hole soldering machine. SMD parts are indeed usually very small. Even those parts filled with large packages, the leads may look as many and thin as cat hair. When the endless bunny headlights panic, they look like It's a cat.

But it is important to step back and understand that not all SMDs are the same. Some of them are small, and only those with god-like power can be hand-soldered, but many devices are packaged in SMD, which is large enough for mortals to use.

So when you see a circuit board covered with components the size of dust particles, don't worry. Few people will try this level of construction, but of course your scribe cannot. (We are waiting for commentators to claim that the thousand-pin BGA and 01005 chip components were manually soldered as expected. However, such a statement is useless without evidence.)

Instead, focus on your SMD package 

deal with. The SMD chip component package is represented by a number related to its size. Confusingly, the scheme has two metric and imperial systems, but the format is the same: length followed by width.

Considering the picture above as well as the PCB and tape measure, it is the bottom surface of the Raspberry Pi B+ model and will be assembled by the robot to automatically place the machine. Most of the components are indeed small, but you will notice that L3 is the black component in the lower left corner, which looks large compared to its neighbors. The package is "1008", 0.1 inches long and 0.08 inches wide. It is still small, but imagine picking it up with a magnifying glass on tweezers. Not bad, you may have dealt with many things in this size range before, do SMD parts look so scary now? The larger components-0805, 1008, and 1206-are surprisingly within the grasp of the average manufacturer.

In a business environment, SMD equipment will be assembled by machines. Glue or solder paste will be printed on the relevant part of the circuit board, and the automatic pick-and-place machine will remove the components from the tape package and automatically place them in the correct direction. Then, the entire circuit board can be soldered at once in a reflow oven or wave soldering machine.

If you are new to SMD, you are unlikely to just put these kits on the bench. Have

with

, But it is safe to say that they are still quite advanced things.

You will also see all commercial kits for desktop SMD builders. Conventional soldering iron hot-air soldering stations or SMD drill bits are very useful, but the price is very high.

The good news is that you don't need any of these special tools to dip your toes in the SMD water. It is almost certain that you already have everything you need, if not, all you are missing is SMD work. If you have the following items, then you are fine:

. Even larger SMDs are still small. Sufficient light ensures that you can see them clearly. A good downward-pointing table lamp is sufficient. 

. Because SMD is difficult to see clearly, it will help if you operate SMD on a bright white surface. Fresh white printer paper on the table can be used as a suitable work area. 

. Unless you are fortunate enough to have amazing eyesight, you will need a decent magnifying glass to use with surface mount components. The "helping hand" type on the stand is appropriate. 

. When soldering, you need to use it to hold down the surface mount component. 

. You will need these to pick up, manipulate and flip surface mount equipment. 

. If you have a standard fine tip soldering iron suitable for use with traditional 0.1 inch pitch through-hole components, then you should be fully equipped.

In other words, there is a special tool that may be worth your consideration. Sometimes it seems that three hands are required to hold the SMD device while soldering, so one or two tools can be found to help. Fortunately, this is something you can build yourself. take a look 

 For example, or your scribe's

.

It is not difficult to imagine that when you look at SMD integrated circuits, the pins are too small and too close, and you may not be able to solder them by hand. The answer is, of course, you just need to see how to weld in other ways.

With through-hole ICs, you can solder each 0.1 inch pitch pin separately. If you manage to build a solder bridge between the two pins and compete to use desoldering pumps or braids, it will be a disaster.

In contrast, when using surface mount ICs, as a mortal, it is almost impossible for you to solder each pin individually, so you don't even have to try. Instead, use an excessive amount of solder to solder the entire row at a time, and then use a braid of solder wire to remove the resulting huge solder bridges, thus making clean and professional work possible. Surface tension and a large amount of flux are your friends, and if you are an experienced through-hole welder, then you hardly need to have any welding skills.

If you can fix it on the board and use a magnifying glass to see it clearly when necessary, then the component can be anything and you can solder it. Try it, you will be surprised!

Therefore, we hope you have convinced SMD doubters to enable you to use SMD yourself. What's next?

Read the topic first. Your favorite search engine should provide it, tutorials can be found from the following locations 

From 

, Or from

 and many more. And look for videos owned by YouTube

But there is no alternative. Find a clipboard with surface mount components of the appropriate size and rework it. Desoldering its components may be difficult, but you should be able to rework the solder joints easily. Coat the pins of the integrated circuit with flux and try to run a mass of molten solder along them, then use a defluxing braid to remove the excess solder. The advantage of the clipboard is that it will not damage it, so you can practice these techniques according to your inner habits until you are satisfied with the newly discovered skills.

Therefore, you can move on and proceed with your first SMD project. well done! What you want to do next is up to you. Design your own circuit and make PCB, buy kit or find

Downloadable PCB files and order by yourself.

No matter what you do, be happy that you have overcome the fear of SMD and are determined to be the first in the queue to try any new technology in the future!

Enough to say: p

When I see that QFN is done with iron instead of hot air, I am always impressed.

Use the "solder mask" method and ordinary soldering iron and solder wire to solder QFN:

Time back in time to make a batch of SMD boards with ordinary soldering iron and solder:

Please enjoy. :)

Not all QFNs have leads that extend to the edge of the package and can be contacted with a soldering iron.

As long as the pad extends to a reachable position and is not covered with solder resist, there is no need to touch the lead.

Fortunately, cold welding should not be performed in this case.

Yes, I also have questions. However, with the right amount of solder paste and flux, you can still try it. In most cases, it will work. Flux, wettability and surface tension are all wonderful things.

Oh, lead (Pb) is also a good thing. Unless you are legally obliged to do so, do not use lead-free solder indiscriminately. Otherwise, you should get any problems, especially when using non-optimal welding techniques.

@ [Chris C.]

I have seen RoHS (lead-free) car battery terminal connectors. You screw them onto the lead battery terminal.

This is not a "just try" thing. It is impossible to work reliably with solder wire. Think about it. If the pad is heated and the solder flows on it, if the solder is below the chip, there will be no contact between the solder contacts and the pad on the chip. Therefore, there is no heat flow. So the joints are very cold.

If there is a *contact between the chip and the pad, the solder will not be able to flow under the chip because the edge of the chip is not connected to the contact point.

*Most* QFN contacts extend to the edge of the chip, which is the JEDEC standard. But not everyone does it. In fact, solder paste and hot air flux are not needed.

Nice video!

You mentioned diluting the flux with alcohol. Which flux and type of alcohol did you use?

You can even desolder the QFN with ordinary iron.

"We are waiting for commenters to claim that a thousand-pin BGA needs to be soldered by hand"

What does "hand welding" mean? Anyone can easily complete a high pin count BGA using StencilQuik and a reflow oven. Yes, the unit cost of StencilQuik has increased by about $10-20, but this is only a small modification of the cost of a thousand-pin BGA.

But the 01005 component is up to you.

At the HaD conference held in San Francisco last fall, a German (sorry, don't remember his name) gave a quick introduction to assembling BGA parts at home. The trick is

. Instead, just apply a coating on the area where the BGA should be fluxed and place it as carefully as possible, then cook the board in your favorite reflow oven/hot plate.

The key is that the right amount of solder has been provided in the form of solder balls on the part. You just need to let it flow onto the mat. The flux and surface tension do the rest.

I haven't tried it, but he made it sound simple.

I have redesigned some DDR (0.8mm) in this way, but if you are making a surface mount board, you should use a template anyway. The more flux you control, the better the reproducibility, and the use of templates to paste paste can make the operation very precise.

Yes, StencilQuick can achieve perfect alignment based on feeling alone, so you don't have to worry about it. This is just a precision cut polyimide adhesive.

Hmm...humorous alert.

It is worth mentioning that WRT manually placed the BGA for reflow soldering. According to what I observed other people did, they found it difficult to align the BGA correctly, so the failure rate was high.

Correct! Because you did not notice the important part in my post.

If you don't want to click the link to view, they are polyimide adhesives with BGA patterns. Yes, you can add solder paste to it, which will definitely help, but the "most important thing" it allows you to do is "feel" the alignment. You align the StencilQuik adhesive to the circuit board, which is very easy to approach perfection-if you tighten it, you can lift it up and try again (if you do tighten, then $10 is usually cheaper than a BGA part).

Then put solder paste on it, and then put the BGA part, and then you can *feel* it falling into the hole of the solder ball.

If StencilQuik’s $10 per cost will disappoint you, then you will find a DIY example

To be honest, another thing I have been curious about is whether you can mill a small metal frame the same size as the BGA and place mounting holes on the board. Then you connect the frame, put the BGA in it, reflow it, and then remove the frame. You are trying to align an object to a part of 1 mm. It shouldn't be that difficult.

You can also feel it on an appropriate board with solder mask. Not so good, but still obvious.

Yes, StencilQuick basically played the role of thickening the solder mask: ordinary LPI masks were about 2 million, while StencilQuick increased by about 4-8 million, totaling about 6-10 million. The height of a BGA ball is usually ~16 mils, so you can see that a Stencil Quick of ~6 mil means it is half the height of the ball, which makes it *really* easy to notice the difference and you really can't tighten it.

Why should you pay attention to? This is the purpose of flux and surface tension, the chips move to the proper position by themselves

The global position is very good, because the surface tension cannot resolve every alignment error ;)

"You are a spammer, hacker or other bad person."

—

I don't like the huge solder speckle/solder braid removal technique. This is very messy and there is a risk of overheating parts or damaging small pins.

The most important technique for SMT welding is FLUX FLUX FLUX FLUX FLUX. There are liquid flux bottles, gel flux tubes, flux pens, a large amount of isopropyl to dilute it as needed, dispenser syringes and brushes, and bottles with dispenser needles.

The second most important technique for SMT soldering is the soldering iron tip. Clean, bright, well-plated and not too small... A fairly large hoof or beveled tip will be very useful.

The third most important trick is to search for "drag welding" on YouTube and hide your welding braid until you learn this technique.

Yes, I have just started surface mounting, and it is all about flux. I use a soldering pen and regular scrap iron, and it works normally. What surprised me was that I was able to do this because I almost sucked in conventional welding.

This one! Even welding very fine pitch parts with good flux quality and iron quality is very easy. I have been using drag soldering! Whenever someone mentions a solder core or breaks it for something, I cringe, because 90% of the time will eventually cause you to lift the traces and pads off the PCB and possibly from you Tear off the pins on the parts!

I cannot emphasize enough the difference between a good set of tweezers! Break it down and spend some money on a good set of clothes, you will not regret it.

In my past work, I often welded 0201 parts regularly, and the tweezers worked. You need a microscope to do most of the work, but it is feasible! Using a desktop magnifier can easily reduce anything to about 0402, but anything smaller does require a microscope to examine joints.

Absolutely, the solder sometimes cools at the edge of the wick and bonds to the pins. When you pick up the wick.... Damn.

That $25 bargaining chip became an expensive rubbish! It’s time to disconnect the magnet wire, assuming there is enough metal to try to fix it... (don’t ask)

Sorry, I don't understand this problem very well-I never happened. One person dragged away the wick and the soldering iron together, and the wick was "melting" all the time...

That's what you should do, but one time you don't do it is when you accidentally tear off the pads on the PCB. This is a good way to solve the error, but I will not use the solder core as a normal and expected part of the soldering process, and you will use this material on each pad.

I try to avoid individual parts smaller than 0402 because they become more difficult to handle. I like resistor arrays because they have the best of both worlds-larger than their single array and save more space. For example, 0302 has two 0201 resistors, but to deal with it is almost like 0402, and you have to place less.

The problem is that in addition to saving space, there are more reasons to make smaller parts.

In many cases, 0402 caps are better than 0603 caps. Its parasitic inductance is about 30% smaller because it is smaller. That is to say, under the equivalent capacitance, the upper limit of 0402 will be higher than the upper limit of 0603 * maintain* 40% of the upper limit.

Unfortunately, you cannot use the "capacitor array" technique here, because the "distance" from the object on which you are placing the capacitor is the cause of parasitic inductance. Therefore, the cap must be as close as possible. This means finer = better.

I really want to avoid using the 0201 upper limit as much as possible, but this extra decoupling can really help when you try to run at 500+ MHz.

When using physically smaller capacitors, be sure to pay attention to the DC bias. When the capacitance value of the 0402 size exceeds 100nF, only a few volts can start to lose a lot of capacitance. In most cases, if I want to be close to 1uF at 5V, I need to use 0805 capacitors.

This is why I don't talk about the cap array.

In a few cases, you happen to get 2 closed pins to take advantage of the array. Anytime you need to break through a few mil jumpers to connect to the capacitor cap, you may forget the capacitor cap. It is best to drop the through hole directly to the power/ground layer and connect the capacitor cap directly to the power/ground Floor. Don't forget that the power/ground plane acts as a very low inductance high frequency capacitor.

The capacitor array can still help decoupling-flip the power and ground connections on one of the capacitors to reverse the current and eliminate some vias and parasitic inductance. A specially designed capacitor array like this (for high frequency decoupling).

Flux is the most important part. And appropriate lead-containing solder. It makes sense, why the TV 40 years ago can still work, but the models from the 2000s are outdated...

The smallest component I soldered is 0402 without a magnifying glass or any other optical aids. I'm short-sighted, so it's easier for me, but to check if the components are soldered correctly, I took photos with Dimage Z5's macro mode. When I soldered TQFP-100 with cheap Chinese flux, I messed up. There are more bridges than Venice. Good flux and concave solder tips solve this problem.

Can people recommend some tweezers? Links to global websites (such as Aliexpress or RS/Mouser/Farnell, etc.) would be great!

I found some different suggestions in a Google search, but no consensus was reached.

It is difficult to determine "good" only by checking online, and there are many choices for the shape of the nib!

I like those who are bent, because they don't block my sight. I like the Prideline curved tweezers near the bottom right corner.

I bought a pair of spare backups from China. Haven't used it yet.

I use Adafruit's ESD bending elbow. Those sparks are too hard.

The tweezers made by Xcelite are very good. I prefer XSST7V for smaller things and XHT-678 for larger things. Amazon's shelf prices are reasonable, or deep-fried prices are high.

Hemostatic drugs may sometimes be helpful, but a good brand suitable for these people has not yet been found.

I bought some tweezers from AliExpress/ebay. The first one is actually pretty good, I still use it. Then for some reason, I wasted 6 more, they are so bad, I should throw them away. I mean instead to buy well-known brands, or buy from well-known distributors.

I discovered that after about 20 years, my solder braid no longer sucks. B ^)

Maybe it's already oxidized...

I think the SMT flux I bought 15 years ago may have crystallized in the container.

Put some flux?

In most cases, you will not encounter a huge problem, I mean, if you are new to SMD, it is not the end of the world.

Yes, flux. And more flux. Therefore, a lot of photos:)

I use Swamp Sander non-acid flux paste BTW, the type used for plumbing. The effect is as good as the expensive SMD flux, which can be easily removed with methylated alcohol.

I thought that non-acidic pipeline flux is feasible, but it seems that most pipeline fluxes are acidic. Flux specifically designed for electronic products is so cheap, I won't risk telling beginners to get close to anything with a "pipe" on it.

Of course, solder mask is the only way to go. But especially on the last pin, some excess solder is easy to gather. In this case, some wicks may help quickly eliminate and control the situation. Yes: using good enough flux is the "start and end" process.

In the early days, I managed to solder to a board so that people could use ordinary iron to manually use surface mount components on the breadboard of the DS2423 device. This part survived. quality education

As it happens, my project aims to provide a cheap entry point for welding equipment. Use Weller's high-quality prompts, but reduce the cost of the workstation.

Or you can buy it without a weak tip

Which flux do you recommend? Is the "unclean" version useful? What should be considered when choosing a flux?

The flux you are looking for is to have flux. Yes, really. A lot of flux.

Okay, there are some preferences (I use Kester RA flux most often in my pen), but it is important to use it.

Importantly, the flux may be corrosive and may conduct electricity, so it should be cleaned. If your boards are present/visible, you will not want to use no-cleaners because they will become hard and difficult to remove (they do not need to be cleaned, hence the name "no-clean"). If it is used on a finished product without a PCB cleaning step, it does not need to be cleaned. Prototype is up to you, but I usually clean the circuit board and use RA flux.

The liquid is not clean! It has annoying and harmful fumes and it is difficult to remove residues. I like water-soluble ones because with a little hot water, an old toothbrush and some ipa (if needed) can be easily washed off. The effect of Kester 2235 is very good, you can buy a gallon of gasoline, the price is reasonable, and the service life is long. RA is also a good choice, very flexible and can be used for almost anything, although it usually leaves a sticky residue.

It seems like a good beer was wasted.

Lol. I see what you are doing there! Moving! If you like to use "IPA" when cleaning, it is not a waste, although you should make sure it is the correct "IPA", otherwise you may be blind!

IPA (as isopropanol) will not blind you, it will be methanol. AFAIK IPA has approximately the same toxicity level as ethanol (something most of us sometimes drink in alcoholic beverages). Of course this does not mean that you should drink it, it is said that it is nothing fun compared to "normal alcohol".

Kester #2331-ZX Flux Pen. Great, especially lead-free solder. Also, for DIY, please do yourself a favor and use regular 60/40 solder.

A lot of light and glasses or head-mounted magnifiers can provide good stereo vision. The magnifying glass is PITA.

Hot the iron. Flux pen pad. Use curved tweezers to press down the part, and then hit a pin or one side of the iron nail. Or tin it first, then hit it with a flux pen, and then the pins will quickly fall into the molten solder. This is a good way to avoid too much solder (this is very effective for me). Unless the beginner on the Chinese rework line is cleaned up, I will never need a solder core.

My product needs to move five 0408 resistors for a customer who purchases 20 to 50 customers at the same time. It takes about 45 seconds in total under the stereo microscope.

With this in mind, we continue to use brass sponges to clean the tips and never "tinned" them. The brass sponge in the heavy pot is another "essential condition" for quick cleaning. Similarly, tapping the sponge and then touching the pins with too much solder will usually pull enough force without any wicks. First touch it with a flux pen to get a beautiful joint.

The same brass sponge can also pull sticky things, fibers and other paste-like substances from the tweezers very quickly.

Currently using Kester#985M "low solid content, no cleaning". good results. I found no toxic fumes. It leaves very little residue, so that I sometimes only clean it once after repairing it, even if it is not really necessary, but I am too persistent. It can be cleaned easily with IPA (isopropyl alcohol). Use with 600°F soldering iron and standard lead multi-core solder. Sometimes solder paste is used, which can be stored in the refrigerator, but has expired and has been reformulated with 91% IPA.

I admit that I still feel a little scared of SMD. My eyes are not good, my hands are not stable, I have to work slower, but in the end I still use most of the 0402/0603 parts in the carpet. Therefore, I still use many through-hole parts. But when SMD is necessary or clearly superior, I will use it without hesitation. The first time I tried to make a reliable solder connection, there was no problem, and no components were damaged by overheating (except for a very small button).

The flux in the picture is sold in the UK for plumbing. It is a non-acidic pulp, and I apply it generously with cotton swabs. After soldering, my circuit board is covered with flux, so it is almost no-clean, but it can be removed with a toothbrush for a few seconds and some methylated alcohol. The small filter PCB shown in the figure does just that.

IMHO, what matters is not the flux used, but the flux used. Use a lot.

some advices.

I have a good magnifying glass with a built-in light source, a separate light source and a separate magnifying glass with different focal points (these are three items). This helps to remove shadows on large blocks, allowing you to use one or another magnifying glass depending on the size or fatigue level, and even double the lens and lighting. Most people should already have a good light source and a second-hand magnifying glass, so getting a magnifying glass/light source combination should not make a big leap.

I use an X-acto knife with No. 11 blade instead of a flat screwdriver (note this). With a little practice, you can skillfully apply solder paste, flux (I use a liquid that I do by myself), up parts, cut traces, and press down parts when soldering (using the back of the blade). If the soldering iron damages the blade or the blade breaks, I will not feel uncomfortable, I just throw it away and replace it. I have used tweezers, but I still put the tweezers somewhere for picking and placing parts, but I am fed up with losing so many resistances and covers, so I just dipped the blade into the flux or paste In the mixture, and then grab the parts as you grab it. A bun-bun because I was too lazy to switch to the tweezers that I still couldn't find, so I could only dig out larger parts. I came up with this idea when I read about Chinese manufacturers using chopped and mashed chopsticks to manually pick and place SMD components. Been doing it for a while, but prefer the blade. Remember, the blade has two sides, please make full use of both sides. If #11 does not match your style, please check other X-acto blade shapes.

If you want to do a lot of SMD work, please buy a stereo microscope. There are some cheap Chinese versions on eBay that work normally. like this:

It will be great.

Get the best soldering iron tip you can find. We use Pace with plugin hints. They are not cheap, but they heat up immediately.

I want a decent stereo microscope as my workbench, which is very useful for removing metal or wood chips from my fingers (obtained in a woodworking shop).

A decent stereo microscope, with a good long (several inches) working length, has paid off in many fields, yes, including pulling wood and metal fragments, and making smd work very easy. do it! I have extra products from edmund Optical and B&L, and the former is actually better (and updated a lot). It takes a while to get used to the curved light path-the hand coordination of the eyes will be confusing at first (not for a while afterwards). worth it. Zoom-priceless. I use a paper background and a cheap halogen gooseneck table lamp with floodlights.

do it! It's worth it for the shards and makes the work of SMD almost easy. Most of them have a good working focal length-around 3 inches. The background of the paper, halogen gooseneck table lamp and floodlight, is worth it. Curved pointed tweezers are cheap when having MPJ. I use Kester 952 flux (diluted with IPA if necessary) and occasionally drag the braid, but sometimes it is not necessary to drag the solder. If you want to lay out your own circuit board, please do the opposite of normal-make the width of the pad 1-2 mils narrower than the pin to ensure that all bridges are not on the pin board, but on the pin. Save the trouble of eliminating any bridging. If you do this often, the expensive MetCal iron is actually worth it – I would not believe there is such an iron (think you are a good customer, because I bought things for me, so I never do it for myself Out of pocket).

Agree, however, this article may not be suitable for beginners.

Cheap Chinese 10-20x stereo oscilloscopes cost between US$100 and US$200. You may be able to build your own LED lights. Anyone who does a lot of SMD work should be equipped with a stereo microscope. Once you use it, you will know that it is worth the money. I cannot strongly recommend these.

You can also easily remove debris from (or someone else's) fingers, fix small accessories, etc.

If you make something small, buy one. You will not regret it.

Even cheap mono USB microscopes cost less than $40, and most still provide good images. For some of us, position is a major factor, and avoiding the stooping posture of certain conventional magnifying glasses will give the bench more time. This is a good basic description, but please avoid the mistakes of the former HAD writer, who laughed at several readers, including me, who said that they always use magnification when working on the circuit board. He is one of your typical theoretical builders and can install 0.01 mm without dragging or bridging. It's nice to see that the editors have eliminated most of this attitude.

My suggestion-please use sharp metal tweezers. When talking about component placement, this is a solution that will never let me down. Plastic tweezers tend to melt and are difficult to sharpen, so you only need to obtain tweezers and a demagnetizer. In fact, 4-5 pairs are obtained. You can give them away, and if a pair is lost, you will have spare parts, so you don’t need to look for one of the most important tools and you don’t have to make things =)

I started in 1206-resistors/capacitors of this shape are still much better than through-holes in terms of taking up space, so the circuit board can be routed more easily. Then I moved to 0805 (still placed manually), it was easy even with a relatively large soldering iron.

You must also manage some more parts. I used to be able to place resistors, capacitors and ICs and everything outside of their packages, and if they happen to be mixed up, I can sort them quickly by color code or number. Now, on many parts, the numbers have no meaning for which part, and their color codes have no meaning. It's better to put them in a package or labelled tape instead of dumping everything you need to use on the bench. The IC is still marked, at least it is easy to figure out. Similarly, you may have to etch a board or fabricate one. No more prototype boards. Most "solutions" are expensive or cumbersome to use.

Well, I still have some through-hole parts, and I only buy through-hole parts for prototyping. In addition, there are simple breakout boards suitable for all types of packages. If you need to prototype on SOT-23 FETs, you can easily solder smaller parts to 2.54 connectors or male-to-male prototype terminals.

The SMT welding method I chose is a standard kitchen-grade hot plate. Open the board, apply solder paste to the board as you like, place the components, then bounce the whole board on the hot plate for a few seconds, and remove it after all the solder paste has melted. Gently place it on a flat surface (I use a ceramic plate), wait for it to cool, and so on, all SMT soldering is done at once!

However, this only applies to boards where all SMT components are on the same side (and a heating board is required *no* magnetic stirring function, unless you wish to observe your components sliding around the board with large metal clumps).

Yes, so far, the easiest way to mount solder is not to do this. Sitting there with a soldering iron and tweezers, soldering, soldering is very cumbersome.

Another voted for the use of electric hot plates. We use them to build prototypes, and even use circuit boards containing only a few components and cheap metal templates for some light-weight production work.

Fear is the way to the dark side. Fear leads to anger. Anger leads to hatred. Hatred leads to Frizinger. "

The only problem is to find the SMD kit. They are really hard to find.

Does anyone have any links?

I use "freezing tubules" to fix my SMD parts. Not my idea, "That is the extinction where I used to work. They are in a small box with a cardboard separator, you can write this value on the screw cap. Check the biological supply site. R's group , A group of C, different packaging boxes of different sizes (although I almost exclusively use 0603)

Whenever you order a part from Digikey, you only need to order 100 cut pieces, and then put the other parts in the freezing tube. You will immediately have a complete toolkit.

One source:

Use a screw cap to obtain the plastic type. Different color upper limit of R and C

My Digikey order last year had a lot of beautiful blue static dissipative zipper positioning bags for many parts, including cutting tape SMT passive components, so I just put them there. My shipping box has the right width for easy storage of bags. :)

Digikey has only recently started shipping in resealable bags. great!

People can buy large cut-resistant tape/ceramic capacitor books on eBay for around $20.

We usually standardize around the following aspects:

0805 is suitable for 1% resistance/LED

Ceramic capacitor 0603

1206 is used for fuse/diode/large capacity capacitor

These sizes are suitable for handling with tweezers and are very tolerant of reflux in household ovens.

Note that if possible, you want to reduce the change in value, as this will reduce the cost of making anything at home and in the factory.

Alcohol-based liquid flux with desoldering braid is indispensable, even with suspicious solder mask (if there is enough gap in the design process), we can easily drag 84-pin QFN solder. If you are using solder paste, you must remove the flux, otherwise you will not be able to check the contact area of ​​the chip for tiny solder ball contamination. Admittedly, I was skeptical at first, but it turns out that QFN bridges have far fewer parts than fine-pitch leads.

It may seem counter-intuitive, but smaller parts (such as 0603) have a higher relative surface tension and will pull the SMT part into place. In addition, when I was a student, people often told us to avoid plating through holes on the ground pad, otherwise the solder paste would reflow into the holes and the chips would not be properly bonded. However, the facts show that usually only about 70% of the actual pads need to be covered with SAC305-based solder, even 5 plated-spaced through holes can significantly improve cooling, and if reflowed, the QFN chip will be less likely to be in alignment During the period, it drifted and pasted too much without alignment. If you want to delay EE, please advocate its recommended practices for all reflow operations. ;-)

I like to see effective ways to avoid using canned death solvents around people. Please note that most RoHS parts will rot in moisture, and unlike plastic, it will not dry out easily. People should use desiccant boxes with aluminum foil bags to extend the 2-year shelf life. In some cases, even too much flux will not help, but I have found that most reflow problems are caused by oxide/debris or solder paste quality.

In addition, if you only need 1 board instead of 3 boards... because most boards have reflow time characteristics, you need to adjust.

Ask the company for a sample kit (if you are not a company, but an engineering student, it may take a long time). Digikey has a kit,

I have to disagree that reflow soldering is an "advanced" skill. This is really easy compared to manual welding of SMD parts. The hardest part is getting a reflow oven, but it's not that difficult-most entry-level hobbyists just modified an oven. I sold my own conversion kit on Tindie, which made it very easy.

With the oven, all that remains is to use tweezers to place the components. I use 0805 as my standard size, it works normally and has a lot of parts. As I become more experienced, I sometimes complete TQFP packaging and QFN. But my first project used nothing smaller than SOIC and 0805.

At this point, what I want to say is that manual PnP and reflow soldering are actually faster than through-hole assembly. After all, all welding is done at once, and no leads are clamped. A circuit board I often make mixes SMD and TH because it handles high voltage. I estimate that in this case (YMMV), the same number of TH parts as SMD will take one-third of the time. All it has to do is gain experience and familiarity.

The last thing I want to say is that I did this for the first time, and I thought it would be a disaster. When I finished placing it, it looked terrible. I almost threw it away, but decided to pass it through the oven. The result is a miracle. It looks bad, but when it first came out it seemed to be made in Shenzhen. Use (surface tension) force Luke.

I even used a fully stocked oven for reflow soldering on more than 100 boards. The only modification is to set a kitchen timer nearby and manually control the profile:) Low 3 minutes, high 2 minutes, turn off and Let cool :)

If you are a beginner at the beginning of SMD welding that this article is aimed at, it is advanced. But yes.

I am the timid beginner, and I want to assure those who are standing in the same place (in my 20s to 20s) that this is easier than they thought.

Hot air and flux, that's all

SMD I say no to you.

Please do not make the mistake of repeating the problematic SMD soldering technique. They are popular, but they are not correct.

First, let's talk about the size of the components. Although it may seem like a large component is needed at first glance, it is not. For example, for chip capacitors, you want to be as small as possible. Larger enclosures are very susceptible to heat-related stress cracking, especially when manually soldered incorrectly. Large components are also expensive. I found that almost everyone can use 0805, and 0603 is the right compromise between cost, size, usability and ease of assembly. 0402 can, but it does require more effort when placed.

Good light source I can agree with. This is very important.

For SMD welding, the magnification is almost always *completely useless*. Inspection after welding is very important. It is also important when making bridge connections and can help with other forms of rework. Welding does not require magnification: you need to be able to operate both hands in a precise manner at the same time, which requires direct visual feedback. You want to grab a component, place it and solder it all at once. If you really have trouble seeing how the components are aligned, you can use magnification, but other than that, it makes the situation worse. Especially the recommended magnifiers, because their optical quality is usually poor and small, so they will always get in the way. In order to check after welding, you want to have a reasonable magnification, ideally a little more than the magnification used for welding.

Slotted screwdriver: Why? do not disturb! You have tweezers! SMT components do not require pressing force, so just use tweezers. I never knew about screwdrivers; they are difficult to handle, and you will most likely end up scratching, cutting or otherwise damaging the components.

Tweezers: Not sure if you need a set. Having multiple tweezers is great, but you can also use a pair of tweezers. They must be of reasonable quality, that is, they cannot bend metal waste. They don't have to be expensive: I spent several dollars with Chinese "Vetus" tweezers, and they are definitely worth the money. I do recommend Aven's EZ-PIK series (I bought it from digikey) because they are inexpensive and color-coded, which makes them stand out and therefore easy to find. Whether you want a straight or curved nib depends on taste, but if you are serious, please provide both models. It's not around $15, and two reasonable tweezers are surprisingly expensive. Over time, I bought some expensive high-quality tweezers for myself. They are better, but not many.

iron. Oh, well, where should I start? Tipping is likely to cause many problems. SMD components are small, so you want to have the most control. Ingenious techniques are very useful for rework, because if you want to connect a wire to a single pin in a 0.5mm pitch package, you only need to heat that pin. They do attract tin because they do not transfer a lot of heat. SMD soldering must be done quickly, because the components are definitely more refined than the lead lead brother. The chisel point is the best. Choose a pad that is slightly wider than the pad of the chip component. I like to use 2mm or 2.4mm wide chisel tips on 0805 and 0603 (sorry, not sure yet). They can heat up the pad quickly, and when the components are tightly packed, they are not clumsy enough to cause problems.

Welding technology: please do not use moisture-proof and moisture-proof methods! ! ! This technique seems popular, but it is one of the worst things you can do. The reasons are as follows: (i) the components have been subjected to great thermal stress; (ii) it is difficult to handle them correctly; (iii) the quality of the solder joints obtained is unqualified.

(I) If you solder all the pins with a large amount of molten solder, it will take a lot of time to heat them. Then, they are heated again with a solder core, which takes a long time. This puts a lot of thermal stress on the components. ICs don't like soldering, so ideally you will keep the thermal stress as low as possible. Using this method, you heat the entire chip twice, each time also requires additional time.

(Ii) The core needs to be heated thoroughly. It is difficult to completely control it. If stuck, it is easy to damage the pads or leads of the component. It is almost impossible to precisely control the amount of solder sucked away. This leads us

(Iii) The quality of solder joints. There are many websites that talk about the quality of SMD solder joints. Go and see them. IPC 610A provides more information, but it is not important. The important thing is: the back and sides of the pin should have a nice rounded corner (heel rounded corner). It is not recommended to solder on the top of the pin or near the component package. With solder wicking, you will first get a lot of solder on the top of the pin and near the package. Then you suck the solder away, usually without leaving heel fillets. Components soldered like this can easily fail to be fixed due to mechanical shock because they are just not properly connected to the PCB.

So, what should you do? Drag welding. There is a lot of information online about how it works. It is very easy. I was anxious at first, but when I tried it for the first time, it worked well on the 0.5mm pitch TQFP. You only need to use the right amount of flux, and try not to use too much solder. The pins and pads take up very little solder, so usually very little solder is enough.

Therefore, the flux is missing from your list. Get the best no-clean gel flux. Use it whenever you encounter a problem, it definitely helps. Yes, good flux is expensive and a bit messy, but it is definitely worth it.

I prefer to use a magnifying glass when soldering, even for through holes. In other words, I totally agree that the magnifying glass displayed is garbage. However, the 5-inch fluorescent rocker magnifier will not hinder you when welding.

Until recently, I used the old Dazor articulated magnifying glass lamp, which was above the work and did not move my hands. Recently, I added an old Bausch & Lomb stereo zoom microscope and I like it very much. Although it is not as wide as a magnifying glass, the clarity is excellent.

It may be a cheap magnifying glass, but it does a good job for my SMD work and will not hinder you. :) I don't use alligator clips.

This article is written in response to some potential customers in my suite business who are afraid of SMD work. Therefore, this is a kind of eating my own dog food, and I must be able to assemble them using the basic kit. I sold them a toolkit that was only used in a reflow oven is useless.

I am an engineer and do some SMD prototyping on a regular 0.1" perforated board. I also own a 5cm^2 small SMD oven with only 80W instead of most 1kW ovens.

For single-sided PCB:

-Put a (micro) dollar solder paste on each pad of the PCB,

-Use tweezers and a magnifying glass to place the component on its pad (don't worry too much about the alignment problem, capillary action will solve this problem),

– Make sure that SO is not at home,

-Turn on the fan to avoid being suspected by the above SO,

– Put the PCB on the ceramic stove,

– Turn on the furnace and wait for the solder to liquefy,

– Turn off the stove and carefully move the PCB to the colder part of the stove,

-carry out,

-So it is not wise.

If there is a better choice to buy resistors/capacitors or buy as needed, does anyone suggest?

If you have a specific design, please buy it as needed, plus some additional things (depending on your confidence). If you are making a prototype and are not sure, get various passive components. Unless you use specific things (buffers, transistors, etc.) in many designs, it makes no sense to get too many extra ICs.

For example, you can get the following information:

Various 0603 parts are also available.

They sell various cutting tapes for resistors/capacitors on Aliexpress. You are buying $2.08 and $3.82 for 36 values ​​x 20 0603 resistors/capacitors. Even for practicing welding, this is a reasonable choice. I don't know what size you want, but 0603 is a suitable size.

This is a store I have visited six times.

It depends on your goals and other factors. I bought large-size SMD resistors and capacitors on Ebay from overseas, and the prices of both are around US$20. Unless these organizers can be expanded, don't waste money on more expensive goods with expensive organizers. Otherwise, you will have to use two organizational systems to store other values ​​to be stored, which can cause confusion. Or merge into a single system, in which case the original organizer is wasted.

In fact, every project still requires me to place an order for things I don't have, so the classification there is not very helpful. But I miscalculated the number of passive components in the order. It's used more than I thought, or more than imagined on the carpet. Found that I need to change a value. In all these cases, having a variety of goods frees me from paying the shipping cost of extra parts in cents, and without waiting for shipping. Various products are worth it to me.

DigiKey offers 5000 0805 1% resistor reels for $13. At such a price, I don’t know why you should buy a cut tape, unless you really want to make a thing, and if you are considering learning to do SMD, it means you intend to do so. More than once.

I need a reel for each value because I need a new one. This led to log purchases. :) In many applications, you only need to keep the value within an order of magnitude-such as a pull-up resistor. Therefore, starting from the decimal value -100Ω, 1kΩ, 10kΩ, 100kΩ, 1MΩ. For 5v LEDs, a 330Ω series resistor is usually required. For 3.3v, it is 150Ω. So far, I have purchased about 12 values, but I have completed many projects.

So far, I have used up one paper plate-0.1µF 0805 capacitor. That is my standard IC bypass capacitor.

And, if you think that there are too many full reels for each resistance value, for small-scale users doing manual assembly, ordering 1000 times on the cutting tape is a good choice to keep inventory. A 1,000-fold bag costs only a few dollars and takes up less space than a full paper tray.

I have been engaged in electronics hobby for 50 years, so I no longer have good eyesight, but SMD is still feasible. Very bright LED lamps (non-halogen lamps can keep heat and ultraviolet light low) are very important. Most of my work is done at one time, and size is not a key factor, so I will try to use through holes on IC breakout boards. I built a rather complicated reflow oven for SMD. I prefer to use an electric stove instead of an oven, because my circuit board is single-sided and the heat can be dissipated to places where there is no need to heat large parts. I use infrared bakeware to minimize the delay. The main problem is solder paste placement. Maybe the quality of the paste I have been using is not good enough, but it tends to be very sticky and does not stick to the pad well. For this reason, I am considering using a small amount of solder resist. For the installation of a single component, I found that when pre-coating the pad, first use a solder core to remove the excess part, and then use no-clean flux and a hot air gun to install the component. Since I usually install a single IC on the through-hole branch board, it works well. I will not use this method for BGA because it is impossible to check. I found that a cheap Chinese USB electron microscope is very valuable. It cannot be used in the welding process but used in post-weld welding inspection.

Any idea of ​​applying solder paste manually or using good solder paste will be welcome.

I find that the paste dispenser is very valuable. If you wipe the board with alcohol first, stick it better. ENIG also helps (thanks OSHPark). It is best if you try to come to the board vertically and try to make a "hershey kiss"-shaped deposition on the mat. For IC, I just put a log on all pads.

However, for QFN, a template must be used, and vacuum placement tools are essential.

My first batch of home construction panels used 0805 components. Use a syringe to apply solder paste and reflow in a cheap oven. This is easily done at home. My latest motherboard requires 0402, which is difficult to do. Although I often perform 0402 at work, I don't like doing it at home because you want a good magnifying glass to check it. You can use the mold for processing, but it is still difficult to rework later.

I have done 0201, only a few 0100 5, I do not recommend them. Hot air is best for small parts. In any manual operation, the circuit board can be preheated, which will minimize the chance of damaging the circuit board. I even used a 500-watt halogen shop lamp to heat the circuit board.

A good soldering iron is a must, and good skills are required. You need the iron to be in contact with the pad and the pin at the same time, and the tips of the hoof and chisel work well. For 2-pin components like resistors and capacitors, you need to use two irons, one in each hand-faster and the surface tension will be aligned with the part.

Compared with SMD, the through hole speed is slower. I can solder SMD devices 10 times faster than through holes. It requires more practice, but it is worth it.

Look at the professionals. Sorry, this is an old Flip camera. That was the ARM11 6410 board that was upgraded with NAND a few years ago.

I remember the first time I tried SMD (without flux), I got stuck in a lot of soldering irons. Later I "discovered" the changes and the situation became much better. Nowadays, when I am really lazy, I use solder paste and hot air or oven. Even I have a tool that can cut templates at any time.

Shameless plug warning:): At this time I am learning how to solder the kit. What are the odds:) Here is more information:

It is a kit that will produce high-precision resistors with 54 0805 resistors (in 1OHM steps 0-1MOHM). All you need is a decent soldering iron and flux (or solder paste and oven, but this is nothing fun: D)

You should publish articles on the development direction of different "open pick and place" projects. I remember Firepick, OpenPnP and LitePlacer. The last one is the only PnP machine that is actively shipped.

How much is your time worth?

I set a price for the home DIY platform and concluded that even the Neoden 240 machine is more economical.

That is, your opportunity cost rate will be lower than $1.50 per hour to build similar projects.

Fast food workers will laugh at you.

;-)

Most amateurs will not benefit from PnP machines. For most hobbyists, placing parts by hand is an order of magnitude faster than setting up a machine.

If you are like me, and you own a Tindie store and do many of the same things, then it's very different. But even I haven't been there yet. On the contrary, for some of my most popular products, I have outsourced the assembly. I still think I might get ahead of time to get my own PnP and set it up.

This is real. true. ;)

Shameless plug, but if you want to keep the PCB stable when soldering SMD parts, try Stickvise. You don’t have to buy it, although you won’t regret it – if you have the tools and time, you can easily make it out of scrap:

Gosh, the second image is crazy. According to modern PCB standards, this is really a hodgepodge.

This is the VFO I obtained in a radio trash sale about 20 years ago. From the perspective of components, it was probably in the early 1970s. Very often. :)

After being forced to weld 0402 and some QFN parts, I developed some "cheater" methods for these parts.

1) The first thing I want to do is to assume that everything I remove is toast and will be exposed to too much heat; it will not be reliable in the future.

2) In the case of using all small SMD passive components (and a large number of ICs), I tinned a pad, and then placed the part while keeping the pad "wet". You are an animal

3) I have always been ready to wind the wire, which will make it ~2" long and wrap 3/4 around the soldering iron. Then, I will form the remaining part into a pointed extension tip and use it to cushion the “thermal inertia of the soldering iron” "Usually, this will cause me to use some very fine solder even on 0201s under extreme pinch injuries.

4) Sometimes I solder a short 30 AWG wire to the pin as an indirect heat transfer path for the soldering iron. I just tinned about 30 AWG wire, glued it to it with a small amount of solder, and then used the wire as a physical fixative and a method of transferring heat to the pins.

Side note: I'm a bit short-sighted, so I can use a relatively close focus to solve many details. I wear glasses most of the time, but when I wear contact lenses to work, because I can take off my glasses, I can't see close-up photos. Therefore, contact can become an obstacle to handling small parts. Get out of trouble

Personally, I don't use a magnifying glass when soldering (the smallest thing I usually do is 0402 on smaller pads). This is my idea. For inspection, this is great. However, the tips here are flux, flux, flux and a good light source. I personally use liquid flux from a syringe. The flux is worthless because you have to press them on the board. Soldering QFNs is not a big problem, as long as the pads extend on the side (so that you can drag them).

I actually do own some 01005 components, but I have never tried to solder them. I should take a moment to try them and see how scary they are.

The worst project I have ever worked on is my picosumo (1.25cm cube robot). The circuit board is 1x1cm and has a lot of 0402 (with additional small pads because the standard in Eagle is very large). Due to a design error, the board even has bridge wires.

By the way, Murata has actually demonstrated capacitors smaller than 01005. They obviously use the metric 0201 (link below), so don't complain that the SMD is too small. It could be worse!

There is an advantage to using through holes that I have not mentioned-if one orders a professional PCB but uses some kind of fast home production method, then fast wiring becomes a problem, because either everything is on a required layer or needs It is processed by riveting/tin wire welding or other methods. On the other hand, vias allow you to get "free vias" on each pin-you can route from each pin on either side, and aligning the sides with the via pad size is usually not required pay attention.

Wow. I do nothing.

I hate flux. The exposed flux surface absorbs moisture and becomes partly conductive. This in turn acts like a huge parasitic capacitor, acting on the traces on the PCB surface, so the less flux I use, the better. Likewise, for modern packages, these have very little spacing between the bottom of the package (such as VQFN) and the PCB surface. Capillary action will cause so much flux to flow into that small cavity.

Side note: I use some of the fluxes mentioned here for cold soldering/silver soldering/low-temperature brazing or wherever you are.

I also use 60/40 solder.

I focus on surface treatment instead of using more flux. I have pee/solution that can eliminate oxidation and pollution.

I drag the fine pitch solder, and finally always leave a solder bridge between the last two pins. Sometimes I use a braid for this, sometimes I tilt the board and apply more solder (actually flux), depending on what I think will minimize the heat to the chip.

Apart from that, I can only add one point, my light box or any spotlight helps a lot. I will try to get pictures.

VQFN 100 0.5 mm

Good light box The bell shape on the light box (5600+Kelvin) is clear.

This is QFP, not QFN (with external leads)

The bent leads in the upper and lower right corners can cause shorts, especially the top shorts.

Without a proper side view, you cannot see if the rounded corners at the heel/toe of the lead are correct. The surface finish of the joint is also not visible.

As for what will cause flux problems later...clean the circuit board after soldering and you will never see any problems.

I learned a lot from one-off, and SMD is not very practical in this regard. I guess? You can't just grab a stripping board and put things together. I think the demand for custom PCB is higher, which is not very practical.

I want to be proven wrong. Is there any good way to use disposable/prototype circuits with SMD components?

For peeling boards, try using SMD resistors and capacitors. 1206 and even 0805 components can be directly welded to the strip by hand, thus providing you with a compact layout.

Generally speaking, if you can spin your PCB quickly but hate all drilling, then SMD is great once. I may modify the PCB design and perform toner transfer etching during the time that I need to perform layout and point-to-point wiring on the ribbon board.

In other words, PCB layout and home manufacturing have a learning curve, and you need to climb first to reap the rewards. The patch panel and splitter of SMD parts allow you to complete the work with less learning.

I think the time to actually learn to make Som PCB is almost over. It may not be that difficult, it is just a matter of actually getting what I need to do.

Sorry, I have provided you with a SMD prototype board. It accepts the most common smd packages, such as SOIC, 0805/0603 RCL, smd switch, 0805/0602 LED, etc.

There are many SMD parts around me, so most of the prototypes I make are SMD. It does not take up too much space, so the wife will not complain too much :)

I think my biggest problem is purchasing parts. Some tqfp microcontrollers have been used for some time. The problem is that most of my inventory is salvaged through punched parts. The results of salvaging smd parts are much less. Most parts are unmarked, so while it is easy to figure out what (in terms of passive components) is easy, determining the value is not so easy. The resistor usually has a code, if not, my multimeter will help. Now, all inductors and capacitors without codes are the bane of my smd experience. My cheap electricity meter cannot measure these. In addition to passive devices, IVE also found it difficult to find data sheets on thin films such as diodes and transistors.

And the other thing that disappointed me was the board of directors. Many of my projects are on various prototype boards. It’s great to have various smd parts in some places, but nine out of ten, I can’t use the parts I want on the board. If possible, I will make the board myself. But all my inventory is wrong. The large and thick double-sided panels I have seem to be too overkill in terms of surface treatment, and I always struggle. One of these days, I will get some lighter single-sided inventory and use jumpers like most sane home etchers.

A few years ago, I tried SMD for the first time, but I never flinched. At that time I only had a tweezers and a soldering iron. Now, I use tweezers, toothpicks, solder paste and electric skillet.

Like most tricky things, the practice is perfect...Anyone is dead HD discs/similar discs? There are many types of sysads that we have worked with, and I used to steal controller boards from there. Desolder some ICs and put them back again. Repeat 10 times. :-). The date granted to me is from the current board with up to three ICs, TSSOP flash memory, SDRAM and less BGA to the present, but many failed HDs are also ancient. :)

"We are waiting for commentators to claim that the Thousand-pin BGA and 01005 chip components were manually soldered as expected, but these statements are useless without evidence."

How about reprocessing 0201 indicators all day? If the BOM of only one board is five numbers, then you have no choice but to learn. In the end, experience and good tools are not that difficult (stereo microscopes such as the mantis are essential).

las, there is no evidence. . Damn NDA:(

I learned to solder smd only with solder paste. It saved me hours of headache. The price of a tube is about 30 dollars, but I have had it for a year and I do a lot of welding work. Use a template with multiple templates, or use a syringe to place the spots on the mat, which are added parts. It has the right amount of flux and solder, all you have to do is hit it with a sharp soldering iron for 2 seconds.

When using conventional solder for soldering, obtain very thin and expensive materials. Also great.

Oh, I do it 0603, sometimes 0402.

This post is at the right time. By the way, the boss of the company asked me if I have any micro-welding experience (lack of confidence), because the main loss of his business is to transport the parts out and fix it through micro-welding, so I told h