Surface Mount Soldering

[ebastler]

But for a beginner in SMD, who owns and knows how to use a soldering iron, and wants to get first experience using SOICs? Why bother buying new equipment and materials, getting a stencil made, and familiarizing yourself with an entirely new process? That creates an unnecessary hurdle, and I strongly advise against it.

And as someone whos done a lot of SMD, i'd say its well worth the effort to tool up with some new gear, and learn the new process.. Because its a skill that is *well* worth learning.. Just as much as learning to hand solder SMD parts.

Jeez, yes -- we are on the same page here. But I recommended not to bother with reflow soldering "for now", i.e. for a new user soldering his very first SMD (SOIC) parts.

If you go back to the third post in this thread, one "mrpackethead" puts ebclr's unconditional recommendation for reflow into perspective, and points to a very nice hand-soldering tutorial video... 

[mrpackethead]

Jeez, yes -- we are on the same page here. But I recommended not to bother with reflow soldering "for now", i.e. for a new user soldering his very first SMD (SOIC) parts.

And i recommend that it is well worth the effort to 'bother'. 

[ebastler]

Jeez, yes -- we are on the same page here. But I recommended not to bother with reflow soldering "for now", i.e. for a new user soldering his very first SMD (SOIC) parts.

And i recommend that it is well worth the effort to 'bother'.

But can we agree that you don't need to (and probably should not) transition to reflow soldering before you solder your first SMD parts ever? Pretty please?

[tautech]

Jeez, yes -- we are on the same page here. But I recommended not to bother with reflow soldering "for now", i.e. for a new user soldering his very first SMD (SOIC) parts.

And i recommend that it is well worth the effort to 'bother'.

But can we agree that you don't need to (and probably should not) transition to reflow soldering before you solder your first SMD parts ever? Pretty please?

Of course not, I quite agree....forget an iron altogether and go vapor phase. 
If only newbies knew what it was. 

[tggzzz]

But for a beginner in SMD, who owns and knows how to use a soldering iron, and wants to get first experience using SOICs? Why bother buying new equipment and materials, getting a stencil made, and familiarizing yourself with an entirely new process? That creates an unnecessary hurdle, and I strongly advise against it.

And as someone whos done a lot of SMD, i'd say its well worth the effort to tool up with some new gear, and learn the new process.. Because its a skill that is *well* worth learning.. Just as much as learning to hand solder SMD parts.

Jeez, yes -- we are on the same page here. But I recommended not to bother with reflow soldering "for now", i.e. for a new user soldering his very first SMD (SOIC) parts.

If you go back to the third post in this thread, one "mrpackethead" puts ebclr's unconditional recommendation for reflow into perspective, and points to a very nice hand-soldering tutorial video... 

I think it is worth learning reflow soldering very soon, even for small boards. The results are fast, easy and surprisingly good; to my surprise I now prefer SMD. The capital cost is minimal: solder paste and a £5 saucepan, and preferably a non-contact "point and shoot" thermometer. See my .sig for my tools and techniques.

For old excellent soldering videos, see https://www.youtube.com/playlist?list=PL926EC0F1F93C1837 The last one is about "flatpack" ICs.

[tooki]

Reflow for SMD is awesome; way easier than using an iron. The $60 for a cheap 858 hot air unit are worth every cent.

[mrpackethead]

Jeez, yes -- we are on the same page here. But I recommended not to bother with reflow soldering "for now", i.e. for a new user soldering his very first SMD (SOIC) parts.

And i recommend that it is well worth the effort to 'bother'.

But can we agree that you don't need to (and probably should not) transition to reflow soldering before you solder your first SMD parts ever? Pretty please?

You don't need to do anything.    But if you want to learn about the soldering method that SMD parts where designed for its a good idea.     

[ebastler]

"packethead" is another word for "blockhead" in English, isn't it?

(Sorry, not a native speaker... But I don't think I was expressing myself in such an unclear manner. Reflow is great, but not for your first SMD part. Bye now.)

[mrpackethead]

"packethead" is another word for "blockhead" in English, isn't it?
(Sorry, not a native speaker... But I don't think I was expressing myself in such an unclear manner. Reflow is great, but not for your first SMD part. Bye now.)

No its not. Its a reference to the fact that at one stage in my life i was the lead architect in a team building networks.     head of packets? 

reflow is one method.  Hand Soldering is another.       Having spent considerable time as a volunteer teachign electronics assmebly at the local maker space,  i'll tell you that the folks to do their first SMD work with a stencil, paste and a low cost toaster actually get something useful working in the space of one evening.

Hand soldering SMD is a skill that takes much more time to develop.  But it is a skill that is very helpful and esssential when you are developing electronics..

[james_s]

I think there's value in both. If one is not in too much of a rush, I would suggest learning to hand solder it first, then when you build something with a stencil and reflow (or more often in my case a toothpick and reflow) it seems very easy.

It's the same as when I was learning to bend neon tubing a number of years ago, which is much,  *much* harder than it looks, practice with the big 14mm stuff and then when you try bending 10-12mm tube it seems so easy.

[tkuhmone]

I bought myself a few Chinese SMD soldering kits to practice ($2 each on EBay), like this one: http://www.ebay.com/itm/SMD-Components-Solder-Practice-Plate-for-Training-DIY-Module-Electronic-Kit-/281916872729

...today I also ordered couple of boards....

set of PCB's arrived today, five boards totally. Some work for the soldering iron :-) Single sided boards and material propably is FR-2.

[TheDane]

I have worked with stencils, and I would only recommend a stencil while doing a project. A bare PCB, which then gets populated and soldered. One PCB by one another.
The soldering iron is for touching up, very few and larger components, and/or doing through hole pin stuff.

A syringe with fresh/good paste and a decent needle can make a pretty line of solder paste, which I find easier to apply rather than fitting a stencil onto an already populated board I'm doing a repair on, smearing paste while holding it flat. It takes me longer, and a possible larger clean up job afterwards is lurking underneath the stencil.
Excess solder I remove with the tip of my tweezers, under a stereo microscope. This works wonders fixing various iPhone internal connectors, and they're not big. The right amount on each pin, the component on top - heat. The solder between the pins flow to the pin, a perfect result. A standard SOIC is easy, when practice is done -  know the trick's, learn and build/make your own new better tricks. Share 

Older paste can be used in a pinch, I have found - the trick is to mix a larger portion of the tired paste well getting it wet/flowing, and then making sure the receiving PCB pads have been completely wetted. Works best with the larger sized stuff, though, but 0201's still solder ok.

[tkuhmone]

set of PCB's arrived today....

Here is attached images of the board. Kit containing of:
Main PCB, single sided
two 16pin SOIC
two 44pin QFP
several 0805/0603 resistors & capacitors
MELF (Most End up Lying on the Floor) diodes
transistors, SOT-23
0603 resistor arrays

[tkuhmone]

I have an access to optical stereo microscope with sturdy base platform...

I took a quick test of the stereo microscope and the optical & mechanical quality looks very good, it offers zoomed view also (Wild M3Z). I need to get practise with eyes-hands-soldering iron combination :-)

[Vtile]

Without purposedly build PCB .. not a joy. I have done it to certain extend in veroboards and similars, but the SOIC (1/20th of inch spacing) and other multilegged parts are just pita to handle with prototyping methods. The best so far is the coilwire (PU coated [better than AWG30 solids what I have tested]) where you can not fit the part to PCB blobs directly.

For flux I use some homebrewed rosin flux (ecoRosin^tm & IPA), which I don't care to soak (clean) away. Depend on the circuit component ofcourse if it makes a difference.

I hop in to the SMD/SMT area when I can buy 3rd hand manufacturing line for pennies. 

[Zero999]

Hand soldering is the way to go for SOICs and smallish PCBs. You will not even need extra flux for 1.27mm pitch pins; the resin core in the solder is perfectly fine. Just make sure that to use sufficiently thin solder: 0.5mm diameter is fine; thicker stuff makes it more difficult to dose it right.

I agree. I've noticed some people recommend extra flux but I find it counter productive. All it does is create more smoke and mess. The only time I've ever had to use extra flux is when the board or components are badly oxidised.

[Vtile]

Hand soldering is the way to go for SOICs and smallish PCBs. You will not even need extra flux for 1.27mm pitch pins; the resin core in the solder is perfectly fine. Just make sure that to use sufficiently thin solder: 0.5mm diameter is fine; thicker stuff makes it more difficult to dose it right.

I agree. I've noticed some people recommend extra flux but I find it counter productive. All it does is create more smoke and mess. The only time I've ever had to use extra flux is when the board or components are badly oxidised.

It might have something to do with the solder used. If it is far from optimum the extra flux might help it to flow. "Dry" solder tend to be "sticky" and mainly forming a globs. This is my experience soldering with far from decent equipment and with old chinese mystery  solder (it were cheap 15 years ago when I got the big spool with my hobby pennies )

[electrolust]

Microscope. Microscope........ (Optical Stereo is best, USB Digital the worst)

Any particular product recommendation?  I've been using a visor/headset glasses but these are limited to about 3.5X magnification and I would like more.  https://www.adafruit.com/product/1345 is interesting, and with TV output I guess the response time is instantaneous but there's still the problem of it being 2D.

[TheDane]

Microscope. Microscope........ (Optical Stereo is best, USB Digital the worst)

Any particular product recommendation?  I've been using a visor/headset glasses but these are limited to about 3.5X magnification and I would like more.  https://www.adafruit.com/product/1345 is interesting, and with TV output I guess the response time is instantaneous but there's still the problem of it being 2D.

Oh boy. Not by name/brand - sorry.
What I have found to be great is a base which has the right height, so you can work comfortably under the microscope without tiring your hands, arms, fingers, shoulders, etc.
Naturally a heavy sturdy construction that doesn't tip, a base working area that suit your needs and dials that move without slack etc.

The 3D view is excellent as it minimizes guesswork on where the tip actually is, minimizing burned/marked adjacent components when placed really close to what is being worked on. There's no delay in visuals, so work pace can be sped up considerably.

I tried to google 'stereo microscope' and then do the pictures view. There's tons of different layouts. Some might work great for you, and not for me - I have worked with a few different types, and typically you get what you pay for quality wise.
Most common difference is viewing angle - some are directly down, I don't like those. I prefer those you 'look into'.


I recommend a microscope with a zoom level starting as low as possible - I find that when working with larger components and a magnification of just 4-5 - there's no room in the viewing area, and I have to move things around. 40x magnification is enough to do 01005 components, and still getting good views - if the lighting is good enough. LED rings are not great imho, the light from a halogen bulb is much better for identifying colors and contours like tracks, solder mask, bare board, component cracks, etc.

- and possibly getting a local one, as shipping glass mounted lenses can (if knocked around too much) cause problems.
 
Good luck, and they do start somewhat cheap - depending on model and features. Buyer beware, try it out if possible before purchase.
PS - I have no experience with the mantis-type

[bson]

Any particular product recommendation?

The cheapest Amscope with 2.5X/5X magnification and boom mounted (so you can swing it out over your work area), with the cheapest CFL ring light.  Directional lights cast shadows, and at such low magnification lighting isn't a big issue.

This will work nicely:
http://www.amscope.com/microscopes/10x-30x-stereo-microscope-on-single-arm-boom-with-ring-light.html
add a 0.3X barlow lens - this will reduce the magnification to 3.3X/10X, widen the field, brighten the view, and triple the working distance (to 300mm at 3.3X).  The light will be a tad dark at 10X but that's a really high magnification for this sort of use and more for close inspection for things like solder crack lines.  I have an older trinocular version of this (2.5X/5X), and trust me - there's very little to be gained from spending more on something better. (http://www.amscope.com/accessories/barlow-lens/0-3x-super-widefield-barlow-lens-for-sm-series-stereo-microscopes-48mm.html; there's also a 0.5X but that would probably yield excess magnification.)  Trinocular if you want a dedicated camera attachment, but frankly unless you are seriously into shooting videos or do classroom instruction, don't bother. I have a trinocular head I never use and in my mind the most expensive tools are those that never get used, no matter how little I paid for them.

The other alternative is to hunt on eBay for a nicely priced used Zeiss, Leica, Nikon, Minolta, etc.  I do strongly recommend low magnification and a boom mount though; even if the glass is superior a dissection type scope will be far less convenient (and at such low magnifications even the Amscope optics are perfectly good).