State of the art SMD/DIY best practices

andrew · 2019-10-26T04:02:23.907Z

v in-depth and informative video. answered a lot of questions I was gonna post here

okyeron · 2019-10-26T04:29:43.769Z

Good stuff on that channel.

I think it’s hard to give advice since technique will/may depend on your own equipment, style, etc.

Like I use a very fine point tip on my iron so I can touch individual pins on a TSSOP, but others might prefer drag soldering instead.

I did just get wise to using a solder sucker to fix big solder blobs on ICs (instead of copper braid) I think that might’ve been in an adafruit video.

okyeron · 2019-10-26T04:34:26.027Z

Another tip

I like this no-clean liquid flux.

Then I have a small syringe with a fine flat tip and just suck up a milliliter or two at a time and squirt out a tiny bit where you need it.

andrew · 2019-10-26T04:47:16.313Z

o how much do I need to think about grounding ? I don’t have a strap :0

nurse_plissken · 2019-10-26T05:20:59.903Z

I’ve never bothered with it. I think it really will only be an issue with real sensitive electronics like fancy micro controllers and stuff, which I haven’t messed with very much. It won’t hurt to have a strap though!

I also do all my DIY work outside on the porch right now, so my practices are probably not state of the art

Cementimental · 2019-11-04T17:15:17.842Z

i remember seeing those in the maplin catalog 20+ years ago and haven’t thought about or used them since so you should be ok

andrew · 2019-11-06T21:10:41.004Z

anyone mind explaining testing strategies ?

I’ve got a multimeter and my first assembled board but honestly I don’t really know what to do with the two lol

just trying to make sure I didn’t make sure I didn’t make any soldering mistakes before I plug in my norns shield

barnaby · 2019-11-07T00:08:57.932Z

The first thing to do is always to test the resistance across the power supply to make sure there are no shorts. To do this, put the meter in resistance mode, connect the positive test lead to the positive power in, and the negative test lead to 0V (repeat across all combinations of pins on a multi-rail powered device, which I guess the norns shield is not). The resistance should start out very low but quickly rise to at least tens of thousands of ohms if not more as the decoupling capacitors charge up. If it stays at or close to 0 ohms, you have a short somewhere, which needs to be dealt with before you power up!

With access to the schematic you can do further testing without power, but typically if there are no power shorts it’s fine to plug in and see if the thing works. Test all buttons, LEDs and whatever, if some don’t work, check their connections, their multiplexing chip, etc.

okyeron · 2019-11-07T00:11:08.081Z

Try to find teat points where you can get known voltages. Usually this means looking at Eagle, isolating/viewing power /ground traces and then testing those points.

With those points you can check for continuity between power/ground - which would indicate a short somewhere.

Norns shield relies on pi for power so you gotta plug it all in to test voltages.

andrew · 2019-11-23T20:22:18.519Z

Approaching - DIY norns shield

IMG_3306961×870 290 KB

srry for the cross-post, but would anyone mind helping out with an IC diagnosis ? (codec is not registering) I hit those legs with my iron again post-photo and I can’t recognize any bridges - but I was having problems getting flux gunk clean so I suspect that could be an issue ? what are the tools and process for doing that properly ?

many thankssssssss

thopa · 2019-11-23T20:35:03.312Z

pin 13 maybe not making contact with the pad?

I know some people won’t agree but it all depends on how much you value your time, and how well you can deal with frustration and debugging, which is also a good thing because you learn a lot!

But if you want to make it as smooth as possible get a stereo microscope…they go around 250€ in amazon but will make your smd soldering so much easier and it actually it’s a lot of fun.

Lots of people do it with proper magnification.

Hope you get it fixed!

okyeron · 2019-11-23T21:31:34.483Z

@andrew - alternately - a cheaper USB microscope like one from Adafruit or this one from Amazon can get you closer to the action.

I use a photography loupe because I have it around from days past looking at film.

For the soldering - see if you can’t clean up the caps - some of those look pretty janky to me (although probably fine).

See about getting some no clean flux if you’re having issues with flux gunk.

For cleanup I use isopropyl alcohol and a toothbrush to do a quick scrub and then wipe the area with a microfiber cloth. Sometimes the cloth leaves behind fibers, but those are easy enough to get with a dry toothbrush.

ermina · 2019-11-23T22:16:13.251Z

i have only done a few SMD projects so this might not be the best advice, but it seems like a little more heat would help the solder flow better. Some joints look like you removed the iron before the pad and component were sufficiently heated. If you have a temperature controlled iron maybe check that the T° is adequate for the solder you are using (usually it’s written on the spool). Maybe someone with real experience could confirm this hypothesis.

Jonny · 2019-11-23T22:28:08.220Z

As @ermina said, I would also say that it looks like a little more heat is needed. The solder does not looks like flowed. If you don’t have a set of tweezers for the small parts that would be a good thing to pick up. Also consider a good iron like a Weller and a few different tip shapes. I picked one up about 10 years ago for my projects and it made a world of difference compared to my terrible radio shack iron. Hakko brass tip cleaner is also handy rather than wet sponges.

andrew · 2019-11-24T00:22:52.293Z

I’ve been keeping the iron in the 350-500 range so I don’t think it’s heat, but I might not’ve hit it long enough. I had some issues with bridging / solder-sucker fun when I first placed it and yea It’s def possible that i didn’t spend enough time getting every pin heated up. gonna spend some more time with it. I’ve got a lamp with a magnifying glass which I believe I’ll stick with for now. I grabbed some toothbrushes though lol

& thank you all !
fwiw random free advice from ppl more experienced than yrself on the internet goes a long way toward what expensive higher education can do. just a thought. it’s nice.

rknLA · 2019-11-24T00:52:35.689Z

As nasty as the stuff is, I’d suggest using way more flux. If any of the solder pulls off with the iron like the left side of C11 or the bottom of C3 in your photo, then it’s not a clean solder joint. The solution is usually either to add more flux or to increase the temp of the iron.

Ithacus:

use leaded solder (flows a lot nicer), and just breath out/blow while soldering

Incidentally, I looked this up a couple weeks ago after switching back to lead-based solder and getting paranoid. It turns out that the health and safety issue of lead-based solder is in the disposal of it, not in its vaporization. Flux is the real culprit for fumes

Recognition, Evaluation, Control – 6 Jan 11

Lead exposure during soldering

“Soft soldering” is a widely used technique in the electronics industry for joining electronic components to printed circuit boards. Traditionally the solder was an alloy of lead and tin, typically…

caelmore · 2019-11-24T01:00:03.555Z

Everyone posting in this thread should probably own a copy of this:

https://www.thonk.co.uk/shop/mtm-book/

I’ve already suggested it as a cheap and useful gift idea for my wife to consider for me this holiday season.

soggybag · 2019-11-24T02:18:36.646Z

I love the recommendation of “best solder we’ve used” .

I have this issue with lead free solder as being hard to un-solder. My guess is lead free solder uses aluminum which lighter than lead. Lead, besides melting at lower temp stays molten a little longer, and the density makes it easier to remove if you heat it up and whack a PCB against a table.

andrew · 2019-11-24T07:52:35.092Z

rknLA:

As nasty as the stuff is, I’d suggest using way more flux.

this rang true !
more flux and toothbrush scrubs eased my pain - thanks folks. the problem ended up being the clock oscillator rather than the IC but I’ve got those joints looking pretty foolproof now

kasselvania · 2019-11-29T02:57:21.395Z

Curious, when using the solder paste method with hot air (not a hot plate) much like one of these: https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.com%2Fulk%2Fitm%2F191343139831

If you’re soldering a two sided PCB, would the heat applies to one side be so hot as to melt the solder of the SMD components on the opposite side?