Hey folks! Lemme pick your collective brains for a minute here...
Waiting for my first ever smd board to arrive from ITead, I was reading up on soldering techniques. Needless to say, a hundred people will have two hundred opinions which continues to hold true when investigating the best soldering tip size.
What do you guys use for 0805 and 0603 and sot23 and such? Basically for anything one could possibly find in a pedal, I'm not talking gaming console rework and whatnot? Conical, chisel, straight, bent? Diameter?
There are many ways to do SMT soldering, however the following two are perhaps the most popular in a DIY home workshop (at least from my limited experience).
1. Iron. First flux your pads with liquid flux or a flux pen (I like the pen). Basically, you will put a small blob of solder on one of your SMT pads on the PCB. Then carefully grab the component with tweezers and melt it down to the blob. This will allow you to place the part so it doesn't move. Then flow the other side, and perhaps touch up the first joint. Make sure you get good "fillets" on each side of the component. A smaller size tip would be helpful but it doesn't really need to be micro sized at all. It is more helpful to use a smaller gauge solder wire. I use small weller tips like the 1/16" ETA screwdriver or conical. It really doesn't make or break any SMT project...personal preference.
2. Solder paste/reflow. This is quite cool but there is a small learning curve. Essentially, you get a syringe of solder paste and put a dab on each pad. It takes a bit of practice to know when is too much/too little. If you are doing medium scale production, you could also use a stencil, but this is more complicated and costly for the actual stencil.
After the paste is in place, put each component with a set of tweezers on the place where it should be. Orientation is important but doesn't need to be exact. After this, you will gently lower your board onto a skillet (hot plate) and turn up the temp. In about 3-4 minutes, the solder paste will turn from grey into silver and magically conform to the pads and even move some of your components around. Once everything has reflowed properly, you can wait about 30 seconds and remove the heat. Shutting off the heat is one way to do this, or if you are careful you can pick up the PCB and remove it from the skillet surface. I like doing SMT this way as it is faster for me and not as tedious as an iron.
You will have some joints which may not have flowed properly or perhaps a solder bridge (especially on ICs) after the reflow. You can touch these up with a bit of #2 desoldering braid and an iron. Check out Sparkfun's Reflow Skillet tutorial for the deets:
https://www.sparkfun.com/tutorials/59 (https://www.sparkfun.com/tutorials/59)
Thanks Rully!
I'm gonna go with the soldering iron for now, as fun as barbecuing boards may be. 8)
I only build a few boards (at best) a month and time-saving is not a concern, especially if it necessitates another (rather large) piece of hardware.
I think I'm gonna go with 1/32 chisel to start things off with and change if I don't end up feeling it.
I wrote up some tips about a year ago, pickdropper has some great comments too
You may find it helpful, but maybe not
Good luck, and enjoy
Couldn't a heat gun be used with the solder paste idea? Granted, it would probably have to be one piece at a time (in order to hold it in place so the heat gun doesn't blow it away) but, at least you wouldn't run the risk of frying the PCB.
In addition to this.... don't most skillets only go up to about 400F? Is that hot enough to melt the paste?
This is copied from an old post I wrote on the subject:
1.) Apply flux to pads
2.) Grab component with tweezer and hold it in place on the board
3.) Put a small amount of solder on the soldering iron tip
4.) Touch the tip of the soldering iron to a component/pad joint
5.) Repeat for the rest of the connections.
After everything is soldered, I often clean up the joints by adding flux to the component pads and reflowing each one very quickly. For this step I do not add any solder to the iron tip, I just make sure the tip is clean.
If you are using a type RMA flux (like I do), you are going to want to clean (wash) the PCB when you are done. There are two good ways of doing this:
1.) Flux Remover - This is the best method, although more expensive. If you have a Fry's nearby, they have good prices on Flux remover. It's handy for a lot of things.
2.) High-Purity Isopropyl Alcohol - This is much cheaper (you can get a bottle at Target or Walgreens for about $1). I have a bottle of this only for cleaning boards. I use a clean soft-bristle toothbrush (not previous used for brushing teeth) that I dip in the alcohol and then scrub the board with. It will take longer with alcohol; you'll likely need to brush it off in 2-3 passes to get rid of the residual stickiness the flux leaves behind, but you can get it.
It sounds like a lot of work, but it's really not. It's just a different process. A cleaned SMT board can look really cool, too (at least I think so). Here is a close-up of a PCB after cleaning:
I have a couple of different tips that I use. For SMT work, I generally stick to round tips. There is one that I have that is very very small and thin, but doesn't work as well if the pad is connected to the ground plane. I nave others that are a bit larger in geometry and transfer heat better.
What sort of iron tip do you find you use most?
EDIT: hah! wrote at the same time... is there a way to delete a post? if there is, I can't find it...
Mostly the smaller conical one, but I don't always change the tip if I have the bigger one on. It really depends on the amount of soldering I have to do.
As I mentioned above, I rarely ever use chisel on a SMT board, although I have on rare occasions.
Interesting. Having no smd experience and seeing how the end of a smd part is perpendicular to the pad, one would (well, I would) think that a chisel tip would fit perfectly at the contact point and certainly transfer heat better to both surfaces than a conical tip?
That's what I thought before I tried to solder small 0805 stuff, but, at least with this size, a small conical tip works exactly as you figure a chisel tip would. You need to heat the pad and the end of the component, but it's not necessary for the tip to fit like a Lego.
OK thanks guys... you just convinced me to go with a small conical tip (I don't have anything smaller than 1.8mm chisel right now). Would 1mm be enough? That's smack in the middle between 0603 and 0805 so I'm guessing I should be fine with that? Or should I go smaller, like 0.8mm (0603 width) or even 0.6?
I just soldered my first QFN and TSSOP devices and it works fine.
Funny thing is that the QFN soldered way easier than the other which is the opposite I thought it would be.
Be not afraid - try it.
Oh I will... i have f(x) board coming in and also my own two smd layouts getting shipped from itead and seeed right about now. Kits bought off ebay, tweezers and flux pen ready to go. Now all I need is a smaller tip and the boards to get here and I'll get started losing parts all over the room. ;D
Quote from: muddyfox on November 21, 2013, 02:47:55 PM
Interesting. Having no smd experience and seeing how the end of a smd part is perpendicular to the pad, one would (well, I would) think that a chisel tip would fit perfectly at the contact point and certainly transfer heat better to both surfaces than a conical tip?
This is what I do. Helps align the part to the pad.
Yeah, Surface mount isn't too bad until you have to do this:
(http://img.photobucket.com/albums/v158/pickdropper/Guitar%20Pedals/TombstoningResistorssmall.jpg) (http://smg.photobucket.com/user/pickdropper/media/Guitar%20Pedals/TombstoningResistorssmall.jpg.html)
:o
I'm sorry folks, I need to keep this going... ::)
Two points I came across in my web preparation that I'm now unsure of.
- apparently it's recommended not to mix fluxes - would I have problems with using Kester 951 noclean pen and what little rosin (resin? never sure which) core there is in 1mm solder?
- noclean fluxes and potential for corrosion later on? Supposedly it's noclean but I'm sure it leaves something on the board and I can't really tell how conductive or corrosive it may be long(ish) term.
I'm sure these questions will seem quite silly a year down the road but right now they are a big deal. ;D
I'm using 0.3mm solder with flux in it (chinese seller on ebay), and a cone tip (I use the same tip for through hole stuff as well).
Sent from my thumbs using Tapatalk!
Does anyone have input on the solder paste/heat gun idea? Would like to get some feedback on it as it seems like a clean, easy way to solder SMD 8)
Quote from: muddyfox on November 22, 2013, 07:59:59 AM
I'm sorry folks, I need to keep this going... ::)
Two points I came across in my web preparation that I'm now unsure of.
- apparently it's recommended not to mix fluxes - would I have problems with using Kester 951 noclean pen and what little rosin (resin? never sure which) core there is in 1mm solder?
- noclean fluxes and potential for corrosion later on? Supposedly it's noclean but I'm sure it leaves something on the board and I can't really tell how conductive or corrosive it may be long(ish) term.
I'm sure these questions will seem quite silly a year down the road but right now they are a big deal. ;D
I can say that I absolutely detest the Kester 951 noclean flux, but that is just my opinion. I FAR prefer the Kester 186 RMA flux.
Quote from: Govmnt_Lacky on November 22, 2013, 12:30:09 PM
Does anyone have input on the solder paste/heat gun idea? Would like to get some feedback on it as it seems like a clean, easy way to solder SMD 8)
I've only used solder paste prior to reflowing in an oven. With passives, it's pretty easy, but with more complicated boards that have a lot of multipin packages, it's better do it with a stencil.
I was surprised at the learning curve with applying paste through the stencil, actually. It's easier than doing it by hand, but takes a bit of practice.
Oven/heat gun is definitely faster for larger boards. For smaller stuff, I prefer hand soldering.
Quote from: pickdropper on November 22, 2013, 01:31:27 PM
I can say that I absolutely detest the Kester 951 noclean flux, but that is just my opinion. I FAR prefer the Kester 186 RMA flux.
Hey, an opinion just as valid (perhaps even more so) than any other! Having never used flux pen, 951 was my best guess.
Having no access to flux cleaners over here (IPA is still debatable, I'm having an argument with the local pharmacy that I'm not likely to win), why would I choose 186 over 951 that's supposedly maintenance-free? Take me to school, Pick! ;D
The main reason is that the 186 flux just works better. I've never found the 951 to be very effective. It's better than nothing, but the 186 is much better at getting solder to wet nicely.
That said, I like to clean my boards anyway , so there is no real downside to using mildly activated flux for me. If you would prefer not to wash your boards, then No Clean might be a better option.
I don't usually like no clean in general, with the exception of boards that go through reflow ovens and cannot be washed. But that is pretty rare and shouldn't apply to guitar pedals.
I see. So if I could score some 70-90% IPA, would that do for cleaning 186?
TBH, I still cringe at the thought of running anything that has to do with electricity through water at any point. I guess I'm just weird that way... :o ::)
Quote from: muddyfox on November 22, 2013, 02:05:16 PM
I see. So if I could score some 70-90% IPA, would that do for cleaning 186?
TBH, I still cringe at the thought of running anything that has to do with electricity through water at any point. I guess I'm just weird that way... :o ::)
I wouldn't use 70%. I can get 91% here for about $1 a bottle. That works, but it will take a while to get all the residue off. I recommend using a soft toothbrush to help it.
Flux remover is better, but more expensive.
Hey whaddyaknow! I just asked around in some real strange places and it seems that I can get IPA
-1L bottle 99,5% IPA for $15 or
-1L bottle 98% IPA for $7
Would that get rid of 186 residue? And at 99,5%, does it even have time to do anything before evaporating? ;D
Quote from: muddyfox on November 22, 2013, 02:30:12 PM
Hey whaddyaknow! I just asked around in some real strange places and it seems that I can get IPA
-1L bottle 99,5% IPA for $15 or
-1L bottle 98% IPA for $7
Would that get rid of 186 residue? And at 99,5%, does it even have time to do anything before evaporating? ;D
Both of those will work fine. I have 99% pure alcohol at work and it works fine as well.
With any of the alcohols, it basically moves the rosin off of the board, so with enough scrubbing it will come clean.
Flux remover is a solvent that dissolves the rosin, so it takes less time, but as I mentioned it is more expensive.
If you don't have access to Flux remover (or want something less expensive), then the alcohols you mentioned are the way to go.
Thanks Pick!
I have 186 ordered from China, I'll just try 951 until it gets here... The school of hard knocks, nothing like it!
Agree with all of pick dropper's comments (and that pic at the top looks fun for sure).
I really like the water soluable fluxes but you really need to be sure to clean, and inside a day is best. Simply wonderful in a production environment with automated cleaners. But it will corrode over time if you don't clean it.
I use no clean as a hobby builder though, simply because I hate cleaning rosin, and water soluable flux wicks under things like crazy.
Btw No clean can cause problems with passives in high freq rf applications, but that would be a really rare thing in typical guitar applications, failing wireless transmitters.
Anyway, glad to see another person dipping their toe in the water!
Interestingly, I'm working with a board at work arhat is RF based and is requires No Clean solder.
The instructions for the board house are "No Clean/ No Wash." In that case , the cleaning process causes more issues than it fixes, which, as you know, is not usually the case.
Quote from: muddyfox on November 22, 2013, 02:05:16 PM
I see. So if I could score some 70-90% IPA, would that do for cleaning 186?
TBH, I still cringe at the thought of running anything that has to do with electricity through water at any point. I guess I'm just weird that way... :o ::)
Most packages (opamps, resistors, diodes, IC's, transistors) are hermetically sealed and therefore waterproof. No concerns there. Pots and switches are another story as they often have openings where the inner workings can be exposed to water...
Quote from: rullywowr on November 24, 2013, 05:21:38 AM
Quote from: muddyfox on November 22, 2013, 02:05:16 PM
I see. So if I could score some 70-90% IPA, would that do for cleaning 186?
TBH, I still cringe at the thought of running anything that has to do with electricity through water at any point. I guess I'm just weird that way... :o ::)
Most packages (opamps, resistors, diodes, IC's, transistors) are hermetically sealed and therefore waterproof. No concerns there. Pots and switches are another story as they often have openings where the inner workings can be exposed to water...
Spot on. I'd would also add trim pots and IC sockets to that list. Sometimes people forget about IC sockets, but getting flux into them can cause problems.
If I have a board with those parts, I usually clean it before installing those, then I add them.
Rf is usually ok with no clean, as much as you can generalize rf, but high frequency rf is where the problems are. Might be that they dialed it in with the residue in place, and if automated should be consistent.
Most parts are usually fine for water soluable fluxes but there are some ic packages that had trouble, and they were installed after. Most parts were not a problem. Automated equipment had driers built in, so it was like magic, put your boards in the basket and daisy fresh at the other side. ;D
Quote from: wgc on November 25, 2013, 03:16:38 AM
Rf is usually ok with no clean, as much as you can generalize rf, but high frequency rf is where the problems are. Might be that they dialed it in with the residue in place, and if automated should be consistent.
Most parts are usually fine for water soluable fluxes but there are some ic packages that had trouble, and they were installed after. Most parts were not a problem. Automated equipment had driers built in, so it was like magic, put your boards in the basket and daisy fresh at the other side. ;D
The board I am dealing with is 2.4GHz. It is sensitive to an awful lot of things. :-)
Well, if you have issues with resistors "changing" values during use, try cleaning the flux off. Might be someone trying to save some $$ at the board house by spec'ing no clean and not knowing any better. Good luck!
Quote from: wgc on November 25, 2013, 12:54:00 PM
Well, if you have issues with resistors "changing" values during use, try cleaning the flux off. Might be someone trying to save some $$ at the board house by spec'ing no clean and not knowing any better. Good luck!
Thanks. :-)
In this case, it was the circuit designer that specified No Clean / No Wash. Despite the instructions, the board house kept washing the boards; current habits die hard, apparently.
We finally got them sorted out.
Hey guys,
as I was finishing up my octaver last night, I scrubbed my board with a soft toothbrush and isopropanol (>98%). That did the trick and removed any solder flux on my board. Unfortunately it also started to dissolve the plastic of the pcb!! Now it looks crappier than before. At least I realised in time.
However, the lesson to be learned here: Go and buy some flux remover instead of using isopropanol.
I should run a mass spec of commercial flux remover, I really want to know which solvent it is, since iPrOH is a rather unaggressive solvent and a soft cleaning agent. Gotta find a cheap way around it with something in the lab ;)
Interesting. Was it a FR-4 board?
I tried some 99% IPA on my smd build and while it didn't remove 100% of the flux residue (didn't even use flux pen, only what's inside the solder wire) it did start eating away on SS12 diode markings. With them being chinese I'm not really surprised but kinda makes you wonder...
Quote from: m-Kresol on December 15, 2013, 08:43:47 PM
Hey guys,
as I was finishing up my octaver last night, I scrubbed my board with a soft toothbrush and isopropanol (>98%). That did the trick and removed any solder flux on my board. Unfortunately it also started to dissolve the plastic of the pcb!! Now it looks crappier than before. At least I realised in time.
However, the lesson to be learned here: Go and buy some flux remover instead of using isopropanol.
I should run a mass spec of commercial flux remover, I really want to know which solvent it is, since iPrOH is a rather unaggressive solvent and a soft cleaning agent. Gotta find a cheap way around it with something in the lab ;)
Hmm, that's interesting. How stiff of a toothbrush was it?
I prefer Flux Remover, but I use isopropanol all the time and generally don't have that issue. Was this an etched board or a fabbed board? And do you know where it was fabbed? I have seen some boards with really poor quality solder masks.
Quote from: m-Kresol on December 15, 2013, 08:43:47 PM
...
However, the lesson to be learned here: Go and buy some flux remover instead of using isopropanol.
I should run a mass spec of commercial flux remover, I really want to know which solvent it is, since iPrOH is a rather unaggressive solvent and a soft cleaning agent. Gotta find a cheap way around it with something in the lab ;)
The MSDS sheets for these should let you know what's in them. I just grabbed the sheets from the manufacturer's websites for the first couple that came up on Mouser. The links are below. They are pretty different from each other, so it would seem it depends on which one you are using.
http://www.techspray.com/descriptions/document/1631_us_eng_msds.pdf (http://www.techspray.com/descriptions/document/1631_us_eng_msds.pdf)
http://www.mgchemicals.com/downloads/msds/english/sds-4140-l.pdf (http://www.mgchemicals.com/downloads/msds/english/sds-4140-l.pdf)
I like to use no-clean flux and the AIM no clean glow core solders. The only time I worry about it is if I've needed to use some solder paste for an SMD part and then I just clean the area with a cotton swab dipped in vodka -- works pretty well and I haven't had any issues.
I agree. Since I switched to no clean flux pen and no clean solder (Kester 251) I haven't had much trouble. I have also had good results using non-chlorinated brake cleaner as an alternative to expensive flux remover. I would suggest testing on other junk parts to ensure this won't melt your PCB or components.
Kester 251? I only see 245 and 275 as no-clean?
Oops, I didn't mean solder paste -- I haven't tried that yet. I meant flux paste, which I have a syringe full of and it helps sometimes with annoying parts that want to move too much (like every time I move). It's no clean too, but there is usually just too much of it left to leave on the board. So, I do the vodka thing with that.
I ended up needing to use some vodka yesterday to clean up a little breakout (SchmartBoard) board for an SC-70 6 pin part. The thing is so small that it wanted to fly away when I brought the tweezers close to it. So, I had to use the flux paste to keep it in place.
I use a flux cored 60/40 solder. It's 0.7mm so need to be careful with how much you use but it works great!
I put a small amout of solder on a pad, place the component and heat the solder and component. Then just solder the other side as normal. There is a good youtube video with this technique, I'll post it later!
Sent from my thumbs using Tapatalk!
Quote from: muddyfox on December 16, 2013, 09:33:01 AM
Kester 251? I only see 245 and 275 as no-clean?
Yep. Kester 50/251 in a 63/73 Pb combination 0.031" diameter I got off eBay for $20 shipped. It is about 5 years old (and subsequently past it's certified 'fresh' date) but works great. I recently switched from the old standby "44' (yellow labeled rosin) to this. I also got a Kester 952 no clean flux pen from Mouser to go with it. I took my old 44 solder and rosin pen and put them in a bag for a rainy day to avoid grabbing the sticky stuff by accident.
What I found about "no-clean"
1) I like the way it doesn't leave a gooey mess on the board
2) The aroma isn't quite as delicious as the old 44 :o
3) Takes a little bit more even heating and fluxing the board helps with this to get a nice shiny joint
Overall, I dig the "no-clean" stuff. Makes a great finished joint and no worries about the gooey mess left over after. It's really a misnomer..."no-clean" because many PCB assembly houses actually still do clean the "no-clean" solder - ironic isn't it? I like the fact it is just a clear residue and doesn't get all sticky.
Quote from: pickdropper on December 15, 2013, 09:35:01 PM
Hmm, that's interesting. How stiff of a toothbrush was it?
I prefer Flux Remover, but I use isopropanol all the time and generally don't have that issue. Was this an etched board or a fabbed board? And do you know where it was fabbed? I have seen some boards with really poor quality solder masks.
It was a medium toothbrush and Madbean's Low Rider PCB, so I don't think that there should be a quality issue with the board itself. I was quite surprised by the effect the alcohol had on the board.
Quote from: m-Kresol on December 16, 2013, 03:44:52 PM
Quote from: pickdropper on December 15, 2013, 09:35:01 PM
Hmm, that's interesting. How stiff of a toothbrush was it?
I prefer Flux Remover, but I use isopropanol all the time and generally don't have that issue. Was this an etched board or a fabbed board? And do you know where it was fabbed? I have seen some boards with really poor quality solder masks.
It was a medium toothbrush and Madbean's Low Rider PCB, so I don't think that there should be a quality issue with the board itself. I was quite surprised by the effect the alcohol had on the board.
Yeah, me too. I cleaned some boards with flux remover and alcohol last night with no issues.
Maybe that particular batch if boards had an issue. It does happen every now and again.
Do you have a scrap board you could try scrubbing with alcohol, just to see if you get the same result?