Electronics > Beginners

Trying to solder with a cheapo hot air gun.

<< < (5/6) > >>

KL27x:

--- Quote ---Get a good pair of tweezers, some paste flux (liquid works, but spreads too far and dries up too quickly for my taste), and a small chisel tip. Tin one of the pads, put the part down, touch the pad you tinned for a second. Part is now attached to the board. Add solder to other pad. If the first pad needs a little more solder, you can add it now. Otherwise, you're done. All of this before your hot air station warms up.
--- End quote ---
This is a good strategy to teach to a beginner. But it's not very good for doing a large number of components.

The reason most people initially prefer a fine pointy tip, like a chisel tip, is because they are afraid of unintentionally touching surrounding board/pads/components. Previous experience with thru hole and solder wire has suggested that this bad, because it leaves bridges and whatnot. Hand in hand with that, they might prefer a gel flux that only goes where they intend the tip to touch. And along with the pointy tip, they find it works a lot better to feed solder wire rather than paint the solder on, because that tip can't hold a solder bead in a usable fashion. The end result is a dance between picking up and putting down iron, tweezers, solder wire, +- pickup tool. In particular, the tweezers and solder wire dilemna is a major problem. If we had 3 arms, this might be the bees knees. But since we don't, the result is highly inefficient. This is all due to the initial starting point, which was chosen due to constraints believed necessary, or at least beneficial, at the time. If you are soldering like this, you are running a self-imposed high precision gauntlet that resembles a game of Operation, when it is not actually necessary. Wave or jet soldering can solder 95% of what you are doing, and the solder touches everywhere.

When you lose your (mostly unnecessary) fear of messing up the surrounding board you may find that there are much larger tips that will get the job done faster and easier. Hand in hand with using these larger tips is spreading the flux around a larger area. You are intentionally putting flux where you don't intend to touch the iron (but will, sometimes). Then it makes most of the surrounding area safe for the larger tip to touch without leaving dry solder strands/blobs. You now only have to worry about avoiding contact with unpopulated pads. Even briefly touching and reflowing pre-existing joints mostly won't matter too much. If they are in the path, you might just go right through them, intentionally reflowing them. That said, there is a hierarchy of components which bridge easier, and you have to take that into account.

The main problem with fluxing a large area is that the flux will tarnish unpopulated pads. It's not too big a deal for component pads, because when they eventually get populated, the flux will eat right through the tarnish. But you have to avoid fluxing ICSP pads and other contacts/fingers, unless you are going to wash the board, after.

With the right techniques, it can be faster and more reliable to hand solder boards than to even use stencil paste and oven (without a PnP machine, anyhow). But a lot  of us are spending 90% time and effort designing the circuitry and pcb, and are only soldering a few boards. Or we are purchasing a single board from someone else's design, and there is no room for error. It's hard to lose the fear, when the ratio of time and effort is so skewed. The PCB is the result of a relatively large bucket of sweat and tears plus money and wait time. It's no big deal to spend more time and effort in the final step that brings it all together. And so the baby elephant grows up and never realizes he can break the rope.

But the beauty of electronics, in general, and PCB's in particular, is the ease and cheapness of which a complicated thing can be replicated, more or less exactly. The Victorian age produced amazing machines to drive complex mechanical algorithms... but to replicate them required tons of labor and hand-tuning of gears and ports and moving parts. A PCB just needs the connections to be made. Furthermore, the manufacturers of SMD components have put a lot of science and experience into the design and spacing of pins/pads to make this easier, by taking advantage of things which you may not yet be realizing in your soldering technique. "Breaking the rope" opens up an area of small quantity production which would be unreasonably expensive to outsource. And a lot of these tricks will be useful in prototyping and reworking.

sourcecharge:
Well, I just got done, soldering the ucurrent boards.

I used that cs hot air solderer, and that solder paste.

Everything went great, and I found that using the syringe to put the paste on the pads really wasn't controlling the amount of paste on them.

So, I laid down a line of paste on the table, and used the exact-o knife to "cut" the right amount of paste for the small teeny tiny itty bitty evil pads which turns out to be about half of what the syringe can blob out.  Of course this is just a subjective observation, and I don't know exactly how much that should be use, but it wasn't much.

After placing the parts, I used hot air solderer at about 300C with a low air flow.

I worked like a charm, except for 1 or 2 small 100k resistors standing up but I was able to get them back in to place while continuing the hot air flow.

After the pads had the paste, and the parts were in place, the whole board was soldered in less than about 2 mins give or take a min.

I really like that method, it was really easy, just kept moving the nozzle around about 1/2'' away from the board, and it was like magic.

Thank you Dave for the great design and easy to make ucurrent.

PS

After I got the boards, I realized that I should have asked for at least the cost of a paste stencil from the pcb manufacturer.  When I was first trying the paste on the practice board, I noticed the amount of paste was too much, so I was thinking of making a 3D printed paste stencil.  After loading it in cura, I realized that I would have to use the 0.2 mm nozzle which would take 2 hours to install, while 1 hour to print.  And that's when I thought, that I could just use the knife, which turned out to be just fine.  Anyways, all I did to make the stencil in cura (the slicer program that makes the 3D printer run), was to take a screen shot of the gerber files with the edges and the pads, copied them to a paint file and colored everything black and white.  Cura can make 3D images of 2D pictures, and it can size it proportionally to the correct dimensions.  I actually don't have any use for it, but if anyone wanted to, they could just use that method.

Thanks again.

KL27x:
A full sized framed stainless steel stencil runs like 150-200.00, IME. I'm sure some manufacturers have cheaper offerings without the frame, or even mylar.

I'm surprised no one mentioned it, but there the 858D hot air station has a huge user base because it consistently works as it's supposed to. And it costs less than the thing you linked. If you want to do SMD rework on a budget, I'd highly recommend it. I shelved a $200 hot air station in favor of the 858D. It works good enough and has lasted tons longer than what the more expensive station got before its first heater died. I don't even care about replacement heaters. When the 858D dies, I'll just buy another one. 

janoc:
I have the Atten 858D but it is not that great station, even if you get the original and not one of the millions of poorly made knockoffs where some even come with live mains on nozzle ...

It is OK for small things  but if you are getting it hoping to rework larger chips, parts with large thermal pads, BGAs and multilayer boards, you are going to swear a lot. It just doesn't have enough power for that task. The Quick 861DW Dave reviewed recently is much better and while more expensive, it is worth it, IMO.

jmelson:

--- Quote from: janoc on July 21, 2018, 01:13:28 pm ---Hot air method is certainly not more "professional". Reflow soldering isn't done with hot air but in an oven.

--- End quote ---
Yes, I hacked a GE toaster oven to do reflow.  It has 4 heating elements, horizontally.  2 above, 2 below the rack.
I got a ramp&soak thermocouple controller.  I found the best control of temperature was by poking the thermocouple into a plated through hole in the board.

I've done several thousand boards with this.

Jon

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod