Doing SMD work by dipping parts in solder paste?

Go To Last Post
27 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have no experience with SMD...
But one hurdle is the hassle with applying solderpaste to each spot.

So I thought: in stead of applying solderpaste with seringes or stencils, would it not be much easier to dip SMD parts in bath of solder paste?

It appears that it is done with BGA's
http://www.youtube.com/watch?v=ECBTjZPoki8

Example of dipping paste:
http://alpha.alent.com/~/media/F...

I have nowhere seen people doing SMD work with dipping. It seems so much simpeler (once you are able to make a controlled thin layer of paste).
So what is the problem?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Great idea... just got to get the flux dispenser...and the robotics equipment to handle components...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A hypodermic syringe is often used to apply a small amount of paste on the PCB pads. They are purchased in vet supplies at farm stores.

It all starts with a mental vision.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I switched to SMD for most projects a few years ago, but only recently tried solder paste instead of classic hand soldering with a soldering iron.

Boy, paste is expensive. I bought a little tiny jar, and then I was surprised when I opened it and even the little tiny jar was half empty when it was brand new!

I had envisioned dragging the micro's leads through the paste to get a little on each lead, then placing the micro on the PCB, and then heating the PCB.

Didn't work at all :(

The paste, at least the stuff I had, was too thick and I could not get it to put a small amount on the leads as I had originally envisioned.

I ended up using a syringe to put some past on the pads for the other components, and ended up hand soldering the chips as I had in the past.

That experiment was a bit of a flop.

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have this paste, http://dx.com/p/lodestar-solderi...

It is dirt cheap, and works fairly well, I do not have experience with other paste to compare though. Paste is too messy to deal with for prototyping work. I used to use paste + stencil + toaster oven but went back to fine tipped iron + solid solder. If you are part of an assembly line then paste would be quite a time saver, but if you just have 1 set of hands then there is no real time saved and you spend extra time cleaning up paste from your hands and work area.

Paste is too thick to apply precisely with a syringe so the best option is a stencil and I find that you have to spin a few stencils so that the perfect amount of paste is applied through the stencil.

Paste also tends to have a shelf life of about 1 year, after that it tends to separate, you can extend the life by putting in the fridge but perhaps it might get mistaken for peanut butter.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Toalan.....

Quote:
you spend extra time cleaning up paste from your hands and work area.

this worries me.....

One should never get paste solder on their hands. The tin balls are so small they can penetrate the skin. one should always use protective gloves when handling solder paste.

I could not find back the article I had on it, but be very careful in handling solder paste. I use rubber gloves that doctors use or paramedics (doc .. nice item to put up for sale here I guess... )

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ignoramus wrote:
Great idea... just got to get the flux dispenser...and the robotics equipment to handle components...

My idea is that a fluxdispencer is not needed anymore, just a plate with a layer of flux.
If you mean by "dispencer" the fluxing disk, yes that is needed of course. But it does not seem to be a very complicated part. And one could think of several do-it-yourself alternatives.

Last Edited: Wed. Apr 3, 2013 - 12:12 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KitCarlson wrote:
A hypodermic syringe is often used to apply a small amount of paste on the PCB pads. They are purchased in vet supplies at farm stores.

My idea is that the syringe is not needed. Just dip the part in a thin layer of flux.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:
I switched to SMD for most projects a few years ago, but only recently tried solder paste instead of classic hand soldering with a soldering iron.

Boy, paste is expensive. I bought a little tiny jar, and then I was surprised when I opened it and even the little tiny jar was half empty when it was brand new!

I had envisioned dragging the micro's leads through the paste to get a little on each lead, then placing the micro on the PCB, and then heating the PCB.

Didn't work at all :(

The paste, at least the stuff I had, was too thick and I could not get it to put a small amount on the leads as I had originally envisioned.

I ended up using a syringe to put some past on the pads for the other components, and ended up hand soldering the chips as I had in the past.

That experiment was a bit of a flop.

JC


I think you best understood what I mean. Sadly the experiment didn't work at all...
Could it be improved by a better (more fluid like) paste? The stuff in the YouTube video looks more like a thick paint. Look in de video at 1:30 where it pulls looong(!) threads of sticky flux...

If we could have some gripper/manipulator for the parts that works on (say 1:5 or 1:10) scale, then the method looks attractive to me.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Many SMD parts seem to be made with a thin coat of solder on them. I wonder if more solder is needed. Has anyone tried just slobbering on some flux and then cooking the board?

I don't know exactly what that coating is. It is particularly noticeable on the big bottom pad of QFN parts. Some QFN chips seem to have more of it that others. I think I could scrape it off with a knife.

One way to get solder on the pads is to drag solder them. This puts a coat of solder on the pads. Then all you need is some flux and heat. The annoying thing about this method is the coat of solder is not flat on top. It is slightly dome shaped. Small parts put on such pads tend to slide off at the slightest provocation.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Look at what a Toaster oven can do ! I've used this and it works fine as wine.

http://www.seattlerobotics.org/e...

1) Studio 4.18 build 716 (SP3)
2) WinAvr 20100110
3) PN, all on Doze XP... For Now
A) Avr Dragon ver. 1
B) Avr MKII ISP, 2009 model
C) MKII JTAGICE ver. 1

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

steve17 wrote:
I don't know exactly what that coating is. It is particularly noticeable on the big bottom pad of QFN parts. Some QFN chips seem to have more of it that others. I think I could scrape it off with a knife.
Don't want the solder on the bottom pad to be too thick; can be shaved with a hot tinned soldering iron tip.
Some QFN pads aren't wrap-around so drag soldering is iffy or improbable.
Soldering a QFN (Quad Flat No-Lead) Package by Hand by CuriousInventor.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

indianajones11 wrote:
Look at what a Toaster oven can do !
Here's a different tack:
Experimentation with Solder Paste and a Toaster Oven by CuriousInventor.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:
indianajones11 wrote:
Look at what a Toaster oven can do !
Here's a different tack:
Experimentation with Solder Paste and a Toaster Oven by CuriousInventor.
He also makes great videos on SMD soldering.

1) Studio 4.18 build 716 (SP3)
2) WinAvr 20100110
3) PN, all on Doze XP... For Now
A) Avr Dragon ver. 1
B) Avr MKII ISP, 2009 model
C) MKII JTAGICE ver. 1

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gdhospers wrote:
It appears that it is done with BGA's
http://www.youtube.com/watch?v=ECBTjZPoki8

If I'm not mistaken, it's just flux that's being applied in the video. BGA parts, as I understand it, come with small balls of solder already applied.

In my mind, the major difference between applying flux and applying paste is that it would be hard to apply too much flux. In contrast, it is very easy to apply too much paste. If you used a "dipping" application method, I think it would be very difficult to control the amount of solder paste that stuck to each pad.

Michael

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

indianajones11 wrote:
He also makes great videos on SMD soldering.
http://www.youtube.com/user/CuriousInventor/

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

crwper wrote:
If I'm not mistaken, it's just flux that's being applied in the video. BGA parts, as I understand it, come with small balls of solder already applied.

Yup. In Ball Grid Array parts, the "Ball" in question is a tiny solder ball. The solder is applied via a lead plating line as a thin hexagonal plate on top of each pad. The part is then reflowed were the lead plate melts and surface tension causes the lead to form a tiny, very precise "ball" on each of the solder pads.

I used to be a process/equipment engineer on the MRC tool set at Intel, which was part of the C4 (controlled collapse chip connect, aka: type of BGA) line at one of their fabs.

 

Clint

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Solder Paste Attempt #2
(Still trying to teach an old dog new tricks)

As noted above I recently did my first PCB with solder paste, and actually hand soldered the chips as I couldn't squirt out a small enough blob of solder paste from the syringe onto the micro's pads. ( I don't have a solder paste template.)

I used an ~ 20 year old frying pan to reflow the solder.

Having watched several of the above links on using solder paste I decided to try again. This time I just placed a small bead (ribbon) of solder paste all along the pads on each side of the chip, as one long ribbon of solder, and did not try to put solder paste on each of the chip's pads separately.

I wasn't sure how well this would work.

It worked great!

I did end up with a few solder bridges. On the first PCB in this batch I removed the bridges with copper braid solder wick very easily.

On the second PCB, which cooked longer as I was busy taking photos, it was harder to remove the solder bridge solder. I think this was because I had cooked off all of the flux.

The one photo clearly shows some "solder balls" adjacent to the caps. I just touched the soldering iron to these to remove them.

I am delighted this technique worked for me. Cleaning up a few solder bridges is much faster and easier than hand soldering each pin of the chips separately.

JC

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Nice result.

Note that even with using a paste stencil if you place components by hand there also will be shorts between pins sometimes. unless you place the components with a machine.

We have a stencil machine and small reflow oven and we also have to remove shorts. This is indeed much less work then having to solder a 100pin device on the board yourself.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jay,

Nice work.

What are the heavy black sections?

Cheers,

Ross

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The looks are not bad at all.
But don't you miss process control?

I also think you need some top heat to get better results. What about a glass on top of the pan?
It seems that regular glass is a very good IR mirror.
That way you get top heat from reflection and you can still see what is going on.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello meslomp, ross, and gdhospers,
...100 pin devices...
Yes, that was one of my motivating factors in trying to use paste. I made an Xmega development board several years ago, ~2009, 100 pins, (about 0.5 mm pin spacing, if I recall correctly), and have never forgotten the experience. I found it difficult to hand solder each pin. I've not used that chip since then just because I did not want to have to do that again until I absolutely had to.

This was obviously a small PCB, just a few components. I can see how with a larger board a template would be helpful, even if there were still a few solder bridges tocorrect.

Process control...
:) You should have seen me researching reflow heating profiles, and then starring at my 20 year of frying pan, with a big black temperature know, and a neon bulb to indicate when the heating element is on. In short, the Process Controls on this project, carefully measured on a 0-10 scale, would be about a Minus 20 !

That said I'll keep the glass cover in mind. Fortunately this isn't a mission critical project!

The Blacked out sections are just electronic marker overwrites on the PCB image to obscure the Project Name on the PCB. It is, however, an Atmel chip!

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Here's another Atmel chip (IIRC an AVR32 UC3A3 in a BGA package):
Our New Soldering Oven by Crypto Stick.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Wow! Nice looking board, and I am continually amazed at BGA packaging.

I've never tried to assemble a BGA device/PCB at home. I'm not sure my PCB layout skills are up to the task, much less my newly acquired reflow soldering skills!

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You might want to try to cut down the amount of paste you are using. You could EASILY use 1/2 the amount of paste and still get perfect joints, with a LOT less shorts. Of course, I am just as guilty of using too much as anybody! It doesn't matter that I KNOW I don't need as much as I end up putting on there, I just can't seem to help adding a little bit more, just in case!

 

Clint

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Clint,

Good point, and part of the learning curve. When one looks at the photo where it is just starting to reflow one can tell there is way too much paste being used.

I'll have to try using a smaller gauge needle to apply the paste next time.

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

From Ben Heckendorn and The Ben Heck Show (YouTube):

Solder Over: Transform Your Toaster Oven In To A Solder Reflow Oven
Disassembles then re-makes a North American model of a toaster oven (non-convection).
Starts with an Arduino Uno but the final is on a protoboard with a mega328.

Surface Mount Soldering Tutorial
Uses a minimal set of tools so looks difficult (doesn't have to be).
Rather than an impractical PCB used only for manual soldering qualification Ben used a Gabotronics Xprotolab Kit by AVR Freaks' ganzziani.

"Dare to be naïve." - Buckminster Fuller