SparkFun Forums 

Where electronics enthusiasts find answers.

Questions relating to designing PCBs
By Philba
#30626
the float, in conjunction with a solder mask makes for automatic alignment of parts. especially for ICs. This is an issue for home made boards with out a solder mask.
By propellanttech
#30629
Peter,

Yes you are correct...tombstoning is the process that a capacitor, resistor, or other similar component either raises to one end, or rolls onto one side (although this is very rare....the side roll).

This is usually from them smt pad being larger than the component soldering location. The surface tension of the molten solder will pull harder on one end than the other......resulting in a tombstoned part. This usually happens from more solder than required to attach the component to the board. A smaller pad would require less solder....and the surface (smaller quantity) tension of the molted solder doesn't lift the part.

This is not a total explaination....because a larger foot print usually extends past the component end(most hand solder pads are longer....but not much wider)....this added area is leverage for the molten solder to help raise the component as well.

You should know that when hot air soldering.....turn the air volume down as low as you can get away with. The air flow will provide an addition lifting sorce for the parts. This could result in some very frustrating moments. I have done some rework to my boards.....and have found out the hard way.

I can't say about the footprints.....I was asking to know if there were different ones for reflow or for hand soldering.

I'm batch reflowing....so I've not had any tombstoning while reflowing total boards.

I'm also very careful about placement.....make sure your parts are pushed down into the solder.......don't go crazy....but make sure they are in contact with all their landings. This helps prevent one side lifting. (But this is theoretical....and my opinion)

James L
User avatar
By bigglez
#30631
Philba wrote:the float, in conjunction with a solder mask makes for automatic alignment of parts. especially for ICs. This is an issue for home made boards with out a solder mask.
Greetings Philba,

Ah yes! I have seen the float when the paste converts. I'm "dotting" the pads with a hand syringe, so I'm not able to control the volume of paste. (I hope to add a pneumatic dispenser in the near future). For good soldering (i.e. no opens or shorts) it takes a very small amount of paste per pad!

If I use an iron, even a very small tip, the iron removes a little solder and can leave a spike. With hot air (and presumably also with IR reflow) the heat is by non-contact and the solder "puddles" around the contact and pad - assuming that a flux provides good surface wetting.

Comments Welcome!
User avatar
By bigglez
#30632
propellanttech wrote: This is usually from them smt pad being larger than the component soldering location. The surface tension of the molten solder will pull harder on one end than the other......resulting in a tombstoned part. This usually happens from more solder than required to attach the component to the board. A smaller pad would require less solder....and the surface (smaller quantity) tension of the molted solder doesn't lift the part.
Greetings James,

Good explanation. So to get controlled paste on the pad (and assuming the pad is the right size) we use a stencil?
propellanttech wrote: You should know that when hot air soldering.....turn the air volume down as low as you can get away with. The air flow will provide an addition lifting sorce for the parts. This could result in some very frustrating moments. I have done some rework to my boards.....and have found out the hard way.
My equipment is primative - a Weller 6966C heat gun that I bought (used) for applying heat shrink. I recently discovered it can melt solder, and use it by hand in small circles around three or four 1206 parts to do the conversion of paste solder. So far I've only used 1206 size, some other parts as noted earlier, and some SOIC28 or TQFP32 devices.

By manipulating the height of the nossel by hand I can direct the air down on the parts, and as you pointed out, the side blast can move the adjacent parts off their centres.
propellanttech wrote: I'm also very careful about placement.....make sure your parts are pushed down into the solder.......don't go crazy....but make sure they are in contact with all their landings. This helps prevent one side lifting. (But this is theoretical....and my opinion)
I'm dotting all the pads with paste on one side of the PCB, then placing the parts with tweezers. I try to get them all lined up and seated on the pads. I was tacking pin 1 of the ICs but now I feel this is not needed (the entire IC can pull to the pad centres if solder on all pads is molten).

Thanks for sharing your experiences! I would never have imagined doing this level of assembly at home. The results have been very good so far, although I seldom make more than three or four of each design and can't readily justify tooling or sub-contractor involvement.

Comments Welcome!
By propellanttech
#30645
Peter,

I use stencils exclusively......I always have a huge number of parts on a board......and those boards will eventually be produced in large quantities.(by me)

I've not tried to place the solder by hand. I'm looking at a pick and place now....and will probably have one by the end of the year. I plan to have a dispenser on it....to do away with stencils all together.

I will probably offer assembly services for surface mount boards.

I'm not planning on wave though....so any through hole parts will be hand soldered.

At the moment I'm using plastic stencils....which tend to be thicker than needed....so more solder is being placed than needed...but it still works very well. I have also gone to the vacuum method of holding the stencil in place.....which was a stroke of genius from someone else on the forum.

I actually heavily modified a very nice toaster oven for reflow. I hated to do the modification, it was a very nice toaster.

But it works perfectly. My boards come out perfect everytime. It even has a fan inside for hot air circulation. (hot air convection oven)

It has taken me some time...but I really have the method down to a science. I did have one board that was over "cooked" when I started. The traces change color slightly. The board worked ok....but you could tell it was just slightly raised too high in temp (or time).

I will warn you about the heat gun method. Becareful....heating just part of the board can cause it to bow or warp. This is not ususally a problem....but can be for some projects.

James L
By Philba
#30651
propellanttech wrote:...
I will warn you about the heat gun method. Becareful....heating just part of the board can cause it to bow or warp. This is not ususally a problem....but can be for some projects.
If by heat gun you mean the cheapo $30 ones, I totally agree. They seem to have 2 settings - nuke and thermo-nuke. However, a hot air station with regulated and controllable heat and flow works really well. The only thing tricky for me was learning to do a pre-heat pass. Once I figured that out, it was really easy and fast. I'm happy with my hot air station.
By propellanttech
#30654
Philba,

That is exactly what I speaking of......I actually have a warming plate I put my board on to do rework. The warming plate heats the whole board to about 125.....and I use my rework station to heat the component area enough to replace the component.

If you only heat the one area...the board most of the time will "draw" and warp.

It can be done with the cheapo guns.....but I'm not brave enough to try it. I just ordered me a rework station.....will pay for itself real fast.(no board to replace)

I got an Aoyue(rework station)....and it's been great.....not a bad price either. Not the fanciest on the market....but it's accurate...and never lets me down.

James L