Does anybody know how well a SMT component position is defined relative to PCB pads? We are using Hall sensor in SOT23 package. Considering that the sensor floats on solder, is there any way to make its location on the PCB repeatable down to 20 mils? Ten mils? Five?? Two??? X-axis? Y-axis?? Any insight would be greatly appreciated.
Hallo Michael,
Might want to consider a sacrifical holder that keeps it lined up until after the reflow.
-- Regards, Joerg http://www.analogconsultants.com
the reflow.
You should look at the accuracy of the Hall sensor as well.
It does not help to position a sensor to plus or minus 2 mills when the sensor is only accurate to plus of minus 20 mills.
If mechanical position is so critical try:
1 - use a through hole part. 2 - use an SMT package that has a reference stud or pin. 3 - route a hole in the PCB for the SOT23 package and mount it inverted. 4 - use high temp glue to hold the SOT23 where the pick and place puts it.
Not an option - no room, the part is selected
I am not sure the mechanical designer will be happy with it
This should work. Thanks!!!!
Keyser Soze wrote:
after the reflow.
As the other guy said, glue is probably the answer. Quite a lot of SMT boards (especially older ones) have the parts glued down, there must be an existing glue product (and dispensing system) for doing this. Chris
It isn't totally guaranteed, though; the package can soften, and the glue can soften, and the surface-tension forces of the cooling solder can be significant. Elastic glues shrink when heated.
Hall sensors are sensitive to stress, too, so having three solder pads pulling on it might give different performance than three solder pads and a glue dot.
One relatively safe approach would be to solder it up, then align the circuit board on the sensitive chip, before drilling the board for its mount/alignment pins (or are you doing that anyway?). On second thought, your fab line probably uses machine vision and printed targets on the PCB for alignment, thus the original question 'how closely aligned to the printed pads'.
And you need to be sure that there is enough solder on the pads to bridge the gaps that the glue will enforce.
In fact, when the SMT was floating on mollten solder, its position could be tightly defined by the surface tension in the molten solder. You would need good control of the volume of solder on each pad to get precise positioning out of this and you'd have to be careful about the geometry of your pads, but this could be a neat way to go.
-- Bill Sloman, Nijmegen (but in Sydney at the moment)
I don't know, but even 0.5mm pitch packages seem to end up in the right place most of the time....
-Lasse
Design the pads so that they are the same width as the leads on the IC - the solder surface tension will pull the leads to the middle of the pad during reflow! This will keep the theta under control, along with one axis.
Thanks a lot to everybody! I will do this: "Let it go" as is (no fancy tricks) and see how repeatable the location will be. My guts feeling is that surface tension will do the trick (I do not know how much difference lead free solder will make). If our mechanical guy finds that position variation is still too high we'll try the glue or positioning jig. Punching the hole in the board and pressing the part in (up side down) is the last resort as it will require mechanical changes beyond the pcb. :o)
Jeff L wrote:
If you make the pad pattern a little bigger than the pin pattern, so that the solder has to stretch sideways just a little to reach, you get much better self-centering.
Cheers,
Phil Hobbs
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.