Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
THT connector solder thieves

THT connector solder thieves

Subscribe to Thread

I need some input on what good solder thieves dimensions sho... - Feb 12, 2003 by D Wanzek

Hi Good Luck, I worked a long time with THT components li... - Feb 12, 2003 by

I agree with DDave, You have something else wrong! .100 dip... - Feb 12, 2003 by RDR

Tell us more about the location of the bridging [ie, front-t... - Feb 13, 2003 by davef

Bridges are front to back. To me this seems to indicate thie... - Feb 13, 2003 by D Wanzek

Given that you are bridging front to back, contast the situa... - Feb 13, 2003 by davef

D Wanzek

#23357

THT connector solder thieves | 12 February, 2003

I need some input on what good solder thieves dimensions should be. I am limited to solving my bridging problem with solder thieves; all process/wave variables have been adjusted to try to solve this problem. There are several of these connectors on this assembly with parallel and perpendicular orientations.

The PCB is standard FR4, 0.063" thick, HASL, SMT with THT and solder mask. The connectors are 0.100" center to center double rows 16-26 leads. The connectors traveling perpendicular to the board direction are causing the most problems. The leads protrude 0.050" though the board. The annular ring is 0.060" diameter on each hole. Hole size is 0.036".

We are planning 0.180" x 0.060" oval thieves. The copper to copper of the thieves to the connector annular rings is 0.020" with 0.010" mask keep out around all pads.

We have tried a couple variations of thieves with minimal success, none of the previous thieve designs were nearly as long as this. This is about as long as the design will allow. There seems to be little documented research on this issue.

Any input is appreciated.

reply »

MA/NY DDave

#23358

THT connector solder thieves | 12 February, 2003

Hi Good Luck,

I worked a long time with THT components like you discuss and boards totally or partially populated.

Dee Machines, Electroverts, Hollis and an other and never had the kind of problems you are listing stayed with us forever. I mean you do have problems from time to time.

I don't think I can help with your request.

I would suggest you bring in a consultant in your area and they might just help you perform some experiments, maybe a flux change and other that will remove this problem.

YiEng DDave

This message was posted via the Electronics Forum @

reply »

RDR

#23361

THT connector solder thieves | 12 February, 2003

I agree with DDave, You have something else wrong! .100 dips with .05" lead protrusion should not bridge (in fact I would think that it would be almost impossible). Tell us about your machine, (laminar flow, A-wave, etc.., size of nozzle, wave height, etc...) flux type, and manuf., flux application method, preheat temp. solder dwell time, and immersion depth for starters.

To your question: Solder thieves are normally (in my case) a duplicate of the pad that is being thieved. My theory is to have the solder bridge located where it is of no concern from a functional standpoint.

Russ

reply »

davef

#23370

THT connector solder thieves | 13 February, 2003

Tell us more about the location of the bridging [ie, front-to-back, back-to-back, front-to-front, etc].

Thoughts are: * Your annular ring around the hole seems large. You have 0.007", consider 0.003" to be the minimum. Your board fabricator can give you advice. * On your oval thiefs, consider tear drops on the trailing edge pads. * On your oval thiefs, consider an additional pad located behind the trailing edge pads.

You're correct. There is very little information on this. Look for advice from: * BIG connector fabricators * Gray beard application types at wave soldering machine suppliers.

reply »

D Wanzek

#23376

THT connector solder thieves | 13 February, 2003

Bridges are front to back. To me this seems to indicate thieves would pull the solder away from the pins. I feel I am correct in that theory.

Yes, it would be a good idea for us to bring in a consultant but I have the opportunity right now to make board design changes to fix this problem. I am trying to give the best possible thieve design so this may be resolved once and for all.

The annular ring is actually 0.012". Reducing this more than we already have is some good input, thanks. I will see if design is willing to do this. The design is due to be released tomorrow.

As far as our process, that is more difficult to explain and a consultant would definitely be worthwhile. We have many limitations due to how business is done... The flux we use is fixed, operators could definitely use more training, product mix is complex (contract manufacturing), wave variables adjustments are limited to dynamics of the wave itself and the profile is good. We have played with all the variables on the wave with minimal results.

Just hoping this thieve design will offer better relief...

reply »

davef

#23378

THT connector solder thieves | 13 February, 2003

Given that you are bridging front to back, contast the situation with the thiefs used to prevent bridging when wave soldering SOIC. If this is reasonable, the thiefs should be located behind the trailing connector pins.

You can prove-out this [and any other theories that you want all at the same time] by gluing pieces of copper foil on the board behind portions of the connector and soldering the test.

reply »