Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


No-Lead solder defect - No solder on pads

Marc Simmel

#26583

No-Lead solder defect - No solder on pads | 12 December, 2003

I have encountered problems with a 90/10 tin-lead plated SMT component wicking all the solder paste off the pads on the substrate. Worst, the defect occurs randomly - adjacent leads may have 'good' joints, though heel fillets are poor. The solder paste is a SnAgCu alloy, and the copper substrate pads are coated with OSP (Entech 106). Co-planarity of the component leads measure within 0.10mm. Identical components with full tin-plated leads or with bright-plated leads do not improve the situation. The only experiment that has worked oxidized the leads before screening and soldering perfectly. We are now trying to adjust the temperature profiles to reduce the heating of the component lead and/or increase the substrate temperature but these may only be 'band-aids'. Does anyone have any similar experiences with no-lead paste or can point to a root cause?

This message was posted via the Electronics Forum @

reply »

#26584

No-Lead solder defect - No solder on pads | 12 December, 2003

Here's two angles to think about. Solder goes to the component and not both the board and the component because: * Component lead is much hotter than the the pad on the board. [This probably doesn't apply in your case, because it doesn't repeat from board to board.] * Component lead is solderable and the pad on the board is not solderable. [This is the likely situation that you face, because either (or both) your board fabricator has poor process control or you are burning-up the pad solderability protection before the solder can take.]

Have you qualified Entech 106 at the higher temperatures necessary for your SAC?

Please reconsider your use of lead-free processes. They are worse for the environment that old-time leaded materials.

reply »

swagner

#26653

No-Lead solder defect - No solder on pads | 17 December, 2003

How long are the OSP boards out of there packaging? Is this a double sided smd process or single? Are the boards stored in nitrogen? Sounds like an osp issue to me!

reply »

Gabriele

#26656

No-Lead solder defect - No solder on pads | 17 December, 2003

Clean or No Clean Solder Paste

This message was posted via the Electronics Forum @

reply »

Marc Simmel

#26699

No-Lead solder defect - No solder on pads | 22 December, 2003

No clean solder paste. One further observation ... the parts have a metal shell and are 'floating' high on the paste. Though there is wetting on the metal shell, it doesn't seem to sink as far as it should and this could be lifting the soldertails from the pads too.

This message was posted via the Electronics Forum @

reply »

Marc Simmel

#26700

No-Lead solder defect - No solder on pads | 22 December, 2003

The boards are vacuum packed in quantities of 50 pieces and used within an hour or two of opening; no nitrogen storage. The boards are processed in a double-sided smd process.

This message was posted via the Electronics Forum @

reply »

#26701

No-Lead solder defect - No solder on pads | 22 December, 2003

Marc: Please be more explicit. * When you say, "parts have a metal shell"; are you talking about the component leads or the component body? What is the metal, what is it appearance, and what are it's implications? * When you say, "the parts have a metal shell and are 'floating' high on the paste", do you mean there is solder on both the pad and on the lead? [Er, what?] * When you say, "this could be lifting the soldertails from the pads too", what do you nean?

reply »

Dean

#26706

No-Lead solder defect - No solder on pads | 22 December, 2003

I have seen this affect on Immersion Tin and OSP. 70 to 90 % of the solder wicks to the lead and forms a "single" homogenous solder mass. This even though the pad was 100 % covered with paste...

Evaluated different solder paste...did find variations on a theme...not all paste is created equal! Can't give much away on this...you'll understand.

Currently not treating this as a defective situation. Wetting spread is poor with High Tin solders anyways...

Excellent evidence of wetting. Intermetallic was formed. toe & heel fillet & side fillet

I have seen this affect recently on Immersion Silver with Tin-Lead solder...can be simulated with poor anti-oxident in the silver bath. However, no evidence of de-wetting was noted

Cross section a sample board. You will see exactly what I'm talking about.

reply »

Marc Simmel

#26715

No-Lead solder defect - No solder on pads | 23 December, 2003

Details, details: 1) The metal shell is part of the component body. The base metal is austenitic stainless steel (304) that has been plated with 90/10 tin-lead over nickel (semi-bright). 2) The entire part (leads and shell) is elevated by the mass of solder underneath it, but to a greater extent than I had expected. The paste screened thickness is 5 mils; when defects are present after reflow, the solder thickness may measure over 4 mils. Parts without solder defects generally have 2 to 4 mils under them. The co-planarity of the component leads is 1 to 3 mils. Co-planarity also checks out after exposure to reflow heat. 3) The volume of solder under the shell is much greater than that of the leads. Though the solder wets onto the leads quite readily, the buoyancy of the shell may lift the leads before the solder wets to the OSP coated copper PCB pads. The major difficulty with this problem is the inconsistency from board to board within the same production lots.

This message was posted via the Electronics Forum @

reply »

#26719

No-Lead solder defect - No solder on pads | 23 December, 2003

A more active flux may help, but if the nickel underplate on the component body is corroded [and not soldering], if could require a VERY active flux.

Just an off-the-wall thought, are you sure that your heat recipe activates the flux properly? Slowing-down the conveyor for a couple of runs could help.

reply »

One stop service for all SMT and PCB needs

Manufacturing Software