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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

Non-wettig on chip cap.



Non-wettig on chip cap. | 6 December, 2001

Pls. help. I�ve problem with non-wetting defect at capacitor termination. I have faced this problem at the 2 kinds of termination.

One is chip capacitor 0603 that its outer termination surface composed with 100 Sn (Lead free) and another one is Stacked metallized polyester film chip capacitor that I now checking the outer surface termination.

This defect do not occurred on particular area or specified product. it�s only happen on this kind of component but all around board. The temperature on the termination were initially measured and there are at 215 Deg C � 220 Deg C. I cannot go any higher temperature because at the other component will exceeded 230 Deg C. After I increase peak temp (before is around 207-213), the failure rates at chip 0603 with 100 Sn surface finish has improved. At the one from ~4 % to 0.4 %. I thought if I apply more peak temp this defect will be decrease. Is it right?

Anyway, at film chip, failure rates hasn�t seem to decrease.(10% to 7%) At big package I found non-wetting, at small package I found tombstone.

Here is the details of my process parameter

a) Profile - ramp rate 0.8 Deg C/sec , Preheat time at 110/150 = 100 sec, time above 183 Deg C = 75 secs, time to 183 Deg C ~ 5 minutes, peak = min 215 � max 220 deg C.

b) Solder paste � I had tried 3 types 1) Delta 738S2 Water soluble 2) Delta 670S1 No-Clean 3) Indium SMQ-92 No-Clean

I would appreciate if any of you guys can help me with suggestion or what should I do ? Thanks very much for your helps


reply »

Scott Davies


Non-wettig on chip cap. | 7 December, 2001

I have used Multicore MP-100 and Heraeus F380 pastes for a while, and experienced very few dry joint problems.

May be worth checking if the pad size and/or spacing are to the correct specification for the components. Check the date codes on the component reels, in case they are really old - may have been sitting around gathering moisture or contamination.

Finally, what thickness stencil are you using, and have any aperture reductions been implemented?

reply »


Non-wettig on chip cap. | 7 December, 2001


I exeperience this problem before and I tried diferent type of solder pastes.The one that help my process is a Kester solder R596L OA this is a water soluble paste that i am very sure is going to help you.You can get the technical data at also make sure that when you use your solder it is at room temperature. Run a profile to your oven with two proves measuring the air to find out if your heaters are out of calibration and make sure to check the manufacture date of the components to see how long the have been in stock.

I wish you good luck and if you have any Questions please e-mail me back and I will be more than happy to help you.

reply »

Brian W


Non-wettig on chip cap. | 7 December, 2001

Before changing solder pastes or jumping through too many hoops, check the melting point of the 100 Sn (pure tin) connection. According to Kester, the melting point of 100Sn solder paste is 232 degrees C. If you do not reach that temperature, you will not wet the joint properly. From what you mentioned about other components, you may have an incompatability between component finishes.

Brian W.

reply »


Non-wettig on chip cap. | 7 December, 2001

We've experienced similar problems with film caps, specifically Panasonic's parts that are alledgedly compatible with both reflow and wave soldering. They are identified with either GB or JB p/n suffixes, which become either GX, GC, or JC for the reflow only (much better results) type parts. Call your supplier for the scoop, if that's what your using.

Of course if your using something other than Panasonic parts, excuse me for butting in. ;)

FWIW, increasing peak temps didn't help us much either with the GB parts.

And one more'll probably get better soldering with an OA paste, but you're not supposed to wash Panasonic's film caps with anything other than alcohol (no water) so unless you're running IPA in your board washer.......

reply »


Non-wettig on chip cap. | 7 December, 2001

I'm with Brian.


Welcome to the lead-free generation. ;-)

The Eurolanders are so blanking stupid, it just makes my head hurt. But then again, it doesn�t take much.

OK, enough pontification. Your profile is setup incorrectly for tin coated components.

For years, we have been fortunate enough to be able to use broad-brush generalities like �60 seconds above 183�C�. 60 seconds [nor your 75 seconds] above 183�C is not necessarily enough for some components, and certainly not with a higher melting solder [Sn100] like you have on your leads. In order to get reliable reflow, you need to be at or higher than liquidous + 20�C for about 5 to 10 seconds. At a peak reflow temperature of 215 - 220�C you are nowhere near that.

Look here for a hand wave on the issues of lead-free .


Check the fine SMTnet Archives for discussions about the causes and solutions to tomb-stoning.

The description of your other issues is vague.

reply »



Non-wettig on chip cap. | 10 December, 2001

Thanks for your advise Now I don't found Non-wetting defect at TIN COATED 0603 after I had increased peak temp and Time above 183 C. At film caps, I still found Non-wetting in high rate at film chip size 2416. I using PANASONIC's part with identified suffic JC. I havn't found Non-wetting at smaller size film cap (suffic JB and JX), but I found Tombstone instead (Solder paste Delta738S2 Water soluble). However I�ve experienced with non-wetting at film chip size 1206 suffic (Solder paste Indium SMQ92 No clean. So I would like to separate the problem at film chip to 2 issues and ask you again. 1. Non-wetting defect at film chip size 2416 suffic JC with Delta738S2. Now I apply additional water soluble flux (Lonco NF#3000) at terminal of part before reflow, failure rate is 0. Have more quantity of flux did heIp to remove oxide? Or what? I don�t know much about incompatability between component finish and flux. Pls. Advise. However I will try others part and other suffic to see the different. 2. Tombstone at film chip size 0805, I set time between 175- 183 (before solder become to molten) almost 80 sec. With slope < 0.1 deg/sec. (Another zone refer to my last post) Stencil thickness 6 mil. Aperture, Pad design and spacing is same as another 0805 size. The pad size is WxH 70x53 and gap 27 and aperture is reduction to 63x46 rectangular shape. The defect rate is not much but I wonder why Tombstone happened only on film caps. (No others parts was found and observed placement is OK.) I thought that the weight of part is 1 factor but I don�t know others reason. May something is from component finish?


reply »


Non-wettig on chip cap. | 10 December, 2001

1) The additional activity of the OA flux probably took care of the oxydation. How do you plan on taking care of the presence of conductive and corrosive residues from it now, though? You're not supposed to wash those parts with water.

2) The tombstoning is a most probably a function of the height of the parts, their mass, and their (questionable) solderability. They're tall, they're light, and they probably aren't wetting at the same time at each end, so they tombstone.

You're not using Panasonic's recommended pad layout if you're using the standard (IPC, I assume, or if not, what?) layout, and you need to to get rid of the tombstoning. It's in their spec.

reply »


Non-wettig on chip cap. | 10 December, 2001

Good for you!!!

Similar to Steve's comments ...

1. NON-WETTING DEFECT: Yes. Your Lonco is a �stand and deliver� flux. Not some limp wrested impersonation of a flux like most nc fluxes.

Without speculating about �incompatability�, recognize that your Lonco is an aqueous flux. Aqueous [water soluble] fluxes are more active than nc fluxes and in most cases, leave corrosive residues that prudent assemblers remove from the products they build.

2. TOMBSTONE: As I suggested in my earlier post on this thread, aside from solder balling, it�s unlikely there has been a topic with more rumination on SMTnet than tomb stoning. Please search the SMTnet Archives to get uo to speed.

reply »

Used SMT machines

ii-feed SMD Intelligent Feeder