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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Skewed components

Views: 5415

#48206

Skewed components | 6 March, 2007

hello

I would like to ask you for advise. I have problem with one of my SMT line and skewed parts at the end of line after reflow. My line is composed: DEK, 3d visual inspection of paste printing, IP3, 8 zones reflow oven. I m builidng there product composed of 3 the same transistors - package DPACK put it at substrate(main layer of substrate is aluminium) one next to the other and all of them are put at metal fixture. Im using Indium NC-SMQ 92J. I checked amount of solder paste at visual control, calibrated IP-3 and before reflow transistors looks fine but after reflow they are skewed. I tried to change profile from Ramp-Soak-Spike to the Ramp-to-Spike but still problem occurs. I also tried increase or decrease peak temp but still problem happend.What is ur opinion? Maybe there is somethin else? Maybe hot stream air is to strong somewhere? that is moving component? Pad for this component is quite big so self alignment parameter maybe critical? do u have any experience in this field? thanks in advance peter

reply »

GS

#48207

Skewed components | 6 March, 2007

It may be to much solder paste under components.

Try different paste doposit schape or thikness

Regards....GS

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

#48208

Skewed components | 6 March, 2007

DPAK align on the tab with leaded solders. They don't align much with unleaded solders, but we know that your Indium NC-SMQ 92J is leaded. So, there may be too much paste on the heat sink tab causing the component to float [or wim] during reflow. If the aperture for the tab is 1:1, consider reducing it to 50 percent.

For more, search the fine SMTnet Archives. For instance: http://www.smtnet.com/Forums/index.cfm?fuseaction=view_thread&CFApp=1&Thread_ID=6071&mc=15

reply »

#48218

Skewed components | 7 March, 2007

We window pane the heat sink pad to reduce the amount of paste for DPACK component

reply »

#48237

Skewed components | 8 March, 2007

Hello, For the most part the ground side width should be reduced, but more than likly you are well into a run of built pcbs? Profiles may or not help in cases of the dreaded DPACK. No issues on any other areas or components so I'd stay away from the oven. A quick fix is to window pane your aperture and start out with a small reduction on top of that. That will correct the scew, but now you will probably see a pull towards the ground. A slight bump in x/y cad placement towards the toe will usually land you a precise placement after reflow.

Kevin

reply »

#48251

Skewed components | 8 March, 2007

If you are in the middle of a run and can't get a replacement stencil there immediately, you can always put a little kapton tape on the bottom of the stencil to mask off a portion of the DPAK pad and reduce the paste deposit. We have had to do this from time to time when the stencil guy (me) makes a mistake.

This of course opens the door to having the tape come off during part of the run and cover some other apertures accidentally. It can work in a pinch, but keep an eye on it.

reply »

#48253

Skewed components | 8 March, 2007

To be honest this product its only a part of bigger PCB, so we have only substrate with 3 FETs on it so no other components at substrate occurs and all substrate are staying at metal fixture. WE have already window pan aperture. WE producing this product for few years and problem with skewed parts starts occuring 2 months ago and its growing up. Tomorrow I m going to do some test at other oven in our company to ensure that oven its not generating this defect and hot stream air doesnt blow up out transistor from substrates.Maybe this is issue.

reply »

#48291

Skewed components | 9 March, 2007

That also allows a channel for degassing and also breaks up paste into sections that tend to balance the forces acting on the part. So basically window paning is even better than you might think.

reply »

Mark

#48295

Skewed components | 9 March, 2007

hello,

try to redesign apperture:- cross-hatched with 0.5mm or 1.0mm web according to pad size

mark

reply »

siverts

#48300

Skewed components | 9 March, 2007

We have seen this issue with our RoHS process also. If that is the case and I if understand you correctlcy; then the advice from KSimpson, Stephen, DaveF + "all others" is totally correct!! We have been under this problem a numeruos numbers from time to times. One thing to try to do; which is not not so easy for everybody as us, since we do a lot of prototyping; is to have the layouter to change the pad layout (foot print) to better fit the actual compomponent to fit on it. Now here is what we have done for totally get rid of the problem once and for all:

1. Do the window pane and reduce the area to 65-75% 2. Use a 0.127mm stencil thickness (It will work on the product you describe). 3. If You have the oppurtunity to send "constructive" feedback to the customer of yours, whitout them going mad; then tell them to make the cooling pad less in length starting from the back. Of course tell them why! You got a problem and they will certainly be happy to know about it and that you are in interaction whith them to solve it. Remember if not medicine or military (assuming that your customer Not has a diiferent req.), the soldering and placement of that particular component is ok if it is whitin 75% of pad. 4. The reflow process is correct that you mention but as a result of this you ending up with skewed comp. but this is not the root course! It is just a result of the soldering process and the diffrent forces involved in that. Keep us in posted.

reply »

siverts

#48301

Skewed components | 9 March, 2007

I mean 75% on pad.

reply »

#48478

Skewed components | 19 March, 2007

Hello I made some test on different oven and problem still occurs. The only progress is that with diffrent thermoprofil ( Ramp to Spike instead of Ramp- Soak- Spike skewed components appears at 1 % of produced parts). With Ramp - Soak-Spike we have problems at 2 % produced parts.

Im wondering what kind of influence has design of stencil for this kind of surface. This dpack is almost rectangular pad 1cm x 1cm and we divided this pad on 4 equal surface. What is ur experience in size of this surface according to skew? ( Now we have for this component 4 surfaces each about 0,25x0,25 cm). Maybe if we divided in more fields this will give better score. What is ur opinion guys?

reply »

Software for SMT

Boundary Scan