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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Very fishy missing solder paste issue

Views: 2089

We have this problem for a long time. It happens to 3 new My... - Jul 15, 2005 by

seaK

#35571

Very fishy missing solder paste issue | 15 July, 2005

We have this problem for a long time. It happens to 3 new Mydata MY-9 machine. When pick and place resistor array 0408 (different value, top and bottom side), the solder paste in the middle two terminals missing occasionally. Thought it's because nozzle heads too big, solder paste is vacuumed, but no solder paste found on machine filter....

Clueless what happened, call Mydata technician, they came in and found no problem with machine.... Guess I could ask a little help here see anyone has good suggestion...

reply »

#35581

Very fishy missing solder paste issue | 15 July, 2005

Questions are: * Are you sure there was solder on the pad prior to placing the component? * Is there solder on the lead, but not on the pad?

reply »

#35604

Very fishy missing solder paste issue | 18 July, 2005

If it were other machines I would suggest the possibility of air kiss blowing the solder away, but mydata's have a release to make sure the air kiss never gets too high. Unless it is not pointing 90 degrees away from the machine like mydata recommends.

If the paste is there before placement, make sure you also check after placement before the oven. Strange things happen in the oven. After the oven check to see if any nearby joints have extra solder.

Once I saw a board misprinted by half an SOIC14 pitch, and after the oven everything was reflowed fine with solder where it was supposed to be except on the SOIC14. One pair of pads has no solder and the pads on the other end had twice as much solder. It was a fairly simple board. I noticed the solder missing from the pads after the oven then checked the boards before the oven and found the misprints.

Stephen

reply »

seaK

#35612

Very fishy missing solder paste issue | 18 July, 2005

Thanks for the response.....

The solder is actually printed okay prior to SMT.... And after placing, found missing solders at middle terminals of 0408 resistor arrays.... Package library is checked, component spec looks fine...

I am suspecting the air kiss scenario that Stephen suggested, but how come only 0408 resistor array....??? Anybody has same problem with Mydata MY-9 machine????

For pictures, you could find by clicking the link http://flash.atpinc.com/public

reply »

aj

#35620

Very fishy missing solder paste issue | 19 July, 2005

Hi,

I looked at the pictures and it looks as if it should be a straight forward solution..but obviously not.

I would focus on the stencil/print more than placement machine. The placement machine simply places the part on the board!#(just one thing to check, I dont know the mydata machines but on fuji there is a setting called z0 which is the position the nozzle touches the board +.3mm..verify this is ok) What is the stencil thickness you are using/pad design? There looks to be a lot of paste on the pads.

aj

reply »


Rob

#35622

Very fishy missing solder paste issue | 19 July, 2005

Have you tried a smaller nozzle? I'm wondering if the air is escaping through the cut outs on the device on the air kiss stage.

reply »

#35624

Very fishy missing solder paste issue | 19 July, 2005

We agree with Rob. It looks like either kiss is blowing the paste around. It could be: * Nozzle id too large. * Kiss is occuring after the head has moved from the component.

0408%20MS%2001.jpg - Paste is not missing on the bottom-center pads. It is on the bottom outside-bottom pads. Notice curious fuchsia color that's in "stipes" on raw solder and dots at the edge of some silk screen legend. There is some raw solder off the bottom pads.

0408%20Missing%20Solder%2003.jpg - Paste is not missing from the upper-right-center and lower-right-center pads. It is on top of the component and on the outside and left-center pads. Paste that originally was on the left-center pads is on the outside left pads. Curious fuchsia color on top of the component and left-center paste.

RN29-1.jpg - Paste is not missing on the top-center pads. It is on the top-bottom pads. Looks like the "kiss" either hit the two top-center pads or was funneled to these pads by the catellation between the pads.

RN31-1.jpg - same as RN29-1.jpg

Rn19-1.jpg - same as RN29-1.jpg

Rn19-2.jpg - same as RN29-1.jpg, but to a lesser degree

Rn20-1.jpg - same as RN29-1.jpg, but to a lesser degree for both RN20 and RN21

Rn21-1.jpg - same as RN29-1.jpg, in both RN20 and RN21 there is a circular pattern of missing paste. Evern the paste pushed up and off the pads fits the pattern. This is driving us the think the "kiss" is moving the paste.

Rn22-1.JPG - similar to Rn21-1.jpg, but not as distinct circular pattern.

Rn28-1.jpg - similar to Rn21-1.jpg, but not as distinct circular pattern in both RN28 and RN28.

reply »

#35626

Very fishy missing solder paste issue | 19 July, 2005

I have seen this problem on Mydata machines before. I am very sure that your missing paste is being caused by the placement machine.

Check these things :

1. There is a small white check valve on the pressure side of the vacuum pump. Make sure it is there. If it is, then check to see if the little ball is missing inside. This is most likely your problem.(This is what regulates "kiss off" pressure if I'm not mistaken) 2. Verify part height in package list. 3. Verify that Z sensor is clibrated and functioning properly in the service program.

Let me know how it turns out.

Regards, Sr.

reply »

CLampron

#35628

Very fishy missing solder paste issue | 19 July, 2005

seaK,

Check your PCB Thickness check in your program. See if it is checking the PCB at this location. Each board the machine will check to find the baseline thickness of each board run. It does this with a tool. I'm thinking this may be it because I have had the exact same problem in the past.

Good Luck

Chris

reply »

#35629

Very fishy missing solder paste issue | 19 July, 2005

CLampron,

There is no place for PCB thickness in a Mydata program. The board thickness is of no consequence.(As long as the thickness is uniform across the entire board) When you run a program, the Z axis touches the surface of the board to determine it's height. This is commonly known as the mount tool test position.

reply »

CLampron

#35630

Very fishy missing solder paste issue | 19 July, 2005

Sr tech,

It is in the PCB header information. F2 from the layout editor then F1 to view through the camera.

Regards

Chris

reply »

CLampron

#35631

Very fishy missing solder paste issue | 19 July, 2005

Sr. tech,

The mount tool test position is what I am refering to. The function of this is to determine the placement plane of the board. The mount plane will change if the board is thicker so thickness is relavent.

regards

Chris

reply »


RDR

#35633

Very fishy missing solder paste issue | 19 July, 2005

We have had this problem with one of our machine although it is not a Mydata. We had to reduce the blow off time for these. Our problem was also these same Rnet packages. the part is proabably picked a little off center, then the blow off hits the pads directly disturbing the paste. This looks to be your problem as all of the misplaced paste is on the same sid eof the component it appears. Use a smaller nozzle if you cannot adjust the air pressure or time with your machine.

Diffrent observation but I am curious if this is a good example of MyData accuracy? Everything is off center is it because of the machine or the placement program?

reply »

#35634

Very fishy missing solder paste issue | 19 July, 2005

How does the mount plane change if the surface of the board is located in the same place every time ?

For example: If you put 2 boards side by side on the Y wagon : one board measures .020" and the other .200" the surfaces of the boards are going to be at the same height due to the way they locate in the board grooves on the Y wagon.

Regardless of whether or not I'm right, the machine measures the board height each time.

At any rate, I don't feel that this is what is causing the problem. It is a blow off/kiss off issue.... most likely.

Regards, Sr.

reply »

#35644

Very fishy missing solder paste issue | 20 July, 2005

Hi,

What's really interesting about your photos is there is solder paste on top of the component.

If it was blowing the solder off the board you would see solder paste splatter away from the component.

But the solder paste is on top of the component, and the missing solder paste is in the middle of the component. So this indicates the nozzle is too big, and there is air running up into the tool between the pins, and it's sucking solder paste up and on top of the component. The nozzle would be largest in the middle of the component, because it's round, and that's the middle two pins where your loosing the paste.

You might also have the component height wrong, so it could be coming down too hard on the pad with the component, so that's worth checking, but it really looks like a over size tool.

Reduce the tool size down to one that's much smaller, so the diameter fits well within the middle of the part between the pins. Because the tool can pick from a wide area on the part, it needs to be smaller than just the size of the part itself.

I hope this helps.

Regards,

Grant

reply »

CLampron

#35660

Very fishy missing solder paste issue | 20 July, 2005

The mount plane is not the same every time. Our machines are conveyorized and the bottom rail is fixed so the mount plane does change with the board thickness. The stand alone table has a V notch for holding the board with the flat edge on the bottom. (See magnetic supports) so a thicker board will sit with a higher mount plane. I disagree that two boards of different thicknesses will have the same mount plane. The function of the mount tool test position is to measure where the mount plane is so the machine knows when to expect to come into contact with the board.

You are correct in that the machine measures the board height each time but why would it need to do this if the mount plane is the same no matter how thick the board is? Would it not come up with the same reading every time?

From the conditions mentioned, I would still bet on the mount tool test position or the vacuum verify being shut off for this component. The blow off is not strong enough to blow off the paste if there is a component on the tip. If it were, you could not possibly handle an 0201 component. If it is a wrong size tip, the vacuum verify would fail the component. Where there is solder paste showing up on top of components, this would indicate physical contact between the paste and the tool.

Just My $.02

Best of Luck

reply »

#35667

Very fishy missing solder paste issue | 20 July, 2005

Hi,

I think the comment about the board height is because the MYDATA checks the board height each time it loads a tool so this is eliminated from being an issue because the machine knows the PCB height. I don't think it's so much what height the PCB actually is, but that the machine has checked the height, so it's eliminated from being a cause of the problem because as board heights change, the machine tracks it.

I am still sticking to the idea the tool is too big. I know the vacuum should fail, but I wonder how good the sensor is, as I have head MYDATA machines pick components with a sucking sound when the tool was wrong for the component, and there was air leaking. I was initially alarmed when I heard it, but the operators said it always did it with that component and it was "normal"

Paste on top of the component could mean the tool is doing the board check somewhere on the PCB where there is paste. This is one setting you need to check on MYDATA's when programming the board, as I have seen the check being done anywhere, because we had forgotten about it.

But in this case there is also missing paste, and it's not splattered away from the component, it's actually on top of the component. So it looks like the paste is being sucked up to me.

Really interesting problem!

Regards,

Grant

reply »

Base

#35678

Very fishy missing solder paste issue | 21 July, 2005

Probably very far-fetched, but I can think of a scenario like this:

- machine picks up a component - between vision and placement the part is lost, but the loss is not detected due to oversize tool, which causes the amount of vacuum drop to go unnoticed (there was massive leakage to begin with) - head goes down to mount, but since there is no component anymore it dips into the paste, and/or sucks up some of the paste and/or blows it away again or spits it out again when applying the air-kiss - result is a gap in the solder paste deposit, some splattering around of solder paste near the original location of the deposits and in some cases some solder paste on top of the retried component that is placed after the initial failure

As said: very far-fetched (lively imagination...) but perhaps worthwhile a check.

... and they lived happily ever after... Hans Christian Andersen

reply »

#35686

Very fishy missing solder paste issue | 21 July, 2005

Hi,

Yes, that's an idea, but from my experience with MYDATA, it would fail test, and the machine would either know it's lost the component, or it would not know, and the component would not be on the pad because the machine would think it's mounted it, and not place it again.

So my moneys still on too large a nozzle letting air in around the pins. Because the nozzle can pick from a wide window on the part based on where the component is in the tape, and the feeder, you really need a tool that's well within the main body of the component, and that excuses the pin area, which is a lot smaller on this type of part. I would keep trying smaller tools in turn until the problem goes away, or the components start to get dropped because the tool is way too small.

I hope you let us know what fixes the problem as I am really interested in knowing!

Regards,

Grant

reply »

Scott B

#35692

Very fishy missing solder paste issue | 21 July, 2005

Have you tried running the machine in stepper mode (Ctrl+S). This will allow you to hold and view the machine at each stage of the placement cycle. This may help determine if it is false placement, vacuum or kiss-off. We find this useful when trying to debug oddball occurences.

reply »

Reflow Ovens thermal process improvement