paper vs emboss (& zig zag placement)

#67376

paper vs emboss (& zig zag placement) | 27 September, 2012

I notice that someone has set all the capacitors & resistors parts as emboss instead of paper tape. Does this make any difference to the pick and placement of the parts,

By the way recently we changed from 1206 to 0603 parts, coming out of the machine all the parts look well placed e.g. nice and parallel with the 0603 part next to it, but when the PCB comes out from the oven, they are all zig zag, could this be (I am only guessing) because the different standards between paper and emboss?

I just thought it would give our products a higher image if the parts were all parallel.

The Machine is Juki KE-750L

#67402

paper vs emboss (& zig zag placement) | 2 October, 2012

Difference in pick height from paper and embossed reel (typically) is 0.2 mm. Try to change your pick height and place height (different type of capacitors have a different height: for 100 nF 0603 and for 4.7 uF 0603 it's a big difference!). Can you show picture with part of your PCB with few SMT-pads and lines)?

JimmyB

#67405

paper vs emboss (& zig zag placement) | 2 October, 2012

What machine? Normally part height in the parts description would dictate placement pressure. In general, paper provides a better centered pick-up but may interfere with the pick-up(paper fibers)

#67407

paper vs emboss (& zig zag placement) | 2 October, 2012

OK I can take a hint when asked for a photo. I did not try to take a photo as I thought the close up image would put the picture out of focus, so this is where a bit of graphics works might help explain the problem.

When I had a look at the 1206 pcb it looked like Example 1

The 0603 had the pads closer together as in Example 2, Not only that the "extra width" was hardly noticeable on the 1206, but the 0603 had a bit more pad. So I guess this is what you jokers had in mind to help me answer my question

Attachments:

smt_twist.png

#67428

paper vs emboss (& zig zag placement) | 3 October, 2012

Attachments:

#67437

paper vs emboss (& zig zag placement) | 4 October, 2012

I have only one incredible idea: this is example of different speed of heat transfer of neighboring pads. I saw it two times during last years, but on lead solder pastes (Sn63Pb37). The reason of this mistakes each times we found two: 1. errors during designing of PCB - wrong pads size and with and direction of trace to pads; 2. wrong profile in reflow ovens (I talk about reflow oven with hor air and without nitrogen). We solved this problem by increasing time of pre-heating of PCB.

Hope, it can help you - or bring another idea to you. Good luck!

#67662

paper vs emboss (& zig zag placement) | 19 November, 2012

Thank you ENIAC for your comments

Here is another problem (JUKI KE-750L) Which applies to 0603 (1608 metric) resistors and only resistors (mind you the board is more heavily populated with resistors)

After the panel has gone through the oven,

When the resistor is placed at 0° (to supply reel) and the two pads will be x1,y1 x2,y1 some resistors at random will lift off the x1,y1 pad (only one resistor on most panels, but can be as much as three resistors now and again on other panels) It will never left off the the x2,y1 pad.

When the resistors are placed at 90° to the supply reel and the two pads might be x3,y3 x3,y4 the resistors will always lift off the x3,y4 pad, never the x3,y3 pad.

(The supply reels have nothing to do with the problem, I just put it in the reference at pad locations.)

The panel has four circuits in a two by two matrix, The problem never comes from the first circuit, it is always be the 2nd, 3rd, 4th circuit.

Also on all the panels we do, the BOC marks are never zeroed, if you put the cursor at the BOC mark coordinates and press camera button, the screen will always be off the to the right side of the BOC mark. (showing the BOC mark off the the left on the visual display.)

So naturally the first circuit coordinates is wrong as well.

One would think (me) that if the first circuit coordinates was correct it would be exactly on the corner of the first circuit. Likewise if the BOC marks was correctly calibrated on the machine the camera would zero in in the middle of the BOC mark.

BY THE WAY: SOME PANELS DO GO THROUGH THE OVEN WITHOUT ANY PROBLEM. EVEN SO IT IS PAIN WITH REWORK.

#67795

paper vs emboss (& zig zag placement) | 8 December, 2012

I hope this picture helps explain the text above if you got confused.

Attachments:

#67796

paper vs emboss (& zig zag placement) | 8 December, 2012

I don't know what is BOC (may be I know it by another name...), but: 1. Check placement coordinates by camera (if your mounter can do it) and tune it if they have a problems; 2. Check your stencil - it can be a mistake during stencil cutting (very small chance if you use laser cutting stencil); 3. Gradient of temperature in your oven - in center of oven temperature is higher compare with the lower zone. Try to decrease temperature; 4. Try to rotate PCB before oven to 90 degrees and compare result.

#67798

paper vs emboss (& zig zag placement) | 9 December, 2012

Is BOC, Beginning Of Cycle?

If it is how boring,

The real root of the problem is, some of the parts are not put on the correct place, some of the misplaced parts wobble over on the bubble of solder going through the oven, but some been so light stand up on a 60° angle.

[all except for the first circuit which is perfectly place components.]

#67799

paper vs emboss (& zig zag placement) | 9 December, 2012

Maybe you need to re-calibrate your mounter?

#67802

paper vs emboss (& zig zag placement) | 10 December, 2012

Thank you ENIAC for your input.

The engineer who did the PCB says all his figures are correct.

I over came the problem by moving the X coordinateness of the 2nd BOC mark (top left on the picture) over just slightly in the file (measured by the distance the part got put out) on the top left corner of the 3rd circuit.

I suspect it could be the overseas people who made the PCB, maybe I am wrong. On another panel part of the silk screen wording was missed from only one circuit on 5*3 circuit panel, the other 14 circuits were all OK.