Centroid/XYRLS/Pick-and-Place

Call it what you may, but surface mount assembly robots need this magic file to determine where to place your components and how to orient them. We call it a centroid. Others may call it something else, but it’s all basically the same. In our case, the basic format is comma delimited, in mils:

Ref designator,     Layer,     LocationX,     LocationY,     Rotation
C1 ,                       Top ,           0.5750  ,       2.1000  ,           90

That’s not too difficult. Most CAD programs will automatically create this file for you. Eagle doesn’t natively, but we have a ULP to do it for you in Eagle (downloaded here). Again, no problems here. Mostly…

I say mostly because, at this point, you are at the mercy of the person who created the CAD library part. Provided they center the origin and follow the IPC for orientation, everything should come out just fine. Unfortunately, we do find parts that don’t follow those rules. We’ll do our best to catch and correct such things here, but for maximum reliability, check you library components to make sure. We find the problem crops up most commonly with passives.

IPC says that zero orientation for two pin passives is horizontal, with pin one on the left. For polarized capacitors, pin one is (+). For diodes, pin one is the cathode. They note that pin one is always the polarity mark pin or cathode. Pin one is also on the left for resistors, inductors and non-polarized capacitors, but left vs. right doesn’t matter so much with non-polarized things. The most common orientation error we see is to have the “zero rotation” 270 degrees off from the IPC standard.

Every now and then we’ll find that someone assumes that since usually the anode on a diode tends to be on the positive side, that the anode should be pin one. Nope. Nope. Nope.

Duane Benson
Is it pulling electrons of pushing holes?

http://blog.screamingcircuits.com/

About Duane

Duane is the Web Marketing Manager for Screaming Circuits, an EMS company based in Canby, Oregon. He blogs regularly on matters ranging from circuit board design and assembly to general industry observations.
This entry was posted in Laying It Out, Screaming Circuits Blog and tagged , , , , , . Bookmark the permalink.

7 Responses to Centroid/XYRLS/Pick-and-Place

  1. John Lambert says:

    Duane,
    Which IPC spec are you referring to? (standards for rotational zero position)
    I would like to review. (so many things to keep track of… )

    The bigger concern .. that your Pick and Place equipment generally has it’s own component outlines/rotational standards/centroids.
    Often, the machine, comes with it’s own libraries (rotational definitions).

    And it gets more complicated… purchasing a LED / DIODE sometimes comes with a choice:
    Do you want the Anode on the right or left side of tape?
    Thus… swapping the component orientation 180 degrees.

    It may or may not be to the same IPC standard (if there is a IPC standard for the device in question)..
    There are a lot of parts (connectors) that are not symmetrical .. centroid is not obvious.

    So match them or keep a internal standard for tracking / converting the differences.
    What ever standard you use.. make sure you understand all the pieces used to produce the final product…
    or
    You can do as most… let the SMT line setup operator figure it out…. (pitch it over the wall)
    Not the most efficient use of company’s resources..
    (Even if you are using a CM, you shouldn’t believe the effort is free – in cost, errors, delays)

    Nearly all new pick and place machines prefer the measurements in metric terms.
    As always.. there are exceptions.
    But generally, millimeters or microns are the native measurement system..
    Not a big deal .. nearly all setup operators have conversion routines available.
    But if you want to minimize confusion, I recommend using metric measurements.

    And I think all Pick and Place machines need a component part number for each reference designator…
    Most P&P machine formats include the part number in the same comma delimited line with designator , rotation, x, y, side.
    Designator, is really not as important as part number.
    After all, you have a x,y location defined, the machine really only needs a part number and rotation angle to complete the task of placing a component.

    Part number.. tells 1) feeder location 2) pick orientation 3) component package 4) component height 5) optical inspection data and 6) electrical test info – some machines test each resistor, capacitor, etc., before placing them on the pcb.

    Very awkward to have to create complete pick and place data file requiring two file sources.
    (designator, centroid, location AND BOM to correlate designator with part number/package)

    Just saying…

  2. Duane says:

    Hi John;

    My reference was IPC-7351B, figure 16-1 in the copy we have. There are a number of other 735* standards related to lands as well. Everything you listed is valid. It’s a jungle out there. The good news is that generally, as you say, the SMT operators can do a lot to fix things, but best practices suggest that it be taken care of before it gets there. We don’t require the part number in our centroid, but many others might.

  3. James says:

    LED manufacturers are putting “dots” on both cathode and anode ends depending upon the manufacturer and device. Grrrr!

  4. As a board designer I find that I have to create the libraries from the manufactures data sheet which rarely has the package illustrated in the orientation of how the component is supplied, which is invariably different as to whether it comes in on a tape or in a tube!

    My CAD package has the Y-axis swapped ( Y is down the screen i.e. 0,0 is top left)
    Its output is imperial, 0 deg orientation is at 3 o’clock with positive increments clockwise with no stock code, and it uses letters for the cardinal angles U=0, Q=90, H=180, T=270.

    I have a macro in Excel to correct the ‘Y’ issue, convert to metric and have real numbers for the angle.
    It also swaps the X-axis as the pick and place machine has 0,0 bottom right.

    I then do a merge of the output file with a stock code file using code written in C# which also changes the orientation according to a file list of rotation correction values, but without knowing how the components are supplied this last part has to be manually verified before loading onto the machine.

    Yes it is a jungle out there, and some of us are “hacking it” for a living!

  5. Marc England says:

    I always create pick and place files with a fiducial as the 0,0 location, and a minimum of two fids in the pick and place data. If you have in house production you have to work together to control both the CAD library and the Pick and Place library, this is well worth the effort as thing run smoothly with no rotational errors. As most components are now metric based using metric all the time realy makes sense in my opinion. The IPC-7351 spec is based on the 0.05mm metric grid, the document The CAD libary of the future” details the benefits.
    You cannot cater for components in tapes at the CAD stage, even though there are moves to standardise the tape orientations from supplier to supplier, I have seen differences between batches from the same supplier! The use of intelligent feaders helps here as, thoygh most machines will allow for the input of the component orientation in the tape and cater automaticly for this.
    For non polorised components I put the rotation out at either0, 45, or 90 degrees, filtering out 180, 225, and 270 etc, cuts down the head rotations, though again with multiple nozzles heads I dont know how effective this is these days for saving time.
    I dont want the SMT operators to fix anything, I want a clean effective transfer of data with the minimal amount of human intervention, this is the only way to achieve the required error free throughput.

    http://ohm.bu.edu/~pbohn/__Engineering_Reference/pcb_layout/pcbmatrix/Component Zero Orientations for CAD Libraries.pdf

  6. John Lambert says:

    Marc,
    I can’t get your link to work..
    Agreed.. less human intervention, the better.
    Only drawback.. it means the data maintenance (pick and place data) has to be done by the pcb layout person.
    And that person has to be informed of relatively simple changes from manufacturers tape and reel packaging!..
    Possibly more effort than the pcb layout person / library manager was expecting.

    Because I am using CADint….
    I can (at any time) define the pick and place centroid and rotational angle used for a library shape, completely separately from the how the part was created/rotated in the design!
    A great feature on this Cad package… allowing for me the make the part any orientation I want, yet still create pick and place data to any standard (IPC , internal, etc…) without changing any pcb layouts or library shapes.
    I assume this feature is available on other CAD packages…

    I can create the part horizontally.. but define the pick and place data being created with an offset of 90 degrees (vertically)…or any variation.
    It also automatically generates centroids based on overall package.
    but allows for explicit centroid definition.

    Some other options for a given library shape (pcb land pattern)..
    – exclude from pick and place data ( example: through hole part or SMT test pad)
    – exclude from BOM (example: smt test pad, fiduicals)

  7. Thanks for sharing a informative video. We are too suppliers of pick & place machines.

Comments are closed.