By Mark Thompson | Published on: January 07, 2019
Today’s column is about eliminating confusion that creates remakes, both from the end user/designer and the fabrication house.
First. Let’s talk about being clear about your intentions. What are some of the things that create miss-communication and lost time and sometimes even remakes at the fab house of your choice.
The first thing is querying the fab house for impedance numbers…
Let us say you have asked for a material type on your drawing that is not either readily available or used by your fabricator, here you should expect the fab house to respond quickly and have all the deviations at once for you to review. This includes any IMPEDANCE width changes, material types or copper weights to be able to produce the part, also any deviations regarding drawing notes, such as WRAP plate requirements that cannot be incorporated due to insufficient space due to the extra etch compensation to meet the WRAP plate requirement. If Your chosen fabricator comes back, and says for instance, the material you have called out is not a standard stocked material for them and you are asking for a 3-day turn. Giving you an alternative is great, but if by changing to the material they suggest, your trace widths or spaces change to the point the part has to be re-layed out, this can be very frustrating for the board designer.
To mitigate this happening, give yourself and addition mil or even 2 mils of space will allow the fabricator to be able to resize the traces to meet your impedance and adjust dielectrics to meet your impedances, if for instance you have designed the part as .1 mm trace AND space, this does not allow the fabricator to simply re-size the traces without affecting the spaces and may even require the fabricator to ask for a lighter starting copper weight to allow for a half a mil change of trace width and still make the part producible. This is where the etch compensation based on starting copper weight is key to remember when designing the part to allow for slight changes to either dielectric distance, starting copper weight or trace width changes to meet your impedances .having gone through all this, if additional deviations are not stated by the fab house due to color of soldermask or material finish type or even material UL rating required this is also frustrating for the designer. Here, you should be communicating with an engineering person at the fab house ahead of time to preclude these deviations to slow down the part.
The second thing is being clear on what you need , if for instance you say initially you ask for half ounce starting copper and the copper weight needs to be increased due to voltage or EMI issues and things have to change , this loses time in the re-lay out of the design to accommodate a larger space to be able to deal with the higher copper weight. The point here being if you have stated one copper weight only to change it, expect you may have to change trace widths or spaces to be able to accommodate these changes, Likewise changing a mask color form say GREEN mask to RED mask , this too may require a complete re-layout due to the slightly thicker red or other colored mask change so what you ask for really should not change from what is originally asked for or you could expect to have to re-layout the design to increase space or trace widths. Again with the example used above at .1mm trace and space, if not all the variables remain the same as what was asked for the fabricator only has but a few deviations that can be made on their end, ALL these negotiations should be made PRIOR to release of the dataset.
Frustrations can occur on both sides (both the end user AND the fabricator when things change based on the information they have received vs what they get once the dataset has been released}
The two main delays we see at the fab level are:
Much of this was never an issue at say .008/.008 and the designers knowledge that those numbers correspond roughly to the dielectrics needed to meet said impedances but TODAY, with trace and space routinely at .004/.004, the deviations from copper weights or mask colors can and will affect the ability to be able to produce your part as you wish it.
Remember what are the things that a fabricator CAN change to meet your desired impedances.
Trace widths (if space allows based on copper weights remembering all copper is typically “compensated” at the cam stage to deal with the known loss at the fabricators etcher}
Dielectrics (if overall thickness can be met by adjusting the dielectrics and if copper weights can be REDUCED on interface layers, let me give you an example, let us say all plane layers are called out as 1 ounce but the trace widths on the surface layers are .004 or below to meet the impedance, Here you may have some push back asking to reduce the copper weight on the plane layer due to having to go to a single ply prepreg construction. As evidence of this for larger than half ounce or 3/8ths oz internal planes fabricators like to use a minimum of TWO plies of pre-preg bearing in mind that even the thinnest preg ply used as a two ply MAY be too thick to deal with the .004 or below trace widths called out on the surface for impedances requiring a re-lay out to increase the space on the surface to be able to increase the trace widths and still meet a 2 ply construction to keep your planes at 1 ounce}
Material type (to get closer to the Dk you were modeling the trace and space at, so no or very little changes need to be made to the trace widths which may also require a new layout}
Copper weight reductions (if for instance a given trace has to be increased digging into your available usable space)
These are really the only things a fabricator can do to be able to get close to your desired impedances on their end without requesting a new design with wider spaces, material deviations and overall thickness.
What are some things a designer can do to help the fabricator if it comes to changes of trace width, material types or overall thickness? Or tolerance requirements?
Well One thing is to NOT use the same trace width for different scenarios on the part, as an example using a .004 trace for SINGLE ENDED 50 ohms and part of 100 or 90 ohm differential pair. Differentiating them by a tenth or even a thousandth of a mil will allow the fabricator to uniquely pick out those structures that need to re resized and not affect the others of the same width. This has been a practice known to designers for years. Additionally, using the same trace width for copper fill area as the impedance traces is NOT a good idea as if one needs to change to meet the desired impedances the fill area would also be affected.
Now let’s talk about mask color changes from what was originally negotiated based on the same .004/.004 design rules for trace and space.
Let us say the original request calls for RED solder-mask and when the part dataset is received it is changed to GREEN mask. The RED mask is thicker and impedances based on RED mask are calculated as RED mask so changing to GREEN mask may mean the .004 traces would have to increase due to the thinner mask color, even as little as half a mil change may throw out your ability to increase the trace as a true .004 trace and .004 space design. Here, even going with a lighter copper weight to start on in an effort to NOT have to re-layout the design with wider space will not work as REDUCING the copper weight means INCREASING the trace width to get the desired impedance.
So what is the bottom line of this column?
As with ISO where we must do what we say and do what we do, As a board designer you need to make sure ALL changes of material type , trace widths or spaces, mask colors or even surface finish from what was originally negotiated with the fab house MUST be negotiated prior to asking for a fabricator to calculate impedances and not have to re-layout the design.
Save yourself some grief by having a conversation with the fabricator about ALL the variables so you will not be asked to re-lay out the design taking additional time.
Copyright © 2013 Prototron Circuits, Inc. All Rights Reserved.