Posted on 17 Jul, 2014 by Richard Barratt
You may have been asked, for instance, to commit to Non-Recurring Engineering (NRE) charges and want to understand better where some of the time charged will be spent? Or perhaps you have been asked to clarify, or provide additional data, to your assembly partner and would like to understand where in the process these issues should be picked up.
Allow this blog post then to shed some light onto those areas of puzzlement. Here we consider the types of checks that should be carried out by your Contract Electronics Manufacturer (CEM) prior to commencing the build process and, if missed, the potential impact on the delivery and cost of your product.
1) Data verification
The first step for the NPI Engineer is to make sure that they are in possession of all of the information needed to build your product. Sounds pretty obvious but what do they really need from you? As a minimum they will need a Bill of Materials (BOM) which details the parts required, and a Gerber file which provides details for the Printed Circuit Board (PCB), such as copper layers and drill size/positions etc.
If you have Computer Aided Design (CAD) data available, in a format such as ODB++, then you should send this to your CEM. CAD data can be imported by your assembly partner to automatically create a programme for their surface mount pick and place machine. Without it, the NPI Engineer will have to create the programme manually by working from the Gerber file. Not only does this take longer than it would if they had the CAD data, but it will also increase the NRE charges and introduce the opportunity for human error.
2) The right tools for the job
Your design is likely to include surface mount components, and as such a solder stencil will be needed during the initial stages of build. You may hear these solder stencils (like those provided by screen printing machine manufacturers such as DEK) referred to as ‘tooling’, when it comes to NRE charges. Prior to the stencil being ordered the NPI Engineer will check that the stepped Gerber file - used to produce the PCB - has been panelised correctly. If it hasn’t, the assembly house will have to spend additional time (which unfortunately is likely to come at a cost) adapting their process and equipment. In some instances, the PCB may have to be scrapped altogether, which clearly adds further cost and delay.
Once these checks have been made the NPI Engineer can then order the stencil. Typically the supplier will send back a file confirming exactly where the apertures within the stencil are due to be cut. This is also then verified by the NPI Engineer for thickness, aperture to pad ratio, fiducial placement, and direction of print; in addition, they will make sure that no plated through hole (PTH) components have found their way onto the stencil incorrectly!
Finally, there will be a visual check carried out on receipt of the actual stencil itself. At this stage, it’s really a case of checking for any damage that may have occurred during transit and that the fiducials have been cut out correctly.
3) Creating the Build Pack
Your EMS Company is likely to use a manufacturing software package, such as those offered by AEGIS, to help create build packs which are used during the manufacturing process. At this stage, the NPI Engineer will be checking that the data contained within these build packs covers everything the production team need. For example, if you have provided any special build or handling instructions, these would now be added into the build pack as dedicated care points.
4) Checking the Programme
Prior to manufacture the NPI Engineer will want to take a look at the details of the surface mount programme that has been generated. They will then look to carry out:
- An offline check for any ‘no fit’ components.
- An offline check for component package size and polarity.
A ‘no fit’ component is one that has been designed into the PCB but is not required on a particular variant. Occasionally, this instruction can either be missed, or ambiguous, and therefore the EMS Company needs to check with the Original Equipment Manufacturer (OEM) so that errors are not made during the production stage.
Finally, to avoid discrepancies between the physical size of the component and the pad laid out on the PCB, the NPI Engineer should step through each placement electronically. Unfortunately it’s quite common to find the wrong component has found its way into a BOM, either through a design change or a typographical error made against the component part number.
Despite all of the pre-build checks there can still be issues that will only really be found during the physical assembly process. Whilst they may take time and effort up front, the purpose of putting these pre-build checks in place is to ensure that issues found during production are minimal – helping to keep the cost of your product on budget and the delivery on time.
Hopefully, this blog post has helped illustrate a number of pre-build checks that should be carried out before the actual NPI build process begins, and their importance to the smooth progress and delivery of your product.
Image by DLR_de