This article briefly describes why it’s important to consider what RF suppression/filtering techniques will be applied to your product as early in the design process as possible.
Let’s say that for whatever reason, you developed a product for the non-residential, industrial market where it is clear that Class A radiated and conducted emissions are required. Sometime after the product meeting, Class A emissions is introduced, a new market emerges with the potential for a lot of sales, but it requires the same product now comply with Class B requirements, which are about 10 dB more restrictive in the emissions allowed to be emitted from the device. This Class A product must be redesigned to meet Class B or somehow made to comply with Class B requirements before it can be shipped into that market.
Let’s also say that you do not want to redesign this already existing Class A product to meet Class B emissions because it’s too cost prohibitive and time consuming (you want to start designing the next new product, not fix the old ones). In this situation, what do you do?
Depending on what the source of the emissions are and where they are emanating from the device, the best approach may be to determine if adding ferrite choke(s) to the cable(s), using an RF absorbing pad(s), applying shielding can(s) to a noisy chip(s) or other area(s) or a combination of all of the above may allow your product to pass Class B.
Here are some issues to consider with applying this “adding on fixes” methodology:
Issues with “Adding on Fixes” After a Design Freeze
- The “ferrite” solution
The sources of emissions/offending signals may be spread out in frequency (some low, some high) and you may not be able to find one solution, like adding a single ferrite choke to a suspect cable, that is able to suppress emissions over both low and high frequencies.
In this situation, you need more than one ferrite choke, and you probably have to select one choke that is made of one type of ferrite material that is only useful for suppressing lower frequency emissions, and another choke made of another type of material that tackles the higher frequency emissions. One problem with this solution is that adding more than one choke may appear “kludgy” and would not likely be accepted by product management or the end customer.
- The “ferrite plus absorber” solution
If the offending frequencies are spread low and high as in the above scenario, an alternative to using two ferrites of differing materials and suppression capabilities would be to use a combination of low-frequency clamp-on ferrite choke for any emissions emanating from cables and a high-frequency RF absorber sheet to suppress emissions at the source, such as those emanating from fast clock chips or microprocessors.
This solution may not appear as kludgy to product management and the customer as the two-ferrite solution described above but may involve more experimentation and design effort to figure out where the best location is to place the absorbing pad to reduce emissions and how to effectively incorporate it on a production basis.
There may also be heat issues to contend with depending on the end-use environment of the product and how hot components may get underneath the absorber pad. Re-doing temperature testing may be in order if this solution works.
Other tests such as vibration, shock, and bump may also be required depending on how the absorber pad solution is implemented. You want to make sure by adding the absorber pad, you don’t create a new problem elsewhere.
- “Other” Solutions
You may be able to incorporate a shield can over offending high-frequency noise source and use a clamp-o ferrite for lower frequency emissions. Reference 2 covers more on the subject of shielding at the PCB level.
Since our original goal was to do as little as possible (not go back into layout), adding a shielding can over offending sources of emissions may not be a viable solution depending on the original design and how much room you have. If room is tight, you may not be able to effectively add a shield can.
No matter what the RF suppression/filtering solutions chosen as add-on fixes to an already production-released product are, the chances of them working on the first try are slim to none. Plan for the “adding on fixes” approach to take a lot of time. It may be easier than relaying out a board, but you still need to figure out what types of ferrites would work best, including size, types of materials, cost, lead-time, etc. The same is true for selecting RF absorber pads.
Had the requirement for Class B emissions been specified and included as part of the original design, there would have been much more freedom to implement more effective solutions sooner and at lower levels of the design, such as at the circuit board level. This “adding on fixes” and hoping for the best approach would have been unnecessary and you would have already started working on your next new design.
References and Further Reading
- In Compliance Magazine. (2019, September 27). Let’s Talk About Flexible Absorber Sheets
- In Compliance Magazine. (2019, August 6). Let’s Talk About Shielding at the PCB Level
- In Compliance Magazine. (2018, November 12). What Every Electronics Engineer Needs to Know About: Filters
- In Compliance Magazine. (2018, August 2). What Every Electronics Engineer Needs to Know About: Shielding