EMC compliance for HDMI Radiated Emissions testing (EMI)
EMC compliance for HDMI Radiated Emissions testing (EMI) in Australia is a subject that many designers face with HDMI been incorporated into many consumer products. At EMC Bayswater we quite often see high-speed data ports with cables that connect to various other products such as Ethernet Local Area Networks (LAN), HDMI etc.
Many times these ports have been developed over years and have defined drivers with suitable filtering and so forth. HDMI poses a similar problem to the computer monitor video signal of the past apart from using faster data speeds and been much more complex.
Most people over a certain age will remember connecting their computer to the CRT or LCD monitor via a multicore cable usually with ferrites on either ends or both. This may have been for the un-intentional generated RF but was more likely used to help reduce the intentional signals in order to comply with the radiated emissions requirements for CE marking, FCC and the old C-Tick etc. These intentional signals included the Pixel Clock frequency of 25.175MHz for old monitors.
Whilst using HDMI 1.0 a maximum pixel clock frequency of 165MHz was specified as per the standard. As HDMI advances and provides higher resolutions, the pixel clock frequency has been increased to cope with the increased required data.
Now HDMI 1.3 has a maximum pixel clock frequency of 340MHz. With advancements and changes in the expected customer demand for products in particular to high resolutions, brings challenges to keep pace with the EMC aspect as well as the functional requirements of HDMI products.
Radiated emissions in accordance with the most used standard for this type of equipment i.e. multimedia equipment is EN 55032 or the other international/national variants such as IEC CISPR 32 AS/NZS CISPR 32 is required for all products falling within the scope of the standard.
Within EN 55032 it requires Radiated Emissions measurements to be performed from 30MHz to 6GHz.
The frequency range is based upon the highest fundamental frequency generated or used within the EUT or the highest frequency at which the EUT operates (this includes frequencies solely used within an integrated circuit).
This basically means that for any product incorporating HDMI 1.0 or above it is most likely that the frequency range will be at least 30MHz to 2GHz if not the entire EN 55032 frequency band (30MHz to 18GHz*).
FCC Radiated emissions testing (FCC Part 15, Subpart B) is very similar however the frequency range is based up the fifth harmonic of the highest frequency of the highest frequency is above 1GHz.
*Satellite receivers may require testing over the frequency range of 30MHz to 18GHz.
Filtering (common mode filters)
In addition to following common EMC design principles and practices with regards to EMI mitigation and good EMC practices such as controlling the EMI source as close as possible to the source and also filtering at the port entry/exit.
Please refer to the links below for a very good application notes and additional information regarding filtering for HDMI emissions. There are other key aspects that may help with achieving EMC compliance.
Use a good quality HDMI cable to start off with including a cable shield well-bonded to the metal connector shell in multiple places. A good quality HDMI cable also has a tight weave with very small openings.
For Radiated Emissions the problem is the common mode noise i.e. the common mode current flowing on the outside of the shield, thus you may require some filtering. Poor quality (usually the cheap ones) with loose shield weaves or cables with pigtail connections to the connector shell, will allow interior signals or common mode currents, respectively, on the outside of the shield.
Ensure a good continuous bond through the transmission line (HDMI cable) from the product to its final termination. Also, the cable length and the impedance matching will factor in how well a cable acts as an antenna.
It is best practice to design (implement filtering, avoid cross coupling, suitably rated components for filtering and other EMC design techniques) to avoid as much RF as possible on the cables in the first place.
Document in the test report the cable used and recommend in the products user manual either using that specific cable or an equivalent cable in order to maintain EMC compliance.
If the enclosure, not a good Faraday box at over the frequency range of the emissions this may need to be looked at. A full metal enclosure with a good earthing arrangement has real benefits.
Firstly the obvious one is that PCB and internal cable radiation will be limited. The only limitation is the shielding effectiveness of the enclosure and the port filtering.
The other key benefit is a good bonding for the connectors and thus also a good bonding for the cable shield.
Changing the actual HDMI signal
We don’t know much about this aspect, however, I’m sure if you are reading this you may be able to provide a better explanation to us!
One online blog (https://community.nxp.com/thread/349979) stated:
The HDMI signal levels can be adjusted to comply with the Eye Pattern. This will have an impact on the EMI.
Read about it here: http://cache.freescale.com/files/32bit/doc/app_note/AN4671.pdf.
Also by simply changing the resolution and refresh rate on the HDMI output will have an impact on the EMI. In my case, it was the pixel clock that was the big noise issue.
In the end, you should always look for the RCM Marking. This would make sure that your product has undergone proper testing.
Source: Courtesy of EMC Bayswater Melbourne, Australia