Posts

Networking Equipment – EMC Radiated Emissions Problem Solving

I had an email from a customer that I’m working on some design consultancy work with, saying that one of their prototype products was having some radiated emissions problems at an accredited lab. Could I take a look?

Absolutely, EMC radiated emissions problem solving is my favourite part of the job! Ironically, it is usually the customers least favourite part!

Thankfully I had a slot free the next week so they bundled their kit into the car for the long drive “Up North” from their base in the South West of the UK.

After some tinkering, the equipment was set up in the chamber for some radiated emissions work. The first scan confirmed the problem levels and frequencies that had been observed at the other laboratory.

The problem areas from their last scan were at 35MHz, 80-90MHz and a broad band between 150MHz and 220MHz.

 

System Overview

The system was housed inside a nice aluminium case that was being used for CPU heatsinking and environmental protection as well as EMC shielding. A rough diagram of the internals shows a main PCB with a large CPU / memory block in the centre and a variety of cables leaving the PCB and the casing.

The main power cable housing also had two debug connections inside the same housing that weren’t being used in the field but were available for updating software and such like.

rwns overview diagram

As is so often the case, this product was in it’s final stages of the development life cycle, meaning that no major design changes were possible. These EMC problems would have to be resolved using easy to fit additional components. Thankfully I have plenty of things in stock to try out.

 

Emissions Analysis

There are two important characteristics about these emissions that show us where to look

  1. They are predominantly broadband, an indication of analogue noise e.g. DC/DC converter / power supply. Sometimes this broadband noise is generated by digital switching but this can be less common.
  2. They are all low in frequency, where large or long structures are the most efficient antennae. This usually means cables.

So power noise and cables…. hmmm…. any good ideas?

 

OK Kids, Let’s Take a Look at the Cables.

In a very sensible move by the designer, both the DC power and Ethernet cables had some common mode filtering on the PCB.

Ethernet magnetics have common mode chokes built into the transformer stack which reduces the noise emitted and increases the susceptibility performance of Ethernet despite the often unshielded twisted pair cables used.

The caveat is that once the cables have left the magnetics that they must be protected from other interference sources. Noise coupling on to these lines is going to be heading straight out of the enclosure using these lines as the antenna. Similarly, if common mode noise gets onto the centre-tap of the output side of the magnetics then this can also cause similar issues.

I have experienced system noise coupling on internally routed Ethernet cables before and it nearly always results in lots of low frequency emissions.

The power cable had a small surface mount Murata filter in place with excellent attenuation at the frequencies of interest.

murata filter characteristics and equivalent circuit

Both the Ethernet and power cables pass through the shielded enclosure with no connection or filtering to the case. In bypassing the quite nice Faraday cage of the enclosure, any noise current on these lines will inevitably appear as radiated emissions and be picked up by the receive antenna..

Now to find out some more info.

 

Radiated Emissions Experiments

First, unplugging the Ethernet cable dropped the emissions significantly from 30MHz to 120MHz.

Secondly, some messing around with ferrite cores on the power cable reduced the 150MHz to 220MHz hump down to sensible levels.

This left a single peak at 270MHz that was traced to noise using the coaxial RF cables to the antenna to radiate.

Lets look at each of the points in a bit more detail:

 

Ethernet

The only practical method of dealing with the Ethernet emissions was to change the bulkhead connector to a metallic screened version and the external cable to a SSTP (Screened Shielded Twisted Pair) type of cable. No exciting analysis here I’m afraid.

 

Details of the Power Cable Noise Coupling

The most interesting coupling mechanism was happening inside the un-screened bulkhead power connector. Thanks to the power filter on the PCB, there was very little noise being conducted back down the cable from this line. However, the debug connections to the CPU are picking up all kinds of noise and carrying that noise to the connector.

rwns cable coupling close up

Disconnecting and bundling the debug cables near the connector cuts the radiated emissions down to next to nothing.

What’s most interesting is that the capacitive coupling region between the power cable and the internal debug cables is so small. The connector is only 20mm long and the cables run parallel with each other for barely any distance. And yet there is enough noise current being coupled onto these lines that it causes a radiated emissions problem.

 

Details of the RF Antennae Noise Coupling

By the time that all of the cables had been filtered or removed, there remained just one emission at 270MHz that was failing the Class B limit. An investigation with RF current probes showed a lack of noise on the main output cables listed above, even when they were screened or filtered appropriately.

A wander round the enclosure with an electric near field probe and spectrum analyser showed a spike in emissions near the RF antenna housing on the side of the EUT.

 

rwns antenna spurious

Checking the antenna feed cables showed them connected to the PCB pretty centrally. Disconnecting the coaxial cables from their mating halves dropped the emissions down to the noise floor.

Even though the noise isn’t in-band for the antennae themselves, they still perform well enough to radiate the noise and cause an emissions problem.

 

Summary of Fixes Applied

The below diagram shows the fixes applied to the EUT to achieve a Class B pass.

rwns applied modifications

Firstly, a fully screened metal bulkhead Ethernet connector was chosen for use with a shielded cable. This isn’t ideal from the installation point of view but is ultimately unavoidable without more significant modifications to the EUT.

Secondly, a Wurth ferrite was equipped around all three of the cables connected to the power bulkhead connector. As detailed above, it is necessary to put the ferrite around all three cables and not just the power to reduce the noise entering the capacitive coupling region around the connector.

Thirdly, a small ferrite was placed around each of the UFL cables at the point at which the antenna cables left the housing. This is a fairly common modification for radiated emissions, one I’ve employed several times before, and there are numerous suppliers of ferrites of various lengths with just the right inside diameter for the type of thin coaxial cable used with UFL connectors.

 

Results

Closing Thoughts

Any time your cable passes through a shielded enclosure with no RF termination at that point, you can pretty much guarantee its going to need some filtering.

Nothing particularly in depth in this analysis of the EUT, but I did find the coupling in and around the power connector particularly interesting.

At the end of the day, the best outcome was a happier customer with a path forward for their product.

 

ram cage being removed from a 2018 mac mini

Apple Multi-Purpose EMC/EMI Shielding

I’ve always been impressed with Apple’s approach to reducing problems caused by EMC/EMI. Making top of the line technology in a compact case means minimising risk and maximising performance.

Let’s look at an example of well considered EMC design and why it is so useful.

 

Even the EMI shielding solutions are stylish

Because their products are charged at top dollar prices, they can afford to (or can’t afford NOT to) put in features like this.

The RAM on the new Mac Mini (thanks to iFixit for the great photos) has its own removable cage, secured to a PCB level counterpart with screws and, no doubt, a decent fit along the edges. What’s interesting is that this shielding system will have multiple functions.

Let’s discuss these below.

ram cage being removed from a 2018 mac mini

Image from iFixit

 

Why is the screening can so important?

Primarily, it will be used to reduce the EMC radiated emissions from the product. The Apple products I’ve had in my anechoic chamber have all been very quiet and this is why I hold Apple in some regard for their EMC design.

Apple will no doubt have tested their design with multiple RAM vendors to satisfy themselves that the design meets the requirements of international EMC standards.

However, were the user to install some non-Apple verified memory modules then the risk of emissions could increase. One can well imagine that Apple will have considered this in their EMC Risk Assessment.

The secondary benefit is more subtle. Take a look at this image.

inside shot of mac mini case with component analysis

Original image courtesy of iFixit, markup by author

The memory modules and their screening can are highlighted in red. Next to it, highlighted in green, is a smaller board level shielding and a UFL antenna connector. (There are another two connectors out of sight underneath the case)

That’s right, Apple have put the most noisy part of the system (RAM) right next to one of the most noise-sensitive (Wi-Fi). What?

 

Noisy Neighbours.

This is not an uncommon problem, especially when trying to compress so much functionality into such a small space.

The Mac Mini is only 165mm square (that’s 6.5″ if you are watching in black and white). The case includes an integrated mains power supply making proximity between electromagnetically incompatible systems unavoidable.

Modern RAM speeds are fast and the Mac mini is no exception. Everymac lists the latest Core i7 model with a DDR4 memory speed of 2.66GHz. That’s uncomfortably close to the Wi-Fi operating band of 2.4 to 2.5GHz.

The interference spectra of a DRAM interface fundamental frequency is generally quite wide band.

If you turn on any form of Spread Spectrum Clocking (SSC) to reduce the peak energy then it can spread over tens or hundreds of MHz. Either way, that puts the edges of the memory fundamental in band for the 802.11 a/b/g/n/ac interface on the Mac mini.

The harmonic emissions of the memory are also prevalent and it’s easy for these to fall in-band of a wireless interface like Wi-Fi. For instance the second harmonic of 2.66GHz is at 5.32GHz in the channel 64/68 region for 5GHz Wi-Fi. Big problems.

 

Improve Performance? The Can Can.

The effect of in band interference on a Wi-Fi interface can be subtle.

At it’s most gentle, there’s a reduction in both performance and range. The modulation, coding type and channel width of the Wi-Fi sets the robustness of the interface to interference.

At the other end of the scale, whole channels can be blocked out entirely.

This intra-system, or platform level interference is pernicious and can be difficult to isolate and track down. Low noise floor spectrum real-time analysers are extremely useful tools here.

Ultimately, segregating the noise source from the receiver, is the only real solution. This can be achieved by physically separating the aggressor and victim (not possible here) or by shielding.

For some companies, the fallout in performance of a couple of Wi-Fi channels is no big deal.

If you are Apple however, then you can’t afford to have dissatisfied customers complaining about poor Wi-Fi speeds. As always, the EMC budget has to be congruent with the product budget and the desired performance.

 

The Last Line Of Defence

Check out the textured surface between the mounting holes for the lid (blue highlight on the above photo). That will be an EMI seal to ensure good contact between lid and case. Not only a nice touch but an important one.

The Wi-Fi antenna is mounted on the outside of the shield so this circular lid actually screens the antenna further from the noisy internal circuitry of the mini.

Well done Apple. I’d love to see your Wi-Fi range testing results… please?

 

 

Off The Shelf and Non-Compliant Power Supplies (from Amazon)

A customer had purchased some power supplies from Amazon UK to get started with the development on their product. And why not? There are lots of cheap products available and everyone has a budget to meet. The chances are that they’ll get damaged, lost or broken anyway.

They were happy with the (perceived) quality of the PSU so approached the manufacturer directly for bulk pricing for volume production. However, the Amazon sample made it’s way to Unit 3 Compliance for EMC pre-compliance testing where the fun began…

infographic comparing two power supplies

Externally, the only way to tell the difference between the compliant and non-compliant versions is a slight difference in the length of the barrel connector and a slightly different shape of strain relief grommet.

These devices are being marketed as the same device on the outside and yet are completely different on the inside!

I’ve not been able to subsequently find this exact power supply on Amazon but there are similar looking variants still available.

 

A Real Problem

Crucially, it’s not just EMC that is being sacrificed. This “race to the bottom” of extracting every last penny from products has more serious consequences.

More dangerously for consumers, electrical safety is also being compromised as shown in this study from Electrical Safety First on Apple chargers.

At a previous employer, an inspection was performed on 50 power supplies (again, bought from Amazon) that one of the project teams had purchased for powering various development platforms within the company. This revealed some serious safety problems (creepage and clearance) resulting in the entire batch being quarantined and scrapped for recycling.

Another aspect to consider – if the manufacturer has two different, almost indistinguishable products then how does your supply chain guarantee that you will receive the correct one? What is to stop the manufacturer from swapping out the more expensive compliant power supply halfway through production?

The principle of caveat emptor still applies. Disingenuous product markings are being used to falsely indicate compliance.

 

What To Do?

The obvious way round this is only to buy small quantity power supplies from trusted suppliers. I know from working with other customers that suppliers like RS and Farnell / Element 14 take compliance seriously. Buying from these sources is more expensive financially but what price do you put on your own safety?

If you are relying on buying a pre-approved power supply always ask for the EMC and safety test reports and the Declaration of Conformity. A supplier who cannot readily supply these readily should be disregarded.

Compare the details in the reports with the physical sample in front of you. Especially for safety reports, photos of the unit are generally included, inside and out. Look for any differences between the two.

Differences in EMC performance are not obvious. The only way to be sure of the quoted performance is to perform some quick tests, conducted and radiated emissions being the two main ones.

 

How We Can Help.

Here at Unit 3 Compliance we can give you some peace of mind that your power supply isn’t going to cause you any issues. Some of the things we do include:

  • Provide full EMC testing for all off the shelf products
  • Electrical safety analysis and testing
  • Help you understand the compromises and
  • We can review test reports and compare to physical samples with an experienced eye
  • Every incoming customer power supply is given a HiPot test as standard to help catch any problems

Please get in touch to reduce your stress levels.

 

Use of an LCD back panel as an image plane to reduce radiated emissions

EMC Radiated Emissions Fault Finding Case Study

I’m really happy to have one of my blog articles featured on Interference Technology.

Problem solving and fault finding EMC problems, especially radiated emissions, is one of my specialities and oddly enough is one of the facets of my job that I enjoy the most. After a successful exercise in helping a customer out with their product, getting the chance to write about it and share it with you is a real bonus.

Fixing radiated emissions is at it’s most challenging when the scope for modification to the unit are limited by the fact there are significant stock of PCBs or components that would require scrapping and redesign. Finding a way to use the existing stock was key in this example as the customer had significant time and money invested into the project. Thankfully I was able to help them out.

Head on over to Interference Technology and have a read through – I even put pictures in! Hopefully it will give you an idea of how I work and the sort of EMC issues that I can help you solve.

Case Study: Poor PC Board Layout Causes Radiated Emissions

 

Case Study: AC Mains Input EMC and Safety Troubleshooting

Many of the customers I deal with are technically savvy and extremely good at designing innovative and clever devices. I’m always learning something new every time I get a different product through the door. Unfortunately it isn’t practical or possible to be good at everything and EMC expertise, especially when it comes to fault finding and problem solving, can be hard to come by. This is where I come in.

I’ve been helping a good customer on a product that they’ve been working with that had some EMC troubles on a prototype design. It had originally been taken to a different test lab where they had performed a mains conducted emissions measurement showing a clear failure at low frequencies. There were a couple of other hard copy scans supplied where a capacitor value had been adjusted to try and improve the emissions but with no effect.

In need of some expertise, they got in touch.

Mains Conducted Emissions Testing

I received the product and quickly set it up in our screened room to perform some EN 55014-1 conducted emissions measurements. Below you can see the first scan result, showing a failure of up to 10dB on the Quasi Peak detector. There’s clearly some room for improvement so let’s analyse the problem and see what we can do.

mains conducted emissions - before

Our starting point for the improvement work

Lower frequency mains conducted emissions are not uncommon and are usually caused by differential mode voltage noise. This is generated by current flowing through the impedance presented by the primary side bulk decoupling and switching circuit. The switching frequencies of the power supply controller are usually in the 30 kHz to 250 kHz range putting it (and it’s harmonics) right in this lower frequency (sub 1MHz) range for this test.

Improving differential mode noise can be done in a number of ways. Removing the noise at source is the approach I advocate, in this case this can be achieved by reducing the impedance of the rectified mains bulk decoupling capacitor. A review of the BOM showed that the units had been built with some general purpose electrolytic capacitors with a relatively high impedance. So the first thing that I did was to swap out these parts for ones from the Nichicon PW series of low impedance electrolytic capacitors.

after fitting low impedance bulk decoupling

Changing the electrolytics to a low impedance variety

The result: a big improvement on the QP measurements, bringing some of them down by around 10dB. The improvement on the Average detector readings was less pronounced, especially around 550 kHz where only a 3dB improvement was registered. It is likely that the HF impedance of the decoupling capacitor is still a problem. One option is to apply a suitably rated high frequency decoupling capacitor in parallel with the bulk decoupling capacitor. The other option is to improve the filtering on the AC mains input to prevent the noise from escaping back down the line.

Filtering for differential mode noise can be provided in several ways. The most common method is to make an LC filter from the leakage inductance of a common mode choke paired with a Class X safety capacitor between Live and Neutral. The leakage inductance is in the tens of micro-Henries whereas the common mode inductance is often a couple of magnitudes larger up in the tens of milli-Henries. Simplistically (there are other effects to consider) a 10uH leakage inductance paired with a 470nF capacitor will roll off frequencies above 100 kHz. Well, let’s try that!

now with added class X cap

Now with an additional 470nF Class X capacitor soldered across the mains input terminals

Performance is improved by around 5dB across a wide range of frequencies; indeed the improvement can be seen up to 15 MHz. This leaves a margin of around 2dB to the average limit line which is perhaps a bit close for comfort and I would generally recommend looking at a little more filtering to bring this down a bit further to allow for variations in production and tolerance of components. Options for further improvements could include a second Class X capacitor to form a pi filter but because of the low impedance of the differential mode noise this approach might not be as effective. Adding some inductance to form an LC filter with the bulk decoupling capacitor is another approach.

However this proved the case to the customer for a PCB redesign to make space for the larger bulk decoupling capacitors and at least one Class X capacitor.

Surge and Safety

Following on from this work, at the customers request, I carried out a full suite of EMC tests on the product to EN 55014-1 (emissions) and 55014-2 (immunity). One thing that I noticed was the sound of an electrical breakdown during the application of a differential mode surge test. Taking off the outer casing, I managed to catch the below arc on camera during a 1kV surge event.

Arcing caught on camera

Snap, crackle and pop.

The arc appeared around the resistor; desoldering and removing it from the PCB showed a couple of points where there was arcing between the resistor body and the trace running underneath it.

Arcind damage to the PCb to surface

Arcing evidence on the PCB

This problem has occurred because the resistor R1 is in series with the Live phase and the trace underneath is connected to the Neutral phase. When mounted flush to the PCB normally, the resistor has only its outer insulation between live and neutral. Reviewing the relevant electrical safety standard for the product requires a minimum clearance (air gap) for basic and functional insulation is 1.5mm. This can be achieved by standing the resistor up on spacers to keep it away from the PCB but then it starts to approach VDR1 and Q4 meaning a considered manufacturing approach is required. This was another incentive for redesigning the PCB.

The take-away lesson from this finding is to consider the Z axis / third dimension when reviewing a PCB as it can be easy to see things purely in two dimensions!

I hope you found this case study useful and that it has given you some tools with which you can improve your designs.

If you need some EMC fault finding expertise then get in touch: I’d be happy to help and I love a good challenge!