This replaces an Excel macro I’ve used in the past with something that can be used in a variety of types test software.
EDIT: now updated to allow addition of specific spot frequencies to the test frequency list
This replaces an Excel macro I’ve used in the past with something that can be used in a variety of types test software.
EDIT: now updated to allow addition of specific spot frequencies to the test frequency list
I had a challenging EMC problem solving project in the lab this week.
A customer making a miniaturised 4 cm^3 buck-boost DC/DC converter for Li-Ion battery charging was having radiated emissions issues. The small size meant that adding common mode chokes to filter the input and output connections wasn’t practicable so a more in depth investigation was required.
Here are the emissions for the EUT without any modifications. The green reference trace is the AC/DC mains power supply being used to power the EUT. It is failing the Class B limit (blue) by some margin.
To investigate the emission radiation source (not the cause yet), I placed large clip on ferrite cores around the DC input cable and the battery output cable to reduce emissions directly from the cables.
This improves some of the frequencies but not all of them. If the radiation was entirely cable related then this would have dropped the emissions significantly. As it hasn’t, we can conclude that the majority of the emissions are coming from the PCB.
The three peaks we’ll focus on are 180MHz, 300MHz and 500MHz.
Next step is to turn on the spectrum analyser and break out the near field probes. I’ve got a selection of commercial and home made probes but the ones I keep coming back to are the give-away probe cards that I have on my exhibition stand at trade shows.
The location of the emissions for the 180MHz and 300MHz emissions was initially puzzling. Mostly it was centred around the drain of Q1. If we consider the operation of the circuit, Q1 is turned on permanently in boost mode with Q3 acting as the normal switching element and Q4 acting as a synchronous rectifier. Where is this switching noise coming from?
Those of you familiar with synchronous switching converter operation will be shouting at the screen right now. Of course, the answer is bootstrapping.
The high side N channel MOSFETs Q1 and Q4 need a gate voltage higher than their source voltage + their threshold voltage to turn on. In this kind of circuit, this voltage is derived from the switching node via a bootstrap circuit.
This explanation on bootstrap circuit operation from Rohm saves me from re-inventing the wheel.
Even though Q1 is nominally on all the time it still needs to perform a switching operation with Q2 to charge up the bootstrap capacitor powering it’s gate driver circuit.
Checking the datasheet, this switching operation takes just 100ns. That’s very fast indeed and explains the source of our switching noise!
The same bootstrap operation is happening to provide the drive voltage for Q4 but because the boost node is continuously switching this voltage is being provided without such a short switching event.
Due to space constraints it wasn’t easy, but I managed to get the microscope out and modify the board to accept a small but high current ferrite bead in series with Q1 drain.
It didn’t take long to narrow down the 500MHz emissions to the boost output diode D1 with a large amount of ringing on the cathode.
The interesting thing about this diode is that it is only conducting for a very brief period in the dead-time between Q3 turning off and Q4 turning on. Dead time between these parts is set at 75ns, again a very short time period. Good for reducing switching losses, disadvantageous for EMC emission.
The part selected for this was a slightly electrically over-rated 40V 1A, SMB packaged part with a reasonable capacitance. Switching 1 amp of current through this part for only a brief period of time before shorting it out and discharging the diode capacitance was causing the ringing to occur.
A ferrite bead was added in series to damp this as the customer wasn’t too keen to head down the rabbit hole of investigating specifying a lower capacitance rated new diode or looking at whether the diode could have been removed altogether at the expense of slightly higher power dissipation in Q4.
Interestingly, this is what the emissions looked like with the diode removed but still with the lower frequency emissions present from the input transistor drain. Note the wideband reduction in emissions above 300MHz.
With both of the ferrite beads in place the emissions profile of the EUT was reduced to meet the Class B limits. With more time the peak at 160MHz could be investigated and further reduced but project time pressures and the customer understandably wanting a “good enough” result meant we concluded this investigation here.
DC/DC converters are often provide a challenging EMC opponent when it comes to radiated emissions. I was glad of the opportunity to work on this project and provide a successful result for the customer. This is the kind of work that I love.
The advantage of being an EMC-consultant-with-a-test-lab combined is that this kind of work can be compressed into hours of work rather than days/weeks oscillating between your lab and the test lab. Problems Fixed Fast!
I hope you found this piece useful, get in touch via the usual channels if you have any questions.
A new customer came to me with their product that was having problems during testing at another laboratory. There were radiated emissions problems (mostly solved with improvements to the ground plane scheme on the PCB) and a very interesting (and challenging) ESD problem which I’ll cover in this blog.
Here was the device exhibiting the problem, a Diodes Inc AP22802AW5-7 “power distribution load switch”. Input VBAT from a stick of AA batteries, SW_PWR from a rotary switch, and output to the rest of the circuit.
The ESD problem was described by the customer:
The EUT stopped working when 4kV contact discharges were applied on discharge point shown. I removed the batteries and I put them [in] again and there was not any response from the sample (no otuput and the green LED remained OFF).
[A second sample] was then tested with the same result, although this time not on the first discharge
Upon inspection both devices had failed due to the load switch (AP22802AW5-7Diodes), with one failing open and one failing short and both becoming very warm.
ESD diode placed on input and output of load switch (with no effect)
ESD diodes placed on all [discharge points] (with no effect)
ESD diode places on VCC close to pullup resistors for [discharge points] with no effect
First thing first was to get the product set up on the ESD table (with a bit of added blur to protect the innocent).
It was very easy to re-create the problem observed at the original test lab with the second contact discharge to the EUT exposed contact point causing the unit to shut down.
In each case, the power switch was failing low resistance from IN to GND. The initial theory was that the device was being damaged by the high voltage punching through the silicon layers leaving a conductive path.
I made a series of experiments to determine the coupling path into the problematic device. Working on the principle that, because of the 15cm distance between discharge point and problem device, that conduction might have been the problem.
Whilst none of these experiments were successful they certainly helped eliminate conduction as the coupling path.
Because of the very high frequency content of the ESD pulse, capacitive coupling is likely going to be the dominant coupling method. Whilst it could couple into the device directly, there was more opportunity for the pulse to couple into the traces connected to the device first. Filtering the inputs eliminates two coupling possibilities
The PCB was starting to get a bit tired from the repeated hot air SMT de-soldering and re-soldering so I swapped to another supplied sample. To be able to operate the unit out of the casing I swapped to a linear DC bench supply instead of the AA batteries.
This proved to be an interesting mode as it allowed me to kill the power quickly. The next set of experiments were in an attempt to reduce the effect of capacitive coupling to the problem device.
Setting the current limit on the DC supply to a fairly low value (about 20% higher than nominal current draw) was a good idea.
When applying the ESD strikes the supply went into foldback as the EUT power input went low resistance. I discovered that quickly turning off the power and then turning it back on effectively reset the failure mode of the device. This proved to be repeatable over several discharges: zap – foldback – power cycle – EUT OK.
What silicon component behaves like this? A thyristor.
This is a phenomena known as “latch up” where the parasitic thyristor structure present in the CMOS process fires due to over voltage… such as an ESD strike for instance!
Because the device is only small the power dissipation caused by the battery short circuit current is enough to “pop” the device through overheating.
Whilst it doesn’t get used very often, my Sony Tektronix 370 curve tracer is perfect for testing components like this.
Here’s the VI curve of an undamaged device. It’s a bipolar voltage between VIN and GND. On the left of centre is the standard forward biased body diode. On the right is the reverse biased breakdown of around 8V.
Now for a damaged device. In this case the current changes quickly for a small applied voltage and there is no non-linear characteristic. Essentially, a short circuit.
Turning up the maximum voltage that the curve tracer can apply and dialling down the series impedance allowed me to simulate the over voltage fault condition and create a latch up condition. This latch up wasn’t permanent due to the bipolar sine wave nature of the curve tracer applied voltage.
However turning up the voltage enough to cause excess power dissipation inside the device did result in the same failure mode using the curve tracer.
I have never encountered a device that is this unusually sensitive to ESD events before. A nearby 2kV discharge on the PCB top layer ground plane was enough to cause the latch up condition.
I noted in the report to the customer that this device had been changed to “not recommend for new designs” by Diodes Inc. I wonder if they identified this condition in the device and withdrew it for that reason.
The customer resolved the issue by replacing the device with a different part and we all lived happily ever after.
It has been a very busy few months at Unit 3 Compliance; it feels like the chamber turntable hasn’t stopped spinning. There has been a wide range of products through the door from prosthetics to video wall controllers, from high spec IoT products to motion sensors, from lighting power supplies to RF amplifiers. I really love the variety of work!
I’ve also had some safety assessment work to carry out on which is always interesting. Disassembling mains transformers to measure the creepage distances inside is fascinating, getting out the angle grinder to hack the laminations apart just adds to the fun.
There have also been a fair amount of design reviews and general consulting work in between. To be able to work with customers right at the start of the project is invaluable as it sets them on the right path without having to find problems further down the line.
I’ve found lots of interesting nuggets of EMC information during this process that I’m looking forward to sharing with you in some future blog posts once I get time to sit down and write them up.
I managed to escape down to Lincoln to speak at the Open Source Hardware User Group oshcamp18 meetup on the subject of EMC testing. The delegates came up with lots of good questions at the end and the audience participation (see below slide) of the talk went down well. Higher! Lower!
Good to see reconnect with some old contacts and make some new ones. The other talks were very interesting also, lots of good work going on in the open source hardware field at the moment.
Lastly, and it snuck past without me spotting it, Unit 3 Compliance had it’s first birthday. It’s been a whole year since I got the keys to the unit. In that time, and with lots of help, it has gone from this:
And finally this:
Here’s to the next 20 years of compliance, I hope to see you on the way.
I had the pleasure of speaking at the local IoT Leeds meetup, held at the newly (and finally) renovated “Platform” venue just outside of the train station. The views over Leeds city centre from our 10th floor suite where the event was being held were impressive.
Paul Wealls of Mokanix had the first slot and gave an interesting talk on their development toolkit for mobile applications. I’m not a software engineer (my code makes grown men weep) but the usefulness of this cross network toolkit was evident and seemed to generate a lot of interest among the attendees. Well worth checking out if that’s your bag.
The synopsis of my presentation, titled “EMC for IoT” was a very high level overview of the European EMC and Radio Regulations for IoT hardware, CE marking, and the importance of spectrum protection through compliance. Because I didn’t know the makeup of the audience beforehand it was tricky to know exactly what sort of content to put in. I opted for more graphics than words which gives more flexibility in the subjects covered.
In the end, in the audience of around twenty people, there were two (and a half) hardware engineers, mostly software engineers but everyone had a technical background. As such I skewed the presentation towards the general but still went into detail where necessary. I really enjoy speaking on EMC matters and it’s great fun to share knowledge with like-minded interested professionals.
The questions received after the presentation were all very intelligent and well thought out. They ranged from instances of counterfeit components causing EMC problems, understanding risk, the costs involved and the use of mobile devices on aircraft. Despite what can be seen as a dry subject the feedback from the presentation was very good.
photo credit Will Newton
It was also great to see my ex colleague Chris who I hadn’t seen for a long time and catch up with him. We both headed to the Leeds Brewery Tap for a pint of Midnight Bell with Will (organiser) and Neil and for a chat that ranged from the local job market to AI before hopping on the train home. All in all, a good evening!
I’m looking to take the talk out on the road for other meetups or organisations so if you know anyone running an technical event where this would fit in then I’d be interested to hear about it.
A customer requested some support with one of their products, an IoT bridge device that takes various sensors and provides telemetry back to a central server using a GSM module. Some of the radio pre-compliance spurious emissions testing had suggested there might be some issues at certain frequencies.
After a couple of hours of radiated emissions measurements in the anechoic chamber and some bench work with some near field probes, I’d developed a pretty good idea of what was going on in terms of where the emissions were coming from and what their radiating mechanisms were.
Interestingly, there was a common theme to all of these emissions…
These features are common to a wide range of similar devices so some notes and a simple drawing (oddly I find sketching like this a good way to relax!) are presented in the hope it will give you some ideas about where your radiated emissions might be coming from.
The sketch shows a keypad board, a CPU board and a battery pack. Some other information is missing to permit a simpler drawing. All of these boards below sandwich together nicely into a plastic case which was the starting point for the investigation.
The problem frequencies identified were a 300MHz narrowband spike and a 250MHz broadband hump. Usually when I see broadband I think “power supply noise” and narrowband I think “digital noise”.
Let’s take a wander around the device.
Capacitive plate near field probing around (A) showed higher than background levels of 300MHz noise around the front panel button board. Since this was a “dumb” board, the noise was probably coming from the main CPU board. The noise emanating from the cable (B) was not appreciably higher but when approaching the CPU/memory the noise increased, the clock line between the memory device and CPU being the highest.
Two possibilities were that there was crosstalk on the PCB at (C) or perhaps inside the CPU itself but without getting into more complex analysis the exact cause is not known. Apart from the power lines, there was no extra HF filtering on the data lines, just a series resistor on the I/O lines of the CPU. The addition of a small capacitor (e.g. 47pF, either 0402 or an array) on each line to circuit ground forms an RC filter to roll off any unwanted HF emissions like this. I generally advocate making provision for such devices on the PCB but not fitting them unless required – better to provision for and not need than to require a PCB re-spin later in the development cycle.
Moving the near field probe around the bottom of the case where the battery lives (D) showed the broad 250MHz hump present on the battery. Unplugging the battery pack made the emissions drop by 10dBuV/m and measuring with a high bandwidth passive probe showed broadband noise present on the outputs of the battery charger (E) from the switching converter. Some low-ohm ferrite beads in series with the battery terminals will help keep this noise on board and prevent common mode emissions from the battery and cables (F).
Lastly, the antenna was unplugged and some other broadband noise was found on the cable (G) at 360MHz, this time from the main 5V DC/DC converter on the main PCB.
So what is the common theme? All the radiation problems stem from cables connected to the main PCB. As soon as you add a cable to a system you are creating a conductor with a poorly controlled return path or “antenna” as they are sometimes known in the EMC department!
Treat any cable or connector leaving your PCB as an EMC hazard. You have less control over the HF return paths in the cable environment than you do on the PCB. Apply appropriate HF filtering to the lines on the cable and remember that even a shielded cable can cause problems.
Sometimes, like the antenna cable, there’s not a lot you can do about it other than practice good design partitioning to keep noisy sources away from the cable and to apply a ferrite core around the cable if it becomes a problem during testing.
I hope you found this useful and that it has given you some pointers for looking at your own designs with a new perspective.
It feels like it has been a busy couple of months here at Unit 3 Compliance with a wide variety of projects coming through the door.
Q1/18 is already shaping up to be busy with some really interesting products booked in for pre-compliance testing and some nice meaty problems to get our teeth into. I’m looking forward to sharing some of the insights I gain from this work with you.
Here’s a quick roundup of what’s been happening…
Our key area of expertise and always the cornerstone of what we do here at Unit 3 Compliance is EMC pre-compliance testing. In the chambers recently we’ve had ticket machines, water boilers, development kits, and a light/motion sensor. Some with problems that we quickly fixed and some sailed through first time.
One particularly interesting product was an industrial lighting system that needed radiated RF immunity testing at 20V/m. This test loves to mess with products by turning on or off semiconductors that were quite happy as they were thank you very much. In this case, there was an transistor based current limiting circuit that, thanks to one of the transistors demodulating the RF carrier, decided to shut down key parts of the circuit. Replacing it with a resistor removed the problem allowing the customers development cycle to continue.
A customer has been leasing the anechoic chamber to make some antenna pattern measurements on a complex microwave antenna system. By loading up the quiet zone of the chamber with extra microwave absorber we were able to provide a highly anechoic (low reflection) environment all the way up to 18GHz.
As part of this exercise we made some rough background noise measurements from 2GHz up to 18GHz revealing very little. This suggests that when we reassembled the chamber in its new home we didn’t leave any gaps!
The vibration shaker and amplifier have been fully commissioned after their move. They’ve been getting a good run in performing a 2g sine sweep test on a large 25kg rack mount power supply.
Jigging equipment onto the vibration table is always a challenge, especially for a large and heavy piece of equipment like this one. I like to use 1″ x 1″ x 1/8″ wall aluminium box section (really stiff and light) along with high tensile M10 threaded bar to clamp an EUT of this size. Smaller EUTs can be easily secured to lighter platforms using hot-melt glue, surprisingly effective!
I always find vibration testing fascinating, especially watching various components come in and out of resonance during a sine sweep test. It’s fun to draw parallels between mechanical and electrical resonance, stiffness, impedance and damping.
In this case we found a large resonance that caused a fracture of the base plate due to excessive motion. We suggested a few approaches to stiffening that area, one of which was implemented and successfully removed the resonance.
One piece of equipment I’m going to be designing soon is an LED strobe lamp that synchronises to the output of the vibration controller so that any flexing in resonant modes can be easily spotted. That will make analysis much easier.
We’ve carried our several sets of schematic and PCB design reviews, from motion sensors to heater controllers, from pump monitors to semiconductor development kits.
Our approach is not only to look at EMC / system level but also to question and educate designers on alternative circuit choices based on our long experience in electronics design. This is part of the value that we give to our customers.
In each case we’ve addressed the circuit design, considering the EMI phenomena and levels that the ports of the design will be exposed to. This is where understanding the tests themselves is so important otherwise the circuit could be susceptible to problems.
We also look at design partitioning in some detail. This is one of the easiest ways to achieve good system level performance (and not just from an EMC perspective) by segregating the design into digital, analogue, power supply and I/O areas with the aim of keeping noise currents where they should be and away from their potential victims.
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.
The main message I took away was that IoT security is achievable but requires a lot of thought up front at the design stage including selection of the primary hardware to enable secure identification of the IoT node. Obviously the level security should be aligned with the value of the asset being secured and that is down to the security risk assessment performed (or not!) by the manufacturer.
First we enjoyed a superb talk by Ken Munro from Pen Test Partners on ethical hacking of IoT devices and how compromised security of one device can lead to a whole heap of problems. Examples included a large DDos botnet made from compromised CCTV DVRs, plain text passwords stored in a Wi-Fi kettle and hacked building management systems being used to mine Bitcoins.
Microchip Technology Inc. gave a talk on their ECC508/608 and demonstrated how it can be used (with Amazon Web Services) to uniquely identify and verify an IoT node in an easily embeddable package. This was a well delivered presentation that covered
Following on from that, there were some good presentations from NXP Semiconductors on their A1006 devices and smartcard based signing ecosystem, ST Microelectronics on their ST32 microcontroller security and ST Safe modules, and lastly Arm on their mbed OS and mbed cloud security features.
Overall it was a well organised event; the Digital Exchange building on Peckover Street has been well converted into a conference and working space. It was nice to meet some new faces as well as catch up with old ones. My only regret is that I couldn’t stay for the fine Bradford curry afterwards as I had a train to catch.
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.
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.
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.
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 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.
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.
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.
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!