In searching for circuit simulations options I’ve learned a few things that are worthy of note.  When it comes to circuit simulation Spice as a tool commonly comes to mind.  Spice was originally developed at Berkley in the 1970s and development ended around 1990.  This version of Spice and the modeling syntax supported is what the current open source Ngspice represents.  I think this is also the level of modeling that LTSpice supports.  When searching for spice models it seems they are most commonly available for Berkley spice or PSpice.  Nearly all spice variants are compatible with the Berkley spice models. 

PSpice is from OrCAD and is a completely different modeling engine than Berkley spice.  PSpice supports modeling methods and syntax that can more closely imitate behavioral aspects of components.  In addition, PSpice models can include nodes for thermal modeling of components.  For some components only PSpice models are available.  HSpice is similar to PSice and has some compatibility with PSpice models.  HSpice is from Synopsys.  Simplis has the re–coded simulation engine that runs much faster than Berkley spice but it doesn’t seem to be compatible with PSpice models.

Since PSpice models are coded in a much different way it is not practical to modify a PSpice model to be compatible with Berkley spice.  The greater modeling accuracy of PSpice would be welcome too.

Xyce is a circuit simulation tool actively developed by Sandia National Labs and available for free.  It is only the simulations engine and so needs to be paired with another tool for visualization.  It has the advantage of a great degree of PSpice compatibility and was designed to run fast.  I’ve successfully imported a number of Berkley spice and PSpice libraries and used Python’s matplotlib to visualize the results.

Interestingly, it doesn’t require a dedicated circuit simulator to do circuit simulation.  I’ve used Python code and simple component models for simulation.  The problem is simple component models do not give a high level of signal accuracy.  The advantage is control logic can be implemented in Python.

Python can also simulate mechanical system whether done directly with code or by interfacing to a modelling engine like OpenModelica through PySimulator.  Scilab with Xcos, Simulink, and Octave are other options for general purpose modelling.

For 2017 the Embedded Systems Conference in the Minneapolis Convention Center was held on November 8 and 9.  This is the first year I’ve gone.  I was surprised by how few attendees there were to the sessions I attended the first day.  There were in the neighborhood of a dozen attendees in rooms that would hold over 50 people. I attended more of the hardware related sessions.  The low attendance may have been somewhat related to this.  Colleagues who attended other sessions noted similar attendance levels.  There was much stronger attendance on the second day with the 4 hour session I attended on migrating to an RTOS being over full initially then losing about a third of the attendees over the four hours leaving about 40 attendees.

I was also surprised to see no semiconductor vendors in the exhibitor area.  As a result I didn’t really get much from the exhibitors.  I spent a little time talking to test equipment representatives, PCB fabricators and stuffers and mechanical component prototypers but didn’t take away much I could use immediately.

The first session I attended was titled PCB Assembly in House on the Cheap.  The presenter discussed low volume and low-cost ways of building and testing PCB assemblies.  To have boards made the domestic example of Oshpark was given and Chinese made boards which can still be delivered quickly can be obtained from Smart Prototyping, Seeed Studio, and Dirty PCBs.  Some of the PCB suppliers will also offer a stencil for your board.  To populate the boards the presenter said a small SMT machine could be used like those from NeoDen but he had good success having high school students hand place components.  He described an optical inspection system he designed using a webcam and open source software to detect missing parts and questionable solder joints.  He also used printed jigs for consistent placement of leaded parts.  For test, he showed systems that used Arduino base and shield boards for IO and an android mobile device for the UI running an app developed in Python.

I attended several seminars given by Rhode and Schwartz representatives.  One focused on power integrity.  It was mainly about accurately measuring the ripple on a DC supply rail.  The presentation featured the advantages of using their RT-ZPR20 probe.  The common method of switching your scope to AC coupling and using a passive probe with a short ground lead has a couple problems.  One the AC coupling circuit inside the scope has a lower signal to noise ratio than DC coupling.  Another is that passive probes can not accurately pass signals above 100MHz.  One way to mitigate this is to solder the coaxial cable from the scope directly to the board.  This has the problem of presenting a high load to the supply rail you’re trying no measure.  It also does not mitigate the need for AC coupling.  The probe being promoted presents a high impedance load and a wide offset range so DC coupling can be used.

I attended another Rohde and Schwartz session on probing methods.  The need for active probes was highlighted when frequencies get into the hundreds of Mhz and GHz.  Loading over frequency is an important probe characteristic to consider.  They showed how their active differential probe with modular tip can mitigate many of the complications associated with signal measurement.

Rohde and Schwarz also offers equipment for IoT device testing.  They have a product that will help you determine how well your device can maintain a wireless connection in the presence other wireless signals that are likely to be present.

In their booth, I asked about high voltage signal measurement.  They suggested their RTH series scopes with isolated channels.  The probes look like traditional passive probes with the exception of having reinforced insulation and high voltage spacing between the ground ring and tip.  This enables high bandwidth measurements at high voltage.  The scope supports remote user interface in the case that a signal under lockout/tagout is being measured.  Under $4K for entry level models seems reasonable.

I attended a panel discussion on sensor fusion.  This is the idea processing data from multiple sensors can provide a higher value than the value that the sum of the information from the individual sensors could give.  Multiple sensors can also allow inferences to be made about an object or process that can’t be directly measured.  The discussion also related to the relationship between edge and cloud computing.  A helpful analogy was given of how the human spinal column will react to the sensation of burning on a finger by ordering the muscles to withdraw the finger.  The brain is not involved.  This is analogous edge computing.  After the finger is withdrawn a person will examine it and decide how to treat it.  This involves the brain and is analogous to cloud computing.

I attended two sessions given by DfR solutions personnel.  One was on selecting connectors and the other was common design mistakes.  It was interesting that they both mentioned the danger in using connectors with red phosphor fire retardant.  In the presence of moisture, it will corrode and cause opens or shorts.  The connector session recommended following the Tin Commandments, the value of gold plating and how much to use and not to mate gold an tin.  He explained ferreting corrosion is when a contact point moves its previous position will oxidize over.  This cycle can continue until the oxide layer becomes too thick and creates an open circuit.  Cheaper materials used in contacts like stainless steel and brass can be problematic.  Both presenters recommended the Toyota practice of test then design rather than design then test.  This ensures your design includes components that are already vetted.  The second presenter emphasized the importance of thermal management and recommended heatsinkcalculator as a useful tool.  I was unsettled by his assertion that ceramic capacitors of the X5R and Y5V dielectric can only be considered reliable for 5 years under moderate bias and temperature before they will start shorting.  I typically use the higher temperature rated X7R capacitors which were not mentioned.

The Jacob Beningo presentation of transitioning to RTOS based systems was helpful in understanding when an RTOS is appropriate and understanding RTOS concepts.  The examples given during the presentation were from FreeRTOS.  While he didn’t endorse FreeRTOS this suggests it is worth considering for a project especially considering there are supported and safety variants available and there is some level of integration with the STM32 Cube library.

The presentation on cyber securing medical devices by Battelle seems much like the effort required for UL 61508 certification.  There is not a hard requirement for passing certification.  It is more about being able to present a credible argument on how your product meets the safety thresholds.  For cyber securing devices the FDA has heightened concern over how firmware updates are handled.  To model threats, the presenter suggested assembling a data flow diagram and applying a threat assessment score to the points where the data is exposed.  OWASP and NIST800-53 are resources for modeling and STRIDE and CVSS are scoring systems.

I’m primarily an embedded engineer.  As such I’m not very familiar with mobile app development to enable the control of an embedded system via Bluetooth.  Much of the development in Bluetooth connectivity is based on the newer 4.0 BLE profiles.  The Anaren app builder that works with their BLE module.  The controls were sluggish and the Bluetooth connection seemed unreliable.  Part of this may be a consequence of the lower data rates supported by BLE as opposed to the higher Bluetooth Classic data rates.  I started their method of exporting the app to a more advanced tool but stopped when I realized it would take man-weeks to learn how to use it.  I also tried ST’s BlueSTSDK.  The example application compiles and functions well but it would take more time than I wanted to deconstruct the app and learn how to build my own app.

My search for a simpler and faster solution led me to an mbed example application.  It uses the HC-05 Bluetooth classic module that implements the Serial Port Profile.  For constructing the app MIT’s App Inventor is used.  I spent about two days building an app that provided a useful user interface for an embedded target.  I’ll suffer my loss of programmer’s pride for using the scratch like programming method for the benefit of the time saved in producing the app.

The expectation for seminars like this is that they are not going to be information treasure troves but there might be a couple nuggets that justify the time spent. The following are take-aways from attending presentations by Cypress, ST, MuRata, and Analog Devices and walking and talking on the vendor floor.

The Cypress presentation focused on PSoC 6. These parts can be had with two core in the part, an M4 and M0. They also have the option for a Bluetooth radio that will be version 5 compliant to come degree. They come with the configurable analog and logic blocks that we’ve heard about since the introduction of PSoC. To some degree the logic can be configured with Verilog code which suggests it might be capable of advanced logic functions. In the past the added development time for peripheral configuration and weak microcontroller has limited the appeal of PSoC. The PSoC 6 might now have a sufficiently capable microcontroller but most applications will probably still not want to incur the additional development time associated with peripheral configuration.

The ST presentation was mostly reiterated the topologies and solutions for offline flyback and buck conversion. Some of their offline converters might have some uniqueness in that they can be applied to either a flyback topology or high voltage buck topology with the way the feedback is implemented. Some of the DC-DC converters can be configured to use the high side as a circuit reference. This may have an efficiency advantage as well as other benefits that were not clear in the presentation. This probably deserves more research to understand the advantages of configuring the converter this way.

Murata has a number of BLE, WiFi, and WiFi + Bluetooth modules based on silicon from several manufacturers. The most intriguing thing was the various software solutions offered. Some of their modules are Linux system on chips. For these they suggested the Android Things platform to make including IoT services more turnkey as opposed to Yocto. For embedded wireless systems on a chip the WICED software platform from Cypress looked interesting. This can be combined with enterprise and cloud services from Ayla and Electric Imp. The IoT software platform concept was so interesting that a little subsequent searching revealed that Anaren builds on the WICED platform with the Atmosphere development platform. It appears to have the potential to make app and embedded development very fast and easy.

John Kruse gave an exceptionally competent presentation on the new medical device standard IEC60601 and Analog Devices’ solutions to meet it. The new standard increases the ESD and EMI withstand requirements while lowering product radiation requirements. The interesting part was that the devices recommended for these harsher than typical conditions where the next generation of ADuM isolators that compete with isolators from other suppliers. Experience has shown that the ADuM series performs better than other solutions when subjected to EMI so it’s encouraging to know the next generation parts will perform even better. John offered an interesting anecdote of using SP724 SCRs for ESD protection because of its high speed and energy handling capacity. John should be kept in mind as a resource for helping EMI/EMC problems.

One thing that stood out in the vendor hall were the large powder core inductors that Vishay and Pulse offered. They could be an option for high power buck converters.

Since there have been a few notices around the obsolescence of Cypress’ PRoC BLE module a conversation with a Cypress representative completed the story.  Cypress is obsoleting the PRoC BLE module but the PSoC 4 with BLE can be substituted since it is the same functionally.  The PSoC 4 will continue to be produced along side the more capable PSoC 6.

An inquiry at the OSRAM booth for high CRI white solutions received the answer that their some of their OSTAR RGBW parts include a high CRI white.  A little additional information gathering revealed that OSRAM will also have a high power tunable white solution called Splitstar.