Sunday, March 15, 2015

Smog Alert

The worldwide push to increase electric motor efficiency that I spoke of in my last blog has some major implications for OEMs that incorporate motors in their products. For OEM's using induction motors, achieving the required IE3 efficiency requirements may mean adding electronics where none had been employed before. All the major semiconductor manufacturers have jumped into this space with components and processors to implement V/F or Field Oriented control inverters. Drive manufacturers, too have off the shelf drives to offer.

Texas Instruments
ST Microelectronics

However, the OEM that is incorporating inverters in their product for the first time must address a problem they may not have had before; electromagnetic interference or EMI. All conventional inverter technologies control the motor by chopping the voltage to the motor stator into a 4Khz to 20Khz pulse width modulated square wave (PWM). One side effect of PWM is that the voltage to and on the stator now has high frequency harmonics that extend well in the radio spectrum. We can think of this as a radio frequency smog raising the noise level of the radio spectrum and making it more difficult for radio based communication to occur.

                          Motor Voltage Spectrum: Conventional Drive

If not constrained, this energy can appear on the motor leads, the power leads into the inverter and even on nearby conductive or magnetic structures that resonate at the harmonic frequencies. Industrial drive manufacturers and system integrators have dealt with this problem for years and have developed a portfolio of tools, techniques and products to solve EMI problems, typically on a case by case basis. However, for the OEM new to inverter development, these solutions may be too costly or too application specific. While the basics of EMI mitigation are the same for the industrial drive and a dishwasher, the business constraints of cost, size and repeat-ability are worlds apart.

At Digital Power Engineering, we had EMI in mind when we developed our Resonant Field Exciter technology (patent pending). Our EMI mitigation approach is to eliminate the problem, or at least make it a lot easier, by not producing the high frequency harmonics in the first place. By using a wound field synchronous motor with a Resonant Field Exciter providing the rotor field energy, the stator voltage need not be pulse width modulated, resulting in a much cleaner EMI excitation spectrum.

                              Motor Voltage Spectrum: WSM with RFE

For applications that turn at a constant, grid frequency related speed, using RFE technology means there is no inverter at all and the EMI footprint is little different than the original induction motor. For applications that are either driven from a DC source or need to be variable speed, the electronics between the source and the motor stator only steer, or commutate, the source power into the motor leads in sync with the motor rotation with no PWM. There is still power switching occurring, but it is at the rotor pole speed, which is typically less than 400Hz and usually 50 or 60Hz.

As a result, EMI mitigation, if required at all, is not only much simpler, smaller and less costly, but more effective over a broader range of installations and environments.

For more information on DPE's Resonant Field Exciter technology drop me an email at or visit our website at

Saturday, February 14, 2015

Staying the Course to a More Efficient Future

A few days ago, I posted a link on Linkedin to a recent IEEE article on U.S. electricity demand ( ).  Domestic electricity per capita has been flat since 2007. Much of this is attributed to energy efficiency efforts, including legislation that raised the bar on acceptable, minimum efficiency levels, including minimum efficiency of the largest consumer of electricity in America; electric motors.

The International Electrotechnical Commission (IEC), an international, non-government, consensus based standards  organization founded in 1906, has established four levels of motor efficiency to date.

Level IE1 was pretty much the worldwide de-facto standard for motors operating under 690V around 1990. In 1992, the U.S led the way by adopting IE2, effective in 1997. Followed, albeit slowly, by the rest of the world. By 2013 most of the world had adopted IE2.

Almost everyone in the U.S. is aware that legislation in 2007 raised electric lighting efficiency standards, ushering in the era of CFC and LED lighting, but few know that the same legislation raised the bar again on electric motor efficiency, to the IE3 level, effective in 2010. This time the world was quicker to follow. By 2017 the IE3 minimum efficiency standard will be the law of the land for new motors manufactured almost worldwide.

At the moment, there is no legislation in process anywhere in the world to implement IE4. At least not yet. But there it sits as the "holy grail" of motor efficiency. Many OEM's that use IE2 motors in their products aren't waiting for legislation. In the pump, compressor or HVAC worlds IE4 provides a significant competitive advantage over the IE2 motors they now have to design out. As long as they have to pull out a white sheet of paper, they might as well go IE4 if they can.

But IE4 technology can be expensive. Induction motor technology alone can't get there without adding electronic drives, and not every induction motor can be driven by an electronic drive. Permanent magnet technologies also need drives and have the hidden variable cost and long supply chain of rare earth magnets. Switch reluctance technology is so different from conventional motor and drive manufacturing that a new manufacturing infrastructure must be built to provide them in volume, and they can be noisy.

At the Motor and Drive 2015 conference in Orlando in January, we at Digital Power Engineering introduced another solution. A technology that takes a motor configuration that's been around for 120 years and is the electric motor of choice above 400 HP, and makes it practical at low HP as a motor that exceeds IE4 efficiency, has no magnets, requires no drive for single speed operation and is compatible with contemporary motor manufacturing. If you or anyone you know uses a motor above 200 watts in their product, they may already know about the need to shift to IE3. Have them drop me an email at and let's start the conversation about an accessible IE4 solution that can provide them a significant competitive advantage.

Monday, January 19, 2015

Eagles, Shields and Sparks: The New World of Electronic Hardware Prototyping

Now that the new year has started, I'd like to share a couple product development trends we've been using at DPE over the last year or two.

At some point in the development of any product, a prototype is built. In electronic products over the last 40 years, I've seen this process morph from classic bread boarding on perforated fiberglass boards, through wire wrap, to PC based CAD and inexpensive, small quantity PCB's ordered over the internet.

Likewise, forty years ago, electronic systems used hardware to compute and make decisions. Today, we use software to do that.

Physical products that still have to interface with the real world are becoming platforms. These are dedicated devices with the hardware necessary to sense and/or manipulate the real world and enough processing power to carry out the tasks. However, what the platform actually does may change over time through reprogramming.

At the same time, the cost of producing an electronic hardware prototype has fallen dramatically. Microcomputer modules such as Arduino, Nucleo and Spark run $20 or less, have the computer power of a $1000 computer board of the mid 1990's, can be as small as a postage stamp and have built in capabilities like Wi-Fi that were unheard of then. In the past, many products required substantial development of a display or other human-machine interface (HMI). Today, most of us are literally carrying around our own personal HMI; our smartphones and tablets. With a microcomputer module as the brain, a smartphone or tablet as the HMI and the hardware necessary to interface with the real world built onto a "shield" (or equivalent), the prototype platform can be ready for software development in as little as two or three weeks.

The result of this evolution is to dramatically shift the entry fee of new products, services and whole industries from capital to labor. The advantage accrues to those teams who have the necessary skills. Think of the process as employing corporate "sweat equity". The more the team skills match those required, the less "sweat equity" is expended and the faster the improvement, product or service hits the market.

If "Shield" prototyping isn't a strong-point for your organization, Digital Power Engineering can help. Depending on the circuit complexity, "shield" prototype turn around can be as little as two or three weeks.

Drop us an email at to start the conversation.

Saturday, July 20, 2013

Open Source Motion

For more than three decades I’ve been a part of an evolution that has combined the microcontroller with power electronics. Over that time, motor drives have evolved into useful building blocks; however they have become “black box” building blocks. While they may use standard communication protocols such as Rs485, Ethernet or Can, each manufacturer created different twists to their product in the name of product differentiation. Drives got smarter, but that intelligence was hidden behind a veil of proprietary user interfaces. External controllers can communicate with the drives for more complex applications, but standalone field programmability is limited. 

Contrast this with the smartphone. It too serves a “black box” function, communication, but by embracing operating systems such as Android to control the phone hardware it flaunts its intelligence and allows users and third parties to create apps that may or may not use the “black box” functionality. Critical phone functions must be accessed through a strict Application Programming Interface (API), but other resources such as memory, USB, Bluetooth, Touchscreen and WiFi are available for general programming by users or third parties through the operating system. To achieve the same flexibility in today’s drives, it is necessary to have an external “brain”, such as a single board computer or PC to run system level software. 

While this is reasonable for stationary applications or small production runs, it can be challenging and expensive for mobile, off grid, standalone or higher volume applications. By incorporating Android, Linux or other “open source” operating systems, drive manufacturers can release the creative potential of their users yet keep the drive operations proprietary. In many systems the drive is already the largest consumer of power and space and an open source drive would eliminate whole cabinets of power supplies, computers and other gear. Several such drives and other hardware can be connected together through CAN or other standard interfaces to form an autonomous local network for flexible coordinated motion. 

In the ad hoc “Maker Culture” there are already examples of open source motion. Perhaps it’s time for the mainstream drive manufacturers to take a look. 

As always, questions, comments and suggestions are all welcome. I can be emailed at 

Maker Movement; 
Embedded Linux; 
Open Source Initiative;

Friday, June 15, 2012

Beyond Tesla

Nicola Tesla was a visionary. In an era dominated by steam engines he envisioned what today we would call a disruptive technology. His AC power distribution system and induction motor ushered in the 20th century, warts and all. Fast foward to 2012, 117 years after Tesla and Westinghouse first harnessed Niagra and we find we can still improve. 60% of the electricity in the United States is consumed by electric motors running at 60 to 70% effeciency. Tesla didn't have the power electronics, materials and tools we have today; but we do. By applying these tools we can increase grid to motor shaft effeciency up to 90% and add unprecedented functionality. That is the mission of Software Defined Power.

While we don't envision our mission quite as disruptive as Tesla's, we do view it as evolutionary. If it moves, we will make it more effecient and more functional. We do this by re-engineering the system piece by piece with the goal of having each piece functioning optimally. This applies not only to induction motor or DC motor systems but also to pneumatic, hydraulic and engine driven mechanical systems.

Challenge us. Give us a call and see what we can do.

Friday, May 25, 2012


Last night I lost a friend.

Tom Emmons, friend for 31 years, engineer, inventor and colleague passed away. Tom was one of those gifted engineers who could visualize the actual flow of electons and fields. Whether it was a high current PCB layout or the solution to an EMI problem, Tom's work was a true work of art.

Beyond that, Tom was a true friend. Always there, whether it was a technical problem to bounce off him or just getting together to let off steam. Always ready to help, even if it meant flying halfway across the country.

I met Tom at a little starup named CPT back in '81. Since then we worked together at Datacard in the late '80's, Aria Corporation in the mid '90's and just recently on some truly breakthrough aerospace technology. Over countless meetings and lab sessions we hammerd out solutions to a variety of engineering and other problems.

Rest well, friend.

Tuesday, May 22, 2012

Cutting the Power Cord

Cordless versions of small electric tools have been around for some time.  As discussed in my previous post, battery improvements and a "clean sheet of paper" design approach have led to significant progress in this area. However, there are power levels where the combination of battery power/ size/ cost just isn't there to make a direct battery version of a product practical. Today these applications are still only available as grid connected devices yet many of these applications are already electronicly controlled.  

What if we could cut the cord on these applications without changing the existing product at all? The answer is that we can by providing electronics to boost the voltage to grid levels.

This isn't a particularly new concept. One often used approach is to install a DC TO AC inverter between the battery and the product. The output of the inverter mimics the 50 or 60 HZ power line at 120 or 220 voltc AC. However, unless the load is an induction motor or has a 50/60 HZ transformer, this approach is ineffecient, unnecessarily costly and electrically noisy.

A better way is to take advantage of the fact that most grid operated electronically controlled systems first rectify the incoming AC to DC. So converting the low battery DC to high voltage DC works.  Put the DCDC converter near the battery, and the wires to the application need be no bigger than their AC counterpart. Size the battery to the mission and the battery size and weight are minimized. Since the DCDC converter can be smart, it can accurately meter battery power and provide an excellent level of safety. All without changing the design of the end product..

At Software Defined Power our DSP controlled DCDC technology can be scaled from peak power levels of a few watts to 5KW and above with effeciences better than 94% and power densities of  16W/cubic inch.

This technology opens all kinds of options for product and system portability. and we love brainstorming options with our clients, so call or drop us an email and start thinking "what if".