Introduction
You probably haven’t heard of Foxtron, but if you follow electric vehicles (EVs), you will soon. This is Foxconn’s EV startup, created in partnership with Yulon Group (YLM), a Taiwanese car manufacturer of NISSAN designs. In 2021, the company introduced three EV models in Taiwan: a sedan Model E (sold by Yulon), a sports utility vehicle (SUV) Model C, and a passenger-transit bus Model T. In 2022, Foxtron introduced a new all-terrain utility pickup Model V and a lower-cost Model B sedan (Figure 1).
Though the company will begin producing these vehicles in Taiwan, the company has plans to roll out production on a global scale. Its bigger plan is to pivot into a hyper-flexible company that focuses on working behind the scenes to become the No. 1 EV manufacturer, even if nobody knows who built the vehicle they’re driving.
The Big Picture
This is organized by a Foxconn consortium called Mobility in Harmony (MIH). The MIH Consortium runs the Open EV Alliance, a new conglomerate built on the principles that Foxconn pioneered in building consumer products like PCs and smartphones. MIH hopes to foster innovation and collaboration to lower entry barriers for alliance partners. As Foxconn passed $216 billion in revenue, there is a lot of confidence that this success story can be duplicated in EVs.
My Perspective
I am writing about this topic from a unique point of view, as I was a chief technology officer in one of Foxconn’s business units for several years—one of the few non-Chinese employed by Foxconn. I was also one of the few Westerners to meet Terry Gou in Taiwan in 1983 when he was just starting out in the electronics component business. At the time, I was managing Hewlett-Packard’s Application Center in Taiwan. Gou eventually became chairman and CEO of Foxconn.
I didn’t hear a lot about Hon Hai, the company that would become Foxconn, until 2008, when I got a phone call from my old HP boss at HP-Taiwan, who had just become the executive vice president of Hon Hai and needed some help in China. From then on, I would learn a lot about Foxconn and Terry Gou.
How Did Foxconn Start?
Foxconn started in the 1970s by making plastic items for TVs. In the ‘80s, when the IBM PC started growing, Foxconn pivoted to making connectors and cables for PCs. This turned out to be a smart move; as the volume grew, so did the gross margin—to over 36%.
As it diversified its manufacturing portfolio, Foxconn became even nimbler and more adept in positioning itself as a leader in whatever it does. Thus, in 2020 when the company entered the EV market, its worldwide sales quickly jumped to $240 billion.
Unlike traditional car builders, Foxconn believes that simple EVs can be built like PCs, as modular elements; to explore that possibility, the company formed the MIH Consortium, which now has over 2,376 members in 64 countries. Through the MIH Consortium, Foxconn will supply startup EV companies with referenced EV designs and all the necessary components to make a new EV (battery, motors, electronics, displays, wiring, software, IC chips, and the body) with full assembly available as an additional service. A real “PC on wheels,” all at high margins for Foxconn.
Foxconn as an ODM
From my vantage point, Foxconn doesn’t hesitate to spend money. It is “vertically oriented” and will develop or build as many component parts for electronics products as it can, including video displays, power supplies, plastic molding machines and cast metal parts, PCBs, flex, connectors, RF/networking, earphones, cameras, etc. Foxconn is the owner of the two largest PCB/flex fabricators in the world: Zhen Ding Technology (valued at over $5 billion) and FIH Mobile Limited (100% captive but over $4 billion in sales)3-4. Expect these two to dominate electronics component production in their EVs, with EV sales projected to total over 40 million vehicles by 2030. Now, that’s a lot of electronics (Figure 2).
Foxconn designs many of the smartphones that other OEMs are marketing in China. They also now build a multitude of robots, many of which are used in their own assembly. This lineup will undoubtably be expanded to include the larger robots used for chassis assembly, painting, and welding. Foxconn has purchased HP’s inkjet module business and even assembles their inkjet printers. Due to automotive demands, Foxconn has purchased a semiconductor fabricator and partnered with NVIDIA for CPU chips, and the company is building a new wafer fab in Malaysia.
In 2001, Intel chose Foxconn to manufacture its Intel-branded motherboards and CPU modules. Next came the smartphone revolution, and it rode that wave with Apple from the very beginning by building iPhones. By keeping its exponential growth under the radar, Foxconn found its way into everyone’s household without becoming a household name.
If Foxconn wanted to finally step out behind from the curtain of obscurity, it had two options: buy consumer brands or move into entirely new industries. It chose to do both. But the real question is: Can all this verticalization for consumer electronics be extended to the EV business?
Can You Build EVs like PCs?
The PC market exploded because Intel essentially gave away its base PC design provided you purchased an Intel CPU; AMD has since followed suit. Now, just a few engineers can put together a working PC. At the time, Microsoft also made Windows software available at a very low price. Compare this situation to the current landscape with EVs. An EV is more than just a motor and a battery; there are many different electronics systems at play, including battery and charging systems, motor control, lighting, steering, braking, air conditioning, wipers, anti-collision and safety systems, entertainment, interior systems, body frame systems, and soon, autonomous 5G communications.
Other factors to consider in putting a car on the road:
To be successful, EV companies will need to differentiate their models to attract customers, requiring even single-EV companies to initiate periodic model changes and updates.
It took 30 years for Taiwan OEMs/ODMs (TSMC, Hon Hai, PCB fabricators) to build the infrastructure necessary to launch this initiative, and all while no one was watching. The Big Three automotive infrastructures (U.S., Germany, and Japan) have been building for 120 years and already have a multitude of component and parts suppliers, as well as the universities needed to educate EV creators. China and Taiwan, in contrast, are just getting started—but Foxconn has already made great strides toward making China and Taiwan major players on the EV stage.
Foxconn plans to ship 500,000 to 750,000 EVs by 2025, mostly to Asian customers; that’s up to $34 billion or $45,000 per vehicle. It plans to take the modular approach used for electronics and robotics to assemble the frame and put all the modules into these EVs.
MIH EV Design LLC, Foxconn’s joint venture with Lordstown (Ohio) Motor Corp. (LMC) will use the Mobility in Harmony (MIH) Open EV Platform to co-design and develop EV programs for the global commercial vehicle market by leveraging the services seen in Figure 3. Foxconn invested $230 million, with $100 million to start up, and gave LMC a $45 million loan “to support its initial capital commitment,” according to an LMC announcement8. Foxconn owns 55% of the new company and LMC owns 45%9.
Foxconn recently acquired a major LMC factory where it will help Fisker Inc. begin producing its PEAR model EV in 2024. An estimated 250,000 PEAR units are slated to be produced per year. Through the partnership, LMC is on schedule to begin producing an all-electric pickup truck, the Endurance, later this year. This will undoubtably be the training plant for automotive engineers as Foxconn and China do not currently train automotive engineers like the U.S., Japan, and Germany.
Strong Competition
Foxconn has just announced a plan to spend $201.9 million to build a battery cell R&D and trial mass production center at the Kaohsiung Hofa Industrial Park in Taiwan, with a goal to begin mass-producing lithium iron phosphate batteries in the first quarter of 2024.
But Tesla, with its multiple new plants and battery operations, has a long head start on the competition. Compare Tesla’s website to MIH’s—Tesla is already taking orders. Current internal combustion engine (ICE) manufacturers have also started working on EVs; 18 companies have introduced EVs to the market. GM has four models on the road, in addition to a very impressive EV architecture that the company has termed the ULTIUM program. GM’s offerings include five power trains for dual-wheel power with five additional power modules for single-wheel drive and dozens of battery configurations (Figures 4 and 6). They have also secured the cooperation of LG Energy Systems to build three new GM battery plants in North America.
What’s Happening Now
There are currently 57 EV startups, with most in Asia, and specifically 20 in China; several companies in China have produced ICE vehicles for many years. I believe many of these startups will flock to Foxconn to get access to its EV Reference Design and to start mastering the many technologies necessary to build a functioning EV. Established ICEs with their own manufacturing, on the other hand, are likely to stay away due to the fear of having to share their hard knowledge with all the newcomers (Figure 5). But issues of product design security and IP will be a major challenge for MIH, as new products must be introduced almost every year to keep the public interested in their product.
Summary
Only time will tell if Foxconn can deliver on its plans. Its success in the PC and smartphone business was aided by the small size of these electronics and the large effort by North America and the EU to outsource electronics manufacturing. Automotive and transportation are now national priorities for these regions, so duplicating the consumer electronics growth may be a tough road to follow.
Times have changed in the last 30 years; innovation is rampant. Already, many are trying organic adhesives to replace spot welding, while newer plastics have strength greater than aluminum with lower weight. Fuel cell technology for large trucks may be miniaturized for autos. The industry is also focusing on small, super-efficient hydrogen-powered electric generators as a means to produce the electricity these EVs need without the weight and cost of lithium batteries. As this industry continues to grow, I’m sure we’ll see a flurry of new ideas take shape in the next few years. Whether Foxconn succeeds in dominating that process remains to be seen.
References
Happy Holden has worked in printed circuit technology since 1970 with Hewlett-Packard, NanYa Westwood, Merix, Foxconn, and Gentex. He is currently a contributing technical editor with I-Connect007, and the author of Automation and Advanced Procedures in PCB Fabrication, and 24 Essential Skills for Engineers.
This column originally appeared in the February 2023 issue of PCB007 Magazine.
03-10-2023
You probably haven’t heard of Foxtron, but if you follow electric vehicles (EVs), you will soon. Foxtron is Foxconn’s EV startup, created in partnership with Yulon Group (YLM), a Taiwanese car manufacturer of NISSAN designs. In 2021, the company introduced three EV models in Taiwan and in 2022, Foxtron introduced a new all-terrain utility pickup Model V and a lower-cost Model B sedan.
View Story01-26-2023
Karl Dietz never wrote on automation and the Smart factory, but these topics have been a priority for larger OEMs since the early '80s. I became involved in automation planning after designing and building Hewlett-Packard’s newest printed circuit fabrication facility in Sunnyvale, California. The journey to a Smart factory is evolutionary; it starts with a thorough business plan that charts a roadmap for your enterprise into the future.
View Story01-05-2023
The printed electronics sector is presently an area of great interest to many in the electronics manufacturing industry. Because of their incredible utility, printed electronics are poised to generate tens of billions of dollars in the coming years. According to IDTechEx, the total market for anticipated and potential printed electronics was near $62 billion in 2019 and expected to grow to nearly $81.8 billion within ten years. Earlier projections of a $300 billion market were floated a few years ago, but the $62 billion figure is still considerable and roughly equal to the current value of the global printed circuit market.
View Story12-08-2022
Karl Dietz never wrote on the topic of palladium as a final finish, but he did write about gold plating as a final finish and had an excellent discussion on copper plating. But palladium now has a renaissance as a final finish. It was very popular in the 1970s, as the only other final finishes were tin-lead reflow-Ni/Au-OSP or immersion tin. Palladium was very popular with the automotive industry then and a major supplier of boards was Photocircuits of Glenn Cove, New York.
View Story10-27-2022
It is hard to believe we have had direct imaging for 40 years, starting with Excellon’s introduction of its DIS-2000 argon laser imager. Since then, companies in 11 countries, ranging from Israel to Norway1, have developed various digital direct imagers. Karl Dietz wrote about this technology several times in his columns2, noting early on the skepticism that laser direct imaging would hit a Golden Age. So, where are we at now?
View Story10-10-2022
Complex build-up HDI technologies continue to expand in applications. Copper electroplating of vias has been perfected but the process is yet another electroplating solution to maintain and can sometimes be a lengthy process. Current conductive paste fills are not as conductive as solid copper but provide reduced cycle time, are still highly conductive and are cost effective.
View Story08-30-2022
Since the first inkjet printer appeared from Hewlett-Packard in 1980, engineers have been trying to use it in printed circuit manufacturing. The first successful application was by HP PCB engineers in 1983 that created an inkjet printer mechanism to serialize each PCB with a unique S/N for traceability. They used one of the UV inks HP developed that worked well on circuit boards but not suited for use on paper.
View Story08-02-2022
In this month’s column, I will discuss optical alignment for pinless lamination stackup, a topic that complements the induction lamination in my November 2021 column. Pin tooling plates have been used for lamination since it first started sometime in the 1960s. I first encountered multilayer stackup when I was assigned to increase capacity for our multilayer output in 1972. This was to accommodate the growth of our computer business. Unfortunately, the explosive growth of our calculator orders in 1973 required that we look for numerous vendors to produce the six-layer logic board in the HP-35 calculator.
View Story07-19-2022
Let’s have a little fun and walk back nearly 70 years into the history of electronics and computers. What was the world’s costliest computer and why? The answer is not today’s supercomputers, nor computers built during World War II. Instead, it lies in a real-time air defense radar system built during the height of the Cold War of the 1950s that had left the U.S. extremely vulnerable to a Soviet bomber attack. This was the beginning of a North American strategic defense system, eventually known as the Semi-Automatic Ground Environment System (SAGE).
View Story05-31-2022
Copper has become a valuable metal, and with the growth of EV has come higher currents needed in PCB with increasing weight of copper in PCBs. This creates the need for increased copper etching and consumption of copper etchants. Today, in an effort to recoup some of that cost, increasingly more extraction and recovery units are being installed in PCB facilities around the world. Annual profit generation from recovering copper and regenerating PCB etchants has the potential to reach six figures.
View Story05-02-2022
In 1965, Gordon Moore predicted that the number of transistors that could be packaged into a square inch of space would double every year for the near future. Although his projection was later revised to every 18 months, Moore’s Law has withstood the test of time for five decades. Today, we are beginning to see obstacles to this type of exponential growth due to the inherent limits associated with silicon lithography, packaging of the devices, and component placement on PCBs.
View Story03-28-2022
I have spent many years in printed circuit fabrication, including nearly 20% of my career in Asia. One problem that concerns all fabricators is the issue of “How many ‘X-outs’ are allowed per assembly sub-panel array? Here are a couple of solutions I have used and encountered in my travels.
View Story03-07-2022
The expansion of IC functionality usually progresses with the shrinking of IC geometries, called “Moore's Law” after Gordon Moore who first coined the phrase. But now that geometries are below 5 nm, the costs and difficulties are creating a barrier to much further advances. So, the solution seems to be to mix IC die on the same substrate as a system-in-package (SiP) that is now called heterogeneous integration (HI).
View Story12-20-2021
Printed Electronics (PE) continues to be a growing technology. But one of the advantages, as well as a drawback is using low-cost substrates, like paper, that cannot take the temperature of solder paste reflow. Also, the inks need to be cured. One current way to cure the printed inks is with ultraviolet radiation curing, such as used with solder mask or legend inks.
View Story11-23-2021
Multilayers have been around about as long as the printed circuit. The industry has always used heated hydraulic lamination presses to produce these multilayers, with the introduction of vacuum assist in the 1980s. But recently, with the encouragement of GreenSource Fabrication, induction lamination has been perfected by Chemplate Materials of Spain. Chemplate had introduced the use of induction-pinning by optical alignment of innerlayers for multilayer stackup in the early 2000s. This was to go with another innovative way to laminate innerlayers together—the Italian CEDAL resistance-foil vacuum-press, which had some early adopters.
View Story10-22-2021
The industry has not had many new structures in the last 60 years. Multilayers have continued to evolve with thinner materials, smaller traces / spaces as well as drilled vias. It’s been nearly 40 years since Hewlett-Packard put their first laser-drilled microvia boards into production for their innovative Finstrate process.
View Story07-05-2017
In the 1990s, the National Bureau of Standards was distributing a popular statistical document, the Handbook 91, written by Mary Natrella of the NBS Statistical Engineering Laboratory. A request by Patrick Spagon of the Statistical Methods Group of SEMATECH to update the NBS Handbook 91, Experimental Statistics, led to the creation of a project team from NIST and SEMATECH to create a new web-based statistical handbook including statistical software.
View Story12-16-2016
New product realization and design for manufacturing and assembly (DFM/A) have now started to become more visible as programs that can improve a company’s time-to-market and lower product costs. Many programs are underway by many companies and what is now needed is a framework to coordinate the application of these programs. This column will cover the interactions of DFM/A and the need for development of a new framework to coordinate the trade-offs.
View Story11-22-2016
From Happy Holden: A long-time printed circuit-industry friend of mine, Martin Tarr, an instructor at University of Bolton, UK, is a leading expert on change. He wrote an excellent tutorial for his university course on electronics manufacturing. With permission from Tarr, I am including a portion of it here as the basis of this column, starting after the graph in Figure 2. But first, a few thoughts of my own.
View Story11-03-2016
It takes more than just a good idea to exploit that brainstorm of yours. Hewlett Packard’s “10-Step Business Plan Process” is the format to present an idea or product in a fashion that will answer most questions that management may have about a product or idea.
View Story10-19-2016
Lean doesn’t have to exist in manufacturing alone. Lean is a fairly recent principle that can apply to all of our goods and services. For those of you not familiar with Lean, I recommend the free E-book "Survival Is Not Mandatory: 10 Things Every CEO Should Know about Lean" by Steve Williams, a regular columnist for I-Connect007.
View Story10-05-2016
After 20 of my columns, readers probably realize that I am an analytical person. Thus, I dedicate this column to metrics—the method of measuring something. I mentioned the four levels of metrics in my June column "Producibility and Other Figures of Merit." I also introduced the five stages of metrics in the second part of the column "Design for Manufacturing and Assembly, Part 2." This column completes the discussion with a look at dimensionless quantities.
View Story09-29-2016
Hopefully, your career has progressed to the point that you are empowered to recruit your own team or a key person for your team. There are always technical people looking for better jobs, but many times, the most talented are busy doing their work and not looking for a new opportunity.
View Story09-22-2016
Semiconductor fabs like to avoid writing custom software to fit all of the idiosyncrasies of individual processing systems. So HP developed PC-10 to handle IC process equipment by separating it into general classes. SECS II was a mandatory prerequisite of the equipment before an interface to PC-10 could be developed.
View Story06-10-2016
Metrics are data and statistically backed measures. It is always expedient to base decisions on data and metrics, for example, in PCB design. These measures can be density, first-pass yield connectivity or in this context, producibility. These measures are the basis for predicting and planning a printed circuit design. But what if a metric doesn’t exist? Then you can create the next best measure, the Figure of Merit
View Story06-01-2016
Learning is not instantaneous! Nor is progress made in a steady manner, but at a rate that is typified by one of two basic patterns. In some cases, plateaus will be seen in learning curves. These are caused by factors such as fatigue, poor motivation, loss of interest, or needing time to absorb all the material before progressing to new. This column will not go into details of how learning is achieved, but will summarize some of these theories.
View Story05-18-2016
The product, and or project (process) life cycle (PLC) is fundamental to a corporation intent on developing new products or processes. It sometimes is called the new product introduction (NPI) process but that is only half of the life cycle. There is product support, enhancement and eventually, obsolescence.
View Story
