Anablog

The Wednesday tech panels also had three tracks. The first was Energy Stewardship, the second was Product Development and the third was Consumer needs. I went to the first session in the Consumer Needs track, a presentation by Tim Stagg from 3M about innovation. It was pretty interesting, and we all know 3M is a very innovative company, drawing ideas from employees, customers and even the occasional manager. Tom seems to be involved in the window film business, at least that is what he referred to the most.

He had a great slide, that I am sorry to say I took a little fuzzy. It shows how product development is both a push and a pull process depending on the circumstance. Click on the image if you want any chance of reading it. One of the great things of Tim’s talk was he mentioned two failures on product launches. He said that they call one the dribble launch— were stuff is just dribbled out due to poor planning, and the other failure they call the “launch and abandon” where you make a big splash and then do nothing. He said any truly new product needs to have a lot more attention because you have to teach the dealers how to sell it and the customer how to use it. That is why it is a new product. He said 3M uses a launch management tool and confirmed it was a proprietary internal 3M tool. He said that you have to get the customer and dealer feedback you need, not what you want to hear. Sounds like corporate group-think is not very popular at 3M. After a little brag about 3M getting 500 patents a year, he explained to an audience question that you really cannot put much stock in the internet. You have to watch people use things in order to understand what is going on. He mentioned that people thought one of their window films suffered from shrinkage but really the cause what the installers were not properly trained in how to apply the film to the glass. All in all a great talk from a great company.

The next talk was in the energy track, a presentation by Michael Wellenzohn of ThyssenKrupp Presta on how to improve mileage by using an electric power steering pump as opposed to a hydraulic one that is continually run even when you don’t need it. I complained yesterday about not being to take pictures and how was I supposed to promote the great SAE and when I was on the show floor I heard that since I was with the press I could get a photo pass. So I go to this presentation and the same guy that stopped me from taking pictures yesterday said I was allowed to take pictures of the presenter but not of the slides. I snuck one in a later presentation but could get none for this one. Gosh this is stupid, if I had a phone camera I could have recorded the entire presentation and no one would be the wiser. Anyway the interesting facts were that power steering has a 20 Hz bandwidth and can use 400 watts idling and 700 watts to 1.5 kW when you are steering. That 400 watts will cost about 1 mpg in a small car. The cool thing about an electric power steering system is that you can turn it off for the vast majority of time when you are not applying a steering input.

I wanted to see a similar presentation about air conditioning by Thomas E. J. Heckenberger of Behr GmbH & Co. KG. I thought the presentation would propose driving the ac compressor electrically but I see from the abstract they only talk about optimizing the system to save 15 to 30% of the energy consumption, which is still quite remarkable. Rather than take in this presentation I went down to the show floor to visit a few more of my analog buddies at the various booths.

I got back upstairs in time to catch a presentation from Delphi about using a virtual environment to prototype ECU (engine control unit) modules. I am sorry; I just don’t see why you would want to simulate something that you could just build in hardware. Maybe Delphi has ridiculous hardware prototyping problems, the presenter said that the hardware might not be ready in time. They may need to read my cover story on prototyping next December, here in Silicon Valley we can whip out a board in a week or two. Since most of the sensors and components should be carryover in any sane automotive project I find it hard to believe you need a simulation environment for anything other than to increase the percentage of software engineers in the building. I became really skeptical when the presenter mentioned that the simulation system runs 50 times slower that real-time, at least until the Matlab and Labview modules kick in, then it runs 150 times slower. I love National Instruments but I have to believe it would be better to have real hardware built and use Labview to evaluate the results rather than create virtual sensors or other tasks. My buddy and former EDN editor-in-chief Steve Leibson who works over at Tensilica tells me big software and digital products really do need a simulation environment but I think it is better to have engineers in a dyno room rather than in carpeted cubicles. Maybe Steve and I can do a story on this and us analog types can see what software simulation is all about.

Since the simulation talk ended early I got to pop over to the energy track for a talk from Wolfgang Huhn of Audi. He noted that standard lighting takes 618 watts, Bi-Xenon uses 313 watts and by 2012 LEDs will allow reduction to 168 watts of lighting energy. Right now HID is higher efficiency than LED but that should change in a few years knowing the way Cree and Osram keep improving things. He noted LEDs now give 100 lumens per watt, in 2012 he hopes for 200 lumens per watt. The theoretical limit, where all the energy gets converted to visible light is 683 lumens per watt. An incandescent bulb puts 4% of the energy into visible light, HID does 15% and LEDs are 15% now and should be 30% IN 2012. Wolfgang mentioned that a Tesla has a 53kWh battery pack and that LED lighting will be essential if electric cars are to become viable. He also has some great animations of how LED headlights could re-aim rather than dim in the face of oncoming traffic and how you can drive LED tail-lights can be brightened in fog so you don’t get hit from behind.

The last talk was from Ivan Ĉelanoviæ of MIT. It was about using Peltier modules to convert heat to electricity. He also discussed using photovoltaic cells to convert IR to electricity. Unlike some MIT researchers, he did not over-hype the work. He noted that these systems were far too large and inefficient for general automotive use. His lab is using the technology in space vehicles and for remote power modules to drive light loads. Ivan explained that the term ZT is a figure of merit for Peltier devices. Modern devices have a ZT of one and there is hope that a ZT of 3 will be developed. Ivan pointed out that increasing the ZT is very hard since it requires a semiconductor that is a good thermal insulator and a good electrical conductor. He also pointed out that the theoretical limit is the Carnot efficiency you can extract from any given temperature difference. The problem with these systems is getting enough delta-T in order to get some real energy. This is the same problem faced by Sterling engines and 6-inch high hydroelectric dams. There just is not that much energy to extract, even if you get it all. Ivan said that using the 39°K difference in radiator water you might get out 230 watts with a ZT of 1 and 430 watts with a ZT of 3. This is predicated on a 40kW radiator where 10kW is sent to the Peltier device. With an exhaust system where you get a 200°K delta-T you might get 720 watts at a ZT of 1 and 1.3 KW at a ZT of 3. That is with a 30kW exhaust heat rejection. Ivan went on to point out that photovoltaic techniques can extract 22% of the energy, but once again, you would need huge temperature differences. He did note that an alternator only has a 30-40% efficiency so there is some chance where thermo-voltaic or thermo-photo-voltaic systems may be able to augment certain circuits in the alternator and thereby increase efficiency. It was refreshing to see an MIT researcher that was practical about the prospects for his research and cognizant that cost is king in the auto industry. Ivan has greatly improved my opinion of the research coming out of MIT.

A slide by Ivan Ĉelanoviæ of MIT sums up the electrical energy you can get from heat.