It's An ENIGMA - Part 3
By Josh Landess
Education of building HEV's
EVW: It sounds like you're into getting students involved, demanding something of them, as you say. Can you indicate if this program is going to be ongoing -- if somebody interested in working on an actual HEV could look to come to SDSU, and if there are other programs around the country that particularly impress you along these lines?
JB: I'd say that the most impressive program in the country right now is the one at UC Davis, and pretty much anyone who's done a web search would know that. They have the two NSF Grants. They have an Office of Transportation Technology Grant from DOE. They have ten years or more of solid program-building up there. They have a Ph.D. Program.
Part of the difficulty of building a program of this complexity to make a real product is that it takes a student, at the undergraduate level, maybe the full junior and senior year before they're really contributing and then they're gone.
So maybe I convince them to stay for the Master's program and they stay another year or two, and they're productive, and they bring the experiences and the training and the knowledge base with them along the way, it develops, and they help other people get up to speed and become part of the team. But if there's no Ph.D. program in a year or two they're gone as well.
The more likely scenario is that the undergraduates, they come in for their final semester, they need to be taught. The few Masters' students I have are working with them to get them up to speed, and they really aren't as productive as they could be because they're teaching.
EVW: Is there an indication that students who have worked on this successful car are in demand now?
JB: Absolutely, mostly because the individuals that have seen the vehicle assume that we have a large program, well-funded, well-supported that produced this fantastic product. And so they come and ask me and they send me promotional items: "Please send your students to me". Or "Can I hire somebody?" And frankly the majority of this effort has been done for a year now and those students have graduated.
The pipeline is somewhat cavetating; we need a few more people in the pipeline. Obviously grants help that: solid money, funding. But there are two student teams now, about four on each team, working on related projects. They just aren't around long enough, embedded in what we're doing, for me to say that they're a shoo-in at a major company doing this kind of big work. If they stayed for a Master's Degree, another year or so, they would be a fantastic resource for any company that wanted to hire them.
Miles per Dollar and Big Red Sticks
EVW: I like to talk about miles per BTU because it gets us thinking more in terms of understanding that if you can improve that, you've improved a lot of different things.
JB: Well I had that approach for a while. And then what I did is I tried to multiply it by the expected market price per BTU of these fuels. And I want to get it in miles per dollar. That's really my goal. If you factor in the infrastructure investment costs of putting a hydrogen fueling center and piping and resources in our country, it changes the whole makeup of things. People do want it in terms of miles per BTU, but the costs are what are going to drive people.
I'm able to rationalize for myself that we're going to use diesel because it's a commodity cheaper than the gasoline we have. Even though it is a cheat in energy terms, you're gaining the advantage of getting that energy more inexpensively. And those costs are related to the economic and ecological prices that people have to pay to deal with the technology.
It's just that we understand that very well now because we've been doing it forever. Add the complexities of hydrogen or vehicle-to-grid, grid-to-vehicle electricity, these new infrastructures people are trying to push. Who know what those costs to society are going to be?
EVW: I somewhat disagree with trying to go to miles per dollar in the sense that you and I both run into a zillion people that just dismiss our ideas with objections based on a sort of short-sided economic, somewhat fallacious argument, you know: "Electricity presently costs this per kilowatt-hour today, therefore I won't listen to you," sort of thing.
JB: Well you're right. And, also, a lot of these people are using faulty data. As I was mentioning, people don't include the total society costs and opportunity costs and other long-term costs in these equations because they don't have any idea of how to estimate them.
The uncertainties are what are going to stop people from trying. But still if you don't go out there with the simplest solution you can and put it in place, we spend twenty years searching for a solution, which is going to happen. We've already spent eight years on PNGV, and we're going to spend another ten years on fuel cells. And what's going to be still driving on the road? Some intermediate stop-gap that made practical sense for people and did not overtax and over-pollute them.
I think it's going to take political will and you know unfortunately that's not my forte, but what I've been trying to do is simply create a big red stick that we can beat people with. And if a politician wanted to pick it up and say "You people are foolish because you're not going after the short-term solutions that are practical and can be implemented now, and here's a school from nowhere that did this." If that's all we do is act as a policy club, I'm happy to do that.
Computer Control of Car Functions
EVW: Terry was talking about one of your dreams of coming up with a computer that would give you data on dollars per mile or something like that as you drove. Depending on the present cost of diesel versus electricity, you could make rational decisions on how to drive the car based on that.?
JB: Absolutely. And/or let the car limit some of your choices. Human beings do not like their choices limited. That's why they're really not ever going to adopt this drive-by-wire technology where they pull into a stream of cars and the cars lock step with each other and caravan down the road in autonomous control. 'Cause people don't want to give up control. And they're still going to not let go of the steering wheel. I'm sorry, it's a long way off. It defeats all purpose of having a car, I think. If you wanted that you'd get on a bus wouldn't you?
EVW: Well I want to get from A to B. As a lot of Southern Californians know, sometimes you can perceive it as not having too much choice about how to do that. A lot of people own a car because they have to, not because they want to.
JB: Yeah but they're really comfortable with that lack of choice. And it's going be an adoption issue. I won't say it will never happen but, boy, it goes against everything that cars are about in people's minds right now. [...]
EVW: The one thing that I keep coming back to a little is the issue of safety in vehicles and the fact that while we're all very into this cool technology for going faster and further on a gallon of fuel. Safety discussions sometimes get sacrificed. And the one thing that the drive-by-wire might help in the long run is to reduce the number of deaths that occur in driving, which is the most dangerous thing many of us do in our daily lives.
EVW: And so even with all you've said, I think that part of the counter-argument would have to be that yes, but lives would be saved.
JB: On the legislatable arena, yes, people as a group are always in favor of saving the other guy from his own mistakes. But we're back to who's going to want to drive a car that is so safe that it's not fun, that you don't feel like you're in control, that you have the nagging suspicion that it's going to somehow fail on you and you're going to crash into a wall. It's the same issue with airplanes. For a long time now there's just been the ability to fly large commercial aircraft by wire. But it's never going to happen because people want to know that there's someone in the front of that airplane. Not a mindless computer somewhere that can have a power glitch.
EVW: I was very struck by your use of, I guess it's carbon fiber in the car?
EVW: I think every technology fan has known for a long time that carbon fiber is very light, very strong. You see it in Indy car races and unfortunately it costs more than regular material. If you were part of a team working for a car company how much really could you consider including that in a real car?
JB: If I had the developmental facilities of a big car company they would definitely be eighty percent composite because I still believe the watchwords that Avery Lovins has put out there for the world. (Even though I think his group is stalled because they can't afford not to hit their goal.)
His watchword is that if you make the car light enough, the engine can be small enough that you can make the car light enough that the engine can be small enough, that the battery storage can be small enough and light enough that you can get your goal. You can hit your hundred, hundred and fifty miles per gallon. It must get lighter.
EVW: So the cost of composites has come down somewhat?
JB: Yes, quite a bit. Well, compared to the days when I worked in the industry. My connection to this whole thing really started because I'm a Composites Trained Scientist-Engineer. That is my area of training and specialization. I worked in aerospace in the grand eighties and that's where the government was funding people to understand these materials better and try to make them produce them.
Secondly, my area of specialization is in manufacturing of composites, dealing with the cost issues, coming up with material components that are more easily produced in interesting structural shapes, keeping the costs lower. So, in this sense, this is the reason that I became involved in this project. As a research effort then it allows me to go on to the next step, which is to go to an all-composite automobile or in this case eighty percent, which is a good estimate.
EVW: The other twenty percent being what?
JB: You can add metals in there. There's no doubt about it.
EVW: For strength at times or...?
JB: No, really it's a matter of economically making thin structures out of composites. They make certain kinds of structures and shapes economically. But, just like any other purist, if that's where you want to go with it, you're never going to have a cost effective perfect solution out of composites. And there was that desire, that holy grail that: "Oh, you know, we're going to an all-composite airplane." The Osprey Tilt Rotor Aircraft was, on the drawing board, an all composite vehicle. But there were components of it that should never have been made out of composites.
EVW: Why shouldn't those specific components not be made out of composites, you said "thin" components...?
JB: Right. [...Detailed Technical explanation....] use sheet metal. I mean, if you're down to eighty thousandths or sixty thousandths or forty thousandths, and certainly at forty thousandths of an inch think, you can't use composites.
Costs of Carbon-Fiber
EVW: Can you give us an idea of where these costs are now and where they were? For example, if I used to watch an Indy race and they'd hold up a carbon fiber part and they'd brag about how many hundreds of dollars costs and how much it might have cost if made out of metal...
JB: This is an important question. And what people don't understand for the most part about composites is that there's a lot of engineering cost that goes in. These Formula 1 people [for example] are pushing the edge. It has to be the lightest part that will meet their needs and be on the car through thick and thin here. If you wanted just a little less performance you could cut the cost by half.
Let me give you a historic benchmark here. When I started this business in '85, the experimental materials we were using were between a hundred and two hundred dollars per pound. And that's just the material cost. Because this is an engineered material, you can multiply the cost per pound again by factors of five. And you're doing this in very small quantities. So we're talking five hundred to a thousand dollars per pound finished cost for structures.
That was in the mid-eighties. [W]here we're at right now and material costs are in the range of oh, I would say, my last recollection was maybe thirty dollars a pound is a good figure to use.
Engineering costs without any economies of scale have come down remarkably because people have gotten more experienced with these things. They've come up with methodologies for making larger structures more cheaply. They use processes like Resin-Transfer-Molding. It is a method that has gotten us much closer to large structures of high complexity that have good properties. And if you go to a lot of the boat hull building now where you're talking about a big structure, the economies are moved away from hand layout where people are out there with wet resin and fiberglass, and they've moved toward this. These methods are making the large structures a lot more affordable because the hand lay-up is less.
The multiplier is now around maybe two to one instead of five to one. So that's just a rough view.
Now if you go with economies of scale you can divide that number by something. And I don't know what that is because no one's really ramped up to high volume production with composites, except the sporting goods people. But I'm trying to stay in the automotive region, car bodies and so forth.
EVW: What about thermoplastics, such as on the Saturn, and fiberglass? I know there's some fiberglass in your car. In my mind, I don't want to say it's "notorious", but fiberglass is not a super strong material that I'm aware of.
JB: Actually this is a misconception that most people have. Graphite and glass reinforcements in the same matrix, epoxy or polyester or whatever, these composites have very similar strengths. But the graphite fiber is five times stiffer. So for the same load it only it only stretches one fifth of the distance that the glass will.
I was one of the first researchers in thermoplastic structural composites. I was in a very early effort at General Dynamics. I can talk about some of those efforts. What's going on in the auto industry really doesn't take advantage of the structural capabilities of composite materials, the thermoplastic side panels and bumpers. They used the thermoplastic matrix in a way to reduce the overall costs. But you're not getting light components out of this. You're getting components that are resilient so that the cost of ownership in crashes and the parking lot dings....
The ownership satisfaction has increased because of the use of composites in the Saturn.
But the vehicles are not any lighter. So if you're going after lightness, which allows you to get the high mileage, this is not the approach.
EVW: These are petroleum industry products, no? ... or somewhat derived from petroleum itself?
JB: You're right. The people that control the goo coming out of the ground will still have their stake in automobile production.
Organizational advantages of the Pro/E design process
EVW: TTell me about the Pro/E software used in the L3 design.
JB: We had an environment, in which basically two people made the decisions about the design and created the detailed level of information generation that resulted in the finished design.
EVW: If we looked at companies like Boeing and Toyota, they'd probably tell us that they have somewhat similar softwares, and dozens of people not only work on them, but work on them successfully and effectively.
JB: They do, and there is a certain set of checks and balances that naturally occurs when you have many minds working on the same problem, but that set of checks and balances doesn't necessarily lead to the highest quality product. If you have dissension, if you have multiple approaches, it keeps there from being absolutely dreadful solutions, because after all, there was a certain level of competence that will come from the mass minds and the mass effort there, but at the same time, there is an expenditure of effort, a breakage of intellectual creation that occurs because there are that many people, you know: "the camel is a horse created by committee", that whole idea there?
Too many individuals who are not totally dedicated to the process as you can find if you get a fanatical one or two in our environment. Too many people would tend to bring down the overall level of creativity. In other words, it's sort of a least common denominator effect of having that many people involved in the decision process.
Certainly you have to have the best one or two people, but if they really want this to happen and you just keep it small and you let the virtue of the new tool sets come to bear on the problem, you can [with one or two people] do the things that would take 20 or 50 people to do and sometimes better. They have a greater combined brain trust it doesn't come to bear on the problem because all of it has to funnel through one or two cadscopes, one or two computer terminals. Why not just put one or two people at those knowledge-generation-terminals and let them do the best shot they can without the noise and confusion of 50 other voices behind it?
EVW: Yet, if I am a manager of a really world class operation, a well-intended operation, at GM or whatever, and I have the benefit of 50 guys who are just incredibly great engineers, really motivated excellent people, I'm going to think the sum of these things is better than the parts.
JB: Well I tell you what. Why don't you take that group of 50 and break it into ten teams of five and since you're now able to create prototypes and get to certain levels of sophistication for 1/5th the price, why don't you have many different activities going on in parallel? They can come together once in a while and meet each other and get, you know, cooperate to compete internally. You're going to spend the same amount of money anyway, you might as well get three or four different high quality designs using the best tools and fewer people in each.
If you don't want to lay them off, if you really believe they are great, restructure the way in which they work together inside the new environment of the computer, and this whole concept, we're working around the idea we're calling it Master Practices Engineering, and it's a new way to think about this because the computer doesn't facilitate design as much as you have to adapt to the stringent discipline that the tools require now. And although you may be a great designer, if you can't work inside the environment, you're going to in some sense detract from and slow up and burden the process of design. I hate to see that.
Larger companies have the ability to share some of these resources more effectively and recently we took the liberty here of a computer aided virtual environment, it's a big wrap-around, ten foot wide, eight foot tall panoramic screen. You can use 3-D goggles and 15 people can sit in front of this thing and interact with this information, so that way you have more of an ability to get all these minds working together.
So we have that added to it, but still on the input end they have all their opinions from seeing the data and then they have to talk some confused underling into typing and entering this data at the desktop. That person, they may have great skills running the tools, running the controls, but they're still going to miss something when those great ideas come filtering through their brains and their eyes and their ears, and try to insert that into the computer.
And so the paradigm is bankrupt, damn it. It is. And until we overcome those two things there are other ways to achieve greatness and I think we've gotten very close to one of those but simply getting a couple of really inspired people, using the tools extraordinarily effectively and not saying "We can't do it".
Combine 50 wage slaves in a company and there may be four or five really inspired people in there, but they get kind of filtered and drowned out by the rest of the individuals who aren't as dedicated as it takes, and don't have the authorship control and the freedom to do it as they see fit, within the guidance of the discipline of the tool.
EVW: The strange case of why several million American engineers who are some of the best engineers in history can't devise better-mileage alternative-fuel cars with greater facility and fewer road blocks is just this weird thing.
JB: Well to put it in perspective, I think this is true in many organizations including the big car companies: individuals making the key top level decisions can't interact with the data or even understand how they tool dictates what can be done, and so they become irrelevant or largely they hinder the process of good decision-making because they don't understand the means of good decision making. They think it's still about "oh well give me five numbers and I'll make a good guess".
No, you have at your disposal every decision made in an integrated fashion in the computer but you, Mr. Decision maker, you hierarchical manager type, you tell people "Give me the bottom line", which is saying throw away all that you know and tell me the difference between this number and that number and I'll choose this number. Well the modern density of data and decisions involved in a really good overall design requires that the managers themselves be fluent in the use of the tool.
Part IV Continued Next Week...
blog comments powered by Disqus