Slashdot Mirror

As evidenced by the other comments here, plenty of us have no idea who he is. He's certainly not "get recognized in an airport" famous.

I agree with that. I'll note also that other than the book I bought at RS 'way back in 1980 or '81, I hadn't heard much of him in the intervening 30-plus years. Good to know, though, that he has stayed in the electronics amateur/enthusiast space and is still doing somewhat relevant publishing on the Web and elsewhere. I think that for those that were in the hardware side of circuits and computers in the pre-electronic-databook era, there are a bunch of authors whose work was instrumental in conveying the experimentalist ethic. Forrest Mims is such an author but there's also Jim Williams, Bob Pease, and a number of others. I would not have recognized any of them in an airport but the names would certainly trigger an "oh, that guy" response.

From a switchmode supply point of view, they are shitty. 75% at 10A

See here for _MUCH_ better parts from their analog competitors http://www.linear.com/product/LTM4620 LTM4620 - Dual 13A or Single 26A DC/DC Module Regulator with integrated magnetics At 10A output, you get close to 90% efficiency and they take 5V or 12V input directly.

Yeah, and the Intel thing will need a double conversion (12V to 2.4V external, then 2.4V to 0.9V? on-chip) so the total efficiency will be even lower. Not an efficiency improvement for the overall system.

Space Micro doesn't list the prices of their components or systems, nor can I find any from anyone else. Honeywell don't list their prices either. Atmel seem to have dropped out of the field. Linear don't list the prices for their space-hardened stuff. Don't see any for BAE either, or Intersil. Empire Magnetics require a lot of personal data before they give you access to even the price classification information. Not the prices, just how they're classified.

You've got to allow for a year's worth of traveling outside of an atmosphere and then operating on Mars for the duration of the mission. This analysis of radiation for manned missions suggests you're looking at 3.5 mSv per day, then 20 rems per year in most of the places of interest.

Converting everything to rads, it's 0.1 rads per mSv and 1 rad per rem, so that's 12.75 rads to get to Mars if you assume a year-long trip, plus 20 rads for the mission, so anything with a rating of less than 32.75 rads is pretty much guaranteed to fail. However, over the course of a two years, the odds of there being a solar flare are not insignificant. To be safe, you want resistance to a further 400 rad. 432.75 rad is within the tolerance of most of the space-hardened components (some components can be taken up to 1000 rad, others up to 10,000). However, the cheapest space components would NOT survive. You're talking high-end on the space scale.

I'm going to figure that the top-line components will cost 100x that of their conventional counterparts, due to the higher-level of precision and QA that are required. It might well be a good deal more. In Russia, you've also got to pay for smuggling decent-grade hardware out of the US, as all of this stuff will be under massive amounts of regulation.

My guess is that the cuts would have saved enough that those doing the cost-cutting could buy second homes in Switzerland.

I have every one of these issues.

I seem best socially in areas I have no knowledge of. If someone is showing me caverns or a forest tour, I make a great team player and go wherever I am led.

But, you get me in an area I know well, and I will fight like the dickens when I think things are not right. If I am on a tour bus, and I know the bridge is out ahead, and know we don't have enough gas to get back, I will holler my concerns until they throw me off the bus.

Even though I know in the corporate world, its not about being right, its about being liked, I still insist - to the point of layoff - that things be done what I think is the right way. Maybe its early Bible training in me - about Joseph becoming second under Pharoah in Egypt because he was honest. Maybe its just hardwired in me. I can't change it no more than a musician can delete his need to make music.

Here I am in McDonalds using their internet, reading Slashdot, and downloading the specs on a WIZNET 5300 chip after discovering it at Saelig.. There is porn all over the net, but it holds no interest to me.

Even though I have no market for this, I want to design a big battery pack using the LLTC6802 battery monitor so I can have enough processing power to do charge balancing functions, make decent charts of each cell and make reasonable predictions of battery health, cell by cell, and make the results available via web server.

The little 8-bit microcontroller won't draw so much juice as to make the whole thing impractical, yet has enough processing capability as to do everything I need as for cell balancing, statistics gathering, and report generation. That little WIZNET chip looks great, I am looking forward to seeing what I can do with it. I have Wireshark, and can see exactly what it is doing on the 'net.

I feel free about talking about such things before a Slashdot audience, as I know a lot of you know exactly what I intend to do.

For me to take my interests and go before some corporate personnel resources type and try to sell it, I already feel judged as too useless for the modern executive-driven enterprise. I already sense they are not looking for someone like me, rather they want some young guy that looks great in suit and tie, and knows a lot of people. The type of people who will lay me off for being obstinate. Why would they value me as much as a $250K/yr guy who hires management types @150K/yr which fill out evaluation forms on how I take instructions - even if everything in me tells me they are barking up the wrong tree?

I do not even look for a job anymore.

I figured I could just learn to live very cheaply and do as much as I can without involving anyone else. Its been my experience once anyone sees me doing anything, they will run to their lawyers to have cease-and-desist letters sent, and I am not strong enough to defend myself.

I do not feel I can work for a big corporation, as I cannot deal with the office politics, nor can I start my own small business, as our Government coins law at the request of the bigger fish that keeps us small fry from spawning and competing in the market.

Legal torts kill just as surely as a bullet. I can die from either blood or economic hemorrhage. Tort guns are legal, but the bullets are way too expensive for me.

I have better sense than to make moonshine, grow pot, or start a business that might upset somebody else who knows how to play off of Congress. All of these activities attract lawsuits.

I have posted enough hopefully to indicate we are not all of the same mold. I feel quite cheated, as I spent so much time in schools, actually DOING the work single-handedly, successfully competing against fraternities who had databases of old tests. When I get in the workplace, I can't hold a job because I try my damndest to be honest and hold my ground. T

I don't know if anyone has mentioned it yet- too lazy to read through the hundreds of responses.
Spice was developed at UC Berkeley. It is the basis of dozens of commercial time domain circuit simulator programs.
You can get a great one for free from Linear Technology- it used to be called switcherCAD, they now call it LTSpice.
Get it here: http://www.linear.com/designtools/software/

So, HP was an instrument company, started with an ingenious application of a light bulb no less. Then they became a computer company sort of by attrition, since they needed machines to control their instruments -- IIRC. Then servers came sort of naturally when they got to dabble with UNIX. Then the core instrument business got spun off as Agilent, pretty much tarring the name of Hewlett and Packard IMHO. Then the PC business gets spun off too. So what remains is servers? What the heck software is HP shipping that hasn't to do with their own hardware? It's becoming more and more of a joke to keep the same name. Their business got nothing to do with Hewlett nor Packard. They're turning in their graves. </rant>

The device which was destroyed wasn't actually an mp3 player, it seems to have been part of a fingerbox with a nice surprise for anyone who sticks their finger in.

The board inside the tin appears to have been this booster, which would love to send 12V/2A through your finger. :)

I found the LTC3108 (or the LTC3801-1 variant) best suited most of my projects, but the LTC3588/LTC3588-1 is better for capturing energy from ambient sound or vibration via a piezoelectric transducer. (Their evaluation kit, which includes all the parts and a selected PZT, is a bit pricey for a hobbyist, so just get some free sample ICs and roll your own)

Their online specs, designs and datasheets provide everything you need to build your own test rig. These chips even come with built-in support for auxiliary capacitor/ultracap storage, including bucking/boosting the ultracap voltage to the programmed output voltage, which I'd expected to have to implement externally. It's all there in one cheap package with a minimal of external components.

I found the LTC3108 (or the LTC3801-1 variant) best suited most of my projects, but the LTC3588/LTC3588-1 is better for capturing energy from ambient sound or vibration via a piezoelectric transducer. (Their evaluation kit, which includes all the parts and a selected PZT, is a bit pricey for a hobbyist, so just get some free sample ICs and roll your own)

Their online specs, designs and datasheets provide everything you need to build your own test rig. These chips even come with built-in support for auxiliary capacitor/ultracap storage, including bucking/boosting the ultracap voltage to the programmed output voltage, which I'd expected to have to implement externally. It's all there in one cheap package with a minimal of external components.

I found the LTC3108 (or the LTC3801-1 variant) best suited most of my projects, but the LTC3588/LTC3588-1 is better for capturing energy from ambient sound or vibration via a piezoelectric transducer. (Their evaluation kit, which includes all the parts and a selected PZT, is a bit pricey for a hobbyist, so just get some free sample ICs and roll your own)

Their online specs, designs and datasheets provide everything you need to build your own test rig. These chips even come with built-in support for auxiliary capacitor/ultracap storage, including bucking/boosting the ultracap voltage to the programmed output voltage, which I'd expected to have to implement externally. It's all there in one cheap package with a minimal of external components.

As for sampling front ends: they don't really cut it due to parasitics -- they can be fast, but they aren't really accurate -- 12 bits of performance would be pushing it I think. Surely if you would be very clever and do, say, FEM of an integrated sampling bridge, and would characterize well all the parasitics and how they can be balanced out, it could probably be done. But using off-the-shelf parts with nothing exotic: I will believe it when I see it done.

The only practical way I know of is with a fast variable gain amp to blank overdrive -- this can readily be done, and this way you can test settling of 18 bit DAC's, and it's really no big deal once you try it and get it working. It doesn't cost much, either.

I'm playing with a blanking front-end that uses JW's variable transconductance amp approach, gated with fast comparators, and so far I've got it to recover within 100ns to 12 bits -- that's on the first try, on a real crappy breadboard. It's still far away from 10ns one expects the recovery to be in a 100MHz scope, but I should be able to cut it down to 25ns or so without doing anything extraordinary. And I don't really have all that much analog tinkering experience. Surely someone who knows what they are doing could get it to work way better and cheaper.

For an oscilloscope to be a truely universal instrument, it should have a minimum number of caveats. Poor signal fidelity (measured in single % - gimme a break), poor overload recovery, no antialiasing protection on many DSOs, ridiculous trigger holdoff times (orders of magnitude worse than on a $100 tek 7K mainframe from the 70s) -- those are the gripes I have with current technology.

Luckily the stuff that used to be out of reach financially is now either affordable or free: you can easily get a dev board with fast FPGA with multipliers on it, 64MB of DDR2 and a USB 2.0 connection for $200 IIRC. The software to do logic design for said FPGA is a free as in beer. You can have a 500MSPS 12 bit ADC for $200, and pretty much a transparent driver for it for 10% more. Fun times, I admit.

This is like big boy electronics rookie mistake 101. Definitely not college electronics 101, as the latter is a rather useless exposition of the former. Read up on some Jim Williams's application notes from Linear Technology -- it's all there. That's how you do experiments in electronics. Pretty darn carefully, checking yourself at every step. JW's app notes in their entirety a required reading for anyone striving to be good at electronics.

As to "Those handy applications" Name one! Everyone talks about how the apps would need to be rewritten but what apps?

I suggest you start with SolidWorks, CoCreate's OneSpace, and Autodesk {Inventor,AutoCAD,*}. However even the simple SwitcherCAD line requires Windows, and it's pretty good to have. Then of course we have PADS, Protel, and plenty more CADs that are designed only for Windows. If you do any development for Windows then Visual Studio is a requirement. Please let me know when you have all that rewritten for Linux.

In case you wonder if there are already Linux clones of the above, the answer is "yes, clones exist, and no, they are not suitable for any business use." Some of them don't even have 1% of the required features. Perhaps you could find a working SPICE simulator; but outside of that, Eagle CAD, to my knowledge, is the best PCB layout tool that runs on Linux, and it's a hobbyist's tool at best.

Thus, laptop batteries - while they may be made with the same technology - are as small and dense as possible.

It's true that a few high-end laptops use costly lithium-polymer cells made to specific dimensions. But open a regular laptop battery and 99% of the time you'll find it uses bog standard roughly AA or AAA-size batteries. The charging circuit (~ $3)and plastic case do cost something to produce, but the manufacturers are buying at wholesale prices and cranking these things out by the millions; they should benefit from massive economy of scale.

A battery assembled in China and rebranded by Dell sells for 2x~3x the retail price of the li-ion cells used to make it. The more expensive the laptop, the bigger the ripoff. Grab the same battery direct from China (say, a fairly reputable Chinese company like Agptek) and it will cost about the same, or even slightly less than the retail price of its cells.

Especially brave or foolhardy adventurers could always pry the dead battery case open and solder in some new cells themselves. Apply too much heat, though, and you end up a bomb. Not to mention that the laptop firmware is sometimes designed to prevent the use of refurbished batteries (each battery pack has a unique ID, and the total capacity of the battery pack as reported by the bios can only go down over time, never up - the battery might be 90% full but the OS will see it as 5% and force a shutdown).

Boost converters are pretty easy to make, these days. But you can also use a diode/capacitor-based voltage doubler (tripler, quadrupler...) and get to the same place, which is often easier because you don't have to worry about finding a high-voltage switcher or building some sort of accurate voltage divider so a boost converter controller doesn't get fried through overvoltage. You can use LTSpice to model a doubler and make sure it's actually going to be able to produce the voltage/amperage you need with real-world components, as well. Or you could wind your own transformer on a toroid, if you're feeling particularly adventurous.

There are a number of SPICE frontends. Right now I'm using the free LTSpice from Linear Technology. It's a professional quality code that the company releases for free since their main product line is actualy electronic components. The software comes with a full library of LT components of course. BUT, it's windows only. There's a Yahoo user group that may be able to answer questions about how well it runs under VMWare or Parallels, It supposedly runs well under Wine. (Winehq.org says "Works well with wine"). So that could take care of Linux and Windows users, but no idea with Mac. I know Mac users are whiny enough, but don't they have a Wine equivalent yet?

FREE! It is Windows only but runs great on Wine. The author supposedly is very supportive of making sure it runs well on Wine.
http://www.linear.com/designtools/software/
It is optimized for analyzing switching power supply circuits so it probably is the fastest spice implementation out there. I have quit using all the other spice based simulators out there in favor of LTSpice. User support can be found on the yahoo group:
http://tech.groups.yahoo.com/group/LTspice/

I've actually been in the same situation myself, two free (as in beer) SPICE derivatives I've found to work well are LTspice and TINA-spice (from linear and Texas Instruments respectively). They are windows binaries but function very well in WINE (in fact the developer(s) for LTspice have designed it to function as well as possible with WINE).

I've mostly used LTspice and it works very well and has a low learning threshold. Of course you can insert spice directives in the schematic to do more advanced functions like basic parameter sweeps as well as monte-carlo simulations and so on and so forth. Check out LTspice's yahoo group for a bunch of documentation.

As far as other recommendations for eagle go I doubt that's what you're looking for as eagle is solely for schematic capture and pcb design, there are no simulation capabilities in it.

LTSpice is free as in beer and works nicely even with more complicated problems. There is only a windows version available, but Linux support with wine should not be a problem. http://www.linear.com/designtools/software/

For simple circuits SolveElec runs on windows and mac, has a very nice user interface and is a good tool for teaching. http://www.physicsbox.com/indexsolveelec2en.html

Don't buy anything until you've played with switcherCAD:

http://www.linear.com/designtools/software/switchercad.jsp

Simulate it, learn it, make it work, then build it...

The guts of a 741 is both a good and a bad place to start learning about [bipolar] transistors. You certainly won't digest it all in one sitting. But it has most of the basic arrangements that are important, and they're relatively cleanly separated. It's got an emitter-follower push-pull output stage, a common-emitter gain stage, a long-tailed pair differential input stage, some current mirrors to set up biasing, and a Vbe multiplier to help bias the output stage; it's really not complicated if you take it in parts and don't feel any particular need to understand it in one sitting. That's true of any non-textbook circuit, though, really.

PICs are a great way to do interesting things, but if you really want to know why your PIC works quite well except when the moon is waxing gibbous, you're probably going to have to learn some analog stuff. You can go a long way without paying attention to the analog side, just as you can do an awful lot of programming without ever looking at compiler output -- but in either case, you're holding yourself back compared to what you could be doing.

Oh, and Jim's scope drawing is probably round because I believe he still uses that scope. Then again, his definition of a computer (page 12) is probably not the same as the poster's ;)

LTspice/SwitcherCAD III

http://www.linear.com/designtools/software/index.jsp

and there is a yahoo group.

http://tech.groups.yahoo.com/group/LTspice/?v=1&t=directory&ch=web&pub=groups&sec=dir&slk=1

Much as you can't learn to program well without looking at programs more complicated than you'll find in any textbook, you need to study real world circuits.

Whether you want to do digital stuff or analog, it's worth your time learning the analog stuff -- digital systems tend to break as a result of the underlying analog problem of circuit design.

For example, Wikipedia has the internal schematic for a 741 op amp along with a decent explanation. Once you understand the function of every one of those transistors, you'll be able to really understand why it has both a gain-bandwidth limit and a slew rate limit, and what the difference is.

The best source of real-world circuits I've found is the application notes and example circuits in data sheets published by manufacturers. Since they need the resultant circuits to work when engineers build them, they don't leave out the random extras that textbooks often do. Does that MOSFET need a gate resistor? A circuit in an app note will probably say, whereas an example diagram might well not.

If your goal is to learn more in general, as opposed to solving a specific problem, I'd pay more attention to the author than exactly what they're writing about. For example, I can't recommend Jim Williams' design notes highly enough -- he's both an excellent engineer and an excellent author. Making Shakespeare a citation is the sort of thing that keeps his writing lively and interesting. Or rating circuit complexity in baby bottles as a measure of how long it took him to design and debug it. And, of course, he often goes into great detail about the *practical* considerations involved in precise, high-speed analog work -- especially as it relates to working at the lab bench, rather than with professionally printed PCBs and the like.

I'm sure others will have excellent textbook recommendations. They're an important part, but only a part. Add some analysis of real-world circuits that you'll find in application notes, and a bunch of fussing around with actual silicon and a scope, and you'll be well on your way.

Much as you can't learn to program well without looking at programs more complicated than you'll find in any textbook, you need to study real world circuits.

Whether you want to do digital stuff or analog, it's worth your time learning the analog stuff -- digital systems tend to break as a result of the underlying analog problem of circuit design.

For example, Wikipedia has the internal schematic for a 741 op amp along with a decent explanation. Once you understand the function of every one of those transistors, you'll be able to really understand why it has both a gain-bandwidth limit and a slew rate limit, and what the difference is.

The best source of real-world circuits I've found is the application notes and example circuits in data sheets published by manufacturers. Since they need the resultant circuits to work when engineers build them, they don't leave out the random extras that textbooks often do. Does that MOSFET need a gate resistor? A circuit in an app note will probably say, whereas an example diagram might well not.

If your goal is to learn more in general, as opposed to solving a specific problem, I'd pay more attention to the author than exactly what they're writing about. For example, I can't recommend Jim Williams' design notes highly enough -- he's both an excellent engineer and an excellent author. Making Shakespeare a citation is the sort of thing that keeps his writing lively and interesting. Or rating circuit complexity in baby bottles as a measure of how long it took him to design and debug it. And, of course, he often goes into great detail about the *practical* considerations involved in precise, high-speed analog work -- especially as it relates to working at the lab bench, rather than with professionally printed PCBs and the like.

I'm sure others will have excellent textbook recommendations. They're an important part, but only a part. Add some analysis of real-world circuits that you'll find in application notes, and a bunch of fussing around with actual silicon and a scope, and you'll be well on your way.

One of the better methods is to use a coulumb counter that attempts to measure the power put into a battery against the power removed from the battery. See http://www.linear.com/pc/productDetail.jsp?navId=H0,C1,C1003,C1037,C1134,P2354 for a typical device. Even using these, we only seem to be able to approach something that doesn't suck.

One of our devices has a tilt sensor, so I may try to impliment the sloshing sound as well as our normal battery icon on the display.

All Li-Ion external batteries have charge controllers built in, and some indeed report their charge status via a serial interface. For a simple one see here.

I was once in your position -- although you seem to have had more early success than I did. I can still remember the first kit that I built that actually *worked*. The key was taking my time and being meticulous about component identification and soldering. After that, I took every opportunity I could to learn how things worked -- you end up stealing a lot as a designer, after all, if it worked for the other guy, it should work for you. Now, everything I take apart, I try to identify the parts and figure out what they do, and *how* they do it. Then, I ask myself: "why did they do it this way, instead of some other way?" You can learn a lot from analyzing others' designs.

Books:
Horowitz and Hill
ARRL Handbook

Tools - Get yourself a really *good* soldering station. I like the old Weller WTCP* series, but anything that comes in two pieces (a transformer/power unit and an iron) should do. Plan on spending US$150 or so on it (or get one used). Same for your pliers and cutters -- spend good money on good tools...you won't regret it.

CAD - for schematics and circuit boards, I like expresspcb.com -- the software is free, but only works with their internet PCB service. I have heard good things about Eagle, too, but you will need to do more work to get a PCB made. EpressPCB is fast and easy. Linear Technology has a free PSPICE -- they call it SwitcherCAD III -- http://linear.com/company/software.jsp
It's free if you register. You can build circuits and simulate them...takes a bit of getting used to, but once you figure it out, you can try different components and see what happens in a simple (or complex!) circuit. Pretty much any PSPICE tutorial book will help you get started. SwitcherCAD and ExpressPCB will also run on Linux under Wine.

Ham Radio groups. These are people who like to play with radios. Occasionally, you will find a "QRP" group, where the guys get their kicks from *building* their own radios and test gear. There's a lot of help available for the beginner. Search for QRP-L, NORCAL and NJQRP on Google. They publish newsletter with circuit designs and descriptions, and do group projects, where you send in a nominal amount (US$20 or so) and get back a PCB and a bag of parts.

You're basically talking about a niche market application, which is not to say that it's unheard of, but if you did find it, it would be horrendously expensive.

Want to build one yourself? The easy way is to take apart a bunch of RadioShack universal wall warts, mount them in an enclosure, wire them up together with AC, add an AC plug, and do some cable management with the resulting output wires. Should take a couple of hours, and you wind up with a big box filled with tightly-packed transformers (watch out for ventilation).

Now, let's say you want to do it right. You'll need a single transformer to take the line voltage down to 32V, or 48V, a diode bridge, then some filtering caps (this is a basic, unregulated DC supply). From there, you will take this 32-48V rail, and use a variable switching converter on each output channel to bring the voltage down to the right level. If you need cleaner power, there are even circuits to take a variable switing converter, and use that to feed a variable linear converter for super clean power.

Some stock switching circuits are available in Linear Tech's Application note on switching regulator circuit collection. My favorite introduction to switching regs is Application Note 25, Switching Regultors for Poets, bless Jim William's heart and drawing abilities.

Hope this helps

A competitive product is the Linear Technology LTC3441 which also dynamically moves from buck to boost to compensate for draining batteries. I used it to provide a stable output voltage regardless of whether my input was 2.7-3.3V or 4.7 to 5.0V

I thought I might teach a class at my local community college on the SPICE circuit analyzer being I have used one for about 15 years. I have several sources for generic paperback textbooks that have been out for years, quite reasonably priced. Linear Technology has generously placed their SWCAD III full-fledged spice analyzer ( albeit pre-loaded with Linear Tech libraries ) for free download.

I looked at this, and saw I could get my students a pretty good understanding of SPICE and have a pretty robust tool for their further studies for about $30. Not bad, I thought, for the skills I did not pick up until I had graduated and was already practicing circuit design. I would have almost killed for one of these.

I figured in all reality, SPICE had not changed much since I had my first operable copy which was translated directly from Lawrence Nagel's "Berkeley 2G.5" back in 1987. ( Several of the guys in the company I worked for rewrote it from FORTRAN to C and linked it to a nifty little graphing utility.) Sure gives you a comfy feeling when you know exactly whats going on. But that one was mostly designed for the XT machine. Believe it or not, its the one I still use the most, because I understand its model structures.

All the time, I had been seeing all these high-priced analyzers on the market - each had some particular feature that made them kinda nice for certain things, but all seemed based off the original core analyzer kernel. And all were pricey. I played with Linear Tech's version, and it seemed quite nice - and if it was gratis, I figured it would be ideal for teaching a class. At least I could get the concepts across without asking the students for a $500 outlay before they even knew what they were doing. Sure, after covering the basics of design, I would discuss switchers too. I didn't plan on getting much into the math side, but just give 'em a heads up on how they work so they won't be too surprised when they see them in real life.

But with all this downsizing, the CC dropped my proposal. I was lucky enough to be doing this kinda stuff when it was being developed and understanding how it worked internally was important and shared, not at all like today's proprietary environment. It annoys me to know I am going to the grave with what I know.

The Linear typical app seems to be much smaller, with fewer external components because it's actually optimized for the application.

The Maxim chips appear to be generic hot-swap IC's not optimized for power-over-ethernet. Using slashdot is an interesting way to publish an app-note and get some publicity.

The Linear typical app seems to be much smaller, with fewer external components because it's actually optimized for the application.

The Maxim chips appear to be generic hot-swap IC's not optimized for power-over-ethernet. Using slashdot is an interesting way to publish an app-note and get some publicity.

Sorry You have to use a adapter connected to the 12V Supply, there is normaly no build in usable DC-DC-Converter in Cars. You may build your own DC-DC-Converter which isnt to comlicated if You use ICs like e.g. the LT1071 http://www.linear.com/prod/datasheet.html?datashee t=191, which is sufficent for small Laptops. You may have to add some filtering and maybe a crowbar-circuit (with fuses) for safty-reasons and to prevent EMI.
Happy Hardware Hacking