Get more transformer outputs by using a grounding bus bar

Train transformers have one pair of screws for each output, which is generally enough for a simple layout, but once you have more than one accessory or building with lights in it, you’ll find it’s difficult to attach all of the wires to the transformer posts.

You can get more on the cheap by repurposing ground bus bars, intended for circuit breaker panels. Read more

Snickering at the Emachine

For several years, I administered a command and control system for the U.S. Air Force. I sat in a datacenter, surrounded by racks jam-packed full of servers, and they kept the building at 64 degrees year round. I quickly learned to keep a jacket handy. I did several things, but mostly patch management.

Our system consisted of a diverse collection of Dell 1U and 5U servers, HP blades, and a couple of Sun SPARC boxes. It was a professional-looking setup, and except for the times we were doing massive system upgrades, the system generally worked as well as it looked.

Then we got a neighbor.

Read more

Weekly roundup: 6 Oct 2010

I used to do a weekly roundup every so often, just doing short takes on stuff that interested me as I found it. I haven’t done that in years; I thought I’d give it a whirl again. I don’t know how often I’ll do it, but it was fun.

Ars Technica says Intel’s neutral stance on Atom in servers is a mistake. Absolutely. A dual-core Atom gives plenty of power for infrastructure servers like Active Directory DCs, print servers, and other similar roles. Atoms could even handle many web server tasks.

Xeons are appropriate for database servers and application servers, but throwing them at everything is severe overkill. A lot of server tasks are more disk-bound or network-bound than CPU-bound.

I worked in a datacenter facility for several years that was literally at half capacity, physically. But they didn’t have enough power or cooling capacity to add much more to it.

The only way anything can be added there is to take something away first. Right-sizing servers is the only way to fix that. If they would yank a Xeon, they’d be able to replace it with several Atom-based servers and get a net gain in functionality per square foot and BTU.

Virtualization, a la VMWare, is an option, but one isn’t necessarily a drop-in replacement for the other.

Or, of course, Intel can sit back and wait for ARM to come in and save the day. ARM provides even more functionality per watt. And even though ARM doesn’t run Windows, it does run Linux, and Samba has reached the point where it can stand in for an Active Directory domain controller.

Is there a market out there for a domain controller that fits in a package the size of a CD/DVD drive and consumes less than 20 watts? I’m sure there is. And if Intel doesn’t want to deliver it, ARM and its partners can.

There may be some resistance to ARM, since some decision makers are nervous of things they haven’t heard of, but it should be possible to overcome that. Maybe you haven’t heard of ARM, but guess what? Do you have a smartphone? It has an ARM CPU in it. That PDA you carried before you had a smartphone? It had an ARM CPU in it. It’s entirely possible that your consumer-grade network switch at home has one in it too. Not your router, though. That’s probably MIPS-based. (MIPS is another one of those scary RISC CPU architectures.)

Put a solid operating system on an ARM CPU, and it can run with anything. I have ARM devices that only reboot when the power goes out. If it weren’t for tornado and thunderstorm season causing the power to hiccup, those devices could run for years without a reboot or power-down.

And speaking of ARM, I have seen the future.

Pogoplug is an ARM-based appliance for sharing files. You plug it in, plug USB drives into it, and share files on your home network and the Internet with it. At least, that’s how it’s marketed. But you can hack it into a general purpose Linux box.

Inside, there’s a 1.2 GHz ARM CPU, 256 MB of RAM, and another 256MB of flash memory. Not a supercomputer, but that’s enough power to be useful. And it’s tiny, silent, and sips power. You can plug it in, stash it somewhere, and it’ll never remind you that it’s there.
I’ve actually considered picking up a Pogoplug or two (they go on sale for $45 occasionally, and the slightly less powerful Seagate Dockstar is available for about $30 when you can find them) to run this web site on. Considering how surprisingly well WordPress runs on a 450 MHz Pentium II with 128 MB of RAM (don’t ask me how I know), I think a Pogoplug could handle the workload.

What stops me? I can build an Atom-based PC for less than $150, depending on what I put in it, and run Turnkey Linux on it. Under a worst-case scenario, Turnkey Linux installs in 15 minutes, and it doesn’t take me any longer than that to drop a motherboard and hard drive into a case. So I can knock together an Atom-based webserver in 30 minutes, which is a lot less time than it would take me to get the LAMP stack running on an ARM system.

But if I had more time than money, I’d be all over this.

A device similar to this with an operating LAMP stack on it ready to go is probably too much to ask for. A ready-to-go image running the LAMP stack, similar in form to the DD-WRT or Tomato packages that people use to soup up their routers, might not be. I think it’s a good idea but it isn’t something I have time to head up.

I don’t think I’ve mentioned Turnkey Linux before. I’ve played with it a little, and I’m dead serious that it installs in 15 minutes or less. Installing off a USB flash drive, it might very well install in five.

And it’ll run pretty happily on any PC manufactured this century. More recent is better, of course, but the base requirements are so modest they aren’t worth mentioning.

I’ve built dozens of Linux servers, but this is fantastic. Spend a few minutes downloading an image, copying it onto installation media, and chances are the installation process will take less time than all of that does.

It’s based on Ubuntu LTS, and comes in literally 38 flavors, with more to come after the next refresh is done.

They haven’t built their collection based on the current version of Ubuntu LTS yet because they’ve been distracted with building a backup service. But that’s OK. Ubuntu 8.04.3 still has a little life left in it, and you can either do a distribution upgrade after the initial install, or build a new appliance when the new version comes out and move the data over.

And if Ubuntu isn’t your thing, or you really want 10.04 and you want it now, or worse yet, Linux isn’t your thing, there’s always Bitnami (

Linux appliances took a little while to get here, but they’re here now, and they work.

Intel\’s Atom mini-ITX board has some interesting possibilities

A story on The Register tipped me off to a small motherboard using Intel’s new Atom CPU. A UK data center is using the chip to power servers, and The Reg asks if it’s madness or genius.

More on that in a minute.It’s an interesting minimalist board. It has a single PCI slot, one DIMM slot, PATA and 2 SATA connectors for storage, and the usual complement of I/O slots. The CPU runs at 1.6 GHz. Newegg sells it for about 75 bucks.

One could use this board to build a minimalist PC, but it would also work well as a cheap upgrade for an old PC. It can bolt into a case designed for an ATX or micro-ATX board. It’s made by Intel, so its quality is likely to be comparable to any board it replaces. And the board consumes about 25 watts of power.

Paired up with some sort of solid-state storage, be it a compact flash card in an adapter or a proper SATA SSD like the OCZ Core, it would be a very quiet, low-power system. Performance-wise, it wouldn’t be a barn burner, but it has more than enough horsepower for word processing, e-mail, web browsing, and other productivity apps. At 1.6 GHz, the Atom doesn’t outrun a Pentium M or even a modern Celeron at comparable clock speed, but it should outrun a sub-2 GHz P4.

I think this thing would be awesome in many business environments. Tasks that would bog it down are the kinds of things you don’t want going on in the office anyway–stuff like 3D gaming, ripping and re-encoding DVDs, stuff like that. The power it would save would be tremendous, especially when paired with an LCD monitor and an SSD.

But I even think it has a place in the server room. For example, my first employer used desktop PCs for domain controllers. The logic was simple: DCs don’t work all that hard most of the day, and by their very nature they are redundant, so if a DC were to fail, it’s not in the same league as your mail server failing. You can grab another desktop PC, stand it up as a domain controller, then start asking questions.

In 1997, when a server cost $4,000 and a desktop PC cost $1,000, this was an obvious place for a college with budget problems to save some money.

I think Intel Atoms would make great domain controllers. They have enough CPU power to do the job, but they sip power, which is increasingly important in datacenters. The PCI slot would limit the type of gigabit NIC you could install, but it should still be OK.

They’d make fine web servers too. They might get bogged down on high traffic sites, but they would have little trouble serving up most corporate intranets, and let’s face it, most people’s web sites aren’t nearly as busy as they would like to think they are. You could always use more than one and load balance them. Besides, it’s typically the database servers behind the web servers that do the heavy lifting. Serving up static web pages isn’t all that difficult of a task, and a 1.6 GHz CPU ought to be up to it.

None of these uses are what Intel had in mind when they designed the Atom–I really think their ultimate goal is to end up in cell phones and PDAs, which was why they sold off their ARM-based Xscale CPU.

But if some enterprising company (or struggling behemoth *cough* Dell *cough*) wanted to build business PCs around these, it would be an easy sell. For that matter, they could stuff two of these boards into a 1U rackmount chassis and sell it as an inexpensive, power-saving alternative to blade servers.

Call me crazy, but having actually administered blade servers, I’d much rather have a bunch of 1U systems with two computers inside the case. Besides costing a lot less money up front, they would be more reliable and consume less power while actually saving space–an HP blade enclosure gives you 16 servers in 10 Us, while my crazy scheme would give you 20 servers in the same space.

Maybe instead of posting this idea where anyone can see it and run with it, I ought to buy a couple of motherboards, take them into my basement and start bending some metal myself. Hmm…

If you think you can do it so much better, then do it yourself

I’ve noticed a disturbing trend lately on the Classic Toy Trains forums. It seems like every time a new issue hits the street, someone has to find an article that has something wrong with it and point it out.It started a few months ago when my friend and mentor Joe Rampolla published an article about adding a capacitor to a toy train to make it stall less often and run more smoothly. The claims, as far as I can tell, were false (I had my longtime friend Steve DeLassus, who has a degree in electrical engineering from Washington University, check them out).

But practically every month since then, someone’s publicly taken issue with something in the magazine.

It’s not about a vendetta against a single author. One issue it was Joe. But last issue it was repair expert Ray Plummer’s advice on repairing a Lionel 2037. This issue it’s the legendary Peter Riddle’s article about getting Lionel’s TMCC and MTH’s DCS (two rival control systems) working together on the same layout.

In the case of each of these articles, the things the author said to do work. There might be an alternative way to do them. But that’s the nature of the hobby. Doesn’t it seem like Model Railroader publishes an article at least once a year about making trees, and not one of those articles has been a repeat since at least 1972 (and possibly 1942)? And if you were to read a complete run of Railroad Model Craftsman, you could probably find another 50 different ways to make trees.

Fifty or a hundred people having different ways to do it doesn’t make the guy who wrote the first article about making trees wrong.

In the case of Ray Plummer, what Plummer said matches what my local repair guy said and did when my Lionel 2037 had problems. When the pilot truck is adjusted within specifications, the 2037 and its many cousins run just fine. Plummer’s critic said the pilot truck is a poor design, and when you lengthen the truck to change its pivot point, it works more reliably.

That’s possible. I don’t know the theory behind pivot points. One of my best friends happens to be a mechanical engineer and maybe he could confirm that for me.

What I can say is that Plummer’s advice preserves the historical integrity and collector value of the locomotive. While modifying the pivot point probably wouldn’t make the locomotive worth any less to someone who just wants to run it, it would make it worth less to a collector.

I can also confirm that Plummer’s advice worked just fine on the locomotive that once belonged to my Dad. It’s almost as dependable as my Honda now.

As far as this month’s article to hit the avalanche of criticism, I don’t use any command control system on my layout and I have no interest in doing so. So I don’t have any experience that would back him up, and neither do either of my engineer buddies.

But I trust Peter Riddle. Riddle has written more than a dozen excellent books about trains. Wiring is a subject that confuses almost everyone, but I’m confident that a fifth grader could read one of Riddle’s books on wiring and understand it, then proceed to wire a Lionel layout effectively. Seriously.

I’ve heard the argument presented in these arguments that if an author is wrong about one thing, the reader loses confidence in everything he says. I don’t buy that argument. Riddle’s advice that the Lionel 1121 switch is a good match for early Marx locomotives isn’t entirely correct. From my own experience I know a Marx locomotive will bounce if it enters the switch from a particular direction.

So do I doubt what Riddle says on the other 95 pages of the same book? No. I also know from experience that the things he says on the other 95 pages work. And I know that even though that Marx locomotive bounces through the switch 33% of the time, it doesn’t derail every time it bounces. So maybe he’s never seen the problem I observed.

I’ll daresay there’s at least one mistake in every computer book I’ve ever read. It doesn’t mean I stop reading computer books. I’ve been wrong once or twice before too. Just ask my boss.

Actually, come to think of it I’d really rather you just took my word on that one.

This criticism bothers me on another level too. Writing an article and getting it published isn’t an easy task. For most people it probably takes about 40 hours’ worth of work. CTT pays $70 per page, and a typical article is 3-4 pages long, so you do the math.

How many people want to spend a week of their lives writing an article only to have some self-styled expert rip it apart in five minutes? Is it worth putting your neck on the line for $300?

Most reasonable people would say no.

I’m sure this is largely an ego thing. Most people regard published authors as special people. So when someone knows something that a published author doesn’t, it must make for some kind of a high.

But the price is also high. How many great ideas languish in the mind of a would-be author, never to see the light of day, because the benefits just don’t outweigh that onslaught of criticism if it happens?

So the next time you catch a mistake in print, that’s great. It means you know enough to be an author. So think of something you know better than anyone else and go write an article and advance the hobby.

Of course, criticism is easier than craftsmanship. Zeuxis made that observation 2400 years ago, and it’s just as true today as it was then. Unfortunately.

Lionel filed for Chapter 11 bankruptcy yesterday

It was all over the various news sites, but Lionel, the train maker, filed Chapter 11 yesterday.

A lot of the news stories got a lot of details wrong.This is the first time Lionel, LLC has filed for bankruptcy. The original Lionel Corporation, which is the company that made the trains your dad’s and grandfather’s friends had, if not the trains they had, in all likelihood, filed bankruptcy three times. The first time was in 1935, when the Great Depression had wiped out most of Lionel’s competitors and the aristocratic J. Lionel Cowen kept on making trains for millionaires’ kids when there weren’t any millionaires left.

Lionel emerged from bankruptcy with a two-pronged approach. On the low end, they started selling $1 windup handcars featuring cartoon characters like Mickey and Minnie Mouse. Today’s hirailers do everything they can to argue that Disney didn’t save their beloved Lionel, but at the very least it made a significant contribution.

Lionel also started paying more attention to hobbyists, making diecast trains that were modeled after real trains, rather than hiring Italian designers to design elaborate, ornate, and some would say gaudy toys that looked like trains. Hirailers say this was what saved Lionel. Whatever.

At any rate, Lionel emerged from bankruptcy and survived the Depression, something only one of its other competitors from the 1920s managed to do. That competitor was Hafner. Heard of it? Probably not. Hafner made cheap windups. Attractive windups, but basically forgotten today. The American Flyer brand name endured, but only because the Coleman family gave it to A.C. Gilbert, of Erector construction toy fame, in exchange for a royalty against future sales. There was also that upstart Louis Marx, who actually made money during the Depression and used some of that money to buy a toy train line, but that’s another story.

Toy production of almost all kinds stopped during World War II because the metal and the production capacity was needed for the war effort. Those who couldn’t make war munitions and other such things made things like bottlecaps. Lionel made nautical equipment. They sold a paper foldup train one Christmas because they were still allowed to do that, but it required the patience of a saint and the coordination of a surgeon to assemble, and once assembled, it served only as a reminder of what kids wanted in the ’30s but the parents couldn’t afford and now that the parents could afford it, it was next to impossible to get (unless you happened to live in New York City and knew about Madison Hardware, which, again, is another story).

So the pent-up demand for toys exploded after the war, and toy trains became a huge fad, giving Lionel, Gilbert, and Marx a license to print money until about 1956, after which the general public decided slot cars would be the next big fad and the people who really liked trains decided they wanted to go to HO scale because they were a lot cheaper, a lot more realistic, and in some cases took up less space.

J. Lionel Cowen decided to retire in 1959 and sold out his share of the company to his grand-nephew. He didn’t mention this to his son, who was only on board because Dad wanted him to be anyway, so he sold his shares too, while they were still worth something. Ray Cohn sought to diversify and make the trains less expensive by using more plastic and less metal, but at best he only slowed the bleeding. By 1967 A. C. Gilbert was being liquidated. Cohn bought the American Flyer brand name and tooling but didn’t have the money to do anything else with it. Later that year, Lionel Corporation filed bankruptcy itself.

Two years later, Lionel decided its chances were better selling toys rather than making them. It sold its train line to General Mills, the cereal people, who also had some toy companies. Lionel opened a chain of toy stores on the east coast, and for a time was the second largest toy store chain in the United States, behind the behemoth Toys R Us.

General Mills kept the Lionel train flame alive, selling O and HO trains branded with the Lionel name throughout the 1970s and early 1980s. In 1979, it even located the old American Flyer tooling and brought those trains back as well. But in the early 1980s it tried to move manufacturing the Mexico, which angered a lot of Lionel fans. It reversed the move, but started looking for a buyer, shuffling the company around within itself and then palming it off to Kenner Parker, and then, in 1986, a Lionel collector who had made his fortune selling real estate in Detroit bought the company and started operating it as Lionel Trains Inc.

Back to Lionel Corporation. Remember them? In 1991, Lionel Corporation found its toy store chain just couldn’t compete with Toys R Us’ economies of scale, and filed for bankruptcy. Starting in 1993, it liquidated. It sold the trademarks to Richard Kughn, the owner of Lionel Trains.

Kughn sold controlling interest in Lionel Trains in 1996 to Wellspring and Associates, a holding company. Rock star Neil Young also purchased a 20 percent share. The new company was called Lionel LLC.

In the late 1990s, Lionel LLC started using Asian subcontractors more and more. Its biggest competitor (and former partner) MTH Electric Trains had been doing the same thing, and undercutting Lionel in price. By 2001, Lionel was manufacturing none of its own trains, and even outsourcing the design of some of them.

In 2000, MTH sued Lionel for misappropriation of trade secrets. A designer who had worked for MTH’s subcontractor did work for Lionel’s subcontractor as well, and the result was what some call an uncanny similarity between some of Lionel’s and MTH’s locomotives. Not having seen them myself–just like many of the people who call the similarity uncanny, no doubt–I don’t know. What I do know is that MTH and Lionel have had bad blood since the early 1990s, and that Lionel’s executives and financial backers conduct themselves in a much more professional manner in public than their MTH counterparts. That may or may not say something.

At any rate, the jurors who saw the evidence agreed with MTH and awarded it $40.8 million, which was more than MTH had sought. Then, on November 1, a judge upheld the jury’s findings and ordered Lionel to stop production of certain locomotives. Lionel then filed bankruptcy two weeks later. Looking over the filing, it’s easy to see why. They have $43 million in assets. They have substantial debt–including $30 million to what appears to be a South Korean subcontractor, not a financial institution like some news reports are saying.

MTH owner Mike Wolf and his buddy, Washington D.C. trash hauling magnate Tony Lash are hopping mad and accusing Lionel of trying to dodge justice by hiding behind Chapter 11. Considering the amount essentially amounts to the corporate death penalty, one would expect them to do whatever they can to appeal. It’s silly to expect a company to not want to stay in business.

In case you haven’t figured it out, my sympathies lie with Lionel. But it’s more out of a dislike of MTH than a love for Lionel, which has basically reduced itself to an importer. It farms out design and manufacturing to the lowest bidder, slaps its name on it, prices it for trial lawyers and heart surgeons, and wonders why nobody buys its stuff.

But MTH is the Rambus of the train industry. Or SCO. Take your pick. Starting in the 1970s, train enthusiasts started using electronics to control them. Eventually these efforts got combined and led to the creation of an open industry standard called Digital Command Control (DCC). Every major maker of HO and N scale trains uses it, and has been using it for years. MTH came along, and while it was developing its own proprietary train control standard called Digital Command System (DCS), patented some elements of DCC. Then it started suing selected companies who made use of DCC.

J. Lionel Cowen was a ruthless businessman and made a lot of enemies, but he had ethics and was proud of them. Not that that’s relevant because the Lionel he founded went out of business in the 1990s and has no direct connection to Lionel, LLC, as much as Lionel, LLC tries to make it look otherwise.

I’d like to see Lionel, LLC survive but I won’t lose sleep over it. Anyone can contract with Asian firms to design and manufacture trains, and to slap the Lionel name on it, all you need is the trademark. If someone other than Lionel, LLC ends up owning that trademark, I don’t see that it makes one iota of difference.

Unless that person happens to be Mike Wolf.

But Mike Wolf isn’t going to steal my hobby away from me. I buy mostly old stuff anyway. Why should I let the Rambus of Trains ruin my fun?

WordPress Appliance - Powered by TurnKey Linux