Inside The Hive: Information Technology, Redneck Style!

Look at the captions on the images... I think that's about all I need to say....

The Server Rack: The Hypervisor and VMs (or, in English, playing with Pretend Computers)

When most people get bored, they watch TV... Me, when I get bored, I waste my time on Wikipedia. Fortunately this introduces me to all sorts of innovative concepts. Unfortunately this almost always costs me money, as I invariably get involved in various oddball projects...
About a month ago I came across something called a hypervisor, or Virtual Machine Supervisory Operating System. This is a concept that I have known about for a very long time, but until recently have mistakenly referred to as VMS. What a hypervisor does is allow a computer to run multiple instances of an operating system, or several discrete operating systems, simultaneously... Mainframes have been equipped with hypervisors for the last thirty years, however this technology has only recently appeared in the microcomputer (PC) market.
In my research (or, more correctly, browsing) I happened upon the VMware website. Years ago I had experimented with VMware (then a PC emulator), with disappointing results. However more recently they came up with a true hypervisor called ESX, which recently has been released as freeware as ESXi. But, as is usual with me and freeware, it isn't exactly free...
Pictured above is my new server: The Hive. I gave it this name because, when it is fully configured, it will be "a swarm of virtual PCs, waiting to be unleashed." It is built from the scraps of at least three dead PCs; the case and hard drive are from my last desktop PC (which died due to a defective front-panel USB connector). I later painted the case with bronze hammertone paint when the Media Center lived in it, the board and memory are from my brother's last PC (the board is his SECOND warranty board, which he said is "dead to him", but with a CPU purchased from EK Computers for $25 works just fine for me). The remainder of the system (video board, fans, power supply) were gathered from my collection of parts.
Unlike my previous "science projects," this one did not go easy... It took no less than a month of experimentation, aggravation, playing games with computers (as opposed to playing games on computers), frustration, and so on... But I finally got it working.
Here is the inside of "The Hive." This is my brother's premium gamer ATX motherboard, an EVGA Socket AM2 nForce 590 SLI board with integrated dual gigabit Ethernet, IEEE 1394 (FireWire), USB 2.0, AC'97 audio, dual PCI-e 16x slots (for graphics cards), Serial ATA (SATA) and a myriad of other features. Of course for ESX server 3i, most of this is basically useless (at least when one is using mostly surplus parts)...
For starters is graphics... ESXi uses the now basically unheard-of resolution of 80x25. This means that the screen only displays 80 columns and 25 rows of text, with only ANSI (or text-terminal) graphics. This renders the long PCI-E 16x slots basically worthless, and I use an ancient SIS 2MB card that I had laying around for the console.
In addition to this the 8.1 surround AC'97 CODEC audio chip is completely useless (ESXi has no support for audio whatsoever), I am unsure as to weather or not 1394 is supported... However many of this board's features are both usable and beneficial to ESXi. For starters, this board supports dual-core processors and dual-channel memory. This allows two applications, or "guest operating systems" to access resources simultaneously. Both the SATA and Dual Gigabit Ethernet features are supported, as is the USB 2.0 interface (for attaching hard disks by USB).
Unlike a typical PC, an ESX server is used entirely over a network. The image displayed on the monitor above is the only thing that ESXi EVER displays on the attached monitor... However the VMware Infrastructure Client (VIclient) gives full access (including a virtual console) to the hosted operating system.












This is a screenshot of VIclient running on my Vista-based laptop. Shown is the console for a Windows 2000 Server virtual machine. On the left of the console is a list of all of the configured Virtual Machines. The ones highlighted in green are currently running. As you can see this single computer is currently running three operating systems, with a fourth shut down. With ESXi the number of operating systems that can be run simultaneously is limited only by the available resources of the host PC; and in all actuality the number of VMs can exceed the available resources of the host PC, at the cost of performance. Configured properly this PC may be able to run nearly a dozen separate operating systems at once!

Well that's all I have on this project for the moment... There may well be more later...



Just a note: the name "LIABnet" on the top picture refers to the computer gaming group comprised of most of my friends and myself, called Losers in a Basement. LIABnet is my part of Losers in a Basement; though all of us are very computer literate and most of us built our own computers, I am the "server guy" of the group, and host the Losers in a Basement Network, or LIABnet.

Hand Tools: The Scythe Pt. 1

As promised this post is about something somewhat lower-tech than PBXes and server computers...
While working on clearing weeds on the railroad, and thinking about hacking down tall weeds with a machete I had joked about using a scythe for a similar purpose. However after doing some research I started to feel that maybe the old-fashioned tools and ways may not be so obsolete as many would believe.
I purchased this antique, American-style scythe about a month ago, and over the course of that month as I have taught myself to set it up, sharpen the blade properly and to properly wield it I am increasingly of the opinion that with the proper skills a scythe can do the work of a bush-hog with similar (if not less) physical effort on the part of the user.
Unlike machetes and similar tools a scythe works by slicing plants; when used properly a full-length cutting stroke (a swath) uses no more energy than a miss. A machete on the other hand requires a "concentrated" swing in order to cut down plants, all but preventing prolonged use.
Since I took this photo I have put the first two coats of linseed oil on the snath (handle) , peened the blade once and applied a coat of linseed oil to the "solid" rust covering the non-cutting portions of the blade (this is to prevent further rusting. In the old days metal tools were intentionally rusted and finished with linseed oil, a process called browning. Of course a "proper" browning job is considerably more involved than simply oiling a rusty tool, see my second note below).

Pictured here are my four scythe blades. The top three blades were given to me by my grandmother. Of these three, in order, one is physically worn out, the next is severely damaged but repairable (I have already partially repaired it) and the third is quite literally in like-new condition and only needs peened and sharpened.
The topmost blade is sadly unsuitable for anything except a decoration. It has simply been worn out, and deserves to be attached to a similarly worn-out snath for display. A sad but distinguished end for any good tool if you ask me... Now I just need to find a worn-out snath with a good blade...
The second blade is severely damaged, but is quite salvagable. All of the cutting edge is in excellent condition, however the entire blade is slightly twisted and (when I aquired it) the tang was bent a full 45 degrees. As an amatuer blacksmith repairing the tang was easy, however I have yet to get a chance to try to un-twist the blade. Once this blade is fully repaired, peened and sharpened it will once again be prime to cut literally acres of grass and weeds on end...
The third blade, which has been stored better (and I believe has barely been used) is in literally like-new condition. All it needs in order to be usable is to be peened, sharpened and oiled.
The fourth and final blade is the blade that came with my scythe. When I first purchased it the blade was nearly destroyed, and required a substantial amount of filing to remove the damage from the cutting edge. However after peening this blade and tuning my snath it is finally able to cut grass again. Even before this it had proved it's weed-cutting abilities, which are now truly spectacular.

A note: In this post I repeatedly talk about "peening" blades. Peening is basically cold-forging. This is the "proper" way to resharpen a scythe blade, as it both greatly reduces wear on the blade and work-hardens the cutting edge (increasing it's sharpenability without making the rest of the blade brittle). Peening the blade literally forces the edge back into the correct profile, minimizing the amount of steel that needs to be removed to obtain a suitably sharp edge.

Another note: Many antique tools are "rusty," however unlike most rusty tools the rust is always small-grained and even. This is because the tool has been "browned," that is intentionally rusted and preserved. A full browning treatment involves cleaning the tool to bare steel, applying a rusting agent (such as salt or salammoniac), allowing it to rust, removing all rust, reapplying the rusting agent, allowing it to rust, removing all rust and so on until the coat of rust is very even and "solid," that is that the rust won't flake off even when struck against a hard object. At this point the tool is finshed with linseed oil, then waxed. In this manner the rust itself provides a means of preventing further rusting. It is amongst the oldest methods of preserving iron and steel, and in some cases is still used today.

That's all I have for now. In part 2 I will show the restoration of the long blade in detail.

The Server Shuffle Pt.2: Servers Shuffled and more PBX stuff

It's finally done.
The first half of the project, installing the Asterisk PBX software and my existing TDM card in the IBM server, went as well as could be expected. The only hiccup involved a dead CMOS battery, which concerning the age of the server (about 8 years) and the fact that it sat in my parents' basement for close to one of those years that isn't surprising. Replacing the rediculously overpriced battery ($5) fixed all of it's problems.
The other half of the project, transferring the Windows 2000 file server from the ICP Electronics computer to the dual Pentium-III workstation box (the former PBX) proved a lot more... interesting.
First, it appears that I had a bad IDE cable; I had as a last resort on the ICP box tried using "tube" cables to increase airflow, and one of the two storage discs apparently failed... After transferring the system to the new computer the problem continued... In frustration I gave up for the night, and the next time I tried the system I discovered that the fan in the switching power supply had failed... Fortunately I had a spare, and since by now I suspected that the hard disk failure was due to a cable problem I replaced the tube cables with standard ribbon cables. Problem solved. Now I just need to finish working that computer through the entire Windows Update cycle and it'll be finished. This computer uses the failed Slot-1 processor form factor, which has decicive advantages over the previous (and current) ZIF form factor, mainly in the area of cooling. The large form factor of the processors (approx. 2.5x5 inches) allows plenty of room for very large heat sinks, which should eliminate all overheating issues... Of course this server could still stand to be upgraded (a redundant power supply would be nice, as would be ECC memory...), but it will work well for now. And probaly for much longer.

And now for PBX stuff...
As I stated in my last post I am now using my IBM Netfinity 6000R as my PBX. This is an obsolete corperate-class server computer incorperating a veritable laundary list of features. The system is highly robust, and capable of "tolerating" a failure of virtually any component (including CPUs and memory modules) save the motherboard and memory board. It incorperates four CPUs, three power supplies and uses error-correction memory. I also added a ServeRAID board, which provides hardware-accelerated RAID functionality. Using a RAID-5 array utilizing three hard disks it is possible to remove one of the disks, shoot it with a rifle, and replace it with an identical disk without the operating system even noticing! In addition to it's redundency this technology also provides a substantial preformance increase, effectively doubling the hard disk access speed.
In addition to all of that this server incorperates the "Netfinity Advanced System Management" board, which is literally a seperate computer dedicated entirely to making sure the server funcions properly. This "service processor" (located in the upper right of the photo) constantly monitors the system and using LEDs located throught the system highlights any defective component in the system... It provides real-time monitoring of any component in the system, a technology that is common now but at that time was only implimented in this class of server.
For analog telephony functionality I have added one (soon two) TDM PCI cards. These cards, also called Zapata Telephony cards (formerally abbreviated to "Zaptel" cards, a name no longer used due to a purely accidental copyright infringement) provide an inexpensive way for a computer to interface with the PSTN. At the most basic level these cards are 4-port modems, however instead of converting data into sound they essentially do the opposite, and convert sound into data. These cards accept four modules each; each module is capable of duplicating either an FXO interface (capable of answering a phone call) or FXS interface (capable of ringing a phone). It is VERY important to note that the interface and signalling on these modules are reversed: Forigen eXchange Station [FXS] represents a telephone, and answers a call; Forigen eXchange Office [FXO] represents the telephone company, and rings a phone. The modules are named for what they are to be connected to, not what they actually do).
What all this means is that using this software, a used server computer, and telephony interfaces aquired via eBay or other online discount sources it is possible to construct a truly full-featured Private Branch eXchange (PBX) for a tiny fraction of the price of a conventional Key Telephone System (KTS) used by many small businesses (as I personally know)... It is possible to create a 24-port Asterisk PBX, with or without digital connectivity (T1 or ISDN-BRI), for under $1,000 (utilizing a used server as it's base). By comparason a 24-port KTS would cost easily ten times this amount, and any other true PBX at least double that!
That's all I have for now... After this high-tech garbage I think the next post will be about something decidedly lower-tech...


BTW: TDM is an acronym for Time Division Multiplexing, a scheme for combining several (usually 23 or 24) analog channels into a single digital channel. The usual use of this scheme is the DS-1. More commonly called a T1, it combines 24 telephone circuits onto a single pair of wires (generally speaking, one of the channels is used for signalling, a scheme called ISDN-PRI). The TDM cards impliment TDM in software (either by de-multiplexing a T1 or by combining as many as 24 traditional analog circuits into a single datastream). By utilizing a Zapata Telephony card this data can be handled entirely by software, allowing applications like Asterisk to interface with the PSTN.

The Server Shuffle Pt. 1

Well I guess it's been a long time since I last posted, though a lot of planning (though not as much doing as I would like) has happened since then. So let's get started with one of my current dilemmas; my server situation...
Owing to the recent heat wave and my unwillingness to air condition an uninsulated house my file server has been experiancing severe reliability issues. This is mainly due to the extreme tight clearances on it's motherboard, and thusly the inability to install better heat sinks on it's CPUs. After some deliberating I decided that it was time to mix it up...
First of all I finally moved my massive IBM Netfinity server over to my house (it honestly weighs over 100 lbs fully assembled!). I made the decision that this huge yet design-constrained system will be my PBX, the current PBX will become my file server, and the existing server will be retired from active use.
The IBM Netfinity 6000R is a corperate-class server (albeit one that's 10 years old). It is a high-reliability, high-stability, highly-overbuilt system featuring 4-way SMP (that's QUAD CPU!), interleaved Error-Correction memory and hardware RAID capabilities. However owing to it's design it is essentially limited to a single consumer-grade hard drive. And seeing as I can't afford to upgrade my storage disks I decided not to relegate this system to that role again. I primarialy purchased it to use as a game server, a role it filled eagerly, however I have a strong feeling that any LAN gaming requiring this robust of a server is probaly long since past... The way I figure I have the thing, I may as well put it to use. Even though it is extremely overpowered for my telephony needs the additional power this system has may be of use if I ever get ambitious...
The existing PBX system is a workstation-grade dual Pentium-III motherboard that I happened upon. This will become the file server, and after some memory shuffling (to maximize it's available memory) it should fill this role happily. And hopefully it will be less suseptible to the thermal issues of my current server...
Which is an ICP Electronics Single Board Computer with an expansion/multiplexing backplane. The entire dual Pentium-III motherboard featuring onboard video, sound, USB and dual Ethernet is crammed onto a single full-length PCI card (about a quarter of the size of a standard motherboard). The Expansion Backplane allows this single-board computer to function as a standard motherboard (and accept both PCI and long since defunct 16-bit ISA expansion cards). For a long time this server proved very satisfactory, however owing to the extremely cramped conditions on the system board, and a design flaw on Intel's part (the CPUs that this board accepts have an extremely small heatsink contact area). Because of this in order to improve CPU heat dissipation I'd either have to upgrade to copper-core heatsinks (and hope that they fit) or place gold (!) wafers between the CPU and the heat sink... Both of these options simply aren't worth the money!
As of the moment I have moved all of the telephony hardware from the old PBX into the IBM machine, but haven't done much else. Currently I am suffering from either a cold or mild smoke inhalation (from cooking over an open fire for over 15 hours over the course of the weekend) and only got two hours of sleep last night. I hope to accomplish more on this project later this week... Until then...

(P.S. I have more planned to post, I simply want to wait between posts to make sure none of it gets overlooked)

Electric Fencing Pt. 2

Over the course of the last two weeks since I purchased the A50LIL fence controller I had it and two others fail on me due to a design flaw... Disgruntled with that model I returned the third defective unit for cash and purchased a second B10LI, identical to my original controller. For me this is a proven, dependable unit that more than satisfies all of my fencing needs, except for the possability of adding a Storm Guard module to the fence controller... However I have a feeling that I will probaly be safe without one...
I installed the new B10 on the fence earlier today, and put the old, "proven," unit back in the blister pack as a backup/secondary controller. I add the secondary because if I decide to use an electric fence around my garden the B10 is capable of running up to a month and a half on four "D" cell flashlight batteries, and requires no shelter from the elements. For common pest control a 6' ground rod driven only 4 feet into the ground is more than sufficient (as I learned when I purchased the first B10). This makes using electric fencing around my future garden easier, as I don't have to string insulated wires across my lawn (which would need to be reeled up every time I mowed).
Anyways, aside from my fence energizer fiasco I have made a few additional upgrades to my fence since the last post. First and foremost I added a lightning arrestor to the control post. Second I added a second 8' ground rod to the grounding system which, especially in dry months, will substantially improve the performance of the fence. Both of these are "invisible" upgrades (the ground rods and their wiring are completely buried, mostly so I don't have to mow around them, and until I finish the control post I'm not going to bother taking pictures of the lightning arrestor).
Third I finally re-strung the lower wire and added the second wire I had talked about in the previous post. The lowest wire is a mere 4" off the ground to exclude vermin, and as you can see already needs weed-wacking. The second wire is about 16" above the ground, and is the main predator control wire. It is between the average nose height of cats (or raccoons) and cyotes, the main predators that I am worried about. Any animal brave enough to climb the chicken wire will still have to deal with the tightly strung top wire, and with over 150 feet of it touching the ground an animal touching both this wire and the chicken wire is in for a nasty surprise!
As you can see both the Mighty Rooster and two of the pullets made a cameo in this picture. I think they have already grown significantly since the last post... With any luck my next post will be about the Henhouse.

Home Farming: Electric Fencing Pt. 1

As stated in my previous post I have opted to use electric fencing both to help contain my birds and to exclude other animals (rodents, feral cats, and true predators such as foxes and cyotes). The principle of electric fencing is simple: An energizer produces a very high voltage (generally between 3,000 and 12,000 volts) in controlled pulses of 1/10,000 of a second once every second. Seperate wires lead to both the fence and a ground rod system. When an animal touches both the fence and the ground simoultaneously, when the fence is energized, it completes the circuit and recieves an electrical shock.
To some electric fencing may seem cruel, however it is far more humane than the alternatives. When properly installed electric fencing is infinately safer than any other fencing system. When an animal is "hit" by an electric fence it recieves a high-voltage, high-current pulse of electricity, however this pulse of electricity lasts only a tiny fraction of a second. This pulse is rather painful (as I personally found out last week), however it contains so little energy that it is physically incapable of injuring an animal or person. Thusly electric fencing is more of a "psycological barrier" than a "physical" one. Conventional fencing, such as barbed wire, relies on either sharp edges or physical restraint of animals, either of which can kill either a person or an animal.
The picture above is of my upgraded electric fence controller setup. After both doing the math and considering the options I opted to purchase an AC-powered electric fence energizer, on the grounds that it would cost about the same as purchasing two batteries, a dedicated charger and a (decidedly substandard) backup fence energizer. The unit I purchased is the Zareba Systems A50LIL (if you ask me the name is rather ironic, since the second "L" means "Large Cabinet"). It is a 2-joule low-impedance energizer which, (unlike my B10LI self-contained unit) can accept Zareba's "Storm Guard" module, a purpose-built surge arrestor intended to mitigate the effects of lightning strikes to the fence. Since my fence controller is located inside my house I plan on installing both this and a true "lightning arrestor" on the fence line itself. Considering the extreme power of lightning I feel that it's worthwile taking every practical measure to mitigate it's effects on my property... As before, the two new danger signs (the two 10KV signs and the Authorized Personnel Only sign) I printed off myself.


After the heavily-insulated leadout wire leaves my house it first stops here: the "Fence Control Post." The post itself is temporary as I plan on replacing it with a pressure-treated 4x4. In addition I plan on adding seperate cut-outs for the top and bottom wires, as well as a third cutout for temporary garden fencing (for when I plant a garden next year).
Though it is not visable in this picture the leadout wire is bent several times to serve as a lightning choke. The short-radius bends discurage the high direct-current lightning stroke from continuing along the fence wire, and finding an easier ground...









Which leads to this: My temporary, "Home-Brew" lightning arrestor setup. Here I pushed a short ground rod into the earth and placed a solidly-bonded grounding wire just outside of the 10KV arcing range from the fence wire, right next to the leadout wire. This was an "emergency measure" I took last night, due to the predictions of severe weather over the next three days. After today's rather absurd lightning storm I felt that this system was well worth my time... Thankfully I don't think I got to find out if it worked or not...
I plan on replacing this system with a true lightning arrestor with a fully-buried 8-foot ground rod as soon as is practical. Here you can also see a new fiberglass rod with two insulators for the new permanant exclusion fence. Hopefully I will get to post about that tommorow...

Chickens Update

Finally, through only partial fault of my own, the chickens are outdoors... Now, the explination:

Yesterday I erected what was intended to be a temporary enclosure for the chickens. I didn't have nearly as much wire as I had thought, and had to improvise a coop for them... After several near escapes I thought I had the birds contained, and left for game night...
I was wrong...

When I came home at 1:00 the first thing I saw when I pulled in was what looked like a dead chicken. Fortunately it was still alive, but very VERY far from where it belonged. It turns out that it and one more bird had ran right under the electrified chicken wire. About an hour later, just as I was getting to bed, I heard a lot of commotion outside, and when I went out with my flashlight... NO CHICKENS!!!

The next morning after I got up I looked around and, sure enough in the pine trees on the north edge of my property I found five of the birds in short order. Soon thereafter I found the rooster in another pine tree by himself. After reuniting him with the hens I left for Tractor Supply to get supplies to make a much larger run for the birds. Here are the results:
The run is about 50 feet long and 15 wide. The chicken wire is topped with an electrified security wire, and the base is surrounded by a second electrified wire. Both wires are energized with a 6,000 volt pulse once every second, which (according to the fencer manufacturer) is enough to control all predators and livestock as large as bison! This fencer is rather overkill for my purposes, but it is a "weed clipper" unit, which means that it has enough power to literally "burn through" weeds. This is both good and bad: Good in the respect that out-of-control spring weeds won't negate the fence, but bad in the respect that the unit is more than capable of starting brush fires in dry weather... I plan on adding a seperate cut-out for the bottom wire, so I can turn it off during drought and red-flag warnings.
This fence was not as cheap as I had thought: All told it cost about $250-300 (mind you I had bought most of the electric fence parts three years ago, which accounts for just less than half of the total). This is the main reason for my advocacy of electric fencing: For what I spent on a 200-foot chicken fence I could've built a 5 mile cattle fence using a purely electric fence. The other reason is that electric fencing is very humane compared to the alternatives (such as barbed wire). The shock is rather painful (as I found out while testing the fence), but causes no actual harm to the animal. Barbed wire, on the other hand...

Now for some pictures...

My custom indoor electric fence energizer setup. It features a Zareba Systems 1/4 joule battery-powered controller currently running off of the battery from my big truck (which is currently out of commision). I printed off the "High Voltage" signs myself.













Four pullets enjoying the evening beside on of my pine trees.












And the Mighty (in his own opinion) Rooster... When the birds escaped last night he had picked his own tree to hide out in, and unlike the hens had actually climbed up into the tree to roost. His wings were the first to be clipped, so I doubt if his tree climbing prowess will be a problem...

The Blue Bird returns to Waterville!

Sorry for the delay, but I have still been busy as of late...
After a long and arduous struggle taking almost six months of time and the countless efforts of over a dozen people, the TLEW's Blue Bird passenger train arrived at the Waterville station for the first time in years a mere 45 minutes behind schedule!
My previous post gave details of the work needed on 5109. In addition to this close to a mile of track needed repairs. Though most of this work was accomplished in the last two months of our maintenance seasonwe still made opening day!
Below is a (hopefully) once in a lifetime photograph of 5109's prime mover (engine), with any luck the scale of repairs requiring all of the doors to be removed won't be nessasary again!

5109

I think it's high time to post about the project that has been consuming the vast majority of my free time for the last several months: Toledo Lake Erie and Western locomotive 5109...
Over the course of the winter I have been involved in several nessasary repairs including replacing fan and drive belts, a severely belated oil change on the locomotive's air compressor, general maintenance and, most importantly, replacing the number 5 cylinder head and sleeve...
By now I have spent more time in the so-called "dog house" then I care to think about... But at this point I am not going to be satisfied until I watch this locomotive hitch up all three coaches and board the final coach for the trip to Waterville...

Unit 5109 is an S4, built by the American Locomotive Company (Alco) in 1953 for the Chessapeke and Ohio railroad (later known as the Chessie System). It is powered by a Seymour and McIntosh (Alco) 539 6-cylinder, turbocharged prime mover rated at 1,000 horsepower. It is a switcher-type locomotive, and thusly it's gearing and traction wiring are tailored for low-speed, high-torque operations.
This locomotive is over 50 years old, and is showing signs of it's age. Over the last three years the compression seal on cylinder 5 failed, resulting in a catastrophic compression leak. Uopn removing that cylinder's head it rapidly became obvious that both the head and the cylinder sleeve would need replaced; both showed a very obvious gap in excess of .020 inches (though this may not seem like much this is a compression seal that during the power stroke is subjected to tens of thousands of pounds per square inch pressure). Work ensued removing the damaged components from 5109 and scavanging replacement parts from one of TLEW's "dead" locomotives, engine 112.
This last weekend we seated the replacement cylinder sleeve, piston and cylinder head, and are striving to finish repairs and test the locomotive, and remove the "dog house" from the locomotive, hopefully before this weekend...
Here are some pictures of the parts involved, just to give an idea of the scale of the work involved...


The head of a single cylinder of the Alco 539 prime mover. Each cylinder displaces 1,595 cubic inches, for a total displacement of 9,572 cubic inches across all six cylinders. The two men in the background should give an idea of the size of this part.









The cylinder sleeve. This massive steel tube is pressed into the cylinder block, and is sealed by four O-rings. I had initially thought it would take two minutes to install the sleeve... In fact it took almost eight hours of excrutiating work. Installing the equally impressive piston (which I do not have pictures of) took a a mere two hours total...













The cylinder sleeve, installed into the cylinder block. The socket and extension at the edge of the picture are a 1-inch drive...

Chickens Pt. 2

Well after owning chickens for two weeks I have realized two things. 1: I in no way regret my decision to get them and 2: they are very messy birds! As of right now I am having to clean out my waterer at least twice if not three times daily because they keep fouling the water with depleted bedding... However the birds are progressing nicely, and have already doubled in size, with the exception of my rooster (pictured) who has almost tripled in size!
These are "Straight-run" chickens, meaning that there is no way of knowing what breed they are. However judging from their current plumage I have at least three if not four breeds. Most of the pullets (hens) are showing dark brown plumage, and are most likely Rhode Island Reds, or a descendant of that breed. The cockerel (rooster) has so far maintained his overall yellowish color and is most likely a white breed. And two of the pullets have noticeably different plumage than the other three, however it is still too early to be sure...
I will keep you posted on the birds' development, and will have pictures as soon as I move them to larger quarters (right now they are borderline overcrowded by the box I am keeping them in, making their care burdensome and any photography options poor at best). This week I hope to move them to larger quarters out in the old bar, I will take and post more pictures then.

Home Farming: Chickens

With my busy schedule I decided that I couldn't take the 2-3 days nessasary to prepare and plant a garden this year... So I decided to go for the other aspect of agriculture that I'm interested in, and to buy some chickens!
This is my flock of six as they looked a week ago. They've already grown, and are well on their way. I plan on erecting a henhouse on the concrete slab located on my property, however I have time to wait on that before these birds are big enough to require such housing.

Magnaflux Spotcheck

Magnaflux Spotcheck is a product that I learned about at work. It is a method of nondestructive testing of metallic objects for surface defects (such as cracks). Since I own overhead lifting equipment I felt the need to aquire the tools nessasary for OSHA-mandated safety checks (even though I am not governed by OSHA, since I don't have any employees). That and this test kit has many other uses...
In essence Spotcheck is a 5-step process involving three chemicals: A solvent-based cleaner, a penetrant and a developer. The solvent cleaner is just that, a solvent. Before purchasing the materials I did my research and discovered that the Magnaflux brand cleaner/remover is nothing but petroleum napthata, which is far more inexpensively available as Zippo lighter fuel, hence the yellow bottle of Ronsonol at the top of the kit.
At the heart of the kit is the Magnaflux Penetrant Dye (the right aerosol can). In essence it is nothing more than a mixture of a blood red dye and the most powerful penetrating oil that money can buy. This mixture is sprayed on the test part, allowed to soak and then wiped off with a rag soaked in solvent to remove any penetrant left on the sufrace.
After this the developer is applied. Basically the developer is merely talcum powder suspended in a powerful solvent (yes just talcum powder, or even finely ground clay, works, just not as well). When sprayed on the cleaned test part it draws the penetrant out of any defects, and the red dye stains the talc. This readily shows any defects in the test part.
I purchased the supplies to assemble this kit for less than half of the cost of the ready-made kit (which includes a can of penetrant, a can of developer, two cans of cleaner/remover, a penetrant marker, a shop rag and a case). The only part that I wish I had was the penetrant pen, other than that the rest of the kit wasn't worth the extra money...

Below is a demonstration of Spotcheck on a known surface discontinuity; where a flat-head bolt meets a plate. The results of this test were not as expected, I will explain below.



Penetrant (the blood-red stain) applied to the test part. I allowed about five minutes for the penetrant to work before the next step.














Surface penetrant removed with a rag soaked with Berryman's B-12 Chemtool. Notice the penetrant left in the "socket" of the allen-head screw working it's way up to the head of the bolt...










Developer applied and allowed to sit for five minutes. The developer applied to bare metal is titanium white; the pinkish color is due to the penetrant contained in the surface rust being pulled back up to the surface. Ideally the bolt should have been surrounded by a red ring, my guess is the lack of a ring is most likely due to both misapplication of the developer and the excessive amount of PB Blaster that I had soaked the bolt with prior to this test (and more likely than not didn't fully remove prior to the test). The entire head of the bolt is highlighted only by the penetrant seen seeping up from the socket as seen in the previous photo.


I want to redo this demonstration at a later date on another known discontinuity, at which point I will revisit this post...

Home Farming

I believe that the content of this post more accurately reflects "Hobbies and Interests" than "Home Improvements," so I will post it here instead of on my Home and Bar Improvement blog.
Within the last few years I have been steadily developing an interest in agriculture; both in the growing of crops and animal husbandry. To this end I have decided to make an attempt at at least some level of homesteading (the practice of growing and/or raising that that one needs to survive).
First and foremost I plan on planting a garden to grow those vegetables that I love the most: Tomatoes, carrots, sweet corn and asparagus. The last of these requires special care and growing conditions, which I have the means to provide... In addition to my planned vegetable garden I plan to plant at least two asparagus rows; one planted with a "local" variety of asparagus from my family homestead of Grelton, and the other with a commercial "High Production" variety.
As for my garden I plan on planting a variety of vegetables I enjoy, including but not limited to carrots, lettuce, string beans and sweet corn. I also have a desire to grow dent corn, but that desire more reflects my desire to make my own beer than that of growing my own food...
In addition to the usual garden vegetables I also would like to grow dent corn (as I already mentioned), grapes (which I would like to aquire from clones of family grape vines) and my own hops (which can be purchased from home-brewing suppliers, however it is too late this year to aquire the hop roots). I have the land and growing conditions to accomidate all of this, and though I am in preference of natural farming practice I am certainly not opposed to the use of commercially-available chemical fertilizers (from my decidedly limited experiance I like TSC-branded 12-12-12 fertilizer).
As far as pesticides go I tend to prefer the deterrance route (electric fencing), however I am an advocate of an old, effective and totally human-safe insecticide. It is made at home by boiling one full-size pouch of chewing tobacco in one quart of water. After boiling the tobacco is strained out and discarded, and the resulting "tea" is sprayed on the desired plants. Cigarette tobacco (which is considerably cheaper) can also be used, however suger will need to be added to make the mixture "sticky." Since the tobacco is prepared for human consumpion it is completely safe to eat, even without washing. However the nicotine contained therein is lethal to most insects. It is notable that this mixture is water-soluable, and will need to be reapplied after watering or rainfall.
As I said I am also interested in keeping livestock. I had originally planned to have more land than I do, and to live in the country. As such I had wanted to keep 5-10 head of cattle, and pasture then over a "rotated range" of 3-10 acres in three "zones." By moving the cattle between these zones it ia supposedly possible to raise saleable cattle without the use of commercial animal feed or feed corn. However with my property it is doubtful that I would be able to grass-feed more than one or two steers, and I don't have either the housing for those steers nor the money to build one.
As such I am considering raising chickens, which are considerably smaller (and thusly easier to feed). I belive that by fencing off a fifth of my property and fabricating a henhouse I could easily house at least a dozen hens.
I have read that by raising a kitten alongside the baby chicks is the best way to deter rodents (as the resulting adult cat will be accustomed to the chickens, and vice versa, yet the cat will remain agressive to any intruding vermin), and if I decide to erect a henhouse this is what I plan on doing, in additon to the double protection of both chicken wire and low-level electric fencing.
Thouch I would like to keep cattle I doubt that it would be possible given the limited amount of land that I own. However a gerden and mini-farm are dreams that I have not given up on...

The Village (of Texas) Blacksmith

It's been a long time since I've posted anything on this subject, so I think it's high time that I did so. Though I have not done nearly as much smithing this winter as I would've liked to I have gotten some accomplished. Here is what I have accomplished so far:
This is the "Hot Set" that I mentioned in a previous post. It is made of S7 tool steel; properly austentized and quenched, and tempered at 500 degrees. Though it looks rough and doesn't follow the traditional profile of a hot-set (which greatly resembles a hatchet) I believe in function over form, and this tool accomplishes it's intended function; that of severing heated metals. Of course it still needs a handle (as it is the recoil of the hammer blows is almost hard enough to break fingers!!!).



This is the beginnings of a large kitchen knife which, owing to my late delivery I may as well say was intended as a gift for my mother. This is a fragment of an old file (1095 tool steel) which I was in the process of seperating the blade and tang when I lost my ambition (both due to burnout and serous flaws in my forge). When I get around to finishing it I see the blade as being 6-8 inches long, and I plan on handling it with either stag (my preferred choice) or hickory wood.



A dual-purpose scraper I made for the TLEW railroad out of rebar (probaly the lowest-grade steel of them all, but adequate for the intended use). The flat end is sharpened at a 45 degree bevel, the flat end is square. Both ends are hardened (a relative term considering the extreme low hardenability of rebar), and the shank is left both unfinished and untreated.

The tools I used to make these: My (considerably undersized) anvil and a pair of cross-peen hammers. The right hammer is a 3 lb. cheap Chinese "Harbor Freight" hammer (whose brother, by the way, I recently managed to break at work!). The right is a 4 lb. Mexican hammer (a considerable improvement over the other) which cost more than both of the previously mentioned hammers combined! I also have a small ball peen hammer (a gift from my father), and I plan on adding a 2 lb. cross peen hammer to this collection. I would LIKE to add a 4 lb. straight peen hammer (identical to the pictured hammers, except the "pointed" end is parallel to the handle), however I think it unlikely that I will find one...

The Server Rack: Private Branch eXchange (PBX)

My latest pet project is once again in the one field of Information Technology that I actually enjoy working with: Server computers.
My latest server is an application system built around the Asterisk Telephony Engine and the GNU/Linux Operating System. The whole project got started by happensance: Whilst searching through images for VMware I found out about Asterisk, an Open Source program for managing telephone connectivity. Despite my minimal use of telephones the subject of Telephony holds a great fascination to me. Owing to my somewhat obsessive nature I got more than a little interested in this software, as it gave me the oppertunity to play with a device that I have always found facinating: the PBX.
Building the physical system turned out to be easy: A trip to EK Computer to find an empty case yielded a nearly complete computer. The case, which I purchased for the price of $15, contained a CD-R drive, a floppy drive, a 450-watt power supply and, very surprisingly, a 2-way SMP Pentium-III motherboard with a pair of 500MHz processors! All that was needed was some memory (which, after some scrounging and a surprise find at Goodwill, turned out to be 512 megabytes) and a hard disk.
Installing the software was the easy part: Digitum (the creators and sponsors of Asterisk) released a specialized Linux distrubution named "AsteriskNOW" specifically tuned for the PBX software... After adding a ATA (Analog Telephony Adapter) card and customizing the configurations I now have a working PBX... However as of the moment it lacks any sort of interface to the outside world, severely limiting it's usefulness...
I plan on connecting the PBX to Skype, through which I can have a real telephone number and basically unlimited calling for about $60... A year! However currently the only way to make this connection relies on unsupported software, which I have yet to get working. Oh well, it's been done before, and I have no doubt that I can get it done eventually.

Note: The cases on the rack are (from left to right) Windows 2000 fileserver, Asterisk PBX, a SCSI drive enclosure (currently just being stored) and a power conditioner for the PBX. The left monitor is connected to the server, the right to the PBX.