The customer was a nation-wide company with branches scattered across the US. Their headquarters decided to buy our VRU for all branches.

When I arrived to install the system I met the branch manager in the middle of their cube-farm. I introduced myself and offered to shake his hand. He ignored my gesture and made it absolutely clear (in a loud voice so everyone in the branch could hear) that he didn’t like to have headquarters forcing him to have any damned computer answering his phones. He then turned around, went to his office and shut the door.

Ummm….

I had several questions for him (not the least of which was “where is your telephone equipment room?”) and needed to brief him on what to expect during the installation process.  I knocked on his door.

He opened the door with obvious irritation, claiming that he was busy, etc. I tried my best to placate him but he wasn’t having any of it. He just kept complaining and telling me that he didn’t want or need the VRU. I had no idea how to handle this guy so I gave up on trying to schmooze him and just stuck to technical issues. After several frustrating minutes I finally found out where I needed to install the system.

After unpacking the equipment and surveying the phone room I had a short list of needs: a chair, a small table for the monitor and keyboard, etc. I had a copy of the letter sent to all branch managers by their headquarters which explained that the branch was required to provide these things. I could see that there had been a small table in just the right spot. All that was left were the marks on the floor, the dust bunnies, and the empty hole where the table used to be.

In the mid-afternoon, I was still waiting for a table. By this time I’d normally be done with the physical installation and would be well on the way getting the system configured. So, I called my boss to give a status update. After relating the morning’s fun, I told him that I might need an extra day (normally it takes two days). My boss was sympathetic but his boss wasn’t. I had to finish on time because of other previously scheduled work, etc. Great. What I didn’t know was that my status update prompted my big-cheese boss to call the customer’s corporate IT manager and tell him about the branch manager’s “resistance”.

At about 3:30, the branch manager came into the phone room with a table. He was subdued but I could tell he was seriously pissed. I had no idea why and I didn’t ask. He told me that I had to leave at 4:30 when the branch closed to customers. This was unheard of; the actual “switchover” from their old phone system to the new one should take place after hours to prevent calls from being dropped. I mentioned this to him but he insisted that I leave at 4:30. OK, whatever. In the hour remaining I completed the physical installation which fortunately included the remote-access line. That evening I did a lot of the system configuration from the hotel room.

The next morning I arrived hoping to finish and get the hell out of there. I still needed to set up dozens of voice mail boxes, call queues, etc. using info that they were supposed to have ready for me when I arrived the day before. Of course, the branch manager said he hadn’t had the time to collect the info and that I had to do it myself by asking each associate, one at a time, but without interrupting their work. Yeah, right.

Collecting the info was “fun” because the manager’s attitude permeated the entire staff. “I’m busy, come back later” was the usual response to my inquiries. Even when they did have time, their kids were constantly interrupting (it was “bring your kids to work” day). While I was doing this, the branch manager came striding across the office, pissed off. He told me that my damn computer system has screwed up his entire phone system. Incoming calls were getting busy signals when no one was on the phone, calls were being dropped,  and other calls were getting crossed with a dress shop two miles down the road.

What the hell?!? “That’s not possible!” It was totally impossible for the system to cross calls with other phone lines. Something else was going on. Besides, I hadn’t connected the damn thing to their phone system yet!

My comment just pissed him off even more. Just then, the operator called the branch manager over to speak with him. She handed him the phone. He said a few words, hung up and looked at me, now *really* pissed.

“Your system just called 911. The police are on the way.” Holy shit!!

He called me into his office and told me to pack up and get out. I told him I would do so, but that he should also know that the system wasn’t even connected so it couldn’t be doing these things. Just then his secretary came in with the police in tow. I said nothing while the branch manager explained to the police that my damned computer system was screwing up his whole phone system. As this was going on, I saw that out the window in the parking lot was a NY Telephone truck next to an open man hole. The cops told me to call 911 when I was finished so they’d know when to stop ignoring the bogus 911 calls. I agreed (even though I knew it couldn’t be my system) then excused myself and left the office with the cops.

I went out to the parking lot. After yelling down the manhole, a NY Tel guy came up. I explained that there were some strange things happening with the phones and asked what there were doing. They were rerouting a trunk line (think of it as a big cable containing lots of separate phone lines). After talking for a couple more minutes, it became obvious that this was the source of the bizarre phone problems but they insisted that their work couldn’t cause 911 calls to be made.  I write down the name of the NY Tel guy, then I left to go talk to the branch manager.

When I got back to the office I spoke to the branch manager. I told him what I had discovered about the NY Tel work. He was dubious, but I gave him the name of the NY Tel guy and requested that he call NY Tel and verify the story. He dismissed me and I left his office. On the way back to the phone room I heard one of the associates scolding their young kid to stop playing with the phone.

Gee, I wonder who called 911?

After a conference call with my company president, the customer’s corporate IT manager and the branch manager I was allowed to finish the installation. When the kids and the NY Tel truck left, strangely enough all of the problems disappeared. Not much else of interest happened after that. It took an extra day and I was happy as hell to leave.

My laptop was 2+ years old when I noticed that the cooling fans were running louder than I thought they should. I decided to try to figure out what was going on.

After consulting the manufacturer’s web site, I found that they give instructions for disassembly for maintainance and upgrades, such as adding more memory, replacing a video card, etc. I downloaded and printed the disassembly instructions.

With instructions in hand, I took it apart. Once I got the main cover off, it was obvious that it was dirty…really dirty.

In addition to finding out why the fans were running so loudly, I also found out why my speakers didn’t sound good anymore…they had a nasty buzz in them:

That's a dirty Speaker

Next I looked at the cooling fans. Here’s what the video card cooling fan looked like (the CPU fan was similar):

Video Card Dirty Fan Blades Detail

To clean the fans, I took them out. When I did this I was astonished to see the amount of debris clogging the airflow through the heat sink fins.

Video Card Heat Sink Debris

CPU Heatsink Debris

It’s worth noting that the amount of crud is probably greatly influenced by the amount your computer is used and the environment in which it is used. I lived on a dusty dirt road in an old house with cats and I smoke. I’m also a bachelor which means dusting for me is a once a year chore. So, YMMV.

Still, it seems like a good idea to at least take a peek and see if your laptop is asphyxiating. Mine was.

Thermite is a mixture of secret ingredients that, when ignited, burns seriously hot and creates boiling hot molten iron. It’s so hot that the reaction gives off UV light like an arc welder. That’s friggin’ HOT!

OK, so it’s neat stuff to burn things with, but what else can you do with it? You can weld iron and steel! (or so I’ve read) Time for an experiment!

I just so happened to have (of course) the necessary secret ingredients, so I decided to try to weld some scrap steel.

Here’s the setup in my home-made fume hood (sorry it’s messy, I use it for lots of different things):

Steel ready for thermite welding

Here’s a detail of the thermite on the joint to be welded:

Ready for fire!

OK, you know you wanna see it…

.

.

.

.

.

FLAME ON!

FLAME ON

DAMN! It’s a piece of the sun! Where are those sunglasses!

MORE FIRE!

That’s better…

Cooling...

Cooling more...

Well, the thermite certainly burned, but did it weld?

No weld, just blobs on the surface

Oh well, no weld.

Still, it was a successful experiment because I learned something:

• Thermite is HOT!
• Did I mention that it’s GDMF FRIGGIN’ HOT!?
• It’s too friggin’ hot to do indoors in a fume hood!

At least I didn’t burn anything that I didn’t want burned.

What’s next…?

I went into the basement near where I thought they were. There were some dead dried up bees on the floor and on the nearby window sills. I still couldn’t see the nest though.

I went back outside and measured from the bee hole to the nearest window, then I went back downstairs and measured off that distance. It was between two floor joists which have insulation packed between them at the outside wall. I figured that they were behind the insulation. I also figure that, given the number of bees coming and going, then nest was probably the size of a grapefruit. I carefully started pulling back the insulation until I started hearing some pissed off bees buzzing. That was confirmation enough for me. I got the hell outta the basement and sat down to think about how to deal with the situation.

Given the high cost of a professional to come out and remove them my sister and I decided to do it ourselves. Besides, it was a probably small nest anyway…

Recalling information from watching nature documentaries such as “Those Annoying Insects” I remembered that bees have different castes, one of which is the warrior whose job it is to protect the hive. I though it would be best to eliminate them first, so I took one of my favorite power tools – the shop vac – and went outside to the bee hole.

Once I get the shop vac set up and turned on, I started banging on the siding next to the bee hole. Sure enough, seriously pissed off bees started pouring out!

HAHA!! Eat my vacuum you nasty little bastards!

It took about 10 minutes but eventually I depleted the nest of the warriors. My shop vac stank like hell.

Before the main assault, we made “bee suits” out of heavy clothing. I made mine from a rain suit for motorcyclists, plus thick gloves and duct tape around the glove / sleeve joint and the shoe / trouser joint. A mosquito netting head cover completed my ensemble. The look was truly bizarre. We didn’t take pictures

After arranging lighting and ladders, I pulled back the insulation. This is what I saw:

Bees Revealed

Uh…it’s bigger than a grapefruit.

I continued to uncover it…with increasing awe and fear until I could appreciate it’s full glory:

The whole nest revealed

Oh $hit…

I’m really glad that I sucked up all of the warriors because that would have been a serious situation with them present. Even without them there were a LOT of pissed off bees buzzing around my face. Let me tell you- even with the netting it was not a comfortable feeling.

I originally thought I’d just be grabbing a grapefruit-sized thing and stuffing it into a garbage bag. That wasn’t going to work with this monstrosity. We “retired” for further consideration.

After discussing things and having a shot or two of courage, we decided to wage chemical warfare. I returned to the nest with an insecticide bomb. The bomb was an interesting thing because it wasn’t aerosol. It was a small cylindrical capsule about 4cm in diameter by 4cm tall. There was an adhesive foil cover on one end that you peeled off to reveal a hole. On the other end were small holes. You then placed this, small holes side down, into a plastic cup that came with it. Finally, you add about 100ml of water and get the hell away from it. I put it directly beneath the hive. After a few seconds, steam began rising which crescendoed as a large billowing column. The bees didn’t like it. I left to prevent being gassed.

After an hour I went back downstairs. All of the bees were dead or near death. There was a pile of dead bees and bee larvae on the floor below the nest. Blea!

I used a spatula to dislodge the nest. Of course, in the process it dumped lots more slimy larvae on me and the floor. Here it is after being removed:

The Nest

For scale, it’s sitting on top of a 48 quart cooler. It was 1/2 a meter long and 1/3 of a meter wide! It weighed about 2kg!

Here are some delightful views from the side:

7 layers of fun

Mmmmm....larvae

After the “glorious” part of the job (removing the nest), the cleanup in the basement was relegated to the shop vac.

Pile o' bees for the shop vac

OMFG did it stink. I had to soak the hoses in hot bleach solution overnight to get that crap out.

So that’s the story of the bee’s nest.

One of the fun things about electronics is that you can build apparatus for experiments.  I built a xenon strobe circuit to investigate high-speed phenomenon. Here’s the circuit:

Flash Circuit Prototype

I wanted to be able to take a picture of a bullet in flight (OK, a pellet from a pellet gun). But first I needed to do some validation on the circuit to make sure that:

1. The flash was sufficiently bright (luminous)
2. The flash duration was sufficiently short
3. Total Luminous Energy was sufficient

The total luminous energy, which is what is needed for the photography to work, is the product of luminosity and duration. Luminosity is a function of the amount of current flowing through the flash tube. Duration is a how long the current is flowing through the tube. Since the total luminous energy is directly related to the amount of energy available in the caps, this means that the caps had to dump their charge as fast as possible. As I discovered, this depends heavily upon a property of the capacitors. More on this later.

First, I had to check the luminosity. Simple. Fire the flash in a darkened room and see what I got:

1st flash test

Not bright enough. Add more caps for more flash energy.

Here’s the second try after adding more caps:

2nd flash test

Seems bright enough to me now.

Next I needed to check the flash duration, which means I had to know how long it needed to be. I derived this from the following computation:

Estimated pellet speed = 350 m/s = 350000mm/s [Just over mach 1. I have a powerful pellet gun.]

Desired maximum “smearing” = 0.1mm

required duration = 35000mm per sec / 0.1mm = 1/3500000 sec

required flash duration = a bit less than 0.3 microseconds.

Wow! That’s short!

My method of validating the flash duration was to photograph a moderately fast event and check the results. The method that I chose was to capture the image of a knife rapidly cutting a fine wire which triggered the flash. I would slam the knife down as fast as possible to cut it. [The reason for the glove is to prevent electrocution. The wire carries a charge equal to the flash caps, which is 310 volts. Shooting it with a pellet wouldn't matter, and I didn't feel like redesigning the circuit to accommodate safely touching it so I just wore gloves.]

Here’s the first attempt:

First test

The first attempt revealed several things.

1. Take the image from the side
2. Focus on the wire
3. The total luminous energy is at least reasonably close to what’s required.
4. Since the blade tip isn’t grossly smeared, I’m at least within two or three magnitudes of the required duration.

Here are the subsequent attempts:

Second test

Significant smearing of the blade. Hmmm…

I altered the circuit by using heavy braided copper (OK, desoldering wick) to connect the flash tube directly to the caps.

Third test

Much better. Lets try again, but add some calibration marks to get an idea of actual timing:

Fourth test

Unfortunately, I seem to have lost the best image of the calibration marks. The white rectangle on the blade is paper printed with horizontal lines of specific width and spacing. There is still smearing of about 3mm (trust me). I couldn’t think of any more mods to the circuit so I had to try to understand why the duration was still too long.

I talked to an electrical engineer friend and he told me about Equivalent Series Resistance (ESR). Basically, it’s the resistance inherent to the capacitor itself. The speed at which a cap can charge or discharge is known as the RC time constant. It is the product of the capacitance and the resistance of the circuit. This means that a low ESR cap has an inherent RC time constant that is shorter than high ESR caps. Basically I had crappy caps. They were just generic electrolytics harvested from an old monitor. That’s why there are such things as photo flash caps. They’re low ESR.

At the time I did the experiment, I never computed the ESR of the caps. I’ll try it now. This is only a rough estimate.

Time Constant = resistance “R”  (in ohms) x capacitance “C” (in farads)

ESR = the “R” from RC time constant computation.

Flash duration = length of smear / speed of knife

I estimate the speed of the knife to be 25 m/s

I estimate the length of the smear to be 3mm.

Duration = 3mm/25000mm per sec = 0.00012 sec = 120 microseconds

(Hmmm… I needed less than 0.3 microseconds? No wonder…)

Capacitance = 25 microfarads = 0.000025 farads

0.00012 sec flash duration = R x 0.000025 farads

Actual ESR = R = 0.00012 sec / 0.000025 farads =  4.8 ohms.

Required ESR = 0.0000003 sec / 0.000025 farads = 0.012 ohms

Ratio of actual ESR to required ESR = 4.8 ohms / 0.012 ohms = 400.

MY ESR WAS 400 TIMES TOO HIGH! Yikes! My caps were total junk!

I didn’t have any low ESR caps. I considered increasing the charge voltage and decreasing the capacitance (decreased capacitance would proportionately decrease the time constant while the increased voltage would keep the total energy the same), but I didn’t have any higher voltage caps in the range needed. I didn’t like the idea of going to 620 volts with the glove thing either. I finally decided to terminate the experiment. I didn’t reach my original goal but I did learn something, which is always my goal when experimenting.

Afterthought: Even assuming that I could get such low ESR high voltage high capacitance caps, it would mean that the strobe output would have to be 400 times as bright too. I’m pretty sure that the little horseshoe tube would explode. Hats off to Dr. Harold Edgerton, who invented the xenon flash tube and did this stuff in the 30′s and 40′s. The man was a genius.

After building many a prototype circuit on breadboards (which is certainly fun), I decided that I wanted to keep some of the more useful circuits around. I’ve used perfboard but it’s just not as cool as a real PCB. Long ago in High school, I did PCB’s using the old Radio Shack etch resist pen to draw the traces but you can never get really professional looking results that way. With my drawing ability, it always looked like a 6 year old drew it with crayons. So, I decided to look into other methods.

I discovered the “laser printer toner transfer” method on the web. There are many web sites with full descriptions so I won’t delve into great detail. In short, you develop the etch mask artwork on the computer then print a mirror image onto paper with a laser printer. Then, using an ordinary iron, you press the paper onto the blank copper face of your PCB. The toner, being thermoplastic, remelts and bonds to the copper. then you remove the paper backing by soaking in water and careful scrubbing. Here is an example of what you get:

1 inch square PCB test mask

I remembered from high school days of PCB manufacture that etching took a long time and wasn’t uniform. A friend of mine is an electrical engineer and I spoke to him about how they did their in-house prototyping. They used a machine that sprayed hot etchant onto the PCB and it finished them very quickly with excellent uniformity. I didn’t want to try to make something so complex so I thought a thank of hot recirculating etchant would work too. So I made an etching tank and used a fish tank heater to warm the etchant and a fish tank air pump to generate fine bubbles via an air stone. I don’t have pics of the entire setup, but here’s the tank:

Home Made PCB Etching Tank

I made it so that one quart of etchant would fit with about 1 1/2 inches to spare. This is important because the bubbles generate a fine mist of etchant, kind of like Coke does when it’s fizzing in a glass. If there’s not enough space above the liquid level to catch most of this mist, you end up wetting the surroundings with etchant. Messy. However, it can’t have too much space because the tank heater, which clamps onto the top edge of the glass, won’t be immersed far enough. So, it’s a compromise. The best option as I discovered is to just cover the top with plastic wrap. The other main consideration in the design was to make it as wide and deep as practical to allow the largest possible circuit board to be etched. This is why it’s only about 1 1/2 inched thick.

So, enough of describing the tank. Here is a result:

Etched test PCB

The finest lines near the top are 15 mil with 5 mil spacing. It worked with this small board but it’s not reliable enough for actual work. Practical limits are about 10 mil with 10 mil spacing. Basically, it worked completely to my satisfaction.

Notes

• The thick boarder around the edge of the board is there for functional purposes. When removing the paper backing after the water soak, the bulk of the removal is done by just peeling the paper off. Without the boarder, the stress of initially tearing the paper away is placed on real circuit traces which are very fragile. The boarder is heavy and is also sacrificial, so it probably won’t come off. Even if it does come off there’s no harm. It also acts as a harmless way to check the ironing before putting it in water. It takes some practice, but by pulling on the paper after the board has cooled, you can tell if the toner is stuck well enough. If not, you can just iron it some more.
• The paper used makes a BIG difference in the quality of the etch mask. The paper to use is the clay-coated ink jet paper. Make sure to print onto the clay coated side. Regular paper doesn’t work well at all because during pressing, the pressure forces the liquid toner further into the paper fibers and bonds to them. Soaking in water allows the unbonded fibers to be removed, but the ones that are bonded can’t be removed without removing some of the toner as well. It’s inevitable that some will be removed in the process of cleaning the paper off of the board. This results in “porous” traces which are a serious pain for soldering. If this occurs in thin traces, it causes open circuits which are hard to find. They also look like crap. Here is an example:

Porous Traces

• When pressing the mask onto the copper, the pressure applied per square inch needs to be high to assure good bonding the the toner to the copper. With this small board, I could easily exert 50 pounds per square inch. As the board size increases the square inches increases, which proportionately reduces the pressure per square inch. So you need to really press hard on larger boards to get good results.

Anyway, that’s my dissertation on PCB manufacture.

Of course, you can’t have a collection of scream art without Edvard Munch’s The Scream:

The Scream by Edvard Munch

This is the one that was hanging in my friend’s office. I thought he needed to switch to decaf.

Francis Bacon - Chronicle Head VI - 1949

Now they start getting scary…

Unknown

droste effect - josh sommers

More Francis Bacon…he’s a weird dude…

Figure with Meat - Francis Bacon

Scream Detail Study for Figures at the Base of a Crucifixion - Francis Bacon - 1944

Unknown

If you have any more disturbing scream art, let me know.

Katana

Components

I like planes to have lots of power. While flying normally, you don’t need to use full throttle. Nice moderate throttle settings give pleasant flying. But if you get into a bad situation, you have the option to use lots of power to pull you out of it.

I chose the Super Tigre .10 Outrunner brushless motor. It’s got way plenty of power for vertical climbs and hovering. Sweet.

Super Tigre .10 Outrunner Brushless Motor

Motor Installation

I had a little bit of an issue with the speed controller. I bought a used one from the shop where I got the plane. He guaranteed it to work. It didn’t. But, true to his word, he replaced it – free of charge. The really cool part (which more than made up for the inconvenience) was that the new one he replaced it with was much more expensive! That’s the way to gain loyal customers. I’ll go out of my way to buy from him.

Anyway, the controller that I ended up with is the Castle Creations Thunderbird 36. Programmable for various things.

Castle Creations Thunderbird 36A Speed Controller Front

Castle Creations Thunderbird 36A Speed Controller Back

The Battery I chose was a used Electrifly BP1500.

Construction

The build instructions are more than adequate, unless you’re a novice or haven’t built anything in years…like me. I screwed up the tail: I glued the rudder in without clamping it. It ‘s off the center line axis of the plane and it’s canted to one side. Not much, but enough to piss me off. I also screwed up the horizontal stab. Measured from firewall to the tip of the stab on both sides and equalized. Unfortunately I should have (like it said in the instructions!) measured from the tip of the rudder to the tips of the stab. Result:  Stab off center and angled. So much for precision aerobatics. I suppose I could add some mixing to the radio setup: rudder -> elevator, rudder -> ailerons, elevator -> rudder, elevator -> ailerons. That would could maybe correct most of the problem. Still, it’s annoying because I know better. I was trying to go too fast instead of carefully planning and rehersing the gluing. Yes, rehersing. I learned to do that on another plane. It really helps identify potential gotchas.

Flying

Ummm…wow. I can’t believe I didn’t turn it into a pile of toothpicks. It’s seriously maneuverable, and it’s fast. I’m holding off on flying it again until I get my thumbs limbered up on a more sedate model. Still, it was really fun. It’ll do anything you ask it to do.  4 point rolls (about all I can manage) are rock solid. It flies like it’s on rails. Snap rolls are insane – about 4/sec if you enter it fast with full throttle.

The design does have a tendency to enter asymmetric stalls (“tip stalls” in RC parlance). This is inherent to the design and is not necessarily a flaw; it’s more of a feature – unless you’re not that good a pilot like me and tend to do slow approaches. I need to work on that. Fortunately the one time I flew it I managed a good landing.

Summary

It’s a really cool airplane. Well constructed. Fun to build and fly. Not for novices.

As for the name, I just liked that particular picture and I wanted something to go with it. The name of my blog is pertinent only in a sense that that’s how I tend to live my life.

On with the show!