floz's corner

floz nietsnezolf

ravings of a tech geek gone awry!




bored?



For two days now, I've opened this window and done diddly squat with it.
I don't think there's much to say, aside from the world is still ending
in its usual way.

Oooh, I know! Nerf guns.

I may have the opportunity soon to mod a Nerf "machinegun". Technically
I suppose this particular nerf blaster is an automatic rifle (yes, it
has rifling in the barrel)... but with a couple of drum mags, it sure
fills the role of LMG well.

So, let's go over the basics of how these things work, and what we can
do to change its performance.

Firstly, there are the flywheels. I will try to use blaster instead of
gun, because that's the right term? Anyway, most automatic nerf
blasters work on a flywheel system. Two opposed flywheels grip the nerf
dart and push it out of the barrel. This means there's also some sort
of automatic loader mechanism. In the case of this one, it's a motor
turning a cam or crank to effect an in/out linear motion. This in/out
linear motion pushes the dart free of the top of the magazine and
JUUUuuust into the grip of those flywheels.

So we have two things going on at the same time. The autoloader
mechanism pushes the next dart out of the magazine, and the flywheels
grab that protruding dart and zip it up to high speeds. Both of these
(in a stock nerf blaster) are performed by brushed DC motors. Brushed
DC motors run at a given RPM per volt applied. This means that if we
increase the voltage going to a brushed DC motor, it will spin faster.
There is, obviously, an upper limit. Mechanically and electrically,
things can only go so fast. Eventually you run into overheated
armatures, heat-soaked magnets, and fried bearings.

Let's think this through a bit further and make sure we're not going to
cook any parts. So we know that we're working with two sets of brushed
DC motors. We know that they're likely 370 size or bigger. We also
know that they're powered by four D-cell batteries (1.5V per in series)
providing peak voltage of about 7V.

I have (due to other hobbies) some 2S (two cells in series) lithium
polymer battery packs. These are rated at 7.4V nominal... however when
fully charged and balanced, the actual pack voltage reads around 8.15V.

These lithium polymer batteries use a standard connector, which I have
spares of. The battery tray in this nerf blaster appears to be at least
partially modular, possibly allowing a reversible install.

Now you should see where I'm going with this. We're going to over-volt
this nerf blaster. I expect a slight increase in rate of fire, and an
increase in muzzle velocity. Basically like fine-tuning the gas system
on a Kalishnikov.

While a simple overvolt is one of the easier mods to do, I would like to
leave the OEM battery tray as in-tact as possible to make this mod
reversible.

Other mods that can be made to an automatic nerf blaster include the
motors, the flywheels themselves, the flywheel holder/cage, etc.
Basically anything about how it shoots can be turned up to ridiculous,
given enough money and time.

If you've ever tinkered with airsoft or now gel-blaster guns, you may be
familiar with overvolt as a mod. In the case of many airsoft guns, the
end result of increased voltage is an increased rate of fire, since the
whole mechanism is a single motor and depends on retracting a
pneumatic piston to fire the pellet.

A little more digging online will turn up 3d printable mods. Things
like flywheel cages to hold brushless DC motors. In this case, the
brushless DC motors used are similar to what would fit in smaller racing
quadcopters. With a ridiculous amount of RPM, I can imagine they can
zip a dart at painful velocities.

Dig a bit further and you'll hit what I consider bedrock in this topic.
A fully 3d printed full-automatic nerf blaster with a mini-gun
equivalent rate of fire and painful muzzle velocities. Yep, that's
right, people build sick nerf machinepistol blasters now with 3d
printing and hobby electronics.

What, you don't believe me? Well fine, here... gaze upon the bedrock of
nerf modding and be amazed.

NERF MOD: THE FDL-2 (3D-Printed, Select Fire, Mag-Fed Marvel)

3d Printed: Nerf Rival Uzi Full-Auto Blaster


-floz

some thoughts on the end of the world



I looked; I saw a pale horse, and his rider was Death, and Hell
followed with him.

-The Bible, book of Revelation (paraphrased)

I've been getting back into the UFO thing lately. I've always been
curious about what might be going on. There are strange things in the
sky, teasing our military pilots. There are strange things reported
throughout history in the sky.

You might find it odd to start a post by paraphrasing the biblical
description of the end of the world... but hopefully I can show that it
makes some sense.

You see, I look at so-called "holy" texts as what they are, books.
Books can be fiction, non-fiction, or sometimes an amalgam of the two.
This means to me that there are probably some accounts of things in
those "holy" books that are historically fairly accurate, while other
things are more likely to be "tall tales told for the benefit of the
reader".

I've heard a variety of theories about the one book in question that I
paraphrased at the start. One theory is that it was witnessing a
thermonuclear conflict from the perspective of someone in that era.
Another theory is that it was witnessing the destruction of humanity
from an outside threat... again from the perspective of someone in that
era.

Circling back around to UFOs. Now we call them UAPs (unidentified
aerial phenomena). Whatever you call it, the definition starts with the
interesting part, it's UNIDENTIFIED. Now, you could argue that any
airplane I see in the sky is unidentified to me, because I don't know
the specific airframe, tail number, or route it's on. However, in a
general sense, I know that it's an airplane. I can hear the engine, see
the lift surfaces, etc. I've personally never witnessed a UFO or UAP
when sober. I have, in the past, taken hallucinogenic plants that
caused me to see them, but that's not really the same.

I worked on RADAR systems for a short while in the US military. I
understand the fundamentals of how a RADAR system works, and some finer
details like beam-forming and phased-arrays. I can say with some
certainty that RADAR is not foolproof. Lots of things can cause a false
return. However, these things tend to display as known false returns.
For example, ground clutter pretty much always looks like ground
clutter. Atmospheric conditions that affect returns pretty much always
look the same for the same conditions.

The RADAR I worked on most was a system designed to identify incoming
missiles, as well as being used to vector warfighters to their target
areas. It is used in conjunction with other data-generating systems
including AWACs to provide a picture of the airspace over a region. For
fear of telling things that I should not, it's a powerful system that
covers a LARGE area. It is configured when deployed by highly skilled
individuals who know even more about it than I do. My point is, it's
not exactly error-prone when being used for its primary mission.

When we see a RADAR return or "track" that does not appear to be a
known return, such as atmospheric conditions, missiles, aircraft, ground
clutter, or other things... It's interesting. What's more interesting
is when a return covers hundreds of miles in a stunningly short period
of time. RADAR antennas tend to rotate at a fixed rate. You get one
return for a given airborne object per rotation. If you plot a return
at 100 miles range, and on the next sweep, it's 100 miles range in the
opposite direction from your site, that means it covered 200 miles in a
matter of seconds. This kind of performance is not impossible, but
rare. Remember, Mach-1 is 700 nautical miles per hour at sea level.
This means that it is moving at approximately 0.194 miles per second.
In order to cover 200 nautical miles in a matter of 6 seconds or less, a
vehicle would need to be going about 720,000 miles per hour, or Mach
1000+.

The example I've just given is simplified for the purposes of being
sensible to read, but reflects some of the things we've seen in the
past. Again, not I personally, but there have been RADAR returns that
performed in this way, which exceeds anything traditional by a big
margin.

Now, let's hedge our bets a bit, let's cut that velocity down to 1/4 of
normal. Let's say the vehicle covered 50 miles in 10 seconds. That's 5
miles per second, or... 18,000 miles per hour. Incidentally that's the
approximate velocity that orbiting objects maintain, such as the ISS.
Doing this kind of speed within our atmosphere raises a lot of
questions, like friction/heat due to air resistance, how in the hell you
got going that fast in the first place, and how do you change vector?

Let's assume the reports that have RADAR data to confirm location and
vector are accurate. This means something (or rather multiple
somethings) have been detected exceeding orbital escape velocity within
our atmosphere. There is no clear explanation in classical physics for
this... in short, it's considered impossible by most sensible people.

What if this thing or its occupants, whatever it might be, has bad
intentions? What if it wants this plant because it's covered in
saltwater? What if it wants this planet for the iron/nickel core?
If it can do 18,000 mph or more in our atmosphere, we humans don't
stand a snowballs chance in hades.


Maybe, just maybe that white horse was a screaming blinding white-hot
craft projecting horrifying imagery to the terrified people below, as it
rained down atomic fire upon anything that dared protrude from the
surface. Or maybe it was an ICBM, who knows?

I looked; I saw a pale horse, and his rider was Death, and Hell
followed with him.


-floz

3d print quality versus speed



Some things you 3d print will take a very very long time. The larger an
object is, or the more complex it is, or even how much plastic it
contains affect print time. There are other factors as well, such as
material type (TPU tends to print more slowly than PLA for example).

There are times you might want to check the overall dimensionality of a
large object, but don't want to spend 18 hours and a considerable
fraction of a spool of filament to make a "test fit" part. This is
where a slicer profile for "minimum viable" is handy.

Here's a quick rundown of my minimum viable profile:

layer height 0.32mm (0.4mm nozzle)
1 top/bottom layer
1 wall
5 to 10% infill depending on geometry
minimal supports
150% feed rate
110% flow rate

Some of these might seem at odds with one another, so I will explain my
thinking behind this setup.

With a 0.4mm nozzle, I'm extruding a 0.4mm diameter "string" of
filament. 0.32mm layer height is just a hair under (0.08mm) that size
and thus will have moderate layer adhesion on my machine. Again, we're
not printing a structural part, just a test-fit.

1 layer for top/bottom skins stops the printer from making a thicker
top/bottom wall for the object than is needed to start printing infill
on. This reduces both print time and plastic consumed.

1 wall width means only make 1 outer shell wall. We're not concerned
with pretty or structurally strong, but rather "pretty close"
dimensionally.

Infill is one of those things that is eagerly debated amongst 3d
printing hobbyists. The straightforward answer of "more infill =
stronger part" isn't always so. We've figured out over the years that
more perimeter walls can do approximately as much for rigidity as
infill. No matter where you stand on the question, less infill = less
extrusion = less time.

150% feed rate means run the gantry at 150% speed. All X Y Z and E
moves are performed at 150% of the slicer's defined speed. I run
60-65mm per second travel speed in the slicer, so this means moving as
much as 97.5mm per second travel speed on the machine.

110% flow rate is entirely an optional thing here for minimum viable,
but I've found that extruding 5-10% extra on every layer helps that tall
0.32mm layer height adhere a bit better. We're not terribly
concerned about a little layer seperation, the resulting part will be
ugly compared to a fine print of the same model, but may take 20% of the
time to complete.

As a final example, I have a model I've designed that takes 2 hours and
45 minutes on this draft setting and 10-12 hours on "fine" with high
infill. This difference in time can help make your rapid prototyping a
little more rapid.

-floz

blinkenlichten fur das auto



More ideas for the light kit.

To catch you up, if you're new here or something... I've designed an
Arduino-based programable LED kit for RC use.

Originally designed for aircraft, the board is pretty small and I'm
planning to make it even smaller while adding more features. I'm
currently evaluating software options.

I see a couple of possible ways to shave this particular yak.

  1. Flash the board as ground, air, or sea... each with programmable
    options.
  2. Flash the board with "client" and ship a programmer app and cable.


I'm liking the second option more and more as the software that is
riding on the vehicle gets more and more feature-rich. This would be
something similar to a Castle-Link but for your lights. Simply plug it
into USB and run the "Flightlites" app on your PC, configure the board,
and then you're good to go.

This means that the app that "programs" the board would need to set at
least the following variables:


So basically, you identify what kind of vehicle it is... boat, truck,
airplane, drone.

Then you select what sub-class; military, civilian, or emergency/special

Then! you can identify what lighting features you want... and heck,
let's shoot for assigning them to specific points in the daisy chain of
LEDs.

I could probably compile a Windows and Mac app from Python to do the
serial comm with the board. Obviously the board itself will need a
"save to EEPROM" function to save settings... otherwise the user will
need to setup each time they swap batteries.

Again, I like this better than flashing an LED on the board with a table
of "flashes 1 time, pause, flashes 3 times = boat, military" and that's
just what options you get. The programmer app would give a lot more
flexibility for assigning different lighting functions to different
parts of the vehicle.

I just gotta quit fuckin' around and playin' video games.

-floz

only you can prevent gremlin hunting



Whenever you work on electronics, there are three fundamental values
that you absolutely must know. These three values make up a simple
algebraic formula that will permit finding any one of the three if given
the other two.

I'm of course talking about Ohm's Law, or...

Current, Voltage, and Resistance

I'm going to propose that one of these values is sometimes stated too
strictly on hobby electronics, and that's Voltage.

There are a variety of hobby electronics, with new kits and widgets
being released daily, almost hourly. Some common ones include AVR
microcontrollers, ARM single-board computers, low-power RF systems,
cameras, servomotors and motor controllers/motors, GPS... just to list
what you might find in a modern hobbyist RC vehicle.

Many of these devices run on a nominal 5 volts DC. Hobby servos get
5V, PWM, and ground from the port where they plug into a radio receiver
or controller. Motor controllers for smaller stuff are provided
whatever voltage they are rated at directly from the battery, and in
many cases will provide 5V and ground for the radio receiver or
controller.

This brings me to voltage sag. Voltage sag is a short
duration reduction in RMS voltage caused by a load in the
circuit. This load can be a short, overload, motor startup,
corrosion on connections, etc.

As an electronics hobbyist myself, I'm well aware that this
happens, and can plan ahead when putting subsystems together. A
simple device that will help for powering microcontrollers and
single-board computers on systems that will experience sag is a
capacitor or two to filter fluctuations in voltage, followed by
a regulator of whatever flavor you prefer. This additional $5
in parts (or less) can help save you a lot of headache when
troubleshooting later.

Microcontrollers and single-board computers can "brown-out"
when their supplied voltage sags. This can cause a wide variety
of issues, including lockups, crashes, faults. They can also
shutdown or reboot, only to draw their own surge current at
boot, causing further voltage sag. All manner of "gremlins"
will plague a system that is floating right around brown-out
voltage for the controlling logic.

A simple voltage regulator IC, some passives, and a filter cap
or two can save you the pain of chasing gremlins. Trust me on
this, I've been tinkering with an updated version of my
flitelights kit for RC lighting. Originally conceived as a way
to quickly and easily put programmable lighting on an RC
airplane, the next iteration will be electrically more solid and
have software that is a bit more mature.

I've transitioned to ground vehicles, because I'm a terrible
pilot. There is a HUGE want for programmable lighting for RC
vehicles of all types, so I'm going to pick development back up
on this one.

Some roadmap ideas for the project to go from its last release
to the next:




As for the software, I've got a few ideas...


So... I need to do some reading and hunting down parts, then get
to work in CAD, then have OSHpark spin off a few dev boards for
me. Once I've got new generation dev boards ready, I can build
a few and try various things with the software.

No idea how long this will take, and I may not have time to do
much in the near future, but I would really like to put more
into this project.

-floz

more tiny little trucks



I had an idea, while sitting in a rather dull meeting.

The monster-truck RC vehicle I'm building based on the frame of a cheap
chinese off-road truck needs a good body shell.

I've 3d printed a 50s era Chevy stepside cab and bed, and it doesn't
look awful... but it's not terribly durable.

WPL makes a Kei truck, if you don't know what a Kei truck is, it's one
of those funny little vans they use for work trucks in most of Asia. A
real-world example is the Suzuki Carry or the Mitsubishi Minicab.

WPL calls their Kei truck model the D-series, and it's available in a
drop-side bed/pickup configuration, or a van body with rear seats.

This got me thinking, back in the day Tamiya made a monster truck they
called "Lunchbox". It was a full size van body over a monster truck
chassis. There is even a re-launch of the model for nostalgia reasons.

Since the D42 (Kei truck van body) is modeled after a Suzuki Carry, a
very Japanese vehicle... it seems more appropriate to dub the truck
"Bento Box" instead of "Lunchbox"

So, if I can't find a paneltruck shell or some other shell I really
like, this little mini-monster might get a Kei truck cab for laughs,
complete with some custom graphics christening it as "Bento Box"...
maybe even in Kanji.

-floz

tiny little trucks



There is a Chinese company that makes little RC trucks. They make a
selection of 1:16th or around that size scale trucks, including a Kei
truck, a rock crawling pickup, and several 6x6 heavy trucks. If you've
guessed that I'm talking about WPL-RC, you're right.

So far I have two of their C-14/24 series chassis and a B-14 series
chassis. The C-series are a four-link suspension design with frame
rails and cross-members, not unlike a full size truck. While the
B-series is a straight-frame, more akin to a heavy truck, with leaf
springs.

I've been working on the first C-series for some time, part by part
upgrading it until it is mostly metal parts and quite quick for a
rock-crawler type vehicle. The other C-series is being built up into a
monster truck, since it's already a four-link setup.

As for the B-series... I removed the heavy truck cab and refitted a
pickup cab, then stuffed a motor suited to 1:10th scale racing into it.
After that, I 3d printed loads of other parts, including a completely
new custom designed front suspension. It literally does donuts until
the motor heat soaks and loses power or the tires come apart. I call it
the hoonigan build, and it's already blown up three rear differentials.

That last bit brings me to the conundrum of this hobby. Which truck do
I build/fix first? The monster truck is sitting on already blown up
axles and stock tires. The hoonigan is sitting with a blown up axle,
and the rock crawler is basically fine.

I want to fix the axle problems with hoonigan I think first. I've been
looking at parts from other models that might be slightly bigger or
beefier. For example, these are 1:16th scale. That means I could
potentially swap a 1:14th or 1:12th scale axle in. Granted, I may need
to machine some parts to make it narrower, but that's all part of
hotrodding.

The monster truck (last I checked and tallied it up) needs somewhere
around $150 in parts to be a rolling monster truck looking vehicle that
can be driven. That's tires, axles, electronics, drivetrain,
etc. Hoonigan needs something more like $40 to $60 in parts
to fix an ongoing problem (maybe for good).

Lastly, there is the topic of body shells. WPL sells a selection
including something like a Land Rover, something like an older Toyota
FJ, similar to Hilux, UAZ, GAZ, M35s, etc etc. These are old "styrene"
shells, like Tamiya used to do with some of their models
(monster beetle, I'm looking at you). That's all well and fine, but I'd
like to run a lexan/polycarbonate shell like a hobby RC car. So far the
closest I've found is shells to fit Traxxas 1:16th stuff. I haven't
purchased one yet, but I will when I find the right one. If you know of
a good 1:16th shell similar to "Grave Digger" (1950s or 1960s panel
truck), drop me a line.


-floz

what a fucking mess you've all made



A lot has happened in the last year.

We swore in a new president in the US.
SARS-CoV2 mutated several times.
I got married.
...the list goes on and on.

I peeked in on my news feed a minute ago, and I noticed an alarming
trend. In spite of, or perhaps due to, the stupidity of a crowd a year
ago, news articles are flush with pieces about it.

Anniversary of an insurrection, and the like.

I don't think we should forget what happened, and we should absolutely
address those responsible... but splattering every news medium with bits
and blurbs about it isn't really the right response either.

I identify as an Anarchist. Meaning I don't believe in the need for a
government or laws to dictate behavior. I can behave myself most of the
time without being told to, most other adults should be equally capable.

However, even though I don't believe in the system that is in place, it
is the system that is in place. "Piss with the cock ya got" as Uncle
Bumblefuck says. If you feel that the system is wrong, you can take
several courses.



I've chosen the last one. You can all have the non-ended stream of
stupidity, vitriol, hatemongering, fearmongering, and hurtful behavior
that keeps flooding from national-level elected officials. I choose to
laugh at them and engage locally.

Anyway, I've rambled enough. My point is, one op-ed piece per news
outlet would be sufficient. Not a dozen on the same damn thing from
every possible fucking angle.

-floz

more feet



So the first spot of peeling showed up today. Right on the end of my
big toe. The instructions say not to peel, so I'm fighting the urge.

I've been soaking my feet each day for 10-20 mins as instructed, and I
can tell a little difference in the callus. They're dried out feeling
and my feet feel a little odd when soaking.

We'll see how this progresses over the next days/weeks. I fully expect
to need to do the second round after this has finished.

-floz

my dogs are mooing



Following up the previous post about better living through chemistry,
I'd like to share something that I'm trying as an experiment.

You see, I get really really bad calluses on my feet. Seriously,
they're like hooves. It's bad, they hurt sometimes, they split
sometimes and hurt more.

Obviously the idea of visiting a podiatrist has crossed my mind. Until
recently I didn't have insurance that would cover such a thing, and it
wasn't really that big of a deal I thought. The thing is, being in a
constant small amount of pain is no more fun than being a lot of pain.
Pain sucks in general, unless you're into that sort of thing... and even
then, medical pain like this isn't fun.

So... my better half suggested a "heel peel". A bit of quick googling,
and it turns out (I'm a guy, I don't know this stuff) there are chemical
peel things for your feet to deal with this very problem.

The stuff inside is pretty safe to use externally, but does come with
some warnings. It works like so...

There are little plastic booties/socks that are sealed up and contain
"magic goo" aka chemical peel soup.

Wash your feet thoroughly beforehand.

You cut the top/neck of these booties/socks open and slip your feet in.

Unless you have exceptionally large feet as I do (12e) and then you're
kinda boned.

You will probably split them, put the whole thing foot and all in a
large plastic zip-top bag to ensure you don't lose the precious goo.

Let your feet soak for an hour in these booties.

Slip them off and discard, wash your feet thoroughly.

Then, the after-care. Each day following you should soak your feet in
lukewarm water for 10-20 minutes to help the peel.

Slowly (over two weeks I think) the dead skin and callus will peel away.
The instructions specify do not try to pick at or peel it, let it fall
off naturally.

So, day 1... my horrible feets. We'll see what is what in a few weeks.
The kit I ordered came with two sets of booties, and said that if needed
after it stops peeling I can go again.

-floz