Friday, 11 September 2020

Boosting broadband speed with satellite co-axial cable

When I moved into this house, there was a lot of extraneous wiring. Phone line extensions that went nowhere, that kinda thing, so I ripped out everything back to the master socket by the front door.

When fibre was enabled, I had a visit from an Openreach engineer, reporting a fault detected on the line - the case was the old master socket had some corrosion, so he replaced it with a new one.

However, I want to move all my network gear to the under-stairs cupboard, and branch everything from there. As I don't use a landline, this would mostly just mean Wi-fi and a few ethernet cables.

There's a not insignificant cost to have Openreach come out and move the master socket, so I'd rather avoid that. It's also not something you can really do yourself - the master socket marks the point in the connection where the providers responsibility for the wiring ends and yours begins. If you tried to move it yourself and screwed up, there's a significant charge to that too.

Just using a phone extension lead works, but slows the broadband speed noticably.

Unfortunately, due to the age of the phone system and it's various evolutions over the years, finding decent, up to date information is not as easy as it should be.
I manged to put together a DIY solution that boosted my broadband speed by approx 50% (~14 Mb download speed with regular phone extension, to ~21 Mb using my solution)

This was done by combining a few other ideas and sources of information.
This is what worked for me. Your results may vary, and usual "don't try this at home" warnings apply.

Moving the master socket without moving it

I found this guide which shows a master socket with 'A' and 'B' connections, which it states, you can run to another socket using CAT5 network cable, which will have a similar effect to moving the master socket, as it bypasses a lot of the filtering of the sockets, and CAT5, being twisted-pair cable, has better shielding qualities than regular phone extension.

But unfortunately, my master socket is the newer type 'NTE5C' socket, so it doesn't have the same connections.. It does have a similar A and B connector, but it's on the provider side of the socket, and is in use, so I don't want to mess around with that.

It give me an idea though. Even if that route is out of the question, a lot of the interference comes from the cabling used. If I were to create a phone extension lead from CAT5, it should still be an improvement to a regular phone line. 

Looking up how to install a telephone extension on the NTE5C type of master socket led me to this video which shows using the front panel to connect an extension using pins 2, 3, and 5.

Pins 2 and 5, according to the first article, are the equivalents of A & B.

So, what if I put in a telephone extension using that connector like in the video, but instead used better cable as per the article? In theory, the speed should improve due to the increase in cable quality, but would lose a bit because a microfilter would be required (because it's not been pre-filtered.)

I also thought, if CAT5 beats regular phone cable because it's twisted pair gives it some shielding, then wouldn't something more shielded so even better? I had a usable length of co-axial cable, the type used by satellite dishes and antennas.

This is a single copper core with a mesh-like shield all the way around it. These would provide the connections to pins 2 & 5 (I used the core for pin 2, the shield for 5, but I don't think it'd matter which way round).

The article does recommend solid core wire, which the core of the coaxial is, but the shielding is not, so I guess there may be some performance trade off there.

This was wired to a regular phone socket under the stairs, in which there's a microfilter and the router.

Initially the speed fluctuated a bit, but once it bedded in, it settled on a speed (as reported by the router) of approximately 21Mb, which is up from the approximately 14Mb that was achieved with the same length of regular phone extension cable.

I'd be tempted to experiment further (ideally with something with twin solid cores and shielded.), as there's still a bit of room for improvement - at the master socket directly, the router could get 26Mb - but for now it's fast enough.

Friday, 21 August 2020

Straight razor from table saw blade

So a while back I replaced the table saw blade in the workshop and got the idea in my head to try and make a straight razor out of it.

I sketched out a quick design and then cut it from some scraps of redwood (the same stuff I used in the ring box project - not sure if that's the correct species, but it's wood, and it's a red-ish colour, so close enough.) on the bandsaw. Fortunately I had two scraps the same thickness, so I needn't worry about the thickness of each of the scales (sides) being identical.

The blade itself was sketched out in marker on the old table saw blade and cut out with an angle grinder. Once the rough shape was done, a mix of bench grinder and rotary grinding tools were used to refine the shape.

The gap between the 2 scales of the handle was padded out using some brass sheet, cut to match the shape of the scales, whilst including cut out space both for the blade to fold in to, and to allow the tang to rotate into when the blade is opened.

This, and the scales were both epoxied together, and then two holes were drilled and brass pins added, for additional strength and their own aesthetic quality.

The third brass pin provides the hinge that the blade will pivot from to make the opening and closing mechanism.

Before the blade and handle were put together, it was easier to finish each part separately.

In the case of the handle, this meant sanding to round off the edges and make it more comfortable to hold, and a wipe out with danish oil.

In the case of the blade this meant a lot of grinding, sharpening and honing to get it to a sharp enough edge to cut hairs. With hindsight although it probably isn't the best steel to use for a blade, it is sharp enough to shave with - it just requires frequent honing.

But, at the end of the day it works, I'm pleased with how it's turned out, and I learnt a bit more about blade making and the dark arts of sharpening and honing along the way. I also now totally get why some people fall into this particular niche rather than just more general metalworking... I'm actually quite tempted to go find some better steel and make another already.

Saturday, 8 August 2020

Tattoo Machine for Leatherwork

This is the fifth of my 'Lockdown Projects' - Projects made during the Coronavirus lockdown, which I limited to one week, and only using materials I had to hand.
Before I start this write up I just want to emphasize this tattoo gun is not intended for use on anything other than leather. Using it for anything else risks a number of heath issues.
With that out of the way, I’ve been intrigued by the idea of using tattooing as a way of patterning it. A quick search shows that it’s not an uncommon practice – particularly it would seem, for leather boots.
The theory goes that given that leather is essentially just skin, it would tattoo in the similar way to a person.
After seeing a home-made tattoo gun in an episode of Orange is the New Black years ago, I’ve had the idea in the back of my head to build one and try it.
During lockdown, it seemed like a thematic fit to finally hack one together.
The body of the machine is from a mechanical pencil. 
The main needle mechanism is a sewing needle, with an empty metal pen refill, and a hook made of a bent paperclip. These bits are epoxied together for sturdiness.
The motor was taken from a handheld fan. A button provided the connection between the needle mechanism and the motor.
The bracket was a small bit of metal from the junk bin - that was probably once part of a printer or something similar I’d previously dismantled.
The motor and bracket were fixed together with cable ties and a final cable tie tensions the mechanism so that there’s less rattle and wobble in the needle.
Initially I used 2xAA batteries for a power supply, but after a bit of testing, it needed more power, so I replaced that with a USB connector.
Obviously I didn’t have any tattoo ink to hand, so I just used leather dye to test.

Monday, 22 June 2020

Context-sensitive macro keypad

I've been doing quite a bit of 3D work recently for both 3D printing and VR software development.

The main software that I use for 3D work are OpenSCAD and Blender.

One annoyance I have is that the short cut keys each use for standard things such as scale, rotate, move, etc. are all different, which makes switching between applications more awkward than it need be.

So my latest electronics project is to create universal short cut keys - a physical keypad that has single button presses for those functions, but then translates them into the relevant keyboard/mouse presses dependent on which application has focus at the time.

The hardware
As we're still under lockdown due to COVID-19, I'm restricted to using only components and tools I have got at home, like with the USB Switch project.

The keypad itself is this mechanical key switch tester.

Unfortunately the eighth switch was lost some time ago. I'll leave the gap there as potential room for expansion in future though.

The controller is an Arduino Pro Micro. I thought about using a different microcontroller, however the real sticking point was my lack of a spare USB to serial adapter.

The Pro Micro helps keep the overall form factor nice and compact, whilst keeping the programming side of things straightforward.

I picked out some LEDs to backlight the keys as there was spacing for them, and the keycaps are translucent. I tried to find all different colours, but my inventory didn't allow for that, so there's a couple of duplications.

The rest of the hardware is just some diodes and resistors, a bit of perfboard and some scavenged wires.

The hardware is wired up like so:
Quite simply, there's two matrices, one for the switches, and one for the LEDs. Because of the missing key, one column in each only has three connections.

The case is just some plastic container that would otherwise have been trash. It wasn't a perfect fit so required some Dremel-based customisation.

The Firmware

The Arduino will stay set up as a USB to Serial device rather than as a keyboard because there needs to be bi-directional communication - the Arduino needs to also receive input from the application in order to control the lights.

This post was a useful resource when putting together the matrices.

The full source is available in the project on GitHub.

The Software
The software is a java desktop app that communicates with the Arduino using jSerialComm. It controls which keys are 'set' and indicates this with the LEDs.

The interfacing with the other applications comes from a thread that uses system calls to the 'xdotool' linux command, reading in the output of that command which is the name of the current window that has focus. There's room here to make the application multi-platform, by implementing a similar Windows command in something like AutoHotKey.

If the window matches the defined rule set, the relevant keys are highlighted and a thread monitors serial input. If that key is pressed, java built-in Robot class sends the relevant commands to the application.

As with the firmware, the source is available on GitHub. It's in a pretty simplistic state at the moment due to the one-week time limit I've placed on these lockdown projects, but there's room to expand and improve in future.

Tuesday, 26 May 2020

Tray Stand / Foldable Table

While lockdown has prevented me from being able to get back to many woodworking projects this year, I did manage to cut a few offcuts to size and sand them in order to fashion a stand for a bamboo tray. The idea being for it to be a temporary table or drinks stand in our small patch of outside space.

The construction is a simple X-frame, with horizontal bars at each of the four points - 2 to support the tray, 2 to act as feet. These are joined to the X-frame using a 3D-printed bracket - the SCAD file for this is up on GitHub.

In order to stop the frame from spreading out and collapsing under the weight of the tray and it's contents, the feet were joined with some denim fabric, courtesy of an old pair of jeans. This allows the stand to be folded up, but providing plenty of rigidity during use.

Sewing things other than leather is a pretty new thing to me, my only prior sewing-related project being the lanyard I made a while back. So I'm the first to admit it's not the neatest work, but it does the job, so can't complain.

With the benefit of hindsight, the hinges allowed a bit too much flexibility, allowing the cross-bars to roll. The fabric on the feet alleviates this at the bottom. On the top, the base of the tray has a small lip which helps latch onto the cross-bars - whilst they still roll, it's impact is minimised and the tray/table works fine.

Thursday, 30 April 2020

Yarn slack winder

This is another of the lockdown projects where I'm trying to keep myself occupied during the Covid-19 lockdown by challenging myself to a project per week, using only materials from my workshop junk bin.

I don't crochet or knit, but my partner does, and I often end up watching TV whilst having been delegated to idly unwinding a ball of wool.

So, apparently, when crocheting or knitting, it's important to make sure that there's some slack between the workpiece and the ball of yarn. This usually means stopping every so often to pull more yarn from the ball. It can lead to inconsistent tension in the workpiece, making the work uneven.

This gave me an idea to create a yarn dispenser that could be hands-free, and unwind small amounts at a time, to maintain slack.

The OpenSCAD model
The design

The basic principle is to repurpose two rubber rollers (grey) from a printer. The bottom one is attached to a motor, which is housed in the case (red), and supported at the other wide by a support (yellow).
The top roller is attached to two mounts (blue) and is free-rolling.

The two mounts will be attached to their respective parts of the case by screws, where the tightness can be adjusted to allow more or less gap between the rollers, to vary the grip depending on the thickness/density of the yarn.
The yarn will be sandwiched between the two rollers, so that when the motor is activated by a foot pedal, it is pulled between the rollers.

To keep the two sides of the frame separate, they were mounted to a small piece of scrap acrylic I found. It's not a perfect size, but it's functional enough for this prototype.

The 3D printed models are pretty basic, but should anyone want to use them, they're up on GitHub.

The electronics

The motor is some generic DC motor that, like most things in the junk bin, was probably pulled from an old printer.

Most of the circuit for this was was salvaged from the old Smoke Machine project.

The SN754410 was de-soldered so that it would function in a more conventional way, powered from a 9v wall wart, and a TS7805 regular to provide 5v.

The circuit. The SW1 switch allows the direction of the motor to be easily switched. SW2 is the foot pedal that will be pressed to feed the yarn.

The electronics were built into an old business card box, similar to the USB KVM Switch.

With the circuit hot glued in place, a piece of a disposable pen was glued to the button to extend it's reach, so that it would sit slightly higher than the top of the box lid. This means the the circuit enclosure also doubles up as the foot pedal - pushing down on the lid pushes on the pen, and in turn presses the switch.

Finished product
In initial testing, it was discovered that the yarn would veer off to the side of the rollers and become tangled. As a quick-fix solution, I used a scrap of leather, punched a guide hole in the middle and tacked it to the input side of the rollers.

Tuesday, 31 March 2020

Simple USB KVM adapter

With the shadow of Covid-19 hanging over our heads, I, like many others, am now working from home.

Having my home PC set up right next to my work set up has been a bit of a pain - two keyboards and two mice competing for desk space. Plus the work keyboard and mouse was a spare wireless set I had kicking about - not really comfortable enough for long-term use.

The monitor I attached to my work laptop was the second monitor I use for my home set-up - thankfully it has two inputs, so that's an easy one-button switch.

I wanted to use my main keyboard and mouse with my work set-up as well as home, but didn't want to unplug/replug each time.

I also wasn't too keen or just ordering a USB KVM adapter (a device which lets multiple computers share keyboard, mouse and monitor), as recently some of our local postal and delivery works have fallen ill, and so I'm avoiding any non-essential deliveries - I don't like the idea of putting others at risk for stuff that's a 'nice-to-have'.

So I figured I'd look through the electronics pile and see what I could piece together.

I did have a PS/2 KVM that I was hoping to adapt, but having read the datasheet for the bus IC that underpins it, it had a low maximum current rating - it might have been fine for just a keyboard or a mouse, but the keyboard I want to use has a backlight, audio controller and USB hub built in, and a brief look with

lsusb -v

showed that they would easily surpass that maximum rating.

Software options are also a no go - whilst they'd been fine when working locally, they won't work with my work's VPN, the the IT guys have enough on their plate without me adding to it.

So, I ended up going for the simplest of options. I had some relays - they were obviously meant for higher power application than this, but found that wired up in parallel, 5V was enough to trigger them. I was planning on powering it from a USB port, so 5V was the target.

A rough schematic of the switch

Unfortunately they had a weird footprint that put one of the pins in between rows on standard perfboard, so I ended up dead-bug soldering them.
A diode across each of them to prevent voltage spikes from the coils, some USB extension leads and a switch later, and I had a 4 channel switch.

I was a bit concerned at the lack of a protective load on this circuit, but adding any additional resistance from the power to the relays would prevent them from switching - as they're really designed for higher power applications, I reason that the coils have enough power requirement to prevent a short.

I have ran it for a few days under close observation and seen no temperature / magic smoke issues, and it's only ever powered when I'm sat right in front of it, so it should be fine. Obviously if this wasn't a "whatever's at hand" project, there'd be much more elegant ways of doing this (bus IC, 4 pole switch, more appropriate relays, etc), but here we are.

A gratuitous amount of cable ties, double sided tape and an old business card box provides structure and an enclosure. I'll probably tidy it up a bit more later as time allows, but for now it's enough that it works.