https://webshed.org/ Recent content on webshed.org Hugo -- gohugo.io en-gb Mon, 03 Jan 2022 21:26:32 +0000 https://webshed.org/projects/harpicolourchemistry/ Mon, 08 Sep 2025 13:53:37 +0100 https://webshed.org/projects/harpicolourchemistry/ Colour change chemistry with Harpic Power Plus and potassium permanganate. https://webshed.org/blog/sonoffs60tpg/ Wed, 01 Jan 2025 14:37:09 +0000 https://webshed.org/blog/sonoffs60tpg/ The Problem I need to be able to remotely power cycle a device in my shed. While I can sometimes go down to the shed and manually reset the device, having the ability to power cycle it remotely would be an advantage. This becomes even more valuable if I can set a periodic power cycle or programmatically power cycle it based on some detected condition. I posted a query on Mastodon asking for personal recommendations for a ‘smart’ plug socket that I could integrate with HomeAssistant or that would otherwise allow me to control the socket’s state (on/off). A Solution? One of the more common suggestions for a device that would do what I wanted was a Sonoff S60TPG. This reasonably priced (£13.32 when I ordered it) device claims to integrate with Home Assistant or can be used entirely locally without cloud integration. It also offers voltage, current, and power monitoring and scriptable control of the socket status. https://webshed.org/blog/fedi-tags/ Sun, 01 Dec 2024 14:26:55 +0000 https://webshed.org/blog/fedi-tags/ I was looking at Neil’s blog the other day and saw a post about adding fediverse tags to pages built with hugo. I’ve just implemented the changes here to add the tags around my content. This post is basically just a test to see if if it worked. https://webshed.org/blog/debugging_a_weatherstation/ Sun, 24 Nov 2024 13:46:09 +0000 https://webshed.org/blog/debugging_a_weatherstation/ I’ve had a Pimoroni weather station running in my garden for around two years. It connects to my public wifi and feeds weather readings over MQTT to a Raspberry Pi running Grafana. It has worked well, with the occasional glitch from time to time. Usually, all that’s needed is to power cycle the weather station to get it working again. Recently, power cycling hasn’t been enough; the weather station has remained resolutely dead. Today, in the middle of Storm Bert, I found some time to take the station down from the 4-metre pole it sits atop to see if I could diagnose its failure. Literal debugging revealed that the Raspberry Pi Pico W, which samples the sensors and controls the weather station’s communication, had many dead flies stuck to it. The flies seem to be fruit flies of some kind - my knowledge of small black flies is lacking. A Pico W on a weather station board with many dead flies attached Many dead flies around the USB connector I cleaned away the flies with a soft brush and some isopropyl alcohol and checked the connections for corrosion. https://webshed.org/projects/leadwhite/ Wed, 16 Oct 2024 17:03:45 +0100 https://webshed.org/projects/leadwhite/ This experiment uses lead, a toxic heavy metal. Several stages of the process make organic lead compounds even more toxic. Seriously, don’t mess with this stuff. It’ll fuck you up. With the warning out the way, let’s see if we can make Lead White, a heavy white pigment composed primarily of basic lead carbonate, 2PbCO3·Pb(OH)2. It was used in art from antiquity until it was first replaced by zinc white, then titanium white starting in the 19th century. Early methods of production Crude lead white can be produced by suspending lead above a pot of vinegar and allowing the acid vapours and carbon dioxide in the air to react. This is relatively slow at room temperature, and as the atmospheric concentration of carbon dioxide is low, the yield of lead white is also low. An improved method developed in the 17th century used horse manure around the vinegar-lead assembly, which provided heat from the decomposing manure and an enriched carbon dioxide atmosphere. https://webshed.org/projects/leaf_ph/ Mon, 14 Oct 2024 15:50:18 +0000 https://webshed.org/projects/leaf_ph/ Probably - but let’s experiment anyway. In the Autumn (Fall in another part of the world), the leaves on the trees change colour from green to not-as-green-perhaps-red. Red organic compounds are somewhat pH sensitive - you’ve probably seen the experiment where the extract from red cabbage is made to change colour with the addition of vinegar or baking soda. Can we do the same with leaves? Extracting the colour from leaves I know from years of experience that alcohol will extract the chlorophyll from green leaves, so I’ll try it to exact the red stuff. Some autumn leaves Chopping the leaves increases the surface area and helps the solvent penetrate the cut edges instead of the waxy, flat surfaces. I ended up stirring the leaves in a warm 50% ethanol solution (25ml of methylated spirit and 25ml of water). chopped_leaves in a beaker with solvent There should have been a video here but your browser does not seem to support it. https://webshed.org/blog/cross_compiling_rust_code/ Fri, 04 Oct 2024 13:27:05 +0100 https://webshed.org/blog/cross_compiling_rust_code/ I wanted to deploy some LLM poisoning code on my server, with the aim of serving up plausible - but nonsensical posts to the various AI scrapers that come by and suck down the entire content of my websites from time to time. Tim McCormack over at brainonfire had some rust code called marco written that takes example text and produces a stream of junk from it, just what I needed. The code builds fine on macOS (my desktop OS of choice) but I couldn’t build it on my Linux based server becuase the version of rust installed was too old. Luckily it’s possible to cross compile from macOS to Linux. This is how I did it. You’ll need homebrew installed rustup target add x86_64-unknown-linux-gnu brew tap SergioBenitez/osxct brew install x86_64-unknown-linux-gnu I added a new target to the Cargo.toml file [target.x86_64-unknown-linux-gnu] linker = "x86_64-unknown-linux-gnu-gcc" Because of some path error I couldn’t debug in the 30 seconds I bothered trying, I just specified the toolchain, etc on the build line in the marko directory. https://webshed.org/blog/lloytronn2405bk/ Tue, 03 Sep 2024 14:04:26 +0100 https://webshed.org/blog/lloytronn2405bk/ The Lloytron N2405BK Calypso I was browsing around Amazon and spotted this rather nice looking radio priced up at £20 with a time limited discount down to £15. I liked the look of the retro styling and wondered if the tuning dial hid a nice variable capacitor or if it was a digital encoder; despite (definitely) not needing another radio I placed an order for one. The radio arrived in a nice cardboard box, inside the box the radio was wrapped in tissue paper. A figure-of-8 power cord was also supplied - a nice touch as these cables seem to vanish shortly after entering my house; I can never find one when I need it. Can we guess what’s inside the radio? The radio takes 4 x AA cells (or mains power) and draws about 80mA listening to FM stations at reasonable volume. This high current draw indicates to me that the radio is probably not completely analogue. https://webshed.org/projects/heavywater/ Mon, 02 Sep 2024 15:44:54 +0100 https://webshed.org/projects/heavywater/ I have a bottle of heavy water; how pure is it? Sometime in the distant past, I acquired a partially used bottle of heavy water. It’s been in storage since I got it; I’ve not really done anything with it. The bottle claims to be 99% pure, but it is also cautioned that the liquid inside is hygroscopic and that the bottle should be refilled with an inert atmosphere. I have no idea if this has ever been done. I don’t think I’ve ever opened the bottle myself. A thread on Mastodon mentioned heavy water, and I realised I had a sample. I thought it might be interesting to see how pure my sample was. This short article documents how I did this. A hand holding a brown bottle with a red cap and a white and red label The label reads Aldrich Deuterium oxide 99 atom percent D A hand holding a brown bottle with a red cap and a white label. https://webshed.org/projects/everreadyskyleaderresurection/ Sun, 30 Jun 2024 21:34:59 +0100 https://webshed.org/projects/everreadyskyleaderresurection/ EverReady SkyLeader The EverReady SkyLeader radio was introduced in 1958 and went though a few revisions with the last versions being turned out in the early 1960s. The early versions had socketed transistors (OC44, OC45) in the RF stages and were point to point wired. I was looking for a repair challenge, and happened up one of these radios on eBay. I bid and won it for £10 plus postage. When it arrived I discovered it was a later model with a PCB and soldered in transistors, but on searching for something else in the shed, I discovered a 1958 vintage version I didn’t even know I owned. Initial observations The hinges on the case are broken and there are two missing transistors, and I’m further checking, one of the other transistors has been pulled out of the socket and collector snapped off (this may be because it was soldered on the backside of the socket on some of these radios. https://webshed.org/projects/daguerreotype/ Sun, 24 Dec 2023 15:50:18 +0000 https://webshed.org/projects/daguerreotype/ Can I make a Becquerel Daguerreotype in a lunchtime? I’ve wanted to make a Daguerreotype for a long time, the problem is you need mercury to develop the photograph, you need silverplate to capture the image on, a plate camera, iodine, thiosulphate and various other materials. My concern for a long time has been getting silverplate to actually create the daguerreotype on, it used to be readily available but I’ve not been successful in acquiring any recently despite looking in many secondhand and house clearance shops. One thing I do have lots of is copper clad PCB material, this has anywhere from fifty to a few hundred microns of copper stuck to either fibreglass or resin bonded paper substrate. PCB material is easily cut and handled, so if I can just find a way to silverplate this then I’ll have a ready stock of material I can make daguerreotypes on. https://webshed.org/blog/computerlogiclab/ Wed, 06 Dec 2023 18:47:27 +0000 https://webshed.org/blog/computerlogiclab/ Update: September 2025, I’ve scanned and uploaded the manual to the Internet Archive. I was given this electronic kit for either my birthday or Christmas in 1986. It’s pretty basic by today’s standards; the only IC is a TTL 7400 Quad NAND Gate. The only inputs and outputs are three Morse-key style switches for inputs, two red LEDs make up the visual output, and a high-impedance crystal earpiece is used for audio output from oscillator circuits. HUMBROL Computer Logic Lab Box Computer Logic Lab Kit Circuits are built by connecting individual components with flying leads pushed into the springs that form terminals on each component. It’s basic, but it works. I quickly worked through all the example circuits in the accompanying book, then used the kit for a few more years, building my own circuits on it and connecting it to other kits and circuits on my bench. https://webshed.org/projects/silverrecovery/ Sun, 09 Jul 2023 17:54:04 +0000 https://webshed.org/projects/silverrecovery/ A local archive was disposing of some microfilm they no longer required (I checked, and it was all duplicated and had also been digitised). It was just going to be thrown away, so I asked for some - partly to test the resolution of the microCT scanner at work, and partly to see if I could extract any silver from it. Two rolls of microfilm After some experimentation I worked out that dilute bleach in hot water would rapid strip the silver from the plastic film material. If I had a tray, I could have run the film though the tray though the bleach bath, instead I just cut up sections of film that had lots of silver (they were dark) and socked them for a while in bleach, until they turned clear. Three containers of bleach solution, the rightmost has most of the silver chloride precipitated out The bleach first seems to breakdown the adhesion between the silver and the plastic film, the silver moves into solution as a dark mass, which then reacts with the bleach to form white silver chloride. https://webshed.org/projects/mediumwave-activeantenna/ Sun, 09 Jul 2023 13:56:31 +0100 https://webshed.org/projects/mediumwave-activeantenna/ The Problem Most medium wave receivers use a ferrite rod type antenna, as this keeps things compact and works fairly well. In a ferrite rod antenna, a tuned coil is wound around a rod of ferrite material, the rod increases the inductance of the coil due to the material’s high magnetic permeability, allowing fewer turns of wire on the coil for the same inductance compared to an air-cored coil. A disadvantage of the ferrite rod antenna is it’s not really suited for picking up weak stations, the antenna is physically small and can’t be positioned up clear away from sources of noise, or just to get the best signal. It is possible to add a small coupling loop to the ferrite rod and connect this to an external wire antenna, but if we’re doing this, why not just make an entirely external antenna, doing away with the ferrite rod? This circuit aims to do just that. https://webshed.org/projects/mwradio/ Sun, 09 Jul 2023 13:51:56 +0100 https://webshed.org/projects/mwradio/ I want to build a good quality medium wave radio receiver, something that can receive some of the more distant stations that become receivable in the evening, but also hold up against the very strong local stations that can literally be received with a just germanium diode connected to a crystal earpiece. Usually when I design and build a radio, I start at the audio amplifier end, because even if nothing else in the radio works, you have an audio amplifier. This time I’m going to start at the antenna and work my way along - for some reason I have a surfeit of audio amplifiers… I want the radio to use an external antenna, rather than rely on the more common internal ferrite rod antennas used in medium wave radios. So I’ll start with a circuit to connect an antenna of random impedance to the ~50 Ω impedance circuits I’ll use in the rest of the radio. https://webshed.org/projects/mediumwave-bandpassfilter/ Thu, 25 May 2023 17:40:55 +0000 https://webshed.org/projects/mediumwave-bandpassfilter/ The medium wave radio band runs from 531–1,602 kHz with 4.5kHz bandwidth in Europe, and 530–1,700 kHz with 5kHz bandwidth elsewhere. A bandpass filter should only pass 500kHz – 2MHz approximately for good front-end filtering. A filter was calculated on rf-tools.com using standard component values and the following parameters: 5th Order Chebyshev Bandpass Conventional, Series First Lower Cutoff Freq. = 500 KHz Upper Cutoff Freq. = 2 MHz Passband Ripple = 0.10 dB The filter schematic diagram The filter schematic diagram Filter parts list 2 x 4.7nF 3 x 2.2 nF 2 x 5.6 µH 2 x 8.2 µH 1 x 10 µH Simulation The filter was simulated in QUCS with these parameters and equations: S-parameter simulation: Logarithmic Start : 25kHz Stop : 40 MHz Points : 1001 Equations: dBS21 = dB(S[2,1]) dBS11 = dB(S[1,1]) Group Delay = -diff(unwrap(angle(S[2,1])),2*pi*frequency) S21 Plot of the filter Plotting S21 vs frequency gives this very nice filter shape. https://webshed.org/projects/raspberrypi/mdns/ Thu, 23 Mar 2023 20:36:26 +0000 https://webshed.org/projects/raspberrypi/mdns/ Setting up mDNS (ZeroConf) If you have a single Raspberry Pi (rPi) on your network, you can connect to it via its IP address (192.168.1.128 for example) or via a local DNS name of raspberrypi.local ssh pi@192.168.1.128 ssh pi@raspberrypi.local If you have multiple rPi on the network, you can either have to remember several IP addresses, or you can use the magic of mDNS (part of a suite of network technologies called ZeroConf) to give them all useful, memorable names. You’ll need to set the wanted name of the Pi in two files (the content of my files are shown), basically replace any occurrence of ‘raspberrypi’ with the name you want. I’m using weather here, because this rPi will be used to collect and store weather sensor data. 1. /etc/hostname pi@weather:~ $ cat /etc/hostname weather 2. /etc/hosts pi@weather:~ $ cat /etc/hosts 127.0.0.1 localhost ::1 localhost ip6-localhost ip6-loopback ff02::1 ip6-allnodes ff02::2 ip6-allrouters 127. https://webshed.org/blog/bbc_radio5_converter/ Sat, 31 Dec 2022 12:59:45 +0000 https://webshed.org/blog/bbc_radio5_converter/ This is from a twitter thread from August 2022 I’ve left it pretty much as the twitter thread, just removing some comments that don’t belong to me. Getting re-acquainted with the deeper, more spider inhabited recesses of my shed, I found this: Sun Aug 21 13:59:33 +0000 2022 Replying to @DTL The topside looks like this: Sun Aug 21 13:59:35 +0000 2022 Replying to @DTL The underneath has a stylised BBC Radio 5 logo Sun Aug 21 13:59:37 +0000 2022 Replying to @DTL Back of the unit. Sun Aug 21 13:59:39 +0000 2022 Replying to @DTL Four screw holes, but only three screws, I suspect someone has been inside this before. Sun Aug 21 14:03:55 +0000 2022 Replying to @DTL The circuit board. A Medium Wave ferrite rod antenna, a TEA 5570 AM/FM Radio IC from Phillips, an NE612 Mixer / Oscillator and a 68. https://webshed.org/projects/arduino_th/ Wed, 23 Nov 2022 14:22:33 +0000 https://webshed.org/projects/arduino_th/ I was contacted by a small city archive to provide a temperature and humidity logging system for their store rooms. Heritage items are kept under standardised conditions in storage to prevent (or at least retard) deterioration, pest damage and mould growth; ensuring those conditions are maintained over long periods of time is essential. Commercial data logging systems exist, but are prohibitively expensive for small institutions. The requirement was for a number of data loggers recording ambient temperature & humidity values every 10 minutes (user configurable). A low-power RF link on 2.4 GHz was investigated for remote reporting sensors, but the RF environment in the archive was inhospitable - thick walls, archive spread over several floors and strong co-channel interference from building WiFi installation. Instead, it was decided to record sensor readings directly on the sensor nodes for periodic collection. Sensors I initially planned to use a combination of DS18B20 1-wire temperature sensor and SHT21 humidity sensor on the data logger, both parts I’m very familiar with. https://webshed.org/main/main/ Mon, 03 Jan 2022 20:29:56 +0000 https://webshed.org/main/main/ https://webshed.org/page/about/ Mon, 03 Jan 2022 19:57:31 +0000 https://webshed.org/page/about/ I’m Dr. David Mills. I have a PhD in Blowing things up with lasers, and now I work with X-rays. This website has been around in one form or another since 2000-11-24, it’s gone from a page created in notepad, which was a pain to keep updated, to one based on Mediawiki which had great promise, but was also a pain to keep updated, it’s now based on Hugo with the Cupper theme and hosted & updated with a bunch of git magic. In theory, all I have do do now is type into a text editor on my computer and type a simple command and the site updates, this doesn’t sound too painful, but who knows… This site languished for a while, because it was a pain to update (see above) and it was just too easy to post to Twitter (see below). I much prefer the idea of writing for my own site, and my own interest, so I’ve brought in new technology and giving it another try here. https://webshed.org/blog/metercal/ Thu, 30 Dec 2021 11:53:20 +0000 https://webshed.org/blog/metercal/ Calibrating an RF power meter I needed a rough and ready measurement of RF power at the -40 to 0dBm range, at frequencies from a few hundred kHz to VHF. I’ve got interested in some radio projects again, specifically making some small signal amplifiers. To check how well the amplifiers work, I need to know what input signal I’m driving them with. I can easily measure their relative gain using a VNA and calibrated attenuators, but had no way to know accurately how much signal I was driving them amplifier with. Ebay RF power meter The meter has sat unused for over a year because it came with zero documentation. Turing it on just yield a frequency selection and somewhere to set the offset of the device. It’s powered from USB and does appear as a USB device on the computer, but there’s no detail on how to talk to it. https://webshed.org/blog/website/thiswebsite/ Wed, 29 Dec 2021 19:06:32 +0000 https://webshed.org/blog/website/thiswebsite/ How this website is setup using Hugo and Github Actions This website uses Hugo to process markdown test created on any machine and pushed to a git repository on github. This lets me update the website from anywhere with a text editor and an ssh client, and removes the need for me to run PHP on the server (for running Mediawiki or Wordpress, as I have in the past). In my opinion, PHP applications offer too many security holes. This guide assumes three things, You have a server running Debian linux You have a Github account that allows Actions You have a “local” computer you can create content on and can push to the github account. These are the steps I took to tie all this together. Step 1. Install hugo on the server. Fairly self explanatory, for the purposes of this I am not using the extended version of hugo, so had no need to install npm or similar, so I don’t cover setting this up. https://webshed.org/projects/heart_blinkenlights/ Tue, 18 Feb 2014 13:58:39 +0000 https://webshed.org/projects/heart_blinkenlights/ Introduction I wanted to make something for my partner for Valentine’s Day 2014. Browsing the local Maplin shop, I came across a Velleman kit in the shape of a heart with approximately thirty 3mm red LEDs. In its standard form, the kit just flashes all the LEDs on and off at a few Hz - it’s just a two transistor astable circuit - nothing too exciting. I bought it and started thinking about how to upgrade it once I had it back in the workshop. First off, I ditched the supplied LEDs switchig the out for some 5mm ultra high brightness ones, unfortunately I ordered 8mm by mistake so some filing and brute force was needed to get them to fit the PCB. I’d wanted to try Charlieplexing for some time, so this seemed like a good opportunity. I had some PIC16F883 devices in SMT form in the spares bin, these are small enough to fit on the back of the heart PCB and will run with no extra parts - that made selection of controller easy - use those. https://webshed.org/projects/2011-vintage-carbon-monoxide-sensor-teardown/ Mon, 13 May 2013 00:00:00 +0000 https://webshed.org/projects/2011-vintage-carbon-monoxide-sensor-teardown/ My British Gas supplied carbon monoxide sensor started bleeping at me, not the usual battery-dying bleep code, but a rather more serious detector-malfunction bleep code. This makes it fair game for a teardown! The PCB The whole of the alarm is implemented on a single PCB, the two most obvious parts being the electrochemical sensor and the piezo sounder. Just peaking out from under the sensor is something that looks rather like a Microchip PIC ISCP port header. fig:Co-alarm-ICSP-port.jpg The Sensor The carbon monoxide sensor is a Figaro TGS5042, this is an electrochemical device that generates an electrical current that depends upon the concentration of carbon monoxide that defuses into the body of the device. As with all sensors of this type, the device has a limited working life - failure being sure to the liquid electrolytes drying out. The 2-D barcode encodes quite a bit of information about this sensor, but I’ve not found anything that can read it. https://webshed.org/projects/raspberrypi-ds1820/ Sun, 03 Feb 2013 20:19:00 +0000 https://webshed.org/projects/raspberrypi-ds1820/ How to use DS18x20 1-wire temperature sensors with the Raspberry PI Original date : 2013-02-03 The Raspberry PI (rPI) has several different serial buses brought out on its GPIO (General Purpose Input/Output) pins, including SPI and I2C, however there is no 1-Wire interface. Luckily, in modern Linux Kernels there is a driver module for bit-banging a 1-Wire interface using a single GPIO pin. In recent [http://www.raspbian.org/ Raspbian “wheezy”] releases GPIO-4 is the pin used. The rest of this article will describe how to use this module and connect DS18x20 series 1-Wire temperature sensors to the rPI. I built my breakout interface directly onto a plastic pin header for testing purposes, but this is hardly the best way to do it. Hardware To interface a DS18x20 sensor to the rPI, the bare minimum hardware you need is a single 4.7K resistor, the DS18x20 sensor and a plug that will fit the rPI GPIO pins. https://webshed.org/projects/testing-low-frequency-losses-in-broadband-transformer-materials/ Tue, 29 Jan 2013 00:00:00 +0000 https://webshed.org/projects/testing-low-frequency-losses-in-broadband-transformer-materials/ FT50-43 and FT50-26 Transformers In the cause of converting a Softrock Lite II receiver to work on the new 472kHz (660m) amateur band, I found I needed a broadband transformer with low loss characteristics at these frequencies. The recommended core material to construct broadband transformers at low radio frequencies is 75-mix ferrite, I had none to hand, so I’ve decided to measure up what I do have handy. Methods Construct a tri-filiar winding of 5 turns on both FT50-43 and FT50-26 cores, keeping the windings as close to identical as possible. Insertion loss (S21) between two pairs of the windings will be measured with a DG8SAQ VNWA. The third winding will be left unconnected. Loss will be measured from 10 kHz to 100 MHz and plotted on a log frequency scale. A spot measurement of S21 at 473kHz will also be made. FT50-26 FT50-43 Results The results are not what I expected to see, I’d been lead to believe that 43-mix material wasn’t much use below around 1MHz. https://webshed.org/projects/ad9850_arduino/ Wed, 22 Aug 2012 17:25:00 +0100 https://webshed.org/projects/ad9850_arduino/ This article was first published on an earlier version of this website in 2012. I’ve re-uploaded here primarily for historical interest. There is a fair bit of information regarding the AD9851 DDS (as used on the NJQRP DDS-60 daughter card) chip with Microchip PICs and Arduino development boards, but not much for the slightly cheaper and lower spec (but still good) AD9850. The AD9850 has no internal reference clock multiplier, requiring a faster reference clock than the AD9851. Not having an internal multiplier means that the code written for the AD9851 will not work directly on the AD9850 - the code will attempt to set registers that do not exist in the AD9850. A no-frills AD9850 DDS module can be obtained from eBay for around £4 at the time of writing (August 2012) and will produce a decent signal from a few Hz to over 30 MHz. A read of the datasheet provides all the information needed to drive the DDS chip and get some RF out. https://webshed.org/projects/in-circuit-esr-meter/ Mon, 09 Jan 2012 00:00:00 +0000 https://webshed.org/projects/in-circuit-esr-meter/ An Equivalent Series Resistance Meter Some time ago I was trying to repair a switched mode power supply (SMPS) in my oscilloscope. I’d been quoted £800 for a new PSU, so trying to fix it first was well worth my while. It’s pretty common for electrolytic capacitors to develop faults in an SMPS; specifically they develop an higher than normal internal resistance. So while the capacitor may hold a charge and measure as the correct capacitance, it will not behave correctly in a filter or PSU circuit. The ESR of a large high voltage, high capacity electrolytic capacitor should be fractions of an Ohm, smaller capacitors have ESRs of a few Ohms typically. As the capacitor degrades in use the ESR can easily climb to several hundred times the normal value, while exhibiting no changes to voltage and capacitance ratings Suspecting the capacitors, and not having a way to measure their series resistance, I set out to design and build an ESR meter. https://webshed.org/projects/ipod-transmitter/ Sun, 05 Jun 2011 00:00:00 +0000 https://webshed.org/projects/ipod-transmitter/ A simple FM transmitter for the iPod similar devices My partner wanted to connect her iPod to her stereo system, but the stereo doesn’t have any external inputs. The simplest solution, avoiding diving into the wiring in the stereo, is to transmit the audio from the iPod to the FM radio in the stereo system. This gave me the opportunity to play with Tetsuo Kogawa’s MicroFM transmitter design as featured recently on Mark VandeWettering’s website. The circuit really is about as simple as it gets, a single transistor is configured as a free-running L-C oscillator somewhere between 88 - 108 MHz (the FM broadcast band) and signals on the transistor base changes - modulates - the oscillator frequency. R4 and R5, the pair of 1k resistors in the schematic are just to passively mix the left and right audio channels from the iPod, producing a mono signal to drive the transmitter. https://webshed.org/projects/144-mhz-qrss/ Wed, 23 Mar 2011 00:00:00 +0000 https://webshed.org/projects/144-mhz-qrss/ 144 MHz QRSS Beacon / MEPT A radio rally purchase in 2010 turned out to be a telemetry or data transmitter on 160-ish MHz. The interesting part of the board was the 12x frequency multiplier and PA. The RF section of the board has four connections to the digital section. These include +5v for the oscillator and frequency multiplier, a second +5v supply for the PA transistor, a modulation input and a signal to switch on or off the first frequency tripler to kill the RF output without stopping the oscillator. Ground is common to both sections of the board. Tele-tx.jpeg The telemetry transmitter with the RF shield part lifted off. Conversion to 144 MHz was surprisingly easy; the oscillator crystal was swapped for a 12 MHz computer grade part and wired in series with a 10-60pF trimmer cap to allow fine frequency adjustment. https://webshed.org/projects/g0rnn_coil_tool/ Mon, 01 Jan 0001 00:00:00 +0000 https://webshed.org/projects/g0rnn_coil_tool/ Original Date : 28 January 2015 This is a fantastically simple and useful tool designed and built by Chick G0RNN (SK) and described to me by Ray G1OJP (SK). Constructed entirely from tinplate from a large tin of baked beans or similar, the tool provides a third-hand making toroid coils easy to wind. There isn’t a lot more to say, the design is pretty self-explanatory from the photographs. The only thing to consider during construction of the tool, is that cut tin-plate will have sharp edges, so should be smoothed with emery paper or similar before soldering. A tinplate tri-wing spike made from an old bean can, soldered to the top of a bean can A toroidal core former is put onto the spike and wire can easily be wound though the core The spike holds the toroid well and lets you easily pass the wire though. https://webshed.org/projects/mercury_pto/ Mon, 01 Jan 0001 00:00:00 +0000 https://webshed.org/projects/mercury_pto/ A dangerous idea - a mercury tuned oscillator! Original date Feb 2011 This experiment uses mercury, brief exposure to mercury isn’t a great problem, but it is a cumulative poison and it attacks the central nervous system. It’s also hard to clean up if you spill any. Don’t try this at home! Some time ago I started to build Steve Webber’s MMR40 radio, which uses a PTO (permeability tuned oscillator). I got the PTO working and then thought about ways to improve it before I finished building the rig. At the time I had access to some very finely graduated glass syringes and I pondered a plunger type tuning arrangement for the PTO. The frequency could be calibrated from the graduations of the syringe. Not wanting to damage the nice glass syringe, I wound a coil on a cheap plastic 5ml syringe and measured the inductance with the syringe empty and with it full of water, saturated NaCl and saturated CuSO4 solutions. https://webshed.org/projects/raspberrypi-multiple-ds1820/ Mon, 01 Jan 0001 00:00:00 +0000 https://webshed.org/projects/raspberrypi-multiple-ds1820/ Multiple DS18x20 1-wire sensors on a Raspberry PI Edited 16/07/2014 to show how to use new version of the code Edited 22/12/2013 to add a link to github Edited 01/01/2013 to fix buggy code, etc In a previous article I showed how to use a 1-wire temperature sensor with the Raspberry PI with minimal interface requirements. The nice thing with 1-wire sensors in that multiple devices can share the same bus. You just wire the sensors in parallel - all GND pins tired together, all DQ pins tied together and all VCC pins (if you use them) tired together. The perl code developed for the previous article can only access a single sensor. To read multiple sensors we need to fist get the device IDs of all sensors on the bus, then read each sensor’s data. On the rPI using the w1 kernel drivers, the file /sys/bus/w1/devices/w1\_bus\_master1/w1\_master\_slaves contains a list of all the device IDs detected on the 1-wire bus. https://webshed.org/projects/pnp80/ Mon, 01 Jan 0001 00:00:00 +0000 https://webshed.org/projects/pnp80/ PNP-80 - An 80m band receiver using only PNP transistors Original date : August 2010 fig:Unknown PNP transistor The idea for this project came when I bought 200 unknown transistors for £1 at a radio rally a few years ago. On getting them home I discovered they were PNP and measured a gain of about 250. I thought it might be fun to build a radio using just these, then put them away and forgot all about it for two years. They have a Motorola logo and part number, but no datasheet I’ve found seems to match these parts. Barring more information, I’ve decided to use these parts as essentially plastic cased PNP BC108 companions. The 80m band is pretty active at any time and the low frequency requires very little from components in a receiver, this is especially suited when you know almost nothing about the parts you’ve set out to build with. https://webshed.org/projects/a-simple-singnal-injector/ Mon, 01 Jan 0001 00:00:00 +0000 https://webshed.org/projects/a-simple-singnal-injector/ A simple signal injector for troubleshooting radio receivers A signal injector is a low-power oscillator that generates a signal containing many harmonics in the audio and low-frequency RF spectrum. This allows you to test a radio receiver’s audio and RF stages. They are very simple to build, and every workbench should have one. The circuit The basic circuit for for signal injector is an Astable Oscillator - two transistors configured to switch the state of the other alternately transistor. fig:Basic circuit diagram of astable oscillator The oscillator’s operating frequency is set by R2, R3 and C1 & C2. where… f is the frequency in hertz. R2 and R3 are resistor values in ohms. C1 and C2 are capacitor values in farads. T is the period. The output approximates a square wave with the duty cycle controlled by the values of R2 and R3. In the case where R2 = R3, the duty cycle is very close to 50%, and the circuit produces a reasonable square wave. https://webshed.org/information/tokocoils/ Mon, 01 Jan 0001 00:00:00 +0000 https://webshed.org/information/tokocoils/ If you’ve ended up here, you are probably looking for information on some random TOKO RCL inductor you’ve found in your junk box. I got fed up of searching on part-numbers only to find people asking the same questions, trying to ID various parts, so I collected together the information I was able to find. This is the information I’ve collected so far: Measured values from random TOKO coils in my collection. Can Marking Type Inductance 154AN-T1005 TK1203-ND Transformer (5 pin) 4.7uH adjustable [Datasheet] 00 714 88304 Variable inductor (2 pins) 0.14uH to 0.13uH [measured] 007 44 88453 Variable inductor (2 pins) 0.22uH to 0.27uH [measured] 00 714 88303 Variable inductor (2 pins) 0.14uH to 0.13uH [measured] 007 44 88463 Variable inductor (2 pins) 0.26uH to 0.22uH [measured] Values from data sheets Part Number Inductance Range uH Q min.