As anyone who’s wanted to measure temperature with a Raspberry Pi knows, it doesn’t offer analog input. This means that the TMP36 sensor which is commonly used with the Arduino doesn’t work. The most popular option for the Pi then is the DS18B20, which is a digital sensor that utilizes the 1-wire interface and is natively supported by the Pi. It’s considerably more expensive than the TMP36, and as is the case when a component is expensive and in-demand, it’s prone to counterfeiting. I recently experienced this firsthand.
Here’s a project I completed last summer that I’m just finally getting around to sharing. It’s an “old-fashioned” bench grinder stand, or pedestal. I call it old-fashioned because I built it for a belt-driven grinder whereas grinders today are typically direct-drive.
The grinder itself is a General Hardware 6″ grinder with no model number. I picked it up at a garage sale for $5. I couldn’t find too much information on it, but what I did find indicates that it may have been manufactured any time between the 1950’s and 1970’s. I didn’t have a grinder, and I had some old utility motors laying around, so I figured it might be a cheap way to have one.
The easy solution would have been to just mount the grinder and a motor to a board and be done with it, but I wanted something a little nicer. After scouring the web for awhile, I stumbled across this webpage which had plans for a pedestal grinder stand, scanned from a November 1971 copy of Popular Mechanics magazine. This was it! Some of the information was cut off, so to be sure I had all the information I needed, I purchased a copy of the magazine off of Ebay, made my own scans, and filed it away until I’d have time to build it. That time finally came when I walked out to the garage one Saturday morning feeling bored but motivated and looking for something to do.
MAKE recently featured a very cool project that I had to build: the Elektrosluch! What’s an Elektrosluch? It’s basically a microphone of sorts that allows you to listen to electromagnetic interference (EMI) which is found all around us in our personal electronics, homes, automobiles, and many other places. The Elektrosluch includes a built-in amp, so listening can be done with headphones, or it can be connected to a recording device for sampling. The tutorial was written by Jonas Gruska, who designed the circuit. It was a lot of fun to build, and overall not too hard. I took my time and checked everything several times and it worked the first time I tried it.
Earlier this year I purchased an Arachnid Labs Re:load Pro, which is an adjustable constant-current load. I’ve always wanted an electronic load for my lab, but didn’t want the spend the money. The Re:load Pro solved that problem as it’s just $125. Sure, it doesn’t sink as much current or have all the options of fancier units, but it does everything I need. For testing panel meters, batteries and LEDs, it’s quite capable. So far, I’m happy.
One of the features that caught my interest when I purchased it was the ability to interact with it via a virtual serial port on the USB interface. I immediately got the idea to develop an application that could control a Re:load Pro, but didn’t have time to work on it. Recently however, I started working on serial port projects at work again, and I finally completed my serial port class, called dsub. I needed to test it, and I thought of the Re:load Pro. It was a perfect device for testing. I set about developing an app, and correcting some bugs in dsub along the way. The result is an improved dsub class, and a small application called Reload Controller which I’m releasing here. (more…)
I’ve long held an affinity for serial ports. They’re easy to understand, easy to setup, and require no special drivers. I’ve worked on several projects over the years that have utilized serial ports, mostly in classic VB applications. Since learning Microsoft C#.Net, I’ve wanted to use it to interface with them. A few years ago I picked up a copy of Serial Port Complete Second Edition by Jan Axelson, and it’s been a tremendous help. Much of what I’ve learned has come from her book and website. I set out to develop my own serial port class based on my needs, and I’ve now finally finished it to the point where I feel comfortable sharing it. It’s called dsub, named for the D-subminiature electrical connector. I’m releasing it under an MIT license so you can download, use and modify the application and source code.
Download dsub here (Visual Studio 2010 project)
dsub uses .Net’s built-in SerialPort class, but adds some additional functionality to deal with multi-threading, error handling, etc. I won’t cover all the details of how it works, or why; for that, you should pick up a copy of Jan’s book and check out the COM_Port_Terminal application available on her website. dsub does differs in several ways from her serial port class, the most important difference being that I use the SerialPort.ReadLine method to get new data from the buffer. As a result, any serial port data that dsub reads will need a defined “end of transmission” character, such as a carriage return or line feed. This can be specified in dsub, so it’s possible to use any character. I did this because all the equipment I deal with sends data this way, and it’s easier to parse the data once I know the transmission is complete. If you have a situation where there isn’t a defined end of transmission character, then dsub won’t work. (Note: Jan’s class does not have this limitation.)
The GUI application that’s included will read data from the selected serial port and display it in a grid. If a field delimiter is specified, it will use that to break up the data into separate columns in the grid. At the bottom of the screen, you can enter text to be sent. There is also a textbox where errors will be displayed. The application implements all the features of dsub so it provides a good example of how to use it.
If you use dsub in your application, let me know! I’d love to hear how it’s being used. I’ll also do my best to answer any questions or address any issues with it.
It was brought to my attention a couple months ago by mnedix that the application I developed for the Analog PC Stats Meter was not reporting the correction Memory usage. I looked into it, and it turns out that the PerformanceCounter I was using in C# ties into the page file, not just physical memory. Apparently it was close enough to the physical memory used when I tested the program initially, because I never caught it. I just released an updated version of the application today that get the actual physical memory percentage used (you can download it here). It utilizes the GetPerformanceInfo Windows API to do this, using code developed by Antonio Bakula. This post at Stack Overflow is what led me to his solution. As the screenshot above shows, it’s now very close to what Windows reports. It’s a little off, maybe due to rounding, I didn’t have a lot of time to dig deeper. It’s close enough, at least for me! In the screenshot, I’ve got a couple textboxes that display physical available memory and total physical memory for troubleshooting.
While I’m posting, I want to point out it’s been approximately a year since I last posted on my blog, but I’m still here. The last year-and-a-half has been quite hectic for me, but I hope to get back into working on projects and sharing them on here soon! I’ve heard from a few people who have enjoyed my posts and used the information I’ve shared to work on their own projects and it’s been great hearing from them. Thanks!
I stopped at an estate sale over the weekend that was taking place just down the street from where I live. I didn’t have anything in particular I wanted to find, so I waited until it was nearly over to see if I could strike some deals. I went in about two hours before it was set to close, and I immediately spotted these vintage anatomical charts being offered for $10. I told the woman it was just weird enough for me to want to buy but… and before I could finish she offered to make it $5 and I was sold. I didn’t even get a good look at them until I got home, but when I did I quickly found it was a great purchase.
These charts were originally a supplement to The New Modern Home Physician, which from the little bit I’ve gathered was an encyclopedic home medical reference first offered in 1934 (then called The Modern Home Physician.) I’m not sure when these charts were produced; one listing I saw for them indicated the 1950’s, but I haven’t found a solid reference to back that up. The envelope refers to the charts as “Manakins”, and they feature different layers of the human anatomy that are able to be removed or folded out to reveal different parts of the body. There is one male and one female. Some of the organs fold out individually to reveal other organ that are otherwise obstructed. All in all, they’re very interesting and beautiful charts with a great vintage charm.
One thing I find interesting is how the sexual organs have been censored, and when I say censored, I mean they’re completely ignored. The male is illustrated with a towel wrapped around his waist, and none of his sexual organs are shown, not even in the internal diagrams. The female is shown fully naked, however she has no vulva, and again, none of the sexual organs are shown in any of the internal diagrams. Despite all of this, the female is pregnant, as her uterus contains a fetus (how she became pregnant then is unclear.) I find it odd that such otherwise wonderfully detailed diagrams do not include such an important aspect of our human anatomy, but given the probable era(s) these were produced and the fact that they were targeted at the home market rather then professionals, it may have been considered the proper thing to do then. To me, it seems rather prudish, but then nowadays all of this information is available in explicit detail on the Internet.
I’ve got more photos/scans of the diagrams after the break, and higher-res copies on flickr. If you know anything more about these, please share! I’ve ordered a copy of the book as it sounds interesting as well.
I recently inherited several tools and and pieces of equipment which I hope to put to good use someday. For now though, I live in an apartment, which means much of this stuff is going into storage. Some of the equipment is heavy and prone to damage, so I need a way to protect it. I decided to to build some small crates for these items, since cardboard boxes won’t be strong enough. However, I don’t want to spend a lot of money doing it, and lumber is expensive, so I figured I’d get what I need from old pallets. They’re a popular source of upcycled wood for DIY projects, which means there’s lots of info on how to reuse them, and best of all, they’re free. Last weekend I set about transforming some old pallets into crates and I’m quite happy with the results.
I love headphones. It’s the only way I can listen to music much of the time; without them I wouldn’t be able to. At work I listen on Grado SR-80’s, which allow external sound to come through so I can still hear my phone ring and be aware of what’s happening around me. At home, I wear closed phones like my Ultrasone HFI-580’s so I don’t have to hear what’s happening around me, allowing me to enjoy my music and movies in peace. With all of this headphone listening, it was inevitable that I would eventually take an interest in headphone amplifiers, and I did. Recently, I built the Objective2 headphone amplifier (actually, I built two – for home and work) and in this article I’ll cover what led me to the Objective2 amp and my experience building it.
In a previous post, I wrote about how I setup a transparent bridge computer, which is able to monitor all network traffic passed through it. It works great, but to make it really useful, it needs some software that can report on the monitored network traffic in a useful manner. I decided to use ntop for this purpose, as it provides powerful reporting on bandwidth usage, which is exactly what I’m after. I’m not a regular Linux user, so I usually take the easy approach and install software through whatever GUI-based software manager is included. When I did this in Linux Mint however, I found the version available was not the latest, which is 5.0.1. I also learned that ntop has since been replaced by ntopng, which wasn’t available through the GUI. I’ve had some college courses in Linux/Unix administration, so I figured I could handle installing it “the hard way”. In this post, I’ll cover how I got ntop 5.0.1 running on my bridge computer.