Past Meetings (Most Recent First):

Dec 12, 2019 - Dan talks about Transistor Amps, Jim Wilber about CERT

Dan started off by talking about a few things- about the visitor to last TechNight, and that next month he will not be running TechNight as he will be at CES in Las Vegas. Also that he finally saw the StarLink satellite bunch this morning at 4AM, low in the eastern sky.

Then Jim went over a little about the Community Emergency Response Team training he just completed at Pepperell, which 3 hams from NVARC participated in, but it turns out it really didn't have a connection with Ham Radio as its intention, really just training citizens to be helpful before first responders arrive, or in situations that are larger that local authorities can handle. Jim went over the training outline, and it was quit involved.

Dan then talked about the different kinds of transistor amplifiers configurations, as in Common Emitter, Common Base, and Common Collector (Emitter Follower). Then other forms of multiple transistor configurations like Darlington, Sziklai, Cascode, and Long-Tailed Pair (diff amp), totem pole driver, and finally the Gilbert Cell multiplier/mixer.
Link to a good website for this that has the math behind it:

http://people.seas.harvard.edu/~jones/es154/lectures/lecture_3/bjt_amps/bjt_amps.html

Nov 14, 2019 - Dan talks more about SPICE

Dan talked about 3 main topics: The Museum of Broadcast Technology in Woonsocket, RI (slide show of pictures taken there), the frequency entry graphics software for the TNRadio, and SPICE circuit simulator, in particular, the free LTSpice by Analog Devices (previously Linear Technology). Dan showed some basic simulations, of an RC, and of the cascode gain stage of the TNRadio IF stages. Simulated time domain distortion, frequency domain gain, input and output impedance, bias point setting, power supply current, etc.

Oct 10, 2019 - Guest Speaker George Whitehead

George spoke at Foxborough, and this was a techie version of that presentation. George workd in the early 1960's on the Apollo Mission communications system, in particular the telemetry system. He presented all the details of how the data from everything on the mission was packed and sent on the Unified S-band radio hardware. He had lots of details on how this system came about and was built.

Sept 12, 2019 - LTSpice, and other things

Dan presented a 1 hour introduction to the free software LTSpice. SPICE stands for "Simulation Program with Integrated Circuit Emphasis", due to its origins as an IC layout design tool. The core of the software is totally text based simulator of circuits, in 2 modes: Time Domain and Frequency Domain. The Time Domain simulator (Transient response) figures out what the voltage and currents are in a network every small time interval from some starting state (initial conditions). The Frequency domain simulator uses the complex impedance of everything in the network to predict the transfer of an AC signal through it. Typically, the frequncy is swept between some specified start and end frequency range in dicrete steps. If you want to know if a circuit will saturate with a certain size signal, you need Transient simulation. If you want to know what it's frequency response looks like, you get that from a frequency domain simulation.

June, July, August - No Meetings, Summer Off

Dan not here, meeting run by Bob KE1JH.

May 9, 2019 - New RTL-SDR Dongle V3

Meeting started with "What Dan bought at NEARFest, then TNR update, then in remaining 45 minutes, we talked about what's inside the new RTL-SDR dongle that many of us bought recently, and ran is with SDR-Sharp software in Windows, showing both 1MHz and 100 MHz FM signal reception in the same dongle.

April 11, 2019 - Show & Tell

Dan not here, meeting run by Bob KE1JH.

March 14, 2019 - The XOR Gate and its Uses

The many uses of the Exclusive OR logic gate: Binary adder, Frequency doubler, Comb generator, Schmidt Trigger buffer, Logic polarity selector, Level translator, Oscillator. Also, Irridium Flares, Electronic Moustraps, and TNR progress. Double-layout on a prototyping board.

February 14, 2019 - Radio Wave Propagation

Presentation by Bob Jackson, physicist and KE1JH. Maxwell's Equations made easy.

January 10, 2019 - Radio Project Part 5

At the 5th of a related series of Tech Nights, we discussed the radio project again.

This meeting we tried out a new screen, a reflecting-bead curved one popular in the 1970s or so. Seemed to work better, though not as big. We also had my 43" HD TV screen set up, so we could show some CAD system views.

For this meeting I had a guest demonstrator for one of the CAD systems, Peter Abdu KB1IXE. Peter is an EE from UNH who works with me at a drone company, Greensight Ag. I was going to demonstrate one electronic CAD system while Peter did another to compare. I did EasyEDA, and Peter did KiCAD (pronounced "key-cad"). Both are free and open-source software packages for capturing schematics and laying out PC boards. We did a little comparison, but ran out of time to do a good job of this.

These two CAD systems are very different in how they run. In my personal opinion, EasyEDA is more intuitive. There are probably about 6 major electronic CAD packages to choose from, varying in price from free to $100K. We had in our audience at the meeting an employee of Cadence, one of the $100K ones, (Bill W1PA) and I asked him why I should pay $100K instead of zero, and he had many very good reasons, mostly about many extra capabilities that some people need, mostly big companies with lots of money. Fortunately for the hobbyist, the free ones are getting pretty powerful. But each of the systems has its champions, and arguments abound over which is better. I'm right, of course, so you should just use EasyEDA. Don't talk to Peter about KiCAD.

Here's an example: If you use computers alot, you probably draw things, in some grpahics software. I like free software so I use Inkscape for simple 2-D drawings, like flow charts or sketches for a report. AutoCAD is famous, Adobe Illustrator used to be big, and lots of others. You find that there is a somewhat standard way of using these, like to move something, you generally point at it with the mouse and click on it to select it, and click and hold to drag it. EasyEDA works like this, KiCAD doesn't. I was frustrated with KiCAD when I first used it, I couldn't get it to do anything, until I learned its paradigm, which is to put the mouse pointer over something you want to work with, and then instead of clicking, you hit a letter on the keyboard. This requires 2 hands, one for the mouse and one for the keys. You can do things very quickly this way if you learn all the keys, like M might be for move, and E might be for edit, and D might be for delete or duplicate, R for rotate, etc. In EasyEDA, I can click and drag it, but if I want to rotate something, I have to click, then hit R, two steps instead of one. So it's a matter of what you want to deal with, or maybe some other advantage of one over the other.

The reason you as an electronic hobbyist should be interested in CAD now, is that in the last year or two the most amazing thing has happened for the hobbyist: Cheap PC boards! What used to be a couple hundred dollars or more is now $10. Good quality, 2-sided with plated through holes and markings and everything you need. These companies are in China of course, but they do amazing shipping for cheap somehow, so for like $20, you get 10 copies of a board that is maybe 3 x 6", in 4 days! It's pretty hard to believe. And the quality is very good, a professional board. Apparently they have figured out how to automate the whole process so you put your CAD files in one end and UPS is standing at the other. As Peter said, it's a good time to be an EE.

At the meeting we then we went on to a part of the radio project. The problem that became obvious when we did the sensitivity or noise figure calculation could be solved with the addition of a gain stage, but it had to go after the mixer and before the filter. Both of these devices are sensitive to their termination impedance. The balanced mixer won't cancel well if it doesn't see a good clean 50 ohms on it's ports, and the crystal filter response shape will change if it does not see proper impedances on it's ports. So we need an amp that looks a lot like 50 ohms on both its input and output. Fortunately, there is such a thing, and I went on to show on the white board how this can be done with hybrid couplers, a device that splits a signal into 2 equal parts, which are 90 degrees in phase different from each other. The usefullness of this comes when you use two of these, one to split a signal into two paths, each with an identical amplifier, then combine the two back into one signal path, again with a 90 degree coupler. In this arrangement, any part of the signal that is reflected back from the two amplifiers will go back through the 90 degree coupler and get shifted another 90 degrees apart again, which makes the total 180 degrees, and the two reflected signals thus cancel out, not affecting the input impedance. So as long as the two amplifiers are identical, they will produce no reflections regardless of how much they each reflect due to imperfect match. In fact, if you want to control the gain of such an amplifier, you can easily do it by just loading (mismatching) the input and output, which would normally screw everything up, but if you do it inside the hybrids and to both sides equally, the hybrids will cancel it out. Really neat circuit. So I have added this amp to the design.

The hybrid splitter itself is rather simple. You can buy these from Minicircuits already made for $10 each, or you can build your own with a single toroid transformer and a couple caps. I will probably design in the Minicircuits parts, and leave space for a home made hybrid later as a cost reduction.

December 8, 2018 - Radio Project Part 4

At the 4th of a related series of Tech Nights, we discussed the radio project again.

The meeting started with an introduction to ham radio digital modes in general, in particular FT-8, where a simple exchange is encoded into the minimum number of bits possible: 28 for a callsign, and 15 for a 4 digit grid square. Then some sync and error bits are added for a total of 174 bits. This is then sent by using a series of one of 8 tones (so 3 bits/symbol), total of 79 tones. At 6.25 baud, this takes about 13 seconds. Then there are 2 seconds for the receiving station to figure this all out before it sends a reply during the next 15 seconds. By agreement, these 15 second intervals are synced to the UTC minute, so everyone transmits and listens at the right time. So you can complete an exchange in 1 minute, which is nice for a contest, or to catch a propagation fade. The real value here, though, is that by sending a fixed tone for a relatively long time (1/6 of a second), a receiver can integrate over that time and pull a signal out of noise pretty effectively. This software decodes perfect messages when you can't really hear a tone there at all by ear. So this is not only a good use of spectrum (50Hz wide signal), but also a very efficient use of power (50 watts gets you lots of states and countries). To top it off, a received signal strength is rather accurately calculated and automatically included as a signal report, so no more "599" stuff, you can actually get a really good idea of how well you are being received, and watch propagation change. The waterfall display it gives you shows all sorts of interesting information, like doppler shift and general fades in and out, and frequency selective fades, as well as how active the channel is. It's neat stuff, and a nice future for ham radio. All the details can be read from the WSJT-X full user's guide, Protocol Specifications section.

Then again, as last month, I introduced John WB2OSZ, who got up to the white board and diagrammed how the WSJT code works to generate tones, and how he modified it for our radio so that our transmitter becomes nothing more than a single signal generator chip with non-linear (FM style) amplifiers after it. He talked about his successful code that modifies the open-source WSJT software to read out the tone freqeuncies directly, so instead of feeding audio into a SSB transmitter, we can just directly drive a signal generator chip that drives the antenna, and have the world's simplest transmitter. John brought in a Raspberry Pi like we intend to use in our radio, and connected it to a signal generator demo PCB with our radio Direct Digital Synthesizer chip on it, set the center frequency to 14 MHz, and drove a piece of wire with about a milliwatt. I brought in my HF radio and laptop with FT-8 on it, tuned it to 14 MHz, and heard the signal. Success was soon to follow when my software decoded a CQ message from John, all received from the signal generator chip, with no SSB transmitter involved! The incoming message was on the projector screen for everyone to see - the first time this was done.

Then we went on to other parts of the radio project. I put up the Gain Distribution table, a spread sheet of signal strength for each stage of the radio, with a new addition, cascaded noise figure. This new information shows how the follow on stages affect the sensitivity and noise floor of earlier stages. Unlike the gain distribution, which you can follow from left to right and see how the signal gets smaller with losses like filters and mixers, and bigger with amplifier stages, with sensitivity you start on the right and work your way left to the input of the radio, with each stage either improving or degrading the noise floor of the receiver. With the current block diagram, we have a problem, which I demonstrated by changing parameters of the stages and watching the effect on the input. We saw the effect of lowering gains and increasing noise figures. This new spreadsheet is now up on the website here on the project page. Note that this is only a picture of the spreadsheet, and not a working example. If you want to play with this, I can give you a copy of the real open-source spreadsheet.

Lastly, I brought in an IF gain demo PC board, that had two of the chips I'm thinking of using for most of the gain of the radio, the AD603. Powering up this board and cranking the gain up to 80dB, then connecting it to the input of my HF radio, you can hear the significant added noise. The question is, will our signals be above this noise? From the Noise calculation spread sheet above, it won't quite make it. The problem is the AD603 has an 8dB noise figure, and we don't have enough gain ahead of it to make up for that. So a design change is needed. That was it for the meeting.

November 8, 2018 - Radio Project Part 3

At the 3rd of a related series of Tech Nights, we discussed the radio project again.

The meeting started with 5 non-radio-project topics: 1) A ham down the street from me has a bunch of stuff for sale, I had a list I made available, 2) I came across a few electronic build-it kits, for kids to learn electronics, they are available if anyone knows of someone who wants to use them, I would be available to make their projects work, 3) I brought in a 1V Tube radio I got at Deerfield last month, and told how I listened to the last World Series game on it. Someone asked how I knew it was the last game... 4) In sharp contrast to the tube radio, I showed a demo PCB with a new 80 GHz radar transceiver on it, a chip with an array of patch antennas (TI IWR1642), and 5) I brought in all the electronics from a large flat-screen TV, including strips of LEDs, and the shift registers that address the rows and columns of the LCD pixels. It's down to like 2 major chips that just do everything. One of the boards had a couple bullet holes in it because the TV was apparently used for target practice.

Then we went on to the radio project. I put up the Gain Distribution table, a spread sheet of signal strength for each stage of the radio. It is now up on the website here on the project page. While we went through the stages I talked briefly about the concerns that each stage has and how it affects the performance of the radio, but there are a lot more details to be discussed than I addressed.

Then I introduced John WB2OSZ, who got up to the white board and diagrammed how the WSJT code works to generate tones, and how we'll modify it for our radio so that our transmitter becomes nothing more than a single DDS chip with non-linear (FM style) amplifiers after it. I brought up the radio block diagram on the screen and showed how we're going to generate the tones on 6 meters, or any other ham band.

October 11, 2018 - Radio Project Part 2

The 2nd of a related series of Tech Nights, we discussed the radio project. Unfortunately, my LINUX laptop seemed to have a hard drive crash and wouldn't boot, after having worked enough to get one slide on the screen. I couldn't deal with it, so I went to the white board. Meanwhile, a new visitor to TechNight showed up, Lee KB1ELE, with a shirt on that said in big letters LINUX, so we asked him if he could look at it. In short order he had it booted from a linux USB drive he happened to have in his pocket, and performed a hard disk repair, and got it all running again! Pretty amazing coincidence.

The meeting turned into a general discussion and question and answer about various parts of the radio project and lots of other things. Toward the end, I explained on the whiteboard how a DDS signal generator chip works, how it creates a sine wave very cleanly from some input clock of a higher frequency, at least double the highest frequency generated. I propose to use 2 DDS chips in the radio. Part of the discussion earlier was how to modify WSJT software to get the tone data so we can generate the 8FSK of FT8 directly with the DDS chips.

September 13, 2018 - Radio Project Part 1

Expected to be the first of a related series of Tech Nights, we discussed doing a project. After discussion about possible project topics, Dan suggested that we design a radio and then build it. This seemed interesting to all present, with the thought of learning about radio design as we go along. Dan listed some suggested goals for the radio project:

1) Cheap - less than $50 not including enclosure or peripherals like AC power supply, keyboard, and screen. Just a PC board, no box.
2) Pretty good quality - totally usable, not contest grade, but a level below that.
3) Small and low power, for portable operation, so it can be used at home or out camping. It would be surface mount technology, but you won't have to solder any of that. It would use a Raspberry Pi as its controller, so that it can run FT8 (and all the WSJT modes). No external computer required. We will write the controller software for this radio, the "front panel".
4) Hardware not too complicated, maybe only one band. I am thinking 6 meters, but it will be buildable to other bands from 160 thru 2 meters. One circuit board for everything, maybe an external power amp.
5) Decent power out on TX, maybe 20 Watts.
6) Maybe it is FT8 and CW only, meaning it does not have to be linear. No SSB (but maybe FM).
7) Useful for local TechNight contests, like maybe a worked-all-attendees, worked all of these radios, or worked all zipcodes of members. Something to get people on and using the radio and having fun with it.

Given this as the start of the project, I discussed some radio topics. We talked about phase noise and spectral purity in transmitters and receiver local oscillators, and some about receiver dynamic range, showing a dynamic range chart from a Harris RF590 radio.

I also talked about a couple of unrelated topics, like new LED light bulbs that create no RF noise (showed one taken apart), and real antenna insulators from a flea market, complete with original clevis and pins! And some details of my move to Fitchburg.

Summer Break, no July or Aug meeting

June 14 - Grady Hydro Station

At the largest Tech Night gathering yet (34 people), we all got a tour of the Grady Research Hydro station, and then John Grady spoke to us for more than an hour about the history of the hydro station, and many other topics about hydro and power distribution. It was completely entertaining to listen to John talk, with his very technical humor and total intolerance for bureacracy and red tape. No one left this discussion early, and it went until 10PM. What a great time it was, to see this really cool big-time technology live (Keith shut down the hydro generator due to low water while we all watched), and understand that it was there only due to someone's purist technical interest, not as a real money maker. Thanks to John and Keith for a great meeting, and for keeping the Nashua River water rights useful.

May 10 - 1V Tubes

This is a repeat of a program done in Feb 2014 at an NVARC meeting.

I presented information on a family of vacuum tubes that was created in the late 1930, and manufactured into the 1970s. The main tubes in the family were the 1T4, 1U4, 1R5, 1S5 and 3S4. They were made for portable battery-powered radios, and so were made to use very little power. They all have in common that their filaments run on 1.5V and pull 50mA, and have directly heated cathodes. The anode voltages were made for the then common "B" batteries at 67-1/2V and 90V. Millions of AM radios were designed with this 4 tube set, along with special radios for 2-way and shortwave. Many WW-II radios used them. I went into some detail on the construction of the tubes, showing cutaway views of tubes I took apart. I went over vacuum tube basics, why there are triodes, tetrodes, and pentodes, and the rare heptode 5 grid tubes used as oscillator/amplifiers. I also went into some detail on the difference between the 1T4 and 1U4, which are 2 RF amplifiers. The 1U4 is a sharp cutoff, the 1T4 is "remote" cutoff. They are made with some small differences in mechanical construction that changes how the electron flow changes with grid bias. Remote cutoff is used for Automatic Volume Control radios as it allows more control over the gain of the tube.

Maybe I will get around to putting that tube radio I started the rest of the way together before I do this again in 4 years.

April 12 - National Grid Engineers

In new format of Q&A, we learned about many aspects of the local power grid from 2 engineers from National Grid, Nathan Walsh and Dan Mungovan. This was the largest crowd ever for TechNight, it was literally standing room only. The discussion lasted 2 hours and could have kept going for another two, but we had to call it a night.

This was a great discussion. Though we started with a planned list of 5 topics, the discussion quickly ran off on some very interesting side topics. Perhaps one of the more interesting things I learned was about the economics of providing high reliability power. There is a big difference between rural and city systems. The classic thought about a rural power line is, why don't you run 2 lines, so that if one fails the other can take over. Turns out that is too expensive to do, as each of those power lines could only be used up to 50%. And rural lines are typically up on poles and easy to repair quickly. In a city the story is different. With higher population density, more power is needed, so they run multiple lines underground, but now a backup line is like adding a 6th line to the 5 that are already required to take the load. Now the non-failed 5 lines only need to take on 1/5 of the load on a failure, and this extra load can be spread to the other lines with load switching. So all the 6 lines carry 5/6 of their rating, not 1/2 if there were 2 lines. When one fails, the others go up to 6/6 (full load on 5 lines) and the load can be carried with one of the lines down. This also allows for maintenance on one line at a time. Makes complete economic sense. Just one of many interesting discussions we had.

Much Thanks to Dan and Nathan for volunteering their time, and we hope to see them back again sometime next year.

March 8 - Minicircuits and RF Components

This time we ironed the sheet for the projector, worked better.

The National Grid people had to cancel due to the storm, so we did Minicircuits. We brought up the Minicircuits web page and went through some of their products, talking about the wide range of components for anything RF that they carry. Mainly amplifiers and mixers, but also couplers and hybrids and splitters and frequency multipliers. Brought in some PC board that had an array of Minicircuits products on them.

This meeting had rather small attendance, due to the North Easter that came the day before. As I write this, we are about to get hit by the 4th one. Been a bad year, unless you like storms.

Feb 8 - Impedance Matching - More Basics (Part 2)

This time we tried a projector onto a sheet, worked OK. Summarized the basics from Part 1, then talked about impedance mismatch effects, and measuring impedance by measuring these effects, like reflected power. Showed a table of Z mismatch vs Return Loss vs VSWR vs Reflection Coefficient gamma, vs percent power to load (table here). Showed some examples of how to measure reflected power, directional couplers, Bird Wattmeter taken apart. Then ran the program Smith to show a smith chart in action, but it kept crashing. Talked a little about Z and S parameters, showed some RLC meters you can buy cheap.

Last slide was unrelated, about a possible bench power supply project, 0-35V 7.5A using an LT1083 kit from china with a rough switcher front end from the datasheet.

Jan 11, 2018 - Impedance Matching - The Basics (Part 1)

I did a simple introduction to energy source impedance and transmission lines, the real reasons that one needs to impedance match. Got to an introduction to the Smith Chart, but didn't get to Z or S parameters, do that next month. References that I gave at the talk are here.

Dec 14 - Marshall Cross - Unusual Antennas of the Cold War

Marshall did a slide presentation (link here) on some unusual things done at low frequencies during the cold war, to maintain critical communications during a nuclear attack.

Nov 15 - Software Defined Radio, Part 2

With our largest crowd yet, we continued from the October 11th meeting, gettting into actually running some hardware and software to make our own software radio, using the GNU Radio software.

We did manage to get something running by the end of the meeting, and demonstrated picking up the aircraft ADSB beacons as they flew overhead, showing a map of the area with planes located.

Dave Goncalves, W1EUJ, a GNU GURU, did a demonstration of reverse engineering a radio signal from a remote control. After using GNU to receive the signal and decode it, he used Gnu to build a transmitter to send it, and managed to successfully control the remote device.

At the end, we ran the software HDSDR and picked up the local 6 meter beacon.

Hopefully, many of you will get your own $20 dongle and get this running at home. If you need any help, feel free to contact me (contact on Who's Dan page). I really hope to offer a project for a good quality transcevier board that can get you on the air with any of this software, cheaper than what's on the market right now.

Oct 11 - Software Defined Radio, Part 1

What do they mean exactly by SDR? Here's what it is and how it works.

Another good turnout of about 15 people, mostly new faces, probably due to the different night.

Defined SDR as digitizing in the signal path and doing at least some of the functions in pure software. Went over history from digital telephony on to your cell phone. Talked about the Icom 7300 and Elecraft radios and their block diagrams. Had a white board discussion of the main advantages and disadvantages. Brought in some software radio boards, and went through a list of them on-line, reviewing capabilities and costs. Mostly for receive only, talking $150 to get Xcvr. Really want to do a group project to build one.

Sept 14, 2017 - Ham Radio Digital Modes     Info page from Sept 14 here.

It's a Ham Radio Phenomenon - As of the release of the new FT-8 mode, all of ham radio is moving to digital. Here's why, and here's how it works. And here's how to set yourself up to run it.
I'll start with a brief summary of my trip to see the eclipse, when I left WSPR running on 20M. This digital mode makes ham radio valuable to the world like nothing before.

Good turnout on the first new Tech Night, including the building owner himself, John Grady. Gave good summary of many of the new digital modes and how to get set up with them. Ended with a live demo of FT8 and WSPR, showing the station at the meeting show up in the world wide database of spots in a couple minutes.

Saw K1JEBs new 10W 10GHz equipment, nice neat assembly work on a Sony directTV dish.