Wednesday, April 28, 2010

BBQs and Upstairs Neighbours

Hi All,

Just a rant about the moron who lives up the stairs from us.

So yesterday we decided "Hey, it's a nice day! Let's get one of those disposable bbqs, get some people together and make an evening of it." So we did, we got some people together, got some food, and fired up the barbie.

The old guy upstairs from us didn't like this. I don't know why he doesn't like us - maybe he was abused as a child, or maybe he's just so close to the grave that he can't stand to see people have fun. Anyway, he stuck his head out the window and told us to put it out or he'd be calling the fire brigade.

Naturally we assumed that no one would be that much of an idiot. Unfortunately we were wrong. Next thing we know four (count em.. 4!!) fire engines are pulling up outside. Just as they were pulling the hoses off the engines and putting on their breathing stuff, we managed to get downstairs and explain the situation. Obviously he'd slightly exagerated the situation to the 999 operator.

Luckily we managed to capture the scene on video

So Mr Upstairs Neighbour, I hope you're happy - you successfully managed to stop our barbeque. I hope you can sleep easy now in the thought that if an orphanage had burned down, all of Edinburgh's fire department were busy dealing with us. I hope they fined your ass for wasting everyones time.

Note: The bbq was eventually extinguished using a small jug of water ;)
read more "BBQs and Upstairs Neighbours"

Tuesday, April 27, 2010

A Simplified CW keying monitor

My homebrew CW station shown at http://www.io.com/~nielw/2tube_xmtr/2tube_xmtr.htm
includes a control (speaker/power supply/TR switch) box. I found, though, that I still needed some sort of CW monitor. The June 1952 issue of CQ magazine describes a simplified CW keying monitor using only 6 parts. Basically it is a neon bulb type audio oscillator keyed by a second neon bulb energized by transmitter RF. I built the circuit into my control box coupling the monitor output to the grid of the 6F6 audio output stage. I added a 12 Meg resistor from B+ to the junction of the two neon bulbs to help reduce motorboating when the circuit is not energized.

It works fine. Now I can silence the receiver on transmit and have the CW monitor to listen to my own fist.
read more "A Simplified CW keying monitor"

Sunday, April 25, 2010

815 Transmitter Coil/L1



L1 of my transmitter calls for ten turns on a 1" coil form spaced over 7/8". I had the form and the wire I needed. Winding spaced turns, though, takes a trick. I first wound the ten turn coil on the form, tightening the wire as much as possible. The turns were not even. It was not pretty. I then wound another "coil" (the red wire) forcing evenything to space out evenly over the 7/8". This I heavily coated with clear fingernail polish. After the fingernail polish had dryed I removed the red spacer coil and recoated with fingernail polish.
read more "815 Transmitter Coil/L1"

Friday, April 23, 2010

Picnic Table QRPing in Decorah, IA








We spent the weekend in Decorah, IA and on Saturday and Sunday I had some spare time for hamming. I used my K1 with a 67' endfed wire up in the tree above the campsite. With the K1's built-in antenna tuner it loaded up on 40, 30 and 20. Sunday afternoon at 3:00 I did finally did have a QSO with Ken, WA8REI/4. He was on the Appalachian Trail at Newfound Gap, NC near Gatlinburg, TN. We were on 14.060, both running battery/5 watt power into end fed wire antennas. With some amount of QSB, signals were around 579. For the bottom of the sunspot cycle I was pretty happy.

See more examples of my QRP operating at http://www.io.com/~nielw/qrp/QRP_Stn.htm
read more "Picnic Table QRPing in Decorah, IA"

Friday, April 9, 2010

Final 815 Transmitter Layout




I've finally committed to a layout and mounted the major parts. From left to right along the front is the xtal (with B directly above it) C1 and C2. Left to right on top is 6V6, L2/L3, 815 and L4. RFC2 is mounted upright behind the 815 and L4. I elevated L2/L3 so that it could be mounted directly above C1. The toggle switch on top of the chassis in the right front corner is to short out L1 when not multipling in the tri-tet oscillator. The extra space along the back is reserved for whatever is needed to get the screen and oscillator voltages right. The spotting switch will probably go in the space between C1 and C2 with a cathode current meter mounted directly above it (if I add a front panel).
read more "Final 815 Transmitter Layout"

Wednesday, April 7, 2010

Final Layout / Drilling Guide


After "playing checkers" for several days tweaking the layout, I'm ready to drill. I usually mark major component locations and holes using masking tape. After I'm done the tape can be removed leaving no markings on the chassis itself.
read more "Final Layout / Drilling Guide"

Tuesday, April 6, 2010

How to make your own UV lightbox for PCBs


Ultraviolet (UV) light boxes can be pretty expensive, some costing upwards of £100. I don't have that kind of money to throw towards making a few hobby circuits, so I decided to build my own out of cheap parts (namely eBay-bought UV LEDs). I thought I'd share with you all my experience.

What is a light box?
A light box is used in a process called photo-lithography to transfer an image from acetate (overhead projector film) onto printed circuit board. This is how to draw all those little lines and circles on printed circuit boards, so that you can stick all the resistors, capacitors and microchips on.

Photolithography consists of the following steps:
1 - Design a circuit on the computer, print the circuit to acetate (OHP film)
2 - UV photo-expose the acetate onto a specially prepared blank circuit board (this is what the light box is for)
3 - Throw in some chemicals
4 - All the copper you don't want on the circuit board melts away, leaving behind some pretty designs

(If you want to know more about photolithography, read the Wikipedia entry here)

Anyway, onto my light box, and how to build your own. You will need the following components:

  • A box - I recommend a £3, 3-litre Really Useful Box® (Maplin: N35BG)>. Keep in mind this is see-through, and if you buy better UV LEDs than I did you may get a bit of a tan

  • UV LEDs (50 or so) - I bought mine off eBay at £4 for 100, just search around

  • Resistors - I used 8x 2W 390R resistors. UV LEDs draw a fair amount of current so I split my LEDs into groups of 6 and used a resistor for each group. YMMV with different LEDs (see the maths section at the end to calculate what you need)

  • Some plain matrix board - This is just strip board without the copper strips - Bought from Maplin, thought I can't find the code on their website

  • A switch, a wall-jack socket (ie. the socket for a 12v power adaptor), some solid core wire
You will also need a soldering iron, solder and a drill.

My light box varies from traditional light boxes because the UV source in mine is in the lid, and the circuit board sits in the base, face up. Traditional light boxes have the UV source in the base, then a sheet of glass to place the circuit on top of, and then a lid. I like to be different ;).

The first step is to figure out how many LEDs you will need. My box gave about 12cm in height from the LEDs to the bottom of the box. With a viewing angle of 25 deg, this meant that I got a circle of about 5cm at the bottom. Just to be safe, I placed my LEDs ~4cm apart. (See the end of this post for all the complicated mathematics)

Next step is to figure out what size to cut your matrix board, so it will comfortably fit in the lid. Mine was about 18x12cm. This still left gaps around the edge (because the lid is irregular around the edges), though as long as I keep the circuit in the middle it comes out OK.

Draw yourself a grid to lay out the LEDs. Try to make this regular so you get even exposure.

Once you have the grid, you should drill 4 holes around the corners of the board. Make sure you avoid the LEDs, and try not to place it in the middle of a line of LEDs, as you will be running wires down that line. Drill 4 corresponding holes in the lid of the box to mount the PCB.

Now push the LEDs through the holes in the matrix board. Remember to lay them with the polarity facing the same way. You can either place them with the two legs horizontally or vertically. This will depend on how you are grouping them together, try to choose a way which will make wiring easier.

(To hold the LEDs in place for soldering I used a little spot of superglue - don't do this - superglue is pretty toxic and the vapour released during soldering can't be good!!)

When wiring the LEDs I placed them in parallel. All of the cathodes (negative legs) can be joined together. The anodes (positive legs) need to be joined together in groups (the size of which we decided earlier). Each group gets its own resistor, then the other side of the resistor is joined to the +ve supply.

Take some wire and strip the plastic coating off it. Cut it into lengths to join the legs of each LED in each group, one wire down each side of the LED. Solder these wires on and your LEDs will be held in place. You can just about see what I mean in the picture, as well as my ingenious use of Lego blocks :)

Now take some more wire, measure out the length you will need to join all the negative tracks together. Strip a little of the black plastic off the end. Measure the distance you will need to travel to the next track and use a knife to cut the plastic here. Slide the rest of the plastic along a bit, and you have a gap to solder on to the next track. Repeat for all the tracks.

Place the resistors on the board and solder one side to the positive track. Repeat the procedure above to join the other side of the resistors together. You now have a circuit board full of UV LEDs. To test it out, use some crocodile clips or pieces of wire, and connect to the positive and negative rails (make sure you connect to the red wire which is before the resistors! you don't want to connect straight to the LEDs and blow an entire group of them!) If all works, you can move on to the next stage.

Drill a hole in the lid for a power switch an the wall jack socket. Wire these to the board, and then screw them into the lid.

Finally, you need to find some posts to hold the board slightly away from the lid. I used little pieces of Lego since they were all I had to hand.

Once it is all screwed together, you have your very own UV light box! I drew up a little test board and printed it on some acetate. I exposed it initially for 4 minutes, but because of my cheap LEDs it didn't work. You will have to use trial and error to find your optimum exposure time, but mine is about 12 minutes (yes I know this is a loooong time).


Here are some photos of the box in action, and the first few circuits I've made.





All the complicated maths stuff
Resistors
Resistors come in different power ratings - 1/4W, 2W, 4W, etc. You will probably want 2W resistors, as these are good compromise between size and power. That being said, you will still need a bunch of them.

Power rating is calculated using the formula P=IV, where P is the power, I is the current, and V is the voltage. I used 2W resistors, at 12V, so by the equation I=P/V, I can draw 2W / 12V, or 0.16666 amps (160mA). This means a maximum of 8x 20mA LEDs.

LED Spacing
To figure out how far apart your LEDs should be, you need to know two high-school maths principles - trigonometry and the square inside the circle.

First step, figure out the radius of the circle at the bottom:
You should use trigonometry (the tangent of the angle by the height of the box - dont forget to divide the angle by 2!)

The equation is: Radius of circle = Height of box * tan(Viewing angle /2)

Now if you are arranging the LEDs in a neat grid, you will need to know the size of the square inside the circle. This is found using Pythagoras' theorm, the hypotenuse being the diameter of the circle.

The equation is: Width of square = sqrt(diameter2 / 2)

This will tell you how far apart your LEDs can safely be to avoid blackspots
read more "How to make your own UV lightbox for PCBs"

Monday, April 5, 2010

Another Power Supply Option

Bryan, WA7PRC, pointed out to me that an "economy" power supply would be a more elegant way of getting the lower B+ needed for the oscillator stage then wasting power in multiple dropping resistors.

From Bryan:

"It's an easy way to obtain high- and low-B+ voltage sources.  In fact, there's more than one way to skin a cat... center tapped or non-center tapped transformer using a fullwave doubler.  In both cases, the output voltages differ by a factor of 2. The doubler, aside from needing a transformer with only a single secondary winding, has the advantage that the output voltage is 2.828 RMS instead of only 1.414 RMS.  So, your secondary can be 177VRMS instead of 354VRMS. Practically speaking, because of resistance losses, the voltage would be somewhat lower under load.  And of course, you'd still need whatever VA rating, so the transformer size/weight would be the same.



With either topology, you'll have a significantly lower low-B+, and should you decide to use a regulator, will not have to dissipate as much power in the series resistor."
read more "Another Power Supply Option"

Sunday, April 4, 2010

815 Transmitter Power Supply

I keep a couple of generic brute force power supplies around to use with my homebrew projects. This one grew out of a much modified and cannibalized WRL transmitter that I found at a swapmeet several years ago. From the back, on the right side is the high voltage section including transformer, choke, filter capacitor and rectifier tube. It will easily deliver the 500 VDC at about 200mA that I need for this project. On the left side is the bias supply. This section has a VR-90 with a pot across it. The exact bias voltage is set and then left alone. The six foot cable shown will plug into an octal socket on the back of the 815 transmitter.
read more "815 Transmitter Power Supply"

Saturday, April 3, 2010

QRP Operations from W0IBM shack

Today was a great day to stay inside in SE MN: windy, overcast, temp in the 30s and snow in the forecast.

The first Tuesday of the month is informal meeting day at W0IBM, Rochester, MN. ( see http://www.w0ibm.org/ ) I took along my K1 as show and tell. 20 wasn't real active but I hooked up to the W0IBM three element yagi and managed a contact with George, KF4UCC, in Chesapeake, VA.

On the 815 transmitter front I've done a rough layout to help orient tube sockets and studied the dimensions a bit more. It still looks like it will fit. I'm just about ready put the drill to that virgin chassis.
read more "QRP Operations from W0IBM shack"