Don't you start this experiment until I've gotten completely through it. I'd hate to lead you down a useless path. This is an educational experiment for me and possibly you to learn something new and exciting.
OK. This is an open experiment, i.e. I'm going to do the writeup as I go along. I'm going to demo, by either success or failure, how to try something without having every nit in place. I'm going to take a Small Wonder Labs PCB for the SWL-20+. The SWL series from Small Wonder Labs is transceiver that sells for $55 with the parts necessary for the single ham band desired. But I'm going to substitute parts in order to create a 15 meter version of the same transceiver, then make mods to bring xcvr up to better standards. I'm thinking that the receiver gain is going to suffer and the transmitter output will fall off significantly due to loss of gain for the increased frequencies necessary for 15 meters.
It is my belief that the 42IF123 IF transformers used throughout will HAVE to be replaced with toroids. Why? Because the 42IF123, even with the internal cap removed, will be frequency limited to under 18 MHz or so. I will demo this when I get to where I need a transformer and show you an experiment that I did years ago....
This experiment requires me to have a relatively large parts supply and hopefully I have enough to not have to stop and order something from online. Usually Mouser.COM. I'm at the point where I've collected a lifetime supply of parts and it is about time that I start using them at a significant rate. This is step one.
We all know that the CW portion of the band starts at 21.000 MHz and upward. I want to cover a 40KHz range (or more if possible). I feel that the gain in the receiver may not be enough, so we'll just go with the flow and then come back with a second version that gets us further along to a better xcvr. My goal is to get this done by January 7th, 2012 for show and tell at the next AZ QRP club meeting in Phoenix. Here goes.
First. Look at the signal path for the receiver section.
ANT · LP FILTER · BP/MATCHING FILTER · FIRST MIXER · IF FILTER · SECOND MIXER · AF PREAMP · MUTE · AF FINAL AMP · PHONES
The VFO/LO feeds one of the inputs to the FIRST MIXER and must be considered also.
OK. Your homework. Go to the following URL and download the manual. Look at the schematic on page 7 of the manual. Follow from the antenna (RF OUT) through the circuit to determine each of the above sections. Do you know enough about the inner workings of a receiver to determine ahead of time which sections we are going to have to modify to get the critter to work on the 15 meter ham band?
I am starting with the SW-20+ as we need the 5.0MHz VFO/LO to get 21.0MHz using a 16.000MHz IF frequency.
If you were going to build this thing and experiment, where would you start? I would and I am going to start with the audio section and work backwards. With the audio section working I can thing test each added section for performance and to make sure I'm on the right track. We'll build test equipment as we go and as needed.
I will list the components for a section and photograph as I install them. I should do this Manhattan Style, but I want to do it this way (even with the high probability of failure) because it looks like fun. I may lose an investment of almost one hundred dollars, but who cares? Education is expensive and the investment will pay off. Guaranteed. Follow along and learn. I do recommend that you do not physically do this until I have finished the entire project. I may just stick my neck out so far and fall flat on my fanny doing this, but it is winter time and I think of stuff that I want to try in the lab. Put your seat back into its upright position and place tray tables into their stored position..... This could be a bumpy flight.
If you really want to follow along closely, then print out pages 4, 5 and 7 of the SW40+ manual and with a highlighter or other marker and mark off each of the parts in the schematic and parts list as I do them here to see if you can follow my train of thought. It will force you to concentrate more. It is a technique that I use on any project that I'm doing just to reduce (note I didn't say prevent or eliminate) errors. It also serves as a blue print for debugging things that do not work. Either due to human error or part(s) failure(s). And it is a sad state of affairs that we now live in a time where we have to worry about getting counterfeit parts that have no chance in hell of ever working where we want to use them.
Here are the photos of the above two segments. This is an advanced project, so I will not explain each parts placement and I will not take a photo for each part as it is installed. I have limited time for this project. You know the drill.
So now comes the moment of truth. When astronomers first use a telescope to observe the Universe, the event is called 'first light'. It is just as exciting to radio amateurs when it is the first time you power up an electronic device or circuit. It is either pass or fail time. What I want to do here is make certain that several things are working correctly and give you a way point to check your work. Don't start this project until I get completely through. hate for both of us to wind up at the same dead end.
First off. Make damn sure that you power up the work you are doing on a clean non-conducting surface. I purchased four nylon screws and nuts at Home Depot for placement in the PCB mouting holes. This keeps the board away from any lose wires and hopefully prevents my accidently shorting something out.
Because I chose a 16.000MHz IF frequency, we will need five 16.000MHz crystals. I happen to have some in my stockpile that I bought from Mouser Electronics online some time ago. See section later on the exact specs on this critters.
Now power up xcvr again. If you have a VE3DNL signal generator or any signal generator capable of outputting 16.000MHz and surrounding frequencies then power it up. You should hear a tone in the headphones if you are using a VE3DNL with 5KHz intervals of output. If are using a tuneable source then start at 16.000MHz and move slowly upwards until you hear a tone in the headphones. It will be weak at first and the tone will start high and get lower, go through zero and then increase both in pitch and in loudness. This proves the mixer is working correctly. You do not need the signal source near the xcvr if you have a random length of wire on the source.
Let's discuss what every experimentor needs in the shop or lab area. First of all you need a crystal tester circuit of some type. I have several. These are used to make sure a crystal is working. They can also be used as a signal source. For the above test, if you have a crystal oscillator circuit, then take another 16.000MHz crystal and put it into the tester and then power it up with the xcvr also turned on. You should be able to hear a signal in the receiver. A tone that may be relatively high in pitch. It helps if there is a way to vary the frequency of the crystal in the crystal tester. Some testers will have a variable cap in series with the crystal.
Another valuable piece of equipment is the VE3DNL signal generator. This was a cheap kit from the NorCal QRP Club for a number of years and then available from other groups for a period of time. Unfortunately, a scan of the Internet using Google shows that no one currently has such a kit available for purchase, which means you have to build one from scratch. The schematic is available on the norcalqrp.org web site. The hard part to find is the 5.12MHz crystal. It is needed in order to divide by powers of 2 to get the 5, 10, 20 and 40KHz spacing. I have the K7QO Marker Generator, which I am about to release to the world at large, as soon as I do one more touchup on it. Stay tuned. It is so cheap and yet so simple that it is a wonder no one thought of it before now. IMHO. :-)
OK. Here are some PDFs of some circuits for you to build, if you haven't already. Nothing complicated and hopefully you have a parts supply that is sufficient for building these critters. They use common parts, except for the VE3DNL puppy.
PDFs to follow shortly in this spot.....
Here we hit yet another bump on the freeway. Time to measure some crystal parameters in order to determine the capacitor values in the IF filter. In order to measure crystal parameters we are going to use the latest G3UUR crystal oscillator in the Fall 2010 QRP ARCI Quarterly. With the circuit I measured the C_m, L_m and ESR for one crystal and am using it to figure out the caps for the filter. Those are used in the following installation steps and I'll get back to write up the details. This stuff takes time and we are shooting for the first Saturday of the New Year. Remember.
At this point it is crunch time. Time to hook up the tune pot and make sure VFO tunes the range of 5.000MHz and up in order that it can hear 5.000MHz + 16.000MHz mixed to get 21.000MHz. If we have an incoming signal from the antenna at 21.001MHz, then we should hear a 1.0KHz tone in the headphones. Otherwise we have a problem Houston.
