Tuesday, February 9, 2016

Simpletransreceiver ~ Transmitter Stages Part 1

Heating Up the Iron ~ Transmitter Stages #1

After some six weeks of being off the air and tending to a family medical emergency, we are ready to start building the transmitter portion. While I was "noodling" over the transmitter stages I realized that many of the circuits used in the receiver stages are simply "ported over" for use in the transmitter. In the lower level stages the one new circuit board is that of the microphone amplifier. According to the LT Spice once again the "home built" Dual Gate MOSFET works quite nicely in that circuit element. So to be true to our initial design premise -- we will continue to use J310 DGM's in the transmitter blocks.
By way of review below is the transmitter block diagram with a special note that the stages with the red asterisk are a direct lift from the Simpleceiver. To that should be added the carrier oscillator. At one time I had thoughts of sharing that with the  in the Simpleceiver. It will now be either a separate oscillator similar to the receiver BFO. [I have already found 4 crystals that are a very close match to the BFO frequency] or it will be CLK2 out of a Si5351
 We'll start first with the microphone stage and once again using our template for the DGM  made from two J310's, is used. Frankly this stage is critical to a high quality sounding station and so we attempted to tailor the audio to the range of about 100 Hz to 4.5 kHz. Your signal should have presence without being "hissy".

Since the rig has been designed for use on 40 Meters the builder will undoubtedly encounter the Spectral Purity Police, many who have never homebrewed a rig, but are running high end SDR boxes with 52 inch LED Displays whose only goal in life seems to be to commenting on how bad your signal sounds. We wouldn't want that.

 We are reposting the schematics for the BFO (we'll now call that the carrier oscillator), the IF amplifier stages and the crystal filter. For the transmitter you must use the 4 Pole Crystal Filter. Please review the earlier post on the building of the crystal filter -- there is some art, science and pure black magic to building a successful filter. Do not believe those who tell you they built one in five minutes --5 hours is more like it.

The low level stages use packaged Double Balanced Mixers, the SBL-1 which are 7 dBm devices, meaning you need 1.414 Volts Peak to Peak to drive these DBM's. Some prefer the ADE1-l (both are from Mini-Circuits labs) which is a 4 dBm device that only needs 1 volt Peak to Peak of RF drive. Caution the ADE1-L is a surface mount device and is small. For the SBL-1 pins 2,5,6 and 7 are grounded. Pins 3&4 are connected together and this is where the audio is injected. Pin 1 is the output pin and Pin 8 is the LO injection Pin. Looking at the top of the SBL-1 package the letter M (as in MCL) is over Pin #2.

There are many purists who believe unless you make EVERYTHING then it is not true homebrew. You can make your own DBM -- I have three videos on youtube that show you how to do it. But you will never achieve the performance of something made under exacting factory conditions. You will also find that a homebrew DBM will need more drive like 2 Volts Peak to Peak. In several of my published articles I have used homebrew DBM's and will sometimes get emails like "Your homebrew DBM design doesn't work!" Well friends it does and often I hear back that the person having difficulty lost track of the windings and it was wired wrong. Save yourself grief and just buy the SBL-1 and get on with it!

The IF amplifier schematic is presented below and much like in the receiver the output has a resistive pad ahead of the crystal filter. Once again we must match the in/out impedances to this this stage. The output of the SBL-1 (pin 1) is at 50 ohms and the input to the DGM is set a 2.2 K thus we have a 1:44 match (2200/50 = 44). Using a FT-37-43 (it must be a ferrite core) a 3 turn primary (50 Ohm side) will match  with a 20 turn secondary to the input --remember turns ratio squared 3^2 = 9 and 20^2 = 400 ---- 400/9 = 44.44444 --so really close! The output side must match 50 Ohms to 150 Ohms which is the input to the crystal filter so we need a 50:150 Ohm match which is 1:3 which is easily done with an 4 to 7 turn match. 4^2 = 16 and 7^2 = 49 --- 49/16 ~ 3. So close enough.

For the second IF amp stage you must match the 150 Ohm out of the crystal filter into the 2200 Ohms input. You can do it with one or two transformers. In the two transformer case it is 150:50 (7 to 4 Turns) and then 50:2200 (3 to 20 Turns). I recommend doing this since you clearly understand the impedance match. Or you could do it with a single transformer where you match 150 to 2200 Ohms which is a 1:14.666 and thus 6 Turns to 23 Turns would get you close 6^2 = 36 and 23^2 =529 ---529/36 = 1:14.7. The resistive pad is not used on the output of the 2nd IF stage.

Following the 2nd IF stage the signal is fed into Pins 3&4 of another SBL-1 which is used as a frequency translator. The LO signal supplied either by an AD9850 or an Si5351 is fed into Pin 8 and the output is taken from Pin 1. The signal out of the frequency translator includes the sum and difference frequencies and so we need to run that through a Band Pass Filter to select the difference frequency ( 12.096 MHz - 5.0 MHz = 7.096 MHz).

Repeated here is the 40 Meter Band Pass Filter that is 50 Ohms in and out. So it simply connects to the output port of the SBL-I which is at 50 Ohms (Pin 1).

Finally is the circuit board layout that can be built with the W1REX MePads. You will note there is space on the board for the SBL-1 Frequency Translator and the 40M Band Pass Filter.

Lot to swallow here but this should keep everyone (so OK probably at this point maybe 3 or 4 of you) busy building the hardware.

Thanks to all who inquired about the XYL -- she is home and returning to her usual old self. The metric here regarding her return is how much she complains that I am a ham radio operator. She has always hated the hobby and now that is going on 49 years!

Pete N6QW