2021 ~ A Chance to Explore the Old and Update with the New!

So OK you Don't Want to Hear About Homebrew SDR Radios!


N6QW Does it Again! Reworking Commercial Radios for use on the Ham Bands!

April 1, 2021 ~ It May Be Working Again!

This is how it looked before N6QW took at turn at the CNC.

Following what I posted earlier, I checked for voltage switching on PTT and for proper bias on Diodes D2 and D3 for Rx and Tx. Harkening back to the good old days, just like the GF who wasn't pregnant (after a close scare) we now see the rig working again. The mystery signal is somewhat back but pretty much a lot less than it was.

I have no rational explanation of what happened; but perhaps the Radio Gods wanted me to understand how the circuit actually works. I do know now; but still do not know what happened.

Pete N6QW

April 1, 2021 ~ What is the problem(s)?

My HF Scout simply stopped working on transmit. The receiver works FB and it will shift over to transmit when you engage the PTT. But there is no RF and nor a waveform signal coming off the board. There are many common elements for receive and transmit so if it's working on receive, we likley can make some assumptions about those common elements being OK for transmit.

So maybe the blog readers have a suggestion or two about what to do.

Here is some background data. IC1 is a double balanced mixer and originally installed and soldered to the board was a TI SN76514C. This has since been upgraded by TI to the TL442C. They are identical and interchangeable. This chip does two jobs with one being the product detector on receive and the second as the balanced modulator on transmit.

Here are some key pin assignments:

  1. Pins 1, 14, 7 and 8 are not connected to anything
  2. Pin 3 is the Audio Output to IC2 which is the Audio Amp
  3. Pin 5 is the BFO Input Pin fed from Q8/Q9. I removed the BFO Crystal and simply fed the Base of Q8 with CLK2 from the Si5351 through a 10nF cap. The output of CLK2 is fed to a 10K pot and the center wiper feeds the signal to the Base through the 10nF.
  4. Pin 11 is unique in that in receive, signals from the IF coming through T3 are fed to Pin 11. That works,
  5. Now in transmit Q5/Q6 are the Microphone amp and the mic audio is likewise backfed through T3 to Pin 11
  6. Pin 13 is only used in transmit and is the 9 MHz SSB signal coming from the Pin 5 BFO mixed with the transmit audio backfed through T3. 
  7. There is audio on Pin 11 on transmit, the BFO signal on Pin 5 and there is DSB output on Pin 13

This is where things go "south" My SG9 board has two additional controls shown on the board where one can adjust the output level on transmit from Q1 and another one that introducses a dc bias level into Ic1 -- a pseudo carrier balance control but not sure.

The output fromPin 11 on transmit is fed to C34 which is also a DC blocking cap because in theory Q11 on PTT provides about a 8VDC bias (from Q11) that is on the same line as the output from Pin11 through C34. The output on the other side of C34 is really low. Now is that an open cap or that there is something shunting the signal to ground downstream. D2 is a steering diode to put the SSB Tx signal on the Crystal Filter while back biasing D1 which is in the receive path. 

There is DC on C39 -- which there should be and hardly any SSB signal on the input pin of FL1. There is virtually nothing coming out of the output pin of the Crystal Filter. Q2 is working OK on receive and so it should work on transmit. Q1 leaves a question as there is no output as there was before. But that might be because of no signal input.

I removed the SN76514C and installed a 14 pin socket. Inserting a TL442c into the socket  -- it works on receive but no transmit. So the issue is not the SN76514C (which I managesd to remove intact and it works in the socket -- but not on transmit. I removed the Q1 FET and installed a J310 -- no Tx output.

So I am left with the fact it works on recive but the suspect components on transmit have to be C34, C37, D3 and D2. 

D3 is may be an issue in that it works OK on receive if it is shunted to ground via a Bias from CH2. All that accomplishes is to ground C37. Thus anything coming from Pin 11 on receive is shuted to ground. Howver If D3 is biased on at all times ,  on transmit that would cause any transmitted signal from Pin 11 to be shunted to ground via C37 and D3. Any DC voltage supplied by CH2 would be blocked from reaching Diode D2 via C37.

Now if on transmit a DC voltage is still supplied to D3 the  9 MHz SSB signal would be shunted to ground via C37 and D3. What I need to look at is if a constant voltage remians on the anode of D3 regardless of receive or transmit. Following my reasoning the anode of D3 should show a voltage on receive but not transmit.Voltage supplied to D3 via R44 & R46 (being fed from the Emitter of Q10) and CH2 but is also supplying Drain Voltage to Q4 via the Emitter of Q10.

So a possible issue is D3. Just looking at the schematic has at least provide a list of things to evaluate. 

Press On --Invictus!

Pete N6QW

March 31, 2021 ~ "Up In Smoke"

This is not a reference to a Cheech and Chong movie; but the most recent fate of my HF Scout SSB Transceiver Project. 

In an attempt to take what I had built so far and move it to the Dentron Brick we experienced some sort of anomaly where now it will not transmit. I think the problem is on the SG9 board. So all workhas stopped until I get things sorted through. (Although the receiver is working FB --so can eliminate common circuits as the fault location.)

Only word word ... Bummer!

Pete N6QW

March 30, 2021 ~ Scout Modifications.

The "Ride Along Signal" was found and now the output is clean!

1st Fit Checks and Initial Fabrication of the Front Panel...  WYKSYCDS

In the original build the Speaker was behind the front panel. Today I made a new front panel and with some aluminum spacers mounted the original panel 1" behind the new panel.  

The new panel contains the Colort TFT and there are several slots so the audio can pass through the front panel. The circle drawn on the new front panel was so I could capture what need to be dispplayed without interference from the circular hole in the front panel. This is where having an engineering degree pays off! 

The new panel will be painted black and the toggle switch selects USB/LSB. The larger knob next to the new panel is the encoder. 

March 29, 2021 ~ 1st Transmitter Tests!

But some of the new technology can suddenly breathe new life into what is frequently called "old boatanchors".

Many radios which were built in the days of old were actually pretty good designs. I came to fully appreciate that point when I modernized a Heathkit HW101. But many old designs suffered from some very common problems and these include:

  • Drifty Analog VFO's. I laugh when I hear the guys on the Collins net state they have had the rig on for at least 1/2 hour so the dift has stabilized.
  • Non-Linear VFO's. Yes many big name radios had VFO scales that were almost logarithmic. You know lots of space between dial divisions at the CW end and the tick marks on phone were like a blur.
  • Heat -- I loved that I could warm up the shack merely by turning on the HW-101.
  • Rats nest wiring so it was hard to find the parts. Or worse gain access to remove parts.
  • Heterodyne and BFO Crystals would drift and so having good sounding signals or being aligned on all bands was a crap shoot.

But Technology has come to the rescue and many (but sadly not all) of these problems can be fixed with low cost readily available bardware. The Arduino and Si5351 can in fact cure the drifty Analog VFO problems as well as putting the BFO spot on the filter pass band. While I haven't seen this -- a Si5351 variant actually has 8 clock outputs --so why couldn't this variant be programmed to provide precise heterodyne injection frequenices. Now we have spot on digital readouts of the operating frequencies but also that precision is maintained for any band.

The new technology will only slightly help improve the heat generation if it is a "toob" radio and the rats nest wiring like in the Hallicrafters SR-150 is what it is.

But some of the earlier solid state radios can really be "spiffed up" if the new hardware is added. So this takes me to the Dentron HF SSB Scout radio I recently purchased. Regrettably there is not much documentation on this radio. What I have been able to piece together is that it was a 100 watt, 5 Channel solid state radio operating on the 4 to 5 MHz region and intended for use in the Civil Air Patrol -- possibly the origin of the name "Scout". 

BTW WF8J heard in the video is in Ohio. 40M was open!

Essentailly Dentron bought a board (SG-9) from a Japanese manufacturer that contained most of the low level stages like Mic Amp, Audio Amp, BFO, Crystal Filter, IF Amp, Product Detector and Balanced Modulator and a Low Level Transmit RF Amp. The Receiver Mixer and Transmit Mixer as well as the Crystal Oscillator and 5 Channel Crystal Deck were on a separate board mostly likely built state side and Finally the RF brick (Driver and Final RF amp) was a separate assembly also built state side.

I saw this as "Golden Opportunity" to rebuild this into a working 40 Meter 100 watt SSB Transceiver. Yes, there is no documentaion, a schematic of only the low level board and a fools rush in mindset.

  1. The 1st task - Replace the 5 channel Crystal function with a Si-5351 and Arduino
  2. The single BFO crystal on the board was at 9.001500 MHz. So to get LSB -- the LO had to be in the 1.8 MHz range. I did that for initial testing. It would be better to have the LO at 16 MHz. But you would have to change the BFO to 8.998500 MHz. [OK Sideband inversion here]
  3. Dentron built the radio so that access to the boards was very limited. I had to disassemble the radio to make some changes. 
  4. I was able to retune the Receiver BPF to 40 Meters but had some issues with the Transmit BPF and the single Dual Gate MOSFET that was the Transmit Mixer stage was pretty funky.
  5. There were some burnt out components in the RF Brick.

So starting with the receiver 1st, I now have the Arduino and Si5351 purring away and after removing the 9.001500 MHz BFO crystal and supplying the BFO from Clock 2 --we have selectable sideband (think FT-8). Bcause I am injecting the LO and BFO signals into places on the boards that formerly had a crystal. I added two 10K pots to the Si5351 ouputs (LO., BFO) and so now I can adjust the amount of injection voltage so that the signal sounds clean on receive and transmit. 

You can tell by ear when the injection is too high -- lots of background noise, audio distortion and too much audio at just a crack of the volume control and some birdies. It is amazing how when you get the right injection levels how the crap in the background disappers and out jumps the signal. You now find that room filling audio is at about 1/3 of the volume control.

The transmit side was of concern -- so I just bypassed the transmit mixer stage and in its place installed a board that contains an ADE-1, the 40 Meter Band pass filter and a single 2N3904 RF amp stage following the BPF. The inputs are the 9 MHz SSB signal from the SG-9 (Japanese Board) and the LO from the Arduino/Si5351. Now the output is at 7 MHz and the signal is clean looking on the scope. There are at least 30 components on the original transmit mixer board and now there are 20 on my revised board. The Bonus -- better signal quality.

The work continues --just getting started. It looks like a mess but soon will be an organized mess.

Pete N6QW

Popular posts from this blog

New Technology for 2020 ~ The Hermes Lite 2.0 SDR Transceiver

21st Century Homebrew SDR SSB Transceiver Project

New Technology for 2020 ~ SDR RADIG #2 Part II