2021 - The Year of Nostalgia. A 60 Year Trip


 A 60 Year Time Trip!


Starting with Today (2/22) -- The N6QW Conversion of the Dentron Scout HF SSB.

3/1/2021 ~ Mystery Likely Solved!

The Linux Mint Machine has a Choice of Bootloader -- Old or New.

I have received several emails stating that my problem with the Nano is the Bootloader! The new Nano's have the newer bootloader installed whereas the IDE 1.8.5 only resolves the older bootloader. Thus the Linux Mint Machine 1.8.12 will do either one.

I was chasing my tail. Why do those guys do that? This is the clue as found in Arduino IDE 1.8.12. This menu is not in IDE 1.8.5 which evidently responds to the old bootloader and not the new bootloader as is likely installed with the recently purchased Nano's. So if you are using a newly purchased Arduino and an older IDE, the answer why you are getting error messages may be this issue. I must ask again --Why do those "Software Weenies" do things like this which drives the user community NUTS??????




Today I opened a new link on my website which documents a project using the MC1496 DBM See Direct Conversion Receiver




The website links in excruciating detail how to build this FB receiver. Also conatined is a link to the work of two hams in the UK, who built this jewel and they show alternative construction methods and many detailed measurements on the performance of the BPF etc. Also in their documentation are two videos one showing SSB performance and the other CW. This is perhaps the most documented DCR you will ever see!

Happy Camper


Pete N6QW  

2/27/2021 --- Strange Happenings
!

I am working on the Color TFT digital LO/BFO for the Dentron Scout and was short on Arduino Nano's. Some arrived this week and that is where the FUN began. 

My main computer (in my home office) has all of the Arduino stuff because that is where I have two back up external HDD's. Don't want to lose those jewels. It has the Arduino 1.8.5 IDE installed. I have not upgraded the 1.8.5, as has been found where later IDE's seemed to have problems with sketches that were developed with earlier IDE's. 

But the problem was the Nano's. The Nano uses the CH340 USB interface and long ago I installed the software so you could program the Nano's to use this inetrface. It worked well! But it didn't work this time! I purchased a three box set of Nano's and all had the problem. One onboard LED just blinks and will not accept the program.

Recently I installed Linux Mint on a homebrew computer out in the shack and also installed Arduino IDE 1.8.12. This version was installed  because I also was able to install the Teensy add on package with 1.8.12. 

My next step: I simply copied sketches (that would not load on the Nano's) from the main computer and using a thumb drive copied them to the Linux Mint machine. Plug in the Nano's on the Mint Machine and the later IDE (1.8.12) and they all took the program and work.





So I ran a bit of a test where I took a programmed Nano (from the 1.8.12 Mint machine) which was verified as working with real hardware and put it on the 1.8.5 machine and tried to reload the same porgram -- error message and  it wouldn't load. 


So has anyone else seen this? There is a bonus to this exercise as I can do all of the software changes on the bench versus taking things back to the home office computer. But now I will have to back up the Mint machine!

I am surmising that new hardware (Nano) is now being updated to match the new IDE software. I am reminded of when earlier Nano's used the FTDI interface which was ripped off by the Chinese. The solution to the rip off was that FTDI went to Microsoft so that during the course of a Microsoft update the knock offs were essentially neutered (bricked). So that is just a data point.

2/26/2021 --- More Update Info.

Just ran across another website where it is stated that the MiniLX from Dentron is built around the Mizhuo SG-9 Board. Boom, that is the name on the board in the Scout HF SSB so no doubt that the schematic for the IF board is indeed correct and proper. We are on a roll! 

Some poking yesterday did not improve finding a location to inject the USB LO signal so that USB/LSB are equal. I then thought that the Crystal Oscillator supplied RF for both the Rx and Tx and so finding an alternative input would have to address the RF LO for transmit. So I think the BFO crystal removal and replace with the Si5351 is the best alternative for USB LSB select. 

Note the SG-9 at the top of the Circuit Board right above the Pot.





2/24/2021 ---- Update on Progress.

The interesting part about dealing with these older radios is that unless you have some tribal knowledge you are like a cooked goose. If you were starting today without having a lot of background in how things were done long ago -- it would be like looking for a needle in a haystack. There are some pluses to being as old as dirt!

Starting with what looked like the Band Pass Filter on the small board behind the MiniLX board I peaked the two IF cans for 40 Meters. There was a dramatic increase in the received signal strength. I reasoned that the BPF had been tuned lower in frequency based upon a quick examination of the Low Pass Filter assembly (shown below). 

Here is the tribal knowledge part that only comes with having built so many LPF's over 60 some years. You get an innate sense of the component values for the various bands. For example having 220 PF's (in the filter) are not values usually found  on 80 Meter LPF's!

The Dentron LPF was likely OK at the time (1980's); but is not good enough today to comply with second harmonic reduction requirements. The LPF assembly consists of two coils and three caps. The three caps include a 1200 PF at either end and a 2400 PF at the center. The Cores are two T-68-2 with 15 turns on each Core. The inductance of that core is 1.2825 uH. Given the values of the Silver Mica Caps -- this LPF likely would cut off in the 5 MHz range. This is a huge clue




I should add the LPF is always in line so it also affects the received signals as well as the transmitted signals. Now that I had some values, I modeled the LPF in LT Spice. Boom given those cap values indeed the LPF Cutoff is around 5 MHz. Pumping a 7 MHz signal into that LPF would sure drop the power output by about 20 dB and it would also impact the received signals.



So to make this play on 40 Meters, I will need to change the LPF and will use the W3NQN filter model for this application as that will improve both the receive and transmitted signals. Specifically the W3NQN are designed to address the 2nd Harmonic issues. Below is some initial work using the W3NQN Model with a bit of tweaking based on what I have in the shack and to reuse two of the original cores. Look specifically at the 2nd Harmonic reduction.






We get the range moved out to cover the 40M band and the 2nd Harmonic is "down in the mud"!

I did note that any adjustments of the BPF placed the cores at the extreme ends of their adjustment. So I may need to adjust some capacitor values across those cores for maximum peak. But just the core adjustment was a significant improvement. There is a second set of BPF transformers used only on the transmit side so they will require peaking when I get to the transmitter modifications.

Now a circle about in that some of the documentation I saw for the HF SSB Scout said it operated over the 4 to 10 MHz range. That info also suggested it was intended for use by the Civil Air Patrol which operates in the 4 -5 MHz range. This is supported by the LPF data. So likely the Dentron had a much more limited operational range geared to the CAP. But not to worry as I think it can make it to 40 Meters by making the approprite adjustments. 

The intial Si5351 tests were done with the approach I used with a Bitx 40 so you could have both USB and LSB simply by switching the LO range ( 2 MHz and 16 MHz). The Si5351 LO is simply plugged into the Crystal Socket and thus the original oscillator transistor is in line with the Si5351 output signal.

But just as what happened with the Bitx40, I found that signals were loud on LSB but seem attenuated on USB. The Bitx40 issue was that I simply plugged into the circuitry that was set up for 5 MHz (LSB) but when I pumped in a 19 MHz LO (USB) the onboard Bitx40 circuitry attenuated that higher LO. I think it is the same issue here in that the crystals used were likely either 5 MHz (9 - 5 = 4) or in the 13 MHz range (13 - 9 = 4). Putting a 16 MHz LO signal in the loop may have to be downstream from the crystal oscillator components. Or just remove the crystal and have the Si5351 supply the LO and BFO signal sources. My 1st choice is to test moving the actual insertion point downstream from the original crystal oscillator as I did with the Bitx40. 

The moving downstream for LO insertion avoids "diddling" with the mainboard. Failing that, more surgey on the main circuit board will be required to bypass the BFO crystal and install the Si53515 BFO signal. The sketch modifications for BFO operationat this point  are in place but simply nulled out.

All this initial work was done using a 16X2 LCD. Today I modified some code I have so that I can use a Color TFT that will add some flash to the rig. Stay Tuned.

***




Yes I bought the Scout and it arrived yesterday. Today I tested the use of the Si5351 for frequency control. I did not mess with the BFO crystal but used the Bitx40 trick I developed: For LSB the LO is set to 2 MHz and for USB the LO goes to 16 MHz. 




I am encouraged by what I see so far. There needs to be some tweaking of the BPF and other circuit peaks -- but many possibilities. WYKCYCDS!

This is the IF stage from the MniLX and as soon as I spotted IC1, I thought about the SN76514 (TL442C) that was a staple in rigs of the 70's and 80's -- Sure enough when I opened up the rig --there was the SN76514 --so it is the same circuit board and now I have a schematic.




 ***
I was thinking about some of the rigs I have owned and built over the last 60 + years and my thoughts centered on how our hobby has changed. In the early days (1930's)  most rigs were simple CW rigs consisting of perhaps just a few tubes -- two or perhaps three "valves' in the receiver and perhaps one or two in the transmitter plus a DPDT Knife Switch and an dipole -- Boom you were on the  air.

Later it seemed the receivers were store bought but transmitters continued to be "homebrewed". But the 1950's signaled a dramatic shift in the hobby as kits were now the venue for getting rigs on the air. If you are an OT, who hasn't built a DX-40 or maybe a DX-100. Perhaps even a Johnson Adventurer. These transmitters were often paired up with a Hammarlund HQ-100 or maybe a hallicrafters SX-99 (a budget station might be a S-38E) or a big box National HRO. 


But that wave soon signaled the entry of the rig in one box -- the transceiver. The very 1st was the Cosmophone 35 soon overshadowed by the Collins KWM-1 and then the flood gates broke open with new starts like SWAN and SBE hit the market.


But our hobby is ever evolving and while we had large scale homebrew early on that shifted away to store bought. But now with a new generation of hams who are software wired, the era of homebrew may see a resurgence especially with tools like the Internet, Amazon is your friend, User Groups, Podcasts and best of all CHEAP Technology.

The chart above is from my recollection of what took place when. There may be some question about actual years (was it 1969 or 1970) but the important point --- it is a continuum of evolvement.

Here is a bit of hardware linkage. Raytheon was the final owner of the Sideband Engineers product line and I think by the late 1960's they were no longer producing radios. The SBE-33 was the 1st hybrid radio to hit the market in 1963. Fast forward to beyond 2000 and we have FLEX Radio and the SDR impact. 

Today Raytheon and FLEX are developing the next generation of communications systems for the US Air Force. The impact of our hobby cannot be overstated. In the span of 60 years we have gone from a "toob" cw transmitter to SDR transceivers that literally fit in the palm of your hand.

What will come next?

73's
Pete, N6QW 

Al Capone went to jail not for murder or racketeering; but for income tax evasion. Now the Trump Tax Returns are in the hands of the Manhattan DA and likely soon the the NY Grand Jury. If I had  Trump in my name or connected to a person with Trump in their name, I guess I would be more than just a little concerned. Teflon can be penetrated -- just ask John Gotti! The Work Continues!

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