Our process has led us to identifying a host of problems in the area of the transmitter module and here is a list of what we discovered:
- Evidently a prior owner was troubleshooting problems in the transmitter section which resulted in cutting wires and tack soldering components everywhere and anywhere. The wire repairs involved what looked like welded back connections. What a jerk!
- Evidently the PA-200 high power board was installed in the transmitter as evidenced by the large mounting screws still left in place and the "homebrew ventilation" holes which were crudely installed in the case top and bottom using a hacksaw. What a jerk! Thehigh power version also required some modifications to the QRP driver board. These included lowering the value of the RF Feedback Network resistor from 150 Ohms to 75 Ohms. Per the instruction you simply tack solder a second 150 Ohm 1 watt resistor in parallel with the normal resistor. You also were asked to remove a 150 PF capacitor across the secondary of the binocular core transformer. You were to save this cap and use in the PA-200 High Power Stage. These changes were in place -- a second paralleled resistor and no cap was in evidence. The PA-200 stage was also missing.
- Evdiently the prior owner must have concluded that there was a problem in the 1st Microphone amp stage (a MPSA12 Darlington) and there was musch evidence that the circuit was rewired in a haphazard fashion. It was awful. In fact the MPSA12 was not the device in the circuit but some NTE replacement part. I ordered some MPSA12 transistors (25 cents each so a buck get you four) and new resistors and capactitors to create a PC circuit board which I would install. See the two photos below. Five minutes with my mill and a peice of scrap PC board and we are in business.
- Found a really curious use of a 7 pin tube socket in the Atlas Tx110. This was the official install as there is a screw on the front panel that has an outline on the silk screen for that screw which then passes through the front panel and through the center cylinder pin of the 7 pin socket where the screw is soldered through that center ground pin on the socket. Several of the pins were removed leaving only three pins in the socket base. This tube socket with the pins facing toward the inside of the chassis and firmly anchored through the center grounding cyclinder is now a terminal strip where components are soldered to each other. The components that were in place when I got the rig were burnt and just looked awful. Later I used one of the tube socket pins to solder to my replacement microphone circuit board as a means of capturing the board so it doesn't move.
- After a lot of poking around I decided it was time to try a power on test with the transmitter. The results were encouraging with 10 Watts out on 75 Meters, about 4 watts on 40 Meters and hardly any on 15 or 10 Meters. That is when I looked at the intermediate driver transistor which supposdely was a 2N3866. It had a heatsink on it that looked like someone had put heatsink compound now was a little brown in color --not a good look. Also the solder pads were disturbed like it had been replaced. I installed a 2N5109 (a good sub for the 2N3866) and boom we got about 7 watts on 40 Meters and 4 watts on 20 Meters and now about two watts on 15 meters and only a watt on 10 meters. But progress. I should mention that I removed the extra 150 ohm resistor required if the high power board is installed and reinstalled a 150 PF Silver Mica across the secondary of the output transformer. For those with inquiring minds, the second resistor in parallel lowers the feedback resistor so there is heavier feedback which reduces the stage gain. The whole idea --don't overdrive the high power final amp if it is installed.
- While the power output has been increased greatly from the initial testing that still leaves the question as to why it is not what you would see from the factory. Now what is clear from the Tx110 manual the power output will drop as you go to the higher frequencies --by close to half between 80 Meters and 10 Meters. So not totally unexpected. But my test number show a far greater drop off as you go higher in frequency.
- More time noodling and looking at the schematic and what was installed. Now there were some parts of the RF amp circuit board where it was clear that nothing had been touched. That is when I looked at the schematic and what was actually installed. The schematic shows a 470 Ohm feedback reistor; but the actual manual states that it is 150 Ohm and that is what is installed. Less feedback does not flatten the gain as much but does increase the stage gain for certain frequency ranges. The 1st transistor in the RF amp stage is an MPS6514 ( a Ten Tec favorite) and it has an interesting feedback resistor network in the emitter circuitry. There are two resistors shown on the schematic and bypass capacitors are across each resistor. Shown on the schematic are two resistors, a 10 ohm in series with a 180 Ohm. Across the 10 Ohm is a 430 PF cap and across the 180 Ohm is a 2200 PF cap. However on my board there was no cap across the 10 Ohm and the pads looked factory. I installed a 470 NF across the 10 Ohm and ran some tests. Boom the poutput on 75 Meters was about the same but on 40 Meters the output was almost 10 watts, and on 20 Meters the output was about 6 watts and even now on 15 Meters the output was 4+ watts. and 10 Meters was now 2 watts. So that one capacitor made a huge difference. I am now looking to the emitter reistor on the 2N5109 which the schematic shows as 4.7 Ohms at 1 watt and the same is actually installed. I am going to test the board using 2 ohms and bypassed with 10 NF. This should now also help raise the output.