Wednesday, September 21, 2016

Taking a Break!

Are you tired of seeing me post about my rigs?

9/28/2016 ~ Some recent updates

I have often said that most of my rigs are experimenter's platforms and the Junk Box Rig stands as an example of that approach. Everyday I look at possibilities for improvement --some changes are in the negative column but many are in the positive direction. Usually I employ the concept of TFMS when "peaking and tweaking" a rig. Oh almost forgot TFMS is an old timer term for Tune For Maximum Smoke. Well in this case when I had the rig outfitted to run 700 Watts out (Yes, Virginia, that is not QRP) I found that I got a report of splatter. One of those Flex Guys saw me pop up on his 60 inch color display and he reported I was very broad and splattering over onto other frequencies.
 
He was absolutely right! If you are running QRP at one watt --there may be splatter that is hardly noticed but magnify that 700 times and you will notice that your signal is broad and spilling over the normal bandwidth. A simple adjustment of the microphone gain can cure that problem. So out comes the scope and I was indeed flat topping and lowering the gain didn't actually impact the PEP output but did immeasurably improve the signal quality and cleaned up the spectral purity. For reference purpose a 1 watt signal is 30 dBm and a 700 watt signal is  58.5 dBm. So you will hear that very large elephant in the room!

Once I reduced the microphone gain, the Flex guy was  happy and so now I was happy too! WE have a responsibility as hams to put out the cleanest signal possible! I am now ever vigilant to not let that happen again. The positive side was that my PNP audio amp was no wimp and I had a gentle reminder about flat topping and splatter.
 
Rummaging through the junk box I found a 2SC2166 on a board and swapped out the 2SC2075.  Now 5 watts out is an easy goal. I also made a change on the driver board where normally the EMRFD circuit has a 2N3904 driving a 2N3866 but in my circuit I have a 2N2222 driving a BD139. In the emitter lead of the 2N3904 (my 2N2222) there is a 22 Ohm resistor to ground. I replaced that resistor with a 200 Ohm pot. The center wiper is connected to ground and the top lead is connected to the 2N2222. This enables me to adjust the stage gain. When I drive the 100 watt intermediate amp ahead of the SB200, I have it set so the Junk Box Rig outputs 1 watt as that is about the max level input to the intermediate amp. In the TFMS QRP mode (without any amplifiers) I can adjust that pot for 5 watts output from the 2SC2166.
 
On 9/27/2016 I made a 400 mile hop to San Francisco from my QTH near Los Angeles running 1 watt out on 40 Meters. Not a lot of signal strength but readable. But I must confess that 40M QSO's to the mid-west are much better running 700 watts!
 
Recap: Beefier RF Transistor, Watch for Flat Topping, Variable RF Output, QRP QSO's.
 
73's
Pete N6QW

9/21/2016 Revised Microphone Amplifier and plot. (See Comments)

Revised Schematic to limit the frequency response.

Output Plot of the revised shematic

Here is a chance to tell me what is on your bench and if you have attempted to build any of my projects.

73's
Pete N6QW

Saturday, September 3, 2016

40M Junk Box SSB Xcvr in a Blue Case

The Blue Junk Box 40M SSB Is Alive!

9/15/2016 ~ Continued -- Listen to the Junk Box Radio...


9/15/2016 ~ Junk Box Experimenter's Platform

Ever wary of unwanted frequency mixing schemes and picking an IF too close (harmonically) to another band, I decided to change out the Yaesu 3.180 MHz Crystal Filter. The second harmonic of the BFO is at 6.4 MHz which is in my opinion too close to the 40M band with just the simple Band Pass Filter I am using. Adding a better BPF (more poles) or changing the IF would of course provide better solutions to my concerns. Still another solution would be to use dual conversion approach to mitigate the problem.

I opted to change the IF frequency as I had a spare 9.0 MHz GQRP Club SSB Filter. The change from 3.180 involved unsoldering two wires and the mechanical removal of the filter from the board. The Z in/out of either filter is 500 Ohms, thus requiring no change in the matching transformers. Insofar as Arduino code changes, about a total of eight entries and that was it. Since the LO is above the incoming there is a sideband inversion so the lower frequency BFO frequency is used to receive LSB (8.998500 MHz) and USB uses the upper 9.001500 MHz Carrier Frequency. The changes went FB and we now have an IF of 9.0 MHz. I now would feel more comfortable in moving the rig to 20 Meters. BTW total elapsed time was about 30 Minutes for the hardware change out and the software modifications / loading on to the Arduino Pro-Mini.

This is the beauty of my approach to building a rig. Changes both in hardware and software can be easily made without a complete disassembly of the radio. It is truly an experimenter's platform.

73's
Pete N6QW

The Junk Box SSB Transceiver is now fitted with a 9.0 MHz Crystal Filter from the GQRP Club


A Close Up of the Junk Box SSB Rig Front Panel -- You gotta love that blue!



9/12/2016 ~ Making Contacts
The one nice facet of our beloved hobby is to make contacts and hopefully by what I have done with the last several transceiver projects is to inspire others to "take up the iron". Perhaps one of the most powerful ways that can be done is by using the rigs on the air. I have actually spent more time on the air in the last several weeks than I have all year long. True much of my operation has been in short bursts like 20 minutes here and there -- but the results have been many fold. Not only have I met some really great people; but the interest level has ramped up with many inquiries about the rigs I have built. Perhaps one or two will be inspired to build their very own rigs.
 
73's
Pete N6QW
 
 
 
9/8/2016 ~ Ramping up the Power!
Yesterday I added some switching circuitry so I could simultaneously key up the Atlas 210X "brick" and my SB200 Linear Amp. I am happy to report I can now put out about 600 Watts into my droopy dipole. This now should give me a shot at working coast to coast with the droopy dipole.
 
Another critical part of the tests were to make sure I wasn't tripping (or ripping up)  the security alarm, telephone system, sprinkler system or the cable TV. All is clean and so the grounding, shielding and good wiring practices have paid off.  
 
I am very pleased at the performance of this radio and now with winter coming on want to do some serious 40M operation. I note that in the early morning hours I am routinely hearing stations in Japan and Indonesia on 40M SSB. So that is another goal to make contact with these station and I feel possible with this rig and the amps. I hardly ever run it at 3 watts so I guess my QRP award is in jeopardy of being revoked.
 
73's
Pete N6QW

9/7/2016 ~ Some observations and updated information

I continue to be amazed at the performance of this "Junk Box Rig". Signal reports affirm its good quality signal and nice sounding audio which does seem to have some added "sparkle" to the sound. I attest the audio reports to spending some time in improving the frequency response of the PNP microphone amplifier. A bit of time with LT Spice pays off in big dividends.
 
You can't imagine my surprise several days ago where about six stations in a row contacted me for details about my rig and how it was built. There is no finer compliment than to hear your station being called by others. With a bit of tuning and some improvements I now get about 115 watts out using the linear amplifier "brick" liberated from an Atlas 210X SSB transceiver. This power level now enables easy QSO's into the mid-west and western Canada with my "droopy dipole" installed on my postage stamp sized Southern California lot. A better antenna would definitely extend the range into the east coast on a regular basis. So the antenna does count. In fact typically the second question after the usual first question about the rig is what kind of antenna are you using?
 
Extensive use of a radio will often identify "issues" and so it was with the Junk Box Rig. I found that after an hour or so of operation the rig would not return to receive after a lengthy transmission and at times even though the rig would go into transmit -- there was no output signal. The problem was traced to the solid state switch. I found that when this occurred the SN7400 IC which was wired as an inverter was "hot"  to the touch. Not just warm; but burn your finger hot. So my fix was to install a small 12VDC SPDT relay in lieu of the switching board. Problem fixed and a simpler solution. The radio no longer suffers from that hang up. But it does beg the question as to why? I will do some further bench evaluation --but in keeping with the theme of this rig --the relay is what I found in the junk box -- and even though the relay is small its contacts are rated for 3 Amps DC at 32 Volts --so plenty of head room.
 
To seem repetitious often when we challenge ourselves to find solutions by using what we have an end result which is often better than simply flipping out the plastic and buying new. I have quite a few boat anchors and aside from the KWM-2, this rig hears better than most of those radios of old. With a bit of scrounging and some horse trading this rig can be built for less than $150. The frequency generation part, BFO, encoder and the LCD can be bought for about $20. The SBL-1 Double Balanced Mixers can be purchased from RF Parts for about $16 and the Iron Powder and Ferrite cores add about $10. The real bang for the buck is the pleasure of saying my rig is homebrew built by me!
 
So what is on your bench???????
 
73's
Pete N6QW

It has been one week since we made the first contact with the 40M Junk Box SSB Xcvr and I am happy to report that the hunk of junk has now been given a proper and fitting home.

This is how it all started!

Using my Harbor Freight metal break I made some brackets to provide support to the front panel

The rig base is a piece of copper clad board 10.5 inches long by 8 inches wide. Here we see the front panel/support brackets installed on the base plate assembly. I used my manual mill to make the front panel.

This was an exercise to determine how all of the boards would fit in the case

The boards are all mechanically installed and the wiring is everywhere.

Here the wiring is about 85% done

Oblique view showing the various boards.

The Junk Box Rig sitting next to its older brother the Nu-Rig. Love that Blue!

The Junk Box Rig is now fully functional.

Front Panel close up!

Innards: Driver Board, Final and Solid State switch

Band Pass Filter, Bi-Directional Amp, Main Board, Si5351 and Microphone Amp.

Left Oblique View.

Right Oblique View: Audio Amp, TR Relay,Linear Amp Switch and Low Pass Filter

Power Amplifier. The device is mounted directly to the bottom copper plate (insulator used). We have 84 square inches of copper heat sink.
 
I now have connected this rig to the "brick" liberated from an Atlas 210X. With 3 watts of drive I get over 100 watts out. Since finishing this rig today I have made 5 contacts running the amp --- all excellent reports. Three of those contacts were into the mid-west.
 
What are you waiting for --get cracking on your junk box rig! Soon I hope to make a video so you can listen to this wonderful radio. This also has been a great experience for me in using what you have -- my 1st thoughts were to buy a chassis --ouch at $30 I figured there had to be another way and there was.
 
73's
Pete N6QW

Friday, September 2, 2016

40M Junk Box SSB Transceiver Microphone Amplifier

Exploring PNP Transistor Microphone Amps


This is a first for me in terms of using PNP transistors for microphone amps. In fact more recently it was a first for me to use a single NPN transistor as a microphone amp. For the longest time my standard building block was the NE5534 for microphone amps.
 
On the air signal reports for the Junk Box rig were very good with the added comment that there seemed to a favoring of the highs with very little in the way of low frequencies. Well Duh, a couple of minutes with an LT Spice Simulation confirmed the why this was so. This post will explore the why.
 
[I should note that I have a bag of vintage PNP audio type transistors that I have often wondered what can be done with these treasures of old? Now I know!]
 
I ginned up the circuit I was initially using and it worked and just moved on but given the reports of the lack of lows I actually simulated the original circuit and found that there was a substantial lack of gain (like 10 to 15 dB) for frequencies below 1 KHz. Above that frequency things were much better.
 
Since there were no LT Spice Library entries for the 2N996 I picked the 2N4403 and then verified the same results with a 2N3906. Two capacitors play a key role in boosting the low end response and those are C1 and C4. Initially I had C1 at 100NF and C4 was 10 ufd and of course we could see that lack of the low end response. With a bit of cut and try these final values of 10 and 100 ufd, really boosts the low end. Now there is less than about a 3 dB change from 300 to 3000 hertz. I used a generator with a 100 millivolt output and swung the frequency from 10 hertz to 5000 hertz.
 
For those picknitters in the readership the + side of the 10 ufd is toward the signal input and the minus toward the base. On the 100 ufd, the + side is connected to the emitter of the device. I installed a socket on this board so I can actually turn this into a PNP transistor checker. For those who are still uncertain about electrolytic cap polarity-- the answer is to purchase and install non-polarized electrolytic capacitors.
 
A little more than the 15 minutes I spent with LT Spice could even further improve the low end so that the circuit is flat to 5 kHz. At this point I say Basta and just move on.


Schematic of the PNP Microphone Amplifier

Finally for the doubting Thomas crowd here is the expected performance of the circuit. I should also caution that R4 is really a 10K Trim Pot and the 220 NF (0.22 ufd) cap is connected to the center wiper. The output end of the 220 NF is paralleled with the audio amp input where both are connected to pins 3& 4 of the SBL-1.

Projected Output from the PNP Microphone Amplifier Stage.

 
The bottom line is that the PNP Microphone amplifier stage is good for about 20 dB of gain over the audio range and confirms the suitability of the device in homebrew SSB transceiver projects. Somehow many of us sort of gravitate to NPN devices --but good circuits can be built using PNP devices.
 
73's
Pete N6QW