The Ham Radio Spirit of Experimentation

The convoluted course of a major improvement to an excellent radio!

By SV9DRU / KI4GIN

This is the description of a modification project which has been posted in a separate article a few days ago. It contains all the details of the original article plus its following revisions, based on different aspects of unforeseen effects that surfaced afterwards...

For the last couple of months, I have acquired a new (to me) Ten-Tec Omni-C. After going over cleaning of contacts, cleaning - lubricating the PTO and aligning the BFO, have been impressed by its excellent performance, it’s very quiet and crystal-clear RX and TX.

It is well built, and of a simple 1 IF analog design, easy to understand and service, qualities almost extinct in present day radios.

All was well and rosy, until the contacts of my local AC grid step down transformer started acting out due to the prolonged dry weather conditions and their suboptimal service.

The result has been severe spark type noise on all bands, 24 – 7.

The NB of the Omni could do a credible job eliminating 90% of the noise in the bands above 30m but has been completely ineffective below 40m. Its function was well below that of other radios at this QTH, far exceeded by the NB of my FT-901DM.

I had read about this particular issue especially with the later design type of the NB in the Omni models, and based on comments on the net, was looking for the alternative design which has the 3 crystals in place of the 1^st IF coil, which is supposedly better, but as easy to find as hen's teeth these days!

Looking at the NB schematic and operation comments from page 60 of the TT-546 owner's manual.

The 9 MHz IF signal. Is split into 2 components, one going through a band pass filter and a Noise gate, whose permeability is controlled via an emitter follower switch transistor (Q2).

The second part of the IF signal is used for the detection and controlled amplification of the noise pulses via IC1 and Q1, which is finally used in controlling the switch transistor (Q2).

I supposed that the lack of effectiveness of the NB was due to insufficient IF level gain in the lower bands compared to above 20m. The Omni has a unique feature of tailoring its sensitivity across the HF spectrum, in order to avoid amplifying the higher ambient noise of the lower bands.

But this low IF level results in the noise pulses presented at the base of the amplifier Q1 to be insufficient for turning it on, and thus fail to stimulate the emitter follower switch Q2.

After much thought, I decided to increase the sensitivity of Q1 by altering its resting base voltage to a slightly (+) potential. This idea came from the supposition that noise pulses brought from IC1 via C9 could not reach the 0.6V potential to turn Q1 on, due to their dissipation via R5 to ground.

Adding a 2.2M resistor from the 9V potential (Q 1 Collector) to its Base, elevates the resting base potential to +0.39V instead of 0V, thus keeping it "primed" for the arrival of the weak noise pulses.

Having a 4M7 trimmer wired appropriately, I tried a 2M resistance, touching it between the appropriate leads of R9 and R5 while receiving...and MIRACULOUSLY the pesky noise was COMPLETELY eliminated!

Additionally, there was absolutely no ill effect in the RX performance for either weak or moderately strong signals. It was like someone had switched the sparks off.

A big AHHA or EUREKA moment intervened.

Played around trying different values for the new resistance and saw a similar benefit from 1M8 to 2M2. The effect waned for R > 3M.

Finally, I ended up installing 2 M 1/4W resistor (2x1M in series), unable to locate a 2M2 one in my spare parts box. What is amazing about this simple modification besides its effectiveness, is its simplicity. One does not even have to remove the NB board from the radio!

With a 30W soldering iron slender tip, I first tinned the 2 respective contacts of R5 and R9, followed by installing the new resistor and securely soldering its ends after tucking them below the leads of the aforementioned resistors. Fig 2

Time required 5 minutes!

The result has to be heard to be believed. The following videos tell the story better than any elaborate text descriptions.

Congratulations, the old faithful OMNI has gotten a new life!

Fig1 The unmodified NB board.

Fig 2 The NB board after the modification.

Fig 3 The schematics *page 60 of the TT546 manual

The following YouTube videos demonstrate the radio listening on 40m weak and moderate strength signals after the modification. Note that S5 in 40m band corresponds to S9 in my other radios.

The function of the NB before the mod, had been IDENTICAL to the OFF position shown in the videos.

https://www.youtube.com/watch?v=9JsnwBTyY6Qos|

https://www.youtube.com/watch?v=35U-BeTcNXk

Any comments or questions are welcome.

Enjoy!

Updated info on the Omni NB modification.

During the night when the extremely strong broadcast stations filled the 40m band above 7200Mhz in Europe, significant intermodulation products appeared in the 40m band only, when activating the modified NB.

Further testing showed that increasing the new resistance value installed between R5 – R9 to 3M3, instead of 2M, led to improvement (but not elimination) of the IMD products while maintaining a slightly reduced effectiveness of the NB.

During the next couple of days, observing the behavior of the NB in the problematic 40m band, I had decided that leaving the NB modification in place with the 2M resistor was worth it giving the best NB effect despite the IMD side effect.

At least that way I could obtain its full benefit during the day in 40m.

No ill effect was noticed in any other band irrespective of propagation.

Upgrading the RX mixer

The NB project takes an unsuspected turn!

Next project was the replacement of the 1N4148 diodes of the Rx mixer with hot – carrier ones. I used the 1N5711 type and the improvement in Rx performance in terms of lowering the Rx noise floor (which was excellent to begin with) and weak signal recovery.

I finished this modification late last night, and satisfied went to sleep...

The following day though, when listening to the local nets, I noticed yet another issue. With the NB activated, while the spark type noise was completely eliminated, the stronger signals had very noticeable distortion.

This had never been an issue before changing the mixer diodes!

Obviously, the higher effectiveness of the 1N5711 diodes had increased the level of recovered signals in the IF path, worsening the strong signal distortion produced by the modified NB.

The next step was to see if I could remedy the situation by increasing the mod resistance back to 3M3 as described above.

The 2M resistor was unsoldered, and I turned the radio on, to start experimenting with different resistance values again.

To my pleasant surprise, the NB now worked marvelously even without any additional resistor at the mod place!

Now that was another moment of amazement!

Over the next couple of days, I had the opportunity of evaluating the performance of the radio under different propagation conditions.

On the NB compartment: It has been transformed from an almost useless feature to one of the best NBs I have ever used in any rig for the past 30 years. Excellent effectiveness, no 3^rd order IMD products induced even amid super strong signals and minimal if any induced distortion of S9 + 40db received signals

From the RX performance perspective: Before upgrading the mixer diodes, the Omni was one of the most pleasant low noise receivers to listen to. Comparable to my FT-901DM. After the diode change, it became so much better, in terms of recovering weak signals buried in background noise.

Crystal clear unmuffled signals just popping out of the band noise!

This probably is the result of the lower noise figure and much lower conversion losses of the hot carrier 1N5711 vs the conventional 1N4148 diodes.

Finally, after going round and round trying to solve this pesky NB problem with the best possible compromise, the perfect solution came from a different source...the improvement of the RX mixer.

Talking about killing 2 birds with one stone!

Now, this whole adventure of a radio modification with its different turns, the thought process, the experimentation and real-world hands-on electronics project, especially for a non-professional in electronics like me has been the most satisfying and fun project!

I am sharing all of its different steps and details, just as an example of how much satisfaction one can extract from the process of tinkering with radio (or any other life process), which far exceeds any experience of using a readymade product, or even implementing an already tried and proved ready modification or solution.

Sadly, with the increased complexity and miniaturization of present-day equipment, and with digital technology substituting analog designs. The average ham radio enthusiast has been reduced to a simple appliance operator – consumer.

And with widespread implementation of the mobile phone and internet technology, anyone can enjoy trouble (and noise) free worldwide communication, reducing the novelty of communicating per-se as an operator only, close to nil.

As soon as the new toy excitement wears out from a newly acquired "state of the art" radio, we rush to buy the next model, full of new and largely useless features, chasing a meaningful and lasting satisfaction which is sadly not to be found in the above process.

This is simply because lasting satisfaction can only be obtained by the mental process of creating something new and enjoying the fruits of your own work.

Robbing the hands-on electronic experiment from our hobby is slowly sucking the life out of it in my view. We need to put some market pressure on the different manufacturers of ham radio gear to bring back kits of analog circuit design.

Of course, we can still experiment with antenna design and installations, but due to space restrictions the process is usually done only once in a blue moon, and we end up leaving the same setup for years at a time.

Elecraft did a credible job with its kit series, but we need more…

73,

Marinos, sv9dru / ki4gin