2-Band Tx

[NZ0I]

I've spent a little time searching for a suitable component to serve as the VHF driver stage in the 2m transmitter. My preference would be to use a high-speed op-amp, perhaps this one:

http://www.digikey.com/products/en?keywords=296-11641-1-ND

The beauty of using an op-amp is the amount of control it would provide over the gain of the driver. Not only should it be straightforward to scale down the CW output power with an op-amp in the driver stage, it would also provide a low-power/efficient way to AM modulate the signal. If I recall correctly, 100% modulating the final output transistor causes about 50% of the transmitter's final stage power to be consumed in the device controlling the collector current. So by doing the modulation in the driver, you can save ~1W in a 1W transmitter.

But, wasting a Watt (or even two) isn't a huge deal, provided we're not needing to conserve battery. And the project's priority is to have a CW transmitter, and the AM option is a convenience for those wanting to use Ancient Modulation. So perhaps we shouldn't care much about the lowered battery life (or -3dB power reduction) for AM. With that reasoning, the following device might be a suitable choice for an unmodulated driver:

http://www.digikey.com/products/en?keywords=599-1009-1-ND

The SBF-5089Z looks robust enough to safely drive the antenna directly (with ~100 mW) in low-power mode... CW only. AM would necessarily require using the final amplifier since that is where the modulation would occur.

Please let me know if you have any comments on this subject.

[NZ0I]

I did another quick search after posting this thread, and discovered the following document: http://www.ti.com/lit/an/slyt102/slyt102.pdf

The reasoning in that document makes me lean more toward an op-amp design, especially since that would support low-power AM and CW on VHF. It also makes me think that an op-amp is a no-brainer for the 80m CW-only driver stage. It also makes me wonder if maybe a design might be possible whereby the 2m driver and final amplifier are shared with the 80m transmitter, cutting cost.

I've been imagining use cases for having a transmitter transmitting on two bands simultaneously. You could, for instance, place five transmitters in the field (plus a homing beacon) all transmitting on two bands at the same time. For practice events, hunters could choose which band they want to hunt on. The transmitters could ID differently on each band, so that one transmitter is MOE on 80m and MO4 on 2m, etc. But, support for simultaneous 2-band transmission would not be possible if the 2m and 80m driver and final are shared.

By sharing more components, the total transmitter cost would be reduced. Also, AM mode should be supported on both 2m and 80m... but the user would have to take care to use AM only in the band segment authorized for its use... where most European ARDF receivers do not operate. So 80m AM probably isn't much of a bonus. So I guess it just comes down to cost, and overall size of the transmitter.

If we don't support having a transmitter on two bands simultaneously, then maybe there isn't much reason to use the Si5351. We won't need multiple clocks, but simply one that can be tuned to 80m and 2m. There might be other, more suitable (cheaper?) clock solutions for a one-band-at-a-time design.

[Gerald Boyd]

This is an interesting application note. Just printed it to do a good look over.

I also support the idea of transmitting on two bands at once. That would make a good homing beacon for events where both bands are in use such as world championships. Also would be good for practice sessions where the hunter could hunt hf of vhf giving more training options.

The fpga transmitter had a simocast mode for 2 and 80.

Jerry

Sent from my iPhone

--
You received this message because you are subscribed to the Google Groups "Receiver Development Platform" group.
To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/2e261109-70cf-4f4e-8132-6e32a2e78139%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Patrick R. Sears]

Hi Charles,

Interesting ideas.  I looked over slyt102.pdf this weekend.  Also, I already had the datasheet for the op-amp (you had mentioned it in an earlier e-mail, different digikey number but basically same part).  I looked at it again.  I haven't decided how to proceed yet on the 2m tx but I'm thinking I may experiment with that op-amp.  I didn't get to 2m work this weekend -- other duties.  But hopefully this coming weekend.  Also, I'm definitely not tied to the Si5351.  In my mind, the Si5351 is really handy if you need to tune across frequencies since the microcontroller can use i2c to control it.  Great for a receiver.  For an ARDF transmitter that only needs a few frequencies from which to choose, there are many more options.

Thanks for the info.

Cheers,

Patrick

--

[Charles Scharlau]

Sounds good, Patrick. But to be clear on the Si5351, I think having the ability to choose an arbitrary frequency, and more than one band, are really good ideas. There are occasions (more frequent in urban areas) where it becomes necessary to move transmitters to a different frequency on short notice, and being able to do so nimbly is a very good thing. Also, the technology for generating an accurate and stable signal on any HF-VHF frequency is not very expensive or difficult to find anymore. 

I also find Jerry's use cases for simultaneous dual-band support to be compelling: homing beacon, and dual-band practice. If a cheaper 2-channel version of the Si5351 (or its equivalent) existed, then that would be a better choice. But the Si5351 isn't a bad fit. Maybe we should support ARDF on a third band. ;-)

One caveat: If you use any synthesized chip to generate a VHF signal from an HF crystal, you will most likely need a TCXO like the Abracon ASTX-H11. A crystal or MEMS oscillator will not be stable enough over temperature (unless you take measures to stabilize its temp), and that drift will get multiplied in the synthesis process, and you will not get a clean-sounding CW Tx signal - though it might sound good enough on AM. The seat-of-the-pants testing of the Fox TCXO and MEMS device (which has stability numbers similar to a precision crystal) seemed to demonstrate that pretty conclusively.

When I had the QSD sampling detector running with the Si5351 I listened to Nadia's Russian 2m AM transmitters (crystal controlled) in CW mode for the first time. I discovered that the Russian design has an annoying chirp at key down of each Morse element. The carrier is getting pulled several hundred Hertz off frequency due to transient effects - poor stability in the oscillator. But you would never know it by listening to the signal with an AM detector. In AM mode you hear only the stable modulation, and no carrier, and no chirp. CW requires better stability, but fortunately that stability is now much easier to achieve than it was back in the 1990's.

Once I've got the OSH Park receiver boards ordered, I'll start fleshing out a dual-band Tx in the shared schematic. You are welcome to reuse any of its design if it suits your purposes. And if you have any suggestions for improving the design, your input is always welcome. As you've seen with the receiver, my schematics evolve and change - sometimes rapidly - so some discussion might be in order regarding the degree of confidence to place in any portion of the design you are considering to reuse. 

To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development-platform+unsub...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/2e261109-70cf-4f4e-8132-6e32a2e78139%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Receiver Development Platform" group.

To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development-platform+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/fb2daba1-8a72-6c16-a676-a4840cd848ad%40gmail.com.

[Patrick R. Sears]

Thanks for the info Charles.  I will study all this.

I did order a TCXO on Dec-13.  The notes I have with me mention it was in response to emails I read in June but I'm not at my main desk so I can't look up what my plans were.  I do remember that I grabbed a chip oscillator that Jerry had used and I think I tested it with a regular crystal instead of the TCXO; I can't remember why.  Anyway, I will definitely go the TCXO route.

Cheers,

Patrick

[Charles Scharlau]

Thinking a little bit about the 3rd ARDF band support. Maybe the transmitter could telemeter data on a 3rd frequency. I believe Jerry's SDR transmitters supported a UHF control channel, was it for remotely turning the transmitters off/on? In these simpler transmitters, the third frequency might be used to send out slow data on an HF or VHF frequency - perhaps sharing the ARDF antenna. If the transmitter were in communication with a timing box (e.g., NFC time register) then it could send back split times for competitors. This might be done using FSK, or some other very simple, slow, robust mode. I'm not proposing adding a receiver to the transmitters; the transmitters would simply send the information automatically. It might be done simultaneously with the ARDF broadcast(s) so that competitors couldn't gain any advantage by listening to the data channel.

I'm not sure how good an idea that is. But the extra Si5351 output channel is available... but would have to be used for HF (not VHF) if the transmitter is transmitting simultaneously on 2m, since the Si5351 doesn't support two different frequencies above 115MHz, as I recall.

[Gerald Boyd]

The uhf chip was going to be used to configure the transmitter. It would only have about 100 meters range.

I wonder if we replaced the CW ID with a FSK data burst if that would count for identification of the transmitter. Then maybe we could send the ID and some data in the last few seconds of the cycle.

Data could be ID,battery voltage ,panic switch state and punches made during the last 5 minutes.

Jerry

Sent from my iPhone

To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/CAK-GiwYwZmgQXOZi8vVytE9yccsWH6oKsyA15heN%3DYTJPTV7hA%40mail.gmail.com.

[Patrick R. Sears]

Hi Charles,

Last night, I finally put an order through for some new components.  Including some ASTX TCXOs and OPA355 op-amp.  They won't be here by this weekend, but next weekend should be much fun.

I found that one of the ASTX sizes matches exactly the size of the crystal in the 80m ARDF transmitters we're currently using.  I have all the parts at home to start building version 5 but haven't started yet.  So I've got a ready made platform for experimenting with the new TCXOs and comparing them with the old parts  :  )   I could even take out all the old crystals and drop in the TCXOs if we decide it's worth it.

Cheers,

Patrick

[Charles Scharlau]

Sounds like fun. Good luck with the experimenting. Keep in mind that the tcxo connection to the Si5351 is a little different from using a crystal. If your version 5 boards are designed to use a crystal, you'll need to cut the trace leading to Si5351 pin 3 as shown:

​

[Patrick R. Sears]

Ah!  Thanks.  This will save me time.

I'll check out what I need to do when I get home.

Cheers,

Patrick