The VXO

[Zozo]

Thought I'd just leave this here, It's intention is purely for a conversation starter.

VXO2.jpg

[Bton-FM]

What's the harmonic output like on the 'London Rig' RF strip - I've never tried it. I imagine the 6p8 capacitor in parallel with the Toko on the output does a fair bit at attenuating the second harmonic since its a harmonic trap at ~200MHz.

Also, where does the 'London Rig' design originate from? I've seen quite a few different PLLs used with it, but the RF part remains largely the same.

The thing that puts me off about this design is the Toko coils that are expensive (well they are compared to air coils!), but I do see some alternatives to them about.

[Zozo]

I'm not 100% sure on where this RF strip originates, but a guess would point me in the direction of (mentioning no names) maybe one or two engineers from Broadcast warehouse in their early days. I say 99% of this schematic is similar to a design that I've noticed on here and around the pirate groups over the years.

I suppose individuals pick and choose the parts of one design and then incorporate ideas of their own to achieve something they want, or possible they just outright copy it.

I do agree with you on the inductors / TOKOs are a little pricey these days. However I do see them on eBay from time to time ( quickly checked now and there's lots ) but radio rallies used to be the best place to pick up hundreds of them for pennies years ago. They do still pop up at rallies to this day.

I really can't remember the harmonic level of the LPF as its been "possible" 15 years since I had this design on a test bench. However it's a sort of series tuned circuit low pass filter with 4 elements, I bet Albert knows a thing or 2 about this method of LPF.

Anyhow I thought this would be an interesting topic to bring to the forum with one of the questions for debate being "what really dictates the stability of a synthesized phase lock loop FM exciter"

[Polecat]

"what really dictates the stability of a synthesized phase lock loop FM exciter"

Pass

[Bton-FM]

I'm not 100% sure on where this RF strip originates, but a guess would point me in the direction of (mentioning no names) maybe one or two engineers from Broadcast warehouse in their early days. I say 99% of this schematic is similar to a design that I've noticed on here and around the pirate groups over the years.

I might be mixing this up with the broadcast warehouse design that was at frequency, but I was recently told it came from some engineers at RFL a long time ago. Although, someone else said it came from Germany?

I do agree with you on the inductors / TOKOs are a little pricey these days. However I do see them on eBay from time to time ( quickly checked now and there's lots ) but radio rallies used to be the best place to pick up hundreds of them for pennies years ago. They do still pop up at rallies to this day.

There are some Chinese variable inductors that I think might make a good replacement. What puts me off is that every other part of my board I can get for ridiculously cheap but I can’t get the TOKO coils for cheap. It’s probably just my own problem that I am a tight bastard LOL!

I might test them soon since I am about to try some more experimenting on clad board. I doubt I will use a doubler in my next boards but I’d like to try a few different RF strips.

[sinus trouble]

Also, where does the 'London Rig' design originate from? I've seen quite a few different PLLs used with it, but the RF part remains largely the same.

You are quite right Bton!

However there is a fundamental which is not confined to just London Rigs! The Oscillator, Pierce, Clapp, Colpitts and so on.... The rest is just amplifier stages.

[sinus trouble]

"what really dictates the stability of a synthesized phase lock loop FM exciter"

Pass

Haha! You crack me up Polecat!

[Zozo]

What's the harmonic output like on the 'London Rig' RF strip - I've never tried it. I imagine the 6p8 capacitor in parallel with the Toko on the output does a fair bit at attenuating the second harmonic since its a harmonic trap at ~200MHz.

Firstly let me apologies as I've just noted you actually mentioned it as a HTF as apposed to a LPF, and your probably very knowledgeable with the understanding of such filters.

I might be mixing this up with the broadcast warehouse design that was at frequency, but I was recently told it came from some engineers at RFL a long time ago. Although, someone else said it came from Germany?

I think there's probably room for a lot of confusion due to the amount of copying / plagiarism or intellectual theft I guess over the years.

Be interesting to see your next copper clad prototype / test and development stage. That takes me back to me spending sometimes hours playing with one particular circuit on clad until I was happy with it. I don't do anything with FM or solid state circuits much now. I'm being nostalgic with Medium Wave again..

[sinus trouble]

"what really dictates the stability of a synthesized phase lock loop FM exciter"

Well i guess the easy answer is? The PLL is only as accurate as its reference (Crystal) but as we all know? Things aint that simple!

To avoid a PLL struggling to control the situation, The VCO needs to be pretty stable free running in the first place!

From my own experience of crystal oscillators? The VXO you posted would most likely work great!

As Bton mentioned? The harmonics would still be present which is not such a bad thing.

[Bton-FM]

"what really dictates the stability of a synthesized phase lock loop FM exciter"

Well i guess the easy answer is? The PLL is only as accurate as its reference (Crystal) but as we all know? Things aint that simple!

To avoid a PLL struggling to control the situation, The VCO needs to be pretty stable free running in the first place!

From my own experience of crystal oscillators? The VXO you posted would most likely work great!

As Bton mentioned? The harmonics would still be present which is not such a bad thing.

Zozo, when you say frequency stability, do you mean frequency stability or stability as in does the rig sprog or not?

If the latter, then it’s down to a multitude of things including but not limited to: decoupling, the use of ferrites, the biasing of the transistors, the impedance marching networks, the loop filter ect...

[Bton-FM]

@zozo
I think it is both a LPF and a HTF, but also doing the impedance matching for the 2N4427. I think it makes more sense to have the filter and the impedance marching separate so that when you adjust the filter you aren’t also changing the impedance transformation. TBH I don’t really know that much about filters, so feel free to add anything. Compared to a lot of people on this forum (including yourself) I have been doing radio for a lot less time, so no need to apologise - I appreciate any knowledge shared!

I might post a few pics of my development on clad board, but it gets quite messy! The design I’ll probably use is closer to an old Dutch design from the 90s (I think the 90s might have been late 80s??) but with more modern transistors. The amplifier stages will be different though, with a 2N4427 on the end.

Also the PLL is going to be a more modern chip as apposed to the the TSA5511 I was using previously.

[Zozo]

"what really dictates the stability of a synthesized phase lock loop FM exciter"

Well i guess the easy answer is? The PLL is only as accurate as its reference (Crystal) but as we all know? Things aint that simple!

To avoid a PLL struggling to control the situation, The VCO needs to be pretty stable free running in the first place!

From my own experience of crystal oscillators? The VXO you posted would most likely work great!

As Bton mentioned? The harmonics would still be present which is not such a bad thing.

Your absolutely correct Sinus in the points you've made. The reason I like and still use VXO's is because they are very stable and incredibly simply to build. Crystals are still susceptible to temperature swings which can alter their frequency slightly, and even the supporting components like the load capacitor can cause it to change with temperature variations, unless it's a High Stab NPO or Silver Mica etc.

I think the main point I was getting at mostly is the designs used today in pirating community aren't either TCXO or Oven Stable Crystals as used in the synthesizers for broadcast equipment, and If you don't need to have the ability to keep changing frequency all the time with the aid of a PLL then you can have a very simple inexpensive driver which has all the stability of a PLL that's using just a "standard crystal" for it's reference.

Another point which makes me favor VXO's is there's less components needed, No synthesizer IC or supporting controller. You can even put the code writing to one side. One of the only down sides to the VXO was ordering Crystals, But now you can order one very cheaply and they get cheaper as you increase you order quantity.(Established pirates long term investment maybe).

The very first 2mtr amateur transceiver I had was the "Trio 7200G" which had no synthesizer circuit at all, It used and you've probably already guessed it, a VXO, which had multiple crystals in a row which were switched in and out of circuit as you changed channels. Even early CB's had VXO's but they used a clever multiplier to achieve more channels with less crystals.

The VXO may have some use even as a cheap band 1 link, you can simple change out the penultimate stage doubler and modify the final stage. Possibly this could maybe something for a beginner to experiment with as a learning project. Either way I think It has been something worthwhile and interesting to discuss, and it may even continue for a little while longer.

[reverend]

Having played with VXO's as FM modulators some years ago, I could never get enough deviation out of them, and that which I could get was not very linear.

I've seen designs for what is termed a 'Super VXO' (see below) which uses two crystals in parallel. This might be able to provide enough deviation when multiplied up to FM frequencies but again I suspect it would not be particularly linear modulation.

c021451d7973559b3422f0a98d9efd88.gif

At the end of the day, a PLL stabilised VCO is always going to offer better modulation performance and even if you don't need to change frequency regularly is not that difficult to make. I did make some 'fixed frequency' PLLs a few years ago which did away with the programmable divider and replaced it with a fixed one. You just chose a crystal to set the transmit frequency. Really easy to build and rock solid. No need to make a broadband design either which means you can eke a bit more gain out of the various stages.

Rev

[Zozo]

Zozo, when you say frequency stability, do you mean frequency stability or stability as in does the rig sprog or not?

If the latter, then it’s down to a multitude of things including but not limited to: decoupling, the use of ferrites, the biasing of the transistors, the impedance marching networks, the loop filter ect...

The first one of the two you've mentioned. I was referring to the true governor of the frequency stability.

Ultimately we need a VCO for a synthesized driver to work along with a synthesizer IC and supporting controller, but the synthesizer solo relies on one vital component, its "reference frequency".

@zozo
I think it is both a LPF and a HTF, but also doing the impedance matching for the 2N4427. I think it makes more sense to have the filter and the impedance marching separate so that when you adjust the filter you aren’t also changing the impedance transformation. TBH I don’t really know that much about filters, so feel free to add anything. Compared to a lot of people on this forum (including yourself) I have been doing radio for a lot less time, so no need to apologise - I appreciate any knowledge shared!

Yes technically you are correct, its still regarded as a Low Pass Filter being a harmonic trap. The name may not be the same, but the goal is still the same. I'm in no way an expert in any area ( I refer to my signature ) but will defiantly do the job, it's simply just adding one extra capacitor to a Pi Network LPF.

To me a final stage would normally consist of a impedance matching network, then a LPF. This network you be something as simple as an (L Network) which is normally an inductor along with a supporting shunt capacitor.

There are obviously other ways of matching system impedances, like the 1:1, 2:1 and 4:1 transformer matching, using a small bifilar wound toroidal. Anyhow after this it would all then go to next stage being the LPF that's been calculated for the system impedance you require "normally 50ohm".

I'd say that a messy clad prototyping means it will probably work just fine, When I've gone back to tidy them up they suddenly stop working. Be interesting what you choose for the synthesizer.

[Zozo]

Having played with VXO's as FM modulators some years ago, I could never get enough deviation out of them, and that which I could get was not very linear.

I've seen designs for what is termed a 'Super VXO' (see below) which uses two crystals in parallel. This might be able to provide enough deviation when multiplied up to FM frequencies but again I suspect it would not be particularly linear modulation.

c021451d7973559b3422f0a98d9efd88.gif

At the end of the day, a PLL stabilised VCO is always going to offer better modulation performance and even if you don't need to change frequency regularly is not that difficult to make. I did make some 'fixed frequency' PLLs a few years ago which did away with the programmable divider and replaced it with a fixed one. You just chose a crystal to set the transmit frequency. Really easy to build and rock solid. No need to make a broadband design either which means you can eke a bit more gain out of the various stages.

Rev

That's an interesting little schematic. I know what you mean with regards deviation and ultimately the pulling of the Xtals frequency in sympathy with the modulation. However this factor can greatly change from crystals to crystal manufactured to certain specifications. I found that crystals cut to operate with a lower series capacitance gives the greatest deviation.

I think a lot of the experimenting with FM VXO has been done with off the shelf crystals that are available to buy or possibly pulled from electrical equipment that may have ordered crystals to a certain spec that's just not suitable in a FM VXO.

It's defiantly been a thought and conversation starter as intended. It's only because after searching all 33 pages in this forum I see it had never been mentioned, and a Anorak section must have all topics covered I think.

VXO's are the primary oscillator in all my AM Valve gear which I like for its simplicity and stability.

[Zozo]

I've seen designs for what is termed a 'Super VXO' (see below) which uses two crystals in parallel. This might be able to provide enough deviation when multiplied up to FM frequencies but again I suspect it would not be particularly linear modulation.

That wouldn't happen to be Mr. P of YouTube? I saw something he did a while back that's FM'ing a VXO. Strangely he doesn't even use Varicap for the Xtal pulling. But he still managed to get a reasonable wide deviation if I remember right.

[yellowbeard]

For me there are too many cons to a crystal oscillator. If you have been assigned a frequency you should not need to move - nobody ever assigned me a frequency but I been told to move plenty of times, so for me a PLL wins. There are only a handful of readily available crystals in the right frequency range and it'd cost £10-£15 to order a custom one and take a couple of weeks to get it, again gimme a PLL. There is quite a few points that require alignment, finnicky compared to a broadband design, but cleaner and simpler. And then there are the issues with modulation. If this was worth it then it'd be available as an option on commercial transmitters for an extra grand or two. I no like - can you tell?

[Bton-FM]

For me there are too many cons to a crystal oscillator. If you have been assigned a frequency you should not need to move - nobody ever assigned me a frequency but I been told to move plenty of times, so for me a PLL wins. There are only a handful of readily available crystals in the right frequency range and it'd cost £10-£15 to order a custom one and take a couple of weeks to get it, again gimme a PLL. There is quite a few points that require alignment, finnicky compared to a broadband design, but cleaner and simpler. And then there are the issues with modulation. If this was worth it then it'd be available as an option on commercial transmitters for an extra grand or two. I no like - can you tell?

This is exactly my thinking - a PLL is so cheap why wouldn't you include one? You can get TSA5511 for about 20 pence and a microcontroller for a bit more, so we are talking under 50 pence maybe for the synthesizer. It's only a couple more components but the benefits it gives you (that you've just mentioned) mainly the ability to be frequency agile definitely justifies 50 pence it costs.

[Electronically]

For me there are too many cons to a crystal oscillator. If you have been assigned a frequency you should not need to move - nobody ever assigned me a frequency but I been told to move plenty of times, so for me a PLL wins. There are only a handful of readily available crystals in the right frequency range and it'd cost £10-£15 to order a custom one and take a couple of weeks to get it, again gimme a PLL. There is quite a few points that require alignment, finnicky compared to a broadband design, but cleaner and simpler. And then there are the issues with modulation. If this was worth it then it'd be available as an option on commercial transmitters for an extra grand or two. I no like - can you tell?

This is exactly my thinking - a PLL is so cheap why wouldn't you include one? You can get TSA5511 for about 20 pence and a microcontroller for a bit more, so we are talking under 50 pence maybe for the synthesizer. It's only a couple more components but the benefits it gives you (that you've just mentioned) mainly the ability to be frequency agile definitely justifies 50 pence it costs.

Yup it costs 50p its a no brainer for the pll. I like how you both think the way I'm thinking.

Sent from my AMN-LX9 using Tapatalk

[Zozo]

For me there are too many cons to a crystal oscillator. If you have been assigned a frequency you should not need to move - nobody ever assigned me a frequency but I been told to move plenty of times, so for me a PLL wins.

That's a far point if you need to move frequency. However I did say about changing out the doubler stage in a previous post and possibly using it else where.

There are only a handful of readily available crystals in the right frequency range and it'd cost £10-£15 to order a custom one and take a couple of weeks to get it, again gimme a PLL.

Could possible be the right one you need, But waiting for crystals doesn't seem to be a problem to me. I've order quite a few custom Xtals over the last couple of years, but I'm using them on 160 / 80 mtr bands for AM transmission. Which is obviously completely different as we don't modulate the oscillator.

There is quite a few points that require alignment, finnicky compared to a broadband design, but cleaner and simpler. And then there are the issues with modulation.

I think you only have to look some early designs from the 1970's to see how modulation is achieved with good results on a VXO, but yes most of those designs needed more alignment as they used lower frequency crystals with more multipliers.

If this was worth it then it'd be available as an option on commercial transmitters for an extra grand or two. I no like - can you tell?

No this would never be good option for commercial broadcast transmitters today, and especially for FM Mode with the technology we have available now. I'm sure you probably already know that crystal oscillators were a commercially available option, and used for more than 40 years with AM Broadcast Transmitters.

Anyhow It's defiantly been the conversation starter it was intended to be, and I've enjoyed everyone's opinions on VXO's in general, and especially using them for FM mode.