Low cost 10GHz Transverter


Andy G4JNT
 

5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


John Quarmby
 

Other of their VCO's around that frequency range are coax cavity based but it doesn't say for this one. The tuning sensitivity is 120kHz/V according to the datasheet, the number quoted below is for frequency pushing.

73

John G3XDY

On 24/11/2022 10:50, Andy G4JNT wrote:

5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


Andy G4JNT
 

Ah....
Must be getting old
Misread the data sheet
Still good though
And pricey



On Thu, 24 Nov 2022 at 10:50, Andy Talbot <andy.g4jnt@...> wrote:
5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


Robin Szemeti - G1YFG
 

I'm still busy trying to work out why you would want a 1 GHz reference frequency in the first place ...

On Thu, 24 Nov 2022 at 11:24, Andy G4JNT <andy.g4jnt@...> wrote:
Ah....
Must be getting old
Misread the data sheet
Still good though
And pricey



On Thu, 24 Nov 2022 at 10:50, Andy Talbot <andy.g4jnt@...> wrote:
5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


--
Robin Szemeti - G1YFG


Andy G4JNT
 

It's used, in the OZ beacon hardware, as the clock for a DDS that generates up to perhaps 400MHz directly, and higher via image products and multiplication.
A DDS allows direct straightforward MFSK modulation, and introduces virtually zero added phase noise onto that from the master clock.   So use a very good 1GHz drive and PN goes as 20LOG(Fout / 1GHz) on that.

Beats the hell out of using the internal PLL clock multiplier in a lot of DDS devices.



On Thu, 24 Nov 2022 at 11:32, Robin Szemeti - G1YFG via groups.io <robin=redpoint.org.uk@groups.io> wrote:
I'm still busy trying to work out why you would want a 1 GHz reference frequency in the first place ...

On Thu, 24 Nov 2022 at 11:24, Andy G4JNT <andy.g4jnt@...> wrote:
Ah....
Must be getting old
Misread the data sheet
Still good though
And pricey



On Thu, 24 Nov 2022 at 10:50, Andy Talbot <andy.g4jnt@...> wrote:
5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


--
Robin Szemeti - G1YFG


Robin Szemeti - G1YFG
 

Ah, noted. I hadn't considered the idea of working downwards to generate LF stuff ... I'm usually trying to work upwards ;)


On Thu, 24 Nov 2022 at 11:37, Andy G4JNT <andy.g4jnt@...> wrote:
It's used, in the OZ beacon hardware, as the clock for a DDS that generates up to perhaps 400MHz directly, and higher via image products and multiplication.
A DDS allows direct straightforward MFSK modulation, and introduces virtually zero added phase noise onto that from the master clock.   So use a very good 1GHz drive and PN goes as 20LOG(Fout / 1GHz) on that.

Beats the hell out of using the internal PLL clock multiplier in a lot of DDS devices.



On Thu, 24 Nov 2022 at 11:32, Robin Szemeti - G1YFG via groups.io <robin=redpoint.org.uk@groups.io> wrote:
I'm still busy trying to work out why you would want a 1 GHz reference frequency in the first place ...

On Thu, 24 Nov 2022 at 11:24, Andy G4JNT <andy.g4jnt@...> wrote:
Ah....
Must be getting old
Misread the data sheet
Still good though
And pricey



On Thu, 24 Nov 2022 at 10:50, Andy Talbot <andy.g4jnt@...> wrote:
5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


--
Robin Szemeti - G1YFG


--
Robin Szemeti - G1YFG


John Quarmby
 

The 13cm beacon at GB3MHZ generates 320.830MHz using the OZ beacon system DDS and mixes it with a doubled version of the 1GHz clock to generate a low phase noise output at 2320.830MHz. Its not a free lunch however, there a few discrete spurii apparent mostly down at -80dBc or lower, one or two of which appear in the narrow band segment around 2320.200.

73

John G3XDY

On 24/11/2022 11:39, Robin Szemeti - G1YFG via groups.io wrote:

Ah, noted. I hadn't considered the idea of working downwards to generate LF stuff ... I'm usually trying to work upwards ;)


On Thu, 24 Nov 2022 at 11:37, Andy G4JNT <andy.g4jnt@...> wrote:
It's used, in the OZ beacon hardware, as the clock for a DDS that generates up to perhaps 400MHz directly, and higher via image products and multiplication.
A DDS allows direct straightforward MFSK modulation, and introduces virtually zero added phase noise onto that from the master clock.   So use a very good 1GHz drive and PN goes as 20LOG(Fout / 1GHz) on that.

Beats the hell out of using the internal PLL clock multiplier in a lot of DDS devices.



On Thu, 24 Nov 2022 at 11:32, Robin Szemeti - G1YFG via groups.io <robin=redpoint.org.uk@groups.io> wrote:
I'm still busy trying to work out why you would want a 1 GHz reference frequency in the first place ...

On Thu, 24 Nov 2022 at 11:24, Andy G4JNT <andy.g4jnt@...> wrote:
Ah....
Must be getting old
Misread the data sheet
Still good though
And pricey



On Thu, 24 Nov 2022 at 10:50, Andy Talbot <andy.g4jnt@...> wrote:
5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


--
Robin Szemeti - G1YFG

--
Robin Szemeti - G1YFG


Andy G4JNT
 

-80dBc spurii - LUXURY !

On Bell Hill I considered -60dBc good.  And more to the point, so did G3VPF operating 20km away LOS, beaming through the beacons to get to the Dx.

That uses a PLL source, so phase noise is not brilliant, but discrete spurii are few.
Phase noise is not harmful, except when very strong signals coexist with weak ones.   PN has to be worse than terrible before it actually affects signal quality using the modes we do for narrow band work.   More significant for the wider band ATVers though.



On Thu, 24 Nov 2022 at 11:55, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

The 13cm beacon at GB3MHZ generates 320.830MHz using the OZ beacon system DDS and mixes it with a doubled version of the 1GHz clock to generate a low phase noise output at 2320.830MHz. Its not a free lunch however, there a few discrete spurii apparent mostly down at -80dBc or lower, one or two of which appear in the narrow band segment around 2320.200.

73

John G3XDY

On 24/11/2022 11:39, Robin Szemeti - G1YFG via groups.io wrote:
Ah, noted. I hadn't considered the idea of working downwards to generate LF stuff ... I'm usually trying to work upwards ;)


On Thu, 24 Nov 2022 at 11:37, Andy G4JNT <andy.g4jnt@...> wrote:
It's used, in the OZ beacon hardware, as the clock for a DDS that generates up to perhaps 400MHz directly, and higher via image products and multiplication.
A DDS allows direct straightforward MFSK modulation, and introduces virtually zero added phase noise onto that from the master clock.   So use a very good 1GHz drive and PN goes as 20LOG(Fout / 1GHz) on that.

Beats the hell out of using the internal PLL clock multiplier in a lot of DDS devices.



On Thu, 24 Nov 2022 at 11:32, Robin Szemeti - G1YFG via groups.io <robin=redpoint.org.uk@groups.io> wrote:
I'm still busy trying to work out why you would want a 1 GHz reference frequency in the first place ...

On Thu, 24 Nov 2022 at 11:24, Andy G4JNT <andy.g4jnt@...> wrote:
Ah....
Must be getting old
Misread the data sheet
Still good though
And pricey



On Thu, 24 Nov 2022 at 10:50, Andy Talbot <andy.g4jnt@...> wrote:
5 to 10kHz/V tuning.   Is that a DRO based unit then?  The data sheet doesn't say.  I can't see an LC based VCO managing to keep that sort of stability and have such a low tuning sensitivity.

Even at > £100 a pop!



On Wed, 23 Nov 2022 at 21:59, John Quarmby via groups.io <g3xdy=btinternet.com@groups.io> wrote:

This is what is used in the OZ7IGY Next Gen beacon platform:

https://www.mouser.co.uk/datasheet/2/94/CVCO55CX-1000-1000-28506.pdf

Not cheap but good PN performance for a 1GHz standard.

73

John G3XDY

On 23/11/2022 21:48, Dominique Dehays wrote:
Ge Robin

I recall a good design by on4iy using a crystek resonator.
If I recall rigth it was in the ghz range and phase noise was rigth
73
Dom


--
Robin Szemeti - G1YFG

--
Robin Szemeti - G1YFG