Welcome to the new FlexRadio Community! Please review the new Community Rules and other important new Community information on the Message Board.
If you are having a problem, please refer to the product documentation or check the Help Center for known solutions.
Need technical support from FlexRadio? It's as simple as Creating a HelpDesk ticket.

Inexpensive GPS reference???

Dave - WB5NHL
Dave - WB5NHL Member ✭✭
edited June 2020 in New Ideas
Just saw this device on the web. Looks like it should provide a GPS-based reference for a Flex?? http://www.force12inc.com/products/gps-locked-precision-frequency-reference-low-jitter-gps-clock-450...

Output drive levels can be adjusted from +7-dBm to +13.3-dBm  10MHz

Looks interesting.

Dave,  WB5NHL


Comments

  • Tim - W4TME
    Tim - W4TME Administrator, FlexRadio Employee admin
    edited March 2017
    It isn't the jitter or accuracy that you should be concerned about, it the phase noise you have to be mindful of.
  • KD4HSO
    KD4HSO Member
    edited February 2017
    Jitter is the integral of phase noise.  Jitter should be a concern since ADC clock jitter determines the noise floor, especially at VHF, but I assume Flex is using the 10 MHz input to discipline the TCXO/OCXO via a narrow loop bandwidth?  At least I hope so; that's where schematics would be helpful to see what oscillator and clock synth you are using.  What's the jitter of the 6500 vs 6700 since the latter is an OCXO?  The phase noise/jitter far outside that loop bandwidth won't affect system clock jitter, so stability and accuracy is the primary concern, hence the GPS.

    That phase noise is similar to my Z3805A at 10 kHz.....but they ought to show the phase noise below 10 Hz.  That's where TCXO begin to **** out and OCXO are much better, at least 20-30 dB lower.  For $260 you can get a Z3805A or equivalent on eBay with a double oven OCXO. At 100 Hz the (linked above) unit has -120 dBc/Hz whereas the Z3805A is -135 dBc/Hz.

    This the the same Chinese vendor I bought mine from (got it in two days).  A suitable supply can be bought from Mouser for $30.

    http://www.ebay.com/itm/HP-Z3805A-GPS-Frequency-Time-Receiver-10-Mhz-1PPS-/271938525721?hash=item3f5...
  • KD4HSO
    KD4HSO Member
    edited February 2017
    N5AC mentioned in another thread that the loop bandwidth for the eternal 10 MHz reference circuit was 10 Hz, so phase noise under 10 Hz offset will be that off the external reference. This is where the OCXO based GPSDO are much better than the TCXO based GPSDO.  So probably better getting a good, used unit with OCXO.

    Ha ha, now look at this.  The Chinese acquired a bunch of surplus OCXO and are building GPSDO around them for $150.  Notice they don't give phase noise specs:

    http://www.ebay.com/itm/10MHZ-OUTPUT-SINE-WAVE-GPS-DISCiPLINED-CLOCK-GPSDO-GPS-Antenna-Power-supply-...

  • Simon Lewis
    Simon Lewis Member ✭✭
    edited July 2017

    I lived in western Germany (JO31) and that's one of the busiest areas of ham populous in the world.

    Huge HF signals from contesters and clubs and massive VHF signals from across EU.

    I used a variety of 10 MHz sources driving HF radios and transverters for VHF-SHF.

    What I found was that anything like the DEMI LO boards with PN in the 70-80 dBc/Hz range you can throw away in high signal situations - outcome .. multiple rx images, terrible CW signals with non T5 notes, poor rx sensitivity and awful tx performance even further out from your carrier.

    I swopped to a Kuhne TR144 transverter and LO's using Kuhne PLL's that have close in performance in excess of 150 dBc/Hz and no more problems, especially using radios with +40 IP's and the TR144 front ends.

    My results show that a half decent GPSDO that has a PN over 110 dBc/Hz will be fine ... and that anything 120 upwards you can sleep well and your neighbours will be fine. Close in you're unlikely to be any issue on the band unless you are a KHz away from a contester using multiple KW's and he's more likely to be a problem if he's using an older radio too - older icom kenwood yaesu radios have terrible phase noise!

    my GPSDO has

    @10 Hz - 105 dBc
    @100 Hz - 125 dBc
    @1 kHz - 140 dBc

    Steer clear of non spec'd PN figures but you will find it nigh on impossible to tell in the real world of the GPSDO has PN of 120 or 150!

    My findings based on real world results ... not lab spec'd measurements!

    Cheers
    Simon ZL4PLM

  • Wim
    Wim Member ✭✭
    edited February 2017
    I just saw another GPS disciplined TCXO from a hackaday project here
    https://hackaday.io/project/6872-gps-disciplined-tcxo

    He has it available here
    https://www.tindie.com/products/nsayer/gps-disciplined-ocxo/

    He shows some pahese noise plots also for it


  • Peter K1PGV
    Peter K1PGV Member ✭✭✭
    edited June 2020
    >I just saw another GPS disciplined TCXO from a hackaday project here

    Wim... I don't see his phase noise plots.  Can you point me to them, please (sorry... yes, I did look at both links).

    I'm curious to compare the phase noise in his $250 TXCO-based unit with that of my Jackson Labs Fury.

    Thanks,

    Peter
    K1PGV
  • Wim
    Wim Member ✭✭
    edited February 2017
    I think I took the Allan plots for the phase noise plots, but isn't the phase noise
    just determined by the DOT050V tcxo he's using

    datasheet is here with phase noise details
    http://www.conwin.com/datasheets/tx/tx395.pdf

  • Peter K1PGV
    Peter K1PGV Member ✭✭✭
    edited June 2020
    > isn't the phase noise just determined by the DOT050V tcxo he's using

    Thanks Wim. I guess we know that the phase noise won't be better than that of the device he's using, that's for sure ;-)

    Looks like a pretty steep curve. I see -100dBc at 10Hz and -130dBc at 100Hz,and at 1K the chip's down near -150dBc... that all seems pretty good for such an inexpensive device.

    Simon (ZL4PLM)'s post seems to indicate that this level of performance is probably good enough.  What do folks think?  I'm trying to educate myself a bit more on this whole topic.

    Peter
    K1PGV
  • Simon Lewis
    Simon Lewis Member ✭✭
    edited October 2015
    -150 at 1 KHz is more than adequate ...

    :)
  • Peter K1PGV
    Peter K1PGV Member ✭✭✭
    edited December 2016
    One follow-up if I may: In terms of receiver performance, do the close-in noise figures (such as -100dBc at 10Hz) not matter? Sorry to be ignorant (in the true sense of the word), but I would think that if the Phase Noise is greater than the MDS this would affect receiver sensitivity. Or am totally off-base here? Thanks for helping to school me on this (I'm a pretty good software engineer, but not much of an RF EE).
  • Simon Lewis
    Simon Lewis Member ✭✭
    edited October 2015
    @ -100 at 10 Hz probably not - but -70 @ 1 KHz for sure!

    and yes poor LO phase noise also affects rx performance too .. it affects MDS, IP etc ... on transverters with really poor phase noise in the LO you can end up with images and rough sounding recovered signals and poor rx performance ... so all round its a bad deal :) 

Leave a Comment

Rich Text Editor. To edit a paragraph's style, hit tab to get to the paragraph menu. From there you will be able to pick one style. Nothing defaults to paragraph. An inline formatting menu will show up when you select text. Hit tab to get into that menu. Some elements, such as rich link embeds, images, loading indicators, and error messages may get inserted into the editor. You may navigate to these using the arrow keys inside of the editor and delete them with the delete or backspace key.