Stable LNB Oscillator solution

Digital ATV - The latest generation, cutting edge ATV - Please discuss it all here.
Forum rules
This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.

Thank you
Post Reply
radiogareth
Posts: 1371
Joined: Wed Jan 06, 2016 9:46 am

Stable LNB Oscillator solution

Post by radiogareth » Thu Nov 12, 2020 4:58 pm

This is another of my 'great gods of surplus giving again' posts, plus a bit of inspiration from Martin, G4FKK.
It started with a purchase of '3.5GHz WiMax antenna and circuitry, a pair' from Mainline Electronics - always worth a browse. I took a chance and bought a pair. On dismantling I found the 16 Element flat-plate was an acceptable match at 3.4GHz (no precise test gear here, so buyer beware....). By adding a suitable SMA a handy portable solution was available. Closer inspection of the control and RF circuitry showed some interesting bits - many far too small for my SMD skills but including a good spec 10MHz TXCO - data found on the www. I unsoldered it, powered it up (3.3V) and listened to it on 10MHz. Very solid. As it has a 'trim' option I added a 10K pot and gave it a tweak. +/- 10kHz so possible to control in a locking setup perhaps? This thought train was triggered by Martin, G4FKK telling me about his GPS disciplined 10MHz TXCO and what he used it for.
Basically he locked a TXCO (10MHz) to a GPS and then used that signal to clock a SI5351 DDS board (under £3 each on eBay, remove the 25MHz crystal). Despite the data sheet saying it was too low a frequency, it works reliably. Common sense says stuff will work below their design frequency, fair enough not much above it although this chip does both! It was then a simple task (for Martin, not me) to write a little Arduino Nano code to send the desired setup/frequencies. I modified the LNB as per the instructions here https://wiki.batc.org.uk/Es%27hail-2_LNBs_and_Antennaes and fed the signal up a suitable coax. It works and the beacon is as solid as a rock. I have wrapped it in some sponge and once it's all reached equilibrium (about an hour), it wanders very little - only kHz on the beacon from cold. It certainly is an acceptable solution with scope for GPS locking - additional investigation is planned.....
If you fancy the same thing, search eBay for "Wifi max Aerial 3.5GHZ transmitter and Receiver PRICE for 2 Bargain"
You get a pair, so two TXCO, 2 panel antennas and lots of SM, including what looks like a pair of low power 3.5GHz transmit chains. The chips are all identifiable and mostly by RFMagic.
Gareth

G4FKK
Posts: 158
Joined: Sun May 05, 2019 12:15 pm

Re: Stable LNB Oscillator solution

Post by G4FKK » Thu Nov 12, 2020 6:40 pm

Just for clarity; I've GPS locked an, already very stable, 10MHz OCXO which is relatively easy to do with a low bandwidth loop keeping everything jitter free and very low phase noise. I've also managed to lock up a Colpitts VCXO with no temperature control but it's a lot harder to keep it in lock when the Wife opens the windows :) I'm going to try one of Gareth's TCXOs which, as it's already pretty good, should be possible to keep locked and clean.

As Gareth mentioned, I've clocked an Si5351 with the locked 10MHz and it's possible to generate clean 25 and 40MHz signals for the LNB and Pluto respectively. There is, however, crosstalk between outputs. For general SSB use on the NB portion of QO100 it works very well and seems excellent receiving DATV too - yet to try it on TX. As the Si5351 boards are so cheap I'd probably use separate units for each frequency.

73, Martin - G4FKK

Post Reply

Return to “DATV - Digital ATV”