Тёмный

Developing a GPSDO from scratch - part 1 

All Electronics Channel
Подписаться 16 тыс.
Просмотров 5 тыс.
50% 1

This is the first part of a series about the development of a GPSDO from scratch.
GPS Disciplined Oscillators have become a cheap way to generate high precision and high stability frequency references on the lab.
The GPS system trilaterates the global position using measures of time difference from the satellites signals. The base of these time measures are high stability frequency oscillators that need to be locked with a global and unique standard.
Even small and cheap hobbyist GPS receivers output a 1 PPS signal that can be used to phase lock a local reference, improving the performance of crystal oscillators.
As the 1 PPS signal is 10 million times slower than the 10Mhz desired output frequency, ovenized crystal oscillators are used to improve short time stability.
Support the channel becoming a Patron
/ allelectronics

Наука

Опубликовано:

 

21 июн 2020

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 22   
@AllElectronicsChannel
@AllElectronicsChannel Год назад
Support the channel becoming a Patron patreon.com/allelectronics
@PowderMill
@PowderMill 3 года назад
Great video! I am in the process of specifying hardware for a public safety simulcast paging transmitter network. Motorola charges over $12900 for their rubidium reference solution and the GPS one is slightly MORE $ Time to build it 😆😷👍🏻🇺🇸
@mattooi4322
@mattooi4322 3 года назад
Love your passion!
@m1geo
@m1geo 3 года назад
Nice project. Subscribed! :)
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
Welcome!! :)
@galileo_rs
@galileo_rs 3 года назад
Ublox module that outputs a 24MHz "PPS" connected to a si5351 will give quite a stable reference but the phase noise is not the best. Alternatively use the MHz range pps output to feed the PLL. The ublox output is jittery when outputting frequencies that are not an integer division of internal clock (48MHz) so chose you frequency accordingly. Lower clock output, better phase noise.
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
Yep.. I'm using the neo6 with 1 PPS output signal. After first lock I switch to a very low freq digital low pass before the phase comparator. I changed a bit the design to use simpler components. FPGAs are no available to everyone. You can see more videos about the project on the channel ! Thanks
@galileo_rs
@galileo_rs 3 года назад
@@AllElectronicsChannel I skimped thru all of them. What is your goal, just stability or do you want to use it as a reference oscillator (low phase noise)?
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
My current goal in experimentation and learning hehehehe.. I got really impressed by the phase noise of the homemade ocxo I designed! But unfortunately I think that the long therm stability is not enough for a gpsdo.. maybe with a properly thermal insulation and metallic case Today I'm characterizing the PWM DAC I made.. the next video will be about it
@galileo_rs
@galileo_rs 3 года назад
@@AllElectronicsChannel A worthy goal in itself. If instability comes from temperature, high thermal mass and insulation should help.
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
Thanks ! I will be posting more about it..
@deepaknarayan9729
@deepaknarayan9729 Год назад
🎉
@FATCAP
@FATCAP 11 месяцев назад
Hi! Thank you for uploading this kind of content I am trying to develop a very similar system and I'm basing my work on the LARS GPSDO (which follows almost the same philosophy as yours). I wonder if you could answer a question: If I wanted to output a Disciplined PPS (not only 10MHz signal), how could I do that? Would you extract that PPS from the DSP or from the OCXO (using several frequency dividers until geting 1Hz) Thank you very much, any kind of help is highly appreciated
@AllElectronicsChannel
@AllElectronicsChannel 11 месяцев назад
Hi! I would divide the high frequency to 1Hz
@pcrengnr1
@pcrengnr1 9 месяцев назад
Thx for pursuing this ultimate frequency standard. I believe the accuracy should be around 10-12 to 10-13. Sorry no superscript. So the control loop is actually two loops. One is an FLL, a frequency locked loop. The other one is a PLL. That gives you the greatest lock in range which is needed. Gregory thx for taking the time in making this video and sharing.
@AllElectronicsChannel
@AllElectronicsChannel 9 месяцев назад
Thanks! This is an old and unfinished project.. hehe. We should get back to it some time soon 😁
@davidtzaig5947
@davidtzaig5947 3 года назад
did anyone replaced the 24mhz or 48mhz crystal thats on the NEO xx with a 20mhz or 40mhz one??? the next stage will be a hcxx buffer and that's it.
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
Why??
@davidtzaig5947
@davidtzaig5947 3 года назад
@@AllElectronicsChannel in order to get a 10Mhz that is clean from jitter directly from the GPS. wouldn't that be already gps disciplined?
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
Hello man! Hum.. the raw output of the NEO have high phase noise (jitter) this is why in GPSDO a very stable and clean local oscillator is slowly disciplined with the GPS reference
@davidtzaig5947
@davidtzaig5947 3 года назад
@@AllElectronicsChannel i didn't see jitter on the oscilloscope when i outputted from the neo7n 12MHZ that is a complete div from the 24MHZ crystal. on 10MHZ i saw jitter. replacing the crystal for 20MHZ may give good results for div by 2 for 10MHZ. just an idea.
@AllElectronicsChannel
@AllElectronicsChannel 3 года назад
Nice idea to give a chance.. for properly jitter measuring you need a spectrum analyzer with good phase noise. I will measure it next time I go to lab
Далее
2DROTS vs RISENHAHA! КУБОК ФИФЕРОВ 2 ТУР
11:31
#1020 Various 10 MHz Oscillators Compared
10:04
Просмотров 11 тыс.
Fundamentals of RF Power Sensors
22:46
Просмотров 6 тыс.
The GPS Disciplined Crystal Oscillator - Part 2
50:27