I will be using a simple 5 digit frequency counter - 1Hz to 50MHz.. It uses a PIC using the built-in crystal oscillator with a trimmer to adjust the frequency.
What can I use as a frequency standard to adjust this to - free or very cheap?

 

The tried and true method is to tune to WWV.

The ARRL had a simple method employing a simple shortwave receiver.

 

What is the frequency of the crystal. Where are you located. I am close to a WWV transmitter so comparing my 10mhz to a world standard is easy with a radio.

 

The crystal is 20MHz.
I am in the UK so I doubt that WWV is an option.

 

There should be an European equivalent to the WWV.
Otherwise you would have to find a GPS-disciplined oscillator. But a used one will cost at a bare minimum $100, maybe more.

 

Hello,

There seems to be a standard frequency at 60 kHz in the uk.
See the attached pdf for more info.
And this wiki link:
https://en.m.wikipedia.org/wiki/Time_from_NPL_(MSF)

Bertus

 

The NPL signal is OOK modulated though so how do I fill in the gaps in the carrier? PLL?
How about this, cheap and ready built, OCXO Module
I see it has a pot labelled 'freq cal'. Does that mean that I would need to adjust this first?

 

Unfortunately, it doesn’t list neither initial accuracy nor the temperature/aging drift.
It could have exceedingly good specs, but there are no guarantees.
Caveat Emptor.

Having said this much, how important is the accuracy requirement to you? And….What is your budget? Because if push comes to shove, you could always send it to a calibration lab, one who has a traceable reference.

 

The WWvb kits about $7 USD
An RTL SDR dongle is good $20
Make a crystal oscillator and ask someone with a good frequency counter to write down exact frequency. $2
A good metrologist just compares an accurate number. From there you can get fancy.

 

Having said this much, how important is the accuracy requirement to you?

The counter will have 5 digits so one digit accuracy would be 10ppm. There would be no point in more accuracy than that.

 

I would not be surprised if you could receive WWV or CHU (Dominion Observatory - Canada) in the UK if conditions are right. We should be at or near the Solar Maximum.

 

In the legacy days of analog NTSC broadcast TV distributed primarily over terrestrial analog microwave and perhaps some dwindling amounts of analog coaxial cable facilities (but not via satellites which could introduce Doppler shifts), and prior to TV stations routinely using digital frame synchronizers clocked to local "close enough" crystals, then the 3.579545 MHz color subcarrier in shows originated by the major national US networks was indeed usable as a readily obtainable reference which itself was derived from Rubidium standards at the networks that were maintained to be ultimately traceable to a major national standard. While this had some operational advantages, it was also secondarily intended for use in disseminating an accurate frequency standard which could be easily recovered in end-user receivers on a wide-scale. Alas those days are in the rear view mirror. I don't follow broadcast engineering much these days, but perhaps something vaguely similar might be remotely conceivable using MPEG PCR, PTP, ATSC NextGen intrinsic data rates, GPS, etc.... lots of uninformed ifs there but it sure is fun for me to think about again. Meanwhile, as I believe someone else mentioned, radio amateurs have done & are no doubt still doing some nice work in this area; maybe Googling the TAPR group and/or ARRL FMT (Frequency Measurement Test) etc. might offer some good insights.

 

This article for a GPS frequency standard (found on Hack-A-Day) may be helpful.

 

@AlbertHall the module you showed is a good option, (see: this, which is the basis for it) but so is a cheap GPS receiver as mentioned by @nsaspook. I have one of those modules and it is really quite good, but NEO 6M GPS receivers are even cheaper now, so either way.

 

@AlbertHall the module you showed is a good option, (see: this, which is the basis for it) but so is a cheap GPS receiver as mentioned by @nsaspook. I have one of those modules and it is really quite good, but NEO 6M GPS receivers are even cheaper now, so either way.

I have ordered that module. Simple, cheap (£11.43) so I will see how it goes.

 

Power up, no lock.



GPS lock and counter warmup.




Sine Wave 1 Vrms output. Still need to design a Master output buffer for 3.3V, 5V with several 50 ohm outputs. The RS232 (DB9) connector provides RS232 NMEA GPS data.
https://hirokun.jp/av/bg7tbl_lcd-gpsdo_man1.pdf

The HackRF SDR has a 10MHz external standard input but it's 3.3V TTL.

 

After a few days the freq counters internal standard has drifted.

The internal OSC check function still reads dead on 10MHz.

 

A junk-box design for a 10MHz signal buffer and TTL 3.3 output from the GPS standard.

Have a few old 300MHz BB3553AM devices that should work for the analog 50 ohm out buffers and a CD74HC132E should be OK (a little slow at 3.3V) as the sine to square and digital buffer.

https://www.alldatasheet.com/datasheet-pdf/pdf/72716/MAXIM/BB3553AM.html

https://github.com/nsaspook/gpsbuf

 

Using a video distribution amp to buffer the 10 MHz analog signal. Works OK into a 50 ohm load but the waveform is a little distorted from what looks like too much input signal. Will open the amp up to see if there's a
input adjustment level or I'll add a attenuator on the input.

 

Bingo!