Trimmable Crystal oscillator circuit

[Boncuk]

I am looking for a crystal oscillator to be used as highly accurate clock reference.

Attached is a photo of a board equipped with a component carrying the describtion 5111D. (Red arrow pointing towards the device.)

According to the photo it has a trim capacitor.

Has anybody seen such a device and if so, where could I find a supplier? I guess it is a japanese part, since all components on this board are japanese.

Thanks in advance

Boncuk

 

[ericgibbs]

hi,
You can buy these in various capacity ranges, eg: 3 to 15pF and 5 to 30pf.

Look here:
https://uk.farnell.com/jsp/search/r...1&Ntt=variable+capacitors&Ntx=&isGoback=false

 

[Boncuk]

Hi Eric,

that just wasn't it. I'm looking for a complete oscillator circuit equipped with a variable capacitor for fine tuning, just similar to the one I posted.

In that case the output frequency has to be exactly 12,800,000Hz, not just about 12.8MHz.

Thanks anyway.

Hans

 

[ronsimpson]

How accurate do you need?
I just picked an oscillator at random and it is rated at +/-1.5 parts per million.
I believe what you are looking for is no longer made. I have some in old frequency counters. Look for voltage controlled crystal oscillators. They have a voltage input pin where you input 0 to 5 volts and get +/- 10parts/million change in frequency. Add a pot!

When a crystal is marked with 12.8mhz it may give you 12,800,013. That is 1 part/million off. In the case of a VCXTAL then you can trim that to 12,8,000,000.

 

[Boncuk]

Hi Ron,

1ppm sounds OK to me. If you can give me any tye number of a VCO I'll try that. The reference clock is needed for a 2.4GHz counter with 8 digits. At mains frequency it is supposed not to display 50 +/-1 60Hz, but e.g. 50.7Hz.

If I stick the oscillator circuit in a quartz oven the accuracy will be zero ppm.

Boncuk

 

[Sceadwian]

You'll never get 0pp in the real world. 200 dollar OXCOs listed on Digikey go as low as .01ppm I doubt you'll get close to that from a home built circuit. You're also going to have trouble calibrating it.

 

[ronsimpson]

Go to digikey.com search for oscillator.
Select oscillators, or VCO
Pick 12.8mhz
There are 5,000 oscillators and 150 VCOs (have fun)

I have a frequency counter with a 10mhz xtal that has a “trim” capacitor. I calibrate with a short-wave radio. The 10mhz osc and WWV radio can be heard at the same time. The difference in frequencies can be heard as a tone. When the difference gets below 30hz it is hard to here. Look at the file strength meter or listen for background noise to here below 20hz. I usually get the field strength meter to swing up/down in 10 seconds. 10,000,000.1 the .1 makes the 10 second beat.

Are you making a 2.5ghz 8 digit counter?
2.400,000,0ghz the fastest you can count. Counting 100 times/ second.
240.000,00mhz 10 counts/second
24.000,000mhz 1 count/second
To display 50.7hz you do not need an accurate counter you need to count for 10 seconds.
In 100 seconds you can count 50.71hz.

 

[Rolf]

No Problem ......

You'll never get 0pp in the real world. 200 dollar OXCOs listed on Digikey go as low as .01ppm I doubt you'll get close to that from a home built circuit. You're also going to have trouble calibrating it.

There is no problem these days with "atomic clocks" all around for reference, to calibrate the xtall.
Keeping a log is most important, make notes of dates, time and what direction the trimmer cap was tweaked. Probably every day to start then two times a week and finely once a month. It is going to be impossible to make it as accurate as an old Micronta stop watch I have that only drifted a few seconds the first year I had it. It is now 5 minutes fast but I have not touched it for over 5 years!
If the clock is going to be used in temperature extremes I would recommend buying an oscillator in an oven. Also a single turn trim cap is hard to adjust, so I recommend placing a tubular cap in parallel with it. It is adjusted with a brass screw and is multi turn.
Remember allso that, as the xtall ages it will change frequency. (drift)
You will spend a lot of time trying to get it absolutely right and in end you will wish you had bought an atomic clock!

Edit: Here is some xtall info: https://www.icmfg.com/index.html

 

[Boncuk]

Hi Ron,

I don't need 5.000 reference oscillators.

Just one or two will do.

No! I don't want to count 2.5GHz. I'll have to stay well below at 1.3GHz. The prescaler I've found can't do more.

Thanks for the good advise though concerning adjustments. Just the same as a BFO.

I found a device which I hope it will be stable enough at **broken link removed** (a subdivision of China Aerospace Company)

The datasheet shows fantastic values: Initial accuracy: +/- 0ppm, trimmable +/- 5ppm, temperature drift +/- 1ppm/K

This is what it looks like.

Regards

Hans

 

[on1aag]

Hi Hans,

I bought some but mine are also voltage controlled.
I found them at the annual radio fleamarket of the
VERON at Rosmalen earlier this year. They were really
cheap, I think I paid € 2,5 a piece. I bought them from:

http://www.radioaktief.nl/arp/

But if you need the regular TCXO's, you can order some
from these distributors.

**broken link removed**

**broken link removed**

Barend Hendriksen is a real crook but he will get the
stuff you need within a week, any stuff !!

The italian is ok I guess, I only bought a couple of
things from him on several occasions when I saw him
at Friedrichshafen.

on1aag.

 

[Boncuk]

Hi on1aag,

the samples you posted are exactly what I need: 12.8MHz

BTW, I found out that the chinese type VTA01 is exactly the same as Vectron's VTA1.

If you have one sample left over I'll be happy if you leave it to me. You can send it to my german friend in FFM.

Are you German? I am!

Regards

Hans