2.087MHZ crystal ? Racal 9900 - UK Vintage Radio Repair and Restoration Discussion Forum

newlite4 · 22nd Sep 2016, 10:44 am

Just looking at the faulty crystal oscillator from a Racal timer counter. The crystal is irreparably damaged, the markings are 2087 with 8214 beneath which I could assume to be a date code. Although the main clock is at 1Mhz, I can only assume that there is some division going on, seems an odd choice of crystal frequency, could this be some sort of standard?
Neil

Dickie · 22nd Sep 2016, 1:40 pm

2087 will probably be the date code. The crystal reference will be 5MHz.

newlite4 · 22nd Sep 2016, 2:55 pm

Thanks for that Dickie, I'll nip over to stores again to see if we have any 5MHz ones in stock.
Neil

newlite4 · 22nd Sep 2016, 3:21 pm

Brilliant, new 5MHz crystal fitted and 1MHz output fine tuned using a 9906 to check, working fine now.
Neil

Dickie · 22nd Sep 2016, 4:24 pm

That's good. If only they were all that easy! But beware there's more to a crystal than just its frequency. It will have a carefully chosen frequency v. temperature coefficent depending on the type of oscillator correction. In an oven-stabilised oscillator (OCXO) the crystal will be cut to have a very small tempco. at about 60C, the temperature of the oven. It will be much worse either side of this temperature. In a temperature compensated oscillator (TCXO) the crystal will be cut to have a relatively constant tempco over its operational temperature range which can be compensated for with straightforward thermistor compensation. In this latter method the compensation network is usually individually trimmed for each crystal. Google SC and AT cut crystals for more information.
So if you have the wrong crystal you will probably get a much worse frequency stability. Which may or may not be important!

newlite4 · 22nd Sep 2016, 5:12 pm

Thanks for that Dickie, I'll bear that in mind,
Neil

22nd Sep 2016, 10:44 am	#1
newlite4 Octode Join Date: Nov 2003 Location: Bath, Somerset, UK. Posts: 1,805	2.087MHZ crystal ? Racal 9900 Just looking at the faulty crystal oscillator from a Racal timer counter. The crystal is irreparably damaged, the markings are 2087 with 8214 beneath which I could assume to be a date code. Although the main clock is at 1Mhz, I can only assume that there is some division going on, seems an odd choice of crystal frequency, could this be some sort of standard? Neil __________________ preserving the recent past, for the distant future.

22nd Sep 2016, 1:40 pm	#2
Dickie Octode Join Date: Sep 2004 Location: St. Albans, Hertfordshire, UK. Posts: 1,478	Re: 2.087MHZ crystal ? Racal 9900 2087 will probably be the date code. The crystal reference will be 5MHz. __________________ Regards, Richard, BVWS member

22nd Sep 2016, 2:55 pm	#3
newlite4 Octode Join Date: Nov 2003 Location: Bath, Somerset, UK. Posts: 1,805	Re: 2.087MHZ crystal ? Racal 9900 Thanks for that Dickie, I'll nip over to stores again to see if we have any 5MHz ones in stock. Neil __________________ preserving the recent past, for the distant future.

22nd Sep 2016, 3:21 pm	#4
newlite4 Octode Join Date: Nov 2003 Location: Bath, Somerset, UK. Posts: 1,805	Re: 2.087MHZ crystal ? Racal 9900 Brilliant, new 5MHz crystal fitted and 1MHz output fine tuned using a 9906 to check, working fine now. Neil __________________ preserving the recent past, for the distant future.

22nd Sep 2016, 4:24 pm	#5
Dickie Octode Join Date: Sep 2004 Location: St. Albans, Hertfordshire, UK. Posts: 1,478	Re: 2.087MHZ crystal ? Racal 9900 That's good. If only they were all that easy! But beware there's more to a crystal than just its frequency. It will have a carefully chosen frequency v. temperature coefficent depending on the type of oscillator correction. In an oven-stabilised oscillator (OCXO) the crystal will be cut to have a very small tempco. at about 60C, the temperature of the oven. It will be much worse either side of this temperature. In a temperature compensated oscillator (TCXO) the crystal will be cut to have a relatively constant tempco over its operational temperature range which can be compensated for with straightforward thermistor compensation. In this latter method the compensation network is usually individually trimmed for each crystal. Google SC and AT cut crystals for more information. So if you have the wrong crystal you will probably get a much worse frequency stability. Which may or may not be important! __________________ Regards, Richard, BVWS member

22nd Sep 2016, 5:12 pm	#6
newlite4 Octode Join Date: Nov 2003 Location: Bath, Somerset, UK. Posts: 1,805	Re: 2.087MHZ crystal ? Racal 9900 Thanks for that Dickie, I'll bear that in mind, Neil __________________ preserving the recent past, for the distant future.