• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Don't know squat about Xtal oscillators...

Status
Not open for further replies.
Hello,

I don't know squat about crystal oscillators and I need to build a VERY accurate time base for a clock project of mine. These are the parameters needed:

1. A center-frequency (fundamental) of 1.00000 MHz.
2. Crystal-controlled for stability with an adjusting device; i.e., trimmer cap, coil or other.
3. Square wave output; 50% duty-cycle; +5-volt maximum output for TTL.
4. +9-volt supply (battery)
5. Low parts count, if possible.

Although I need it to have a CF of 1.00000MHz, I actually need it to oscillate slightly above that frequency, hence the adjuster device. The reason for this is because cutting or grinding or whatever crystal companies do to make a crystal in the exact frequency I need would cost me over $150.00 for just ONE crystal!!

Anyway, any and all help is greatly appreciated. Thank you. <ckd>
 

Sceadwian

Banned
He said 1.00000 so I'm going to say that means 100ppm. It's the same way crystal notate their frequency typically. I think the biggest problem is going to be calibrating the clock in the first place, and he does need to define 'slightly above' that frequency because crystal oscilators are not known for their pullability. You can use trimmers in place of the load caps but I think that only gives you a few ppm's of frequency adjustement. Depending on your exact clock requirements you may need to use a PLL. Also thermal stability becomes a problem at that point so you may need to do an oven controlled crystal.

Why do you need such a precise clock?
 
Last edited:

JimB

Super Moderator
Most Helpful Member
I don't know squat about crystal oscillators and I need to build a VERY accurate time base for a clock project of mine. >
OK, have a look here Euroquartz - quartz crystals, oscillators, crystal-controlled oscillators, vcxo and learn a bit about crystals and oscillators.


These are the parameters needed:
1. A center-frequency (fundamental) of 1.00000 MHz.
2. Crystal-controlled for stability with an adjusting device; i.e., trimmer cap, coil or other.
3. Square wave output; 50% duty-cycle; +5-volt maximum output for TTL.
4. +9-volt supply (battery)
5. Low parts count, if possible.
All possible, but define "low part count", 5, 10, 20 ??
If you must have an accurate 50% duty cycle, consider using a higher frequency and dividing down to 1Mhz using a divide by 2 circuit as the last divider in the chain. eg if using a 10Mhz crystal, divide by 5 and then divide by 2 to get 1 Mhz.

Although I need it to have a CF of 1.00000MHz, I actually need it to oscillate slightly above that frequency, hence the adjuster device. The reason for this is because cutting or grinding or whatever crystal companies do to make a crystal in the exact frequency I need would cost me over $150.00 for just ONE crystal!!
1Mhz is a stock frequency, most suppliers will have it in stock for a few $/£/Euro.

Crystals are specified with certain "circuit conditions", for fundamental mode crystals this is usually 30pF in parallel with the crystal, a trimmer capacitor as part of that 30pF allows for fine adjustment of the frequency.
Specify the frequency you want, not some frequency you wish to try to "pull" on to 1.000000Mhz.
150$ sounds a bit steep even for a custom made crystal, in the past I have bought crystals to my spec for less than 20$.

JimB
 

JimB

Super Moderator
Most Helpful Member
Having had a bit of a think about this, a good solution would be to use a 4060 with an 8 or 16Mhz crystal.

The 4060 http://www.electro-tech-online.com/custompdfs/2009/04/74hc4060.pdf has the active components for an oscillator, you just need to add your crystal and a couple of resistors and capacitors.
The 4060 also has several divide by 2 stages, just take the output from the divide by 8 or divide by 16 output as appropriate.
Et voila!
An accurate 1Mhz signal, 50% MSR, battery powered (5V), direct drive to TTL, with a low component count.

JimB
 
Last edited:
Hi guys,
Thanks for the responses. To clarify, I'm already dividing the adjusted 1.00000MHz signal three times to get down to about 1 second. I actually need a frequency of 1.002737916MHz which will give me 1.002737916Hz. This is the frequency needed for astronomy. Accuracy, I guess, is a matter of not having a ready way of correcting time errors. I know they're going to happen anyway, but I'd like to minimize them, hence the term "very". Also, "low parts count" = "low money". Thanks again. <ckd>
 

JimB

Super Moderator
Most Helpful Member
I actually need a frequency of 1.002737916MHz which will give me 1.002737916Hz.
Well here we go again!

A prize example of a problem which is not completely specified.

Why could you not have said that in your original post, rather than:
=
captainkirksdog said:
Although I need it to have a CF of 1.00000MHz, I actually need it to oscillate slightly above that frequency
Puts a whole different view on the problem.

JimB
 

Diver300

Well-Known Member
Most Helpful Member
OnSpec Oscillators Ltd will also do crystals at any frequency.

1 MHz or so is a very low frequency for an AT cut crystal. All 1 MHz crystal oscillators are made with crystals at higher frequencies, usually 8 or 16 MHz.

You can also buy oscillators at custom frequencies, or VCXOs if you want.

BTW, $150 is very reasonable if it includes delivery to Uranus!
 
Last edited:

ronsimpson

Well-Known Member
Most Helpful Member
A common oscillator (crystal frequency) is 12.031125.
A divide by 12 gives 1.00259375. almost 1.002737916MHz
Also look at 11.046mhz and 53.125 (these do not divide down well)
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top