I have a receiving circuit built for 262KHz. Antenna, filters and amplifier stages all work great, however my PLL center frequency can drift by as much as ~4KHz. The frequency is drifting due to temperature variations. I can blow on the circuit and it moves +/- 900Hz.
I use a PLL chip that requires an RC network to produce a frequency at 2x the desired 'lock' frequency (262KHz = 524KHz), since this PLL chip has a (x/2) divider network inside.
Resistor is a tunable Pot and the Capacitor is fixed value. I already changed out the Pot for one with better thermal tolerance, and it helped, but there's still an issue.

I guess my initial question is; How much PLL drift is 'average'?
Secondary question; Should I consider a more reliable oscillator for the PLL, such as a crystal?

Thanks in advance.

PLL: http://www.ti.com/lit/ds/symlink/lmc568.pdf

 

Hello,

What type of capacitor do you use?
There are capacitor types that change the value with temperature changes.

Bertus

 

As Bertus noted, a more stable capacitor may help.
The oscillator stability is mainly determined by the RC stability.
Some standard ceramic capacitors have as much as 1500 ppm/°C change in capacitance (≈400Hz/°C @ 262kHz).
Use an NPO ceramic, which has a typical maximum drift of ±30 ppm/°C (≈8Hz/°C @ 262kHz), or a low temperature coefficient film type such as polyphenylene sulfide.

If that still isn't sufficient then you may need to go to a crystal controlled oscillator.

 

Hey Guys,
It's a NP0 (SMD) capacitor.
Capacitance change/Temperature coefficient is 0±30ppm/°C.
It seems relatively stable, to me.

I think I just need to switch to a reliable crystal like you said. There aren't any 524KHz crystal off the shelf (Online).
I really don't want to add a custom made crystal ($$$) to my circuit unless I absolutely have to. Do you guys think a crystal going into a frequency divider chip and then into the PLL would be an advisable setup?

Much appreciated.

 

Hello,

My experience is that styroflex film capacitors are quite stable.
styroflex capacitor

Bertus

 

Does the centre frequency drift far enough for the loop to lose lock or otherwise affect the receiver performance?
The PLL should adjust the VCO to follow the input frequency so as long as it can lock it should be OK.

 

Albert,
Yes, it drifts far enough away to lose lock but it mostly changes the maximum receiving distance so much so that the receiver must be placed very close in order to pick up the transmission and not surrounding noise (since it's analog). Anytime my center freq is where I want it (524KHz) everything works great. I'm dealing with a very low-power transmission (A Pill tx'ing from inside of the body), so I don't have much distance to play with. I'm dealing with a maximum of 25" distance, and when the PLL drifts down, say 600Hz, that max distance for a PLL lock changes to 18-19".
If the PLL drifts far enough away from center freq, It will start picking up noise and just gets messier and messier the more it drifts.

 

I have ordered many off frequency crystals........same price as any other.

Edit: I have ordered crystals from here for over 40 years.

http://www.jancrystals.com/

 

BR-549,
Under 1MHZ? The Couple of manufacturers I have talked to have overpriced me for a sub 1MHz crystal with a PPM range for PLL's.

 

You can email them like I do. You can always divide down.

 

I shot them an Email, thanks!

 

I think my question has been answered, I'm going to hunt for a crystal solution. Will let you guys know how it works out.
Thanks for your time everyone.

 

Give them your phone number.......and ask one of their engineers to call you. Tell them you have a custom app.

It's been a while......since I used them......but I've been in their labs. Very accommodating and easy on the price.

Edit: Even for one crystal.