Low Frequency 3rd Overtone Crystal Osc

I know third overtone is used for higher frequency oscillators, mainly because fundamental are hard to make at higher frequencies. I need a

120 kHz crystal and can't find one without ordering a full reel. 40 kHz is easy to find however. How practical is it to run an X cut crystal at the third overtone?

I plan to bend it anyway to make it into a VCXO, so if the overtone is slightly off from three times the fundamental, that should be ok.

--

Rick C
Reply to
rickman
Loading thread data ...

Just tweak the resonating capacitors, make sure the amp has enough bandwidth, and almost any fundamental crystal will gladly jump to the 3rd harmonic. 120 ***K** Hz? That's actually fairly low, I'm no expert, but I think those are cut differently than an RF X-cut crystal. I suspect tuning fork design. So, that may not respond so well to the 3rd harmonic trick.

Jon

Reply to
Jon Elson

Maybe I have my cuts mixed up, but I thought the tuning fork crystals used an X cut. I guess X cut is used for other frequencies and shapes as well.

I've written to Epson about sampling a 120 kHz part. There is one listed at Digikey, but not stocked and minimum 250 piece reel. I suspect both the reel and the sample will have to wait some time for them to do the whole shebang, like getting a part into a fab line queue.

We'll see. I really don't want to have to play with overtones.

--

Rick C
Reply to
rickman

60kHz is easy to find--60kHz x 2 = 120kHz. Or divide down from high up.

Cheers, James Arthur

Reply to
dagmargoodboat

Making xo's yourself is always hazardous. I prefer to buy them all built and tuned, xo's or vcxo's or ocxo's. They are super cheap nowadays, and they always oscillate.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

Do you know of a simple frequency doubler circuit for a sine wave or a square wave? Why is 60 kHz of value?

Actually I don't think I want 60 kHz on this board.

--

Rick C
Reply to
rickman

True. They don't always have good close-in phase noise, though, and you have to watch out for the programmable ones!

Cheers

Phil "Designed a lot of XOs BITD" Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

A BJT fed with sufficiently high amplitude will give you a collector current with usable harmonics up to the 10th. A parallel LC with suitable Q will easily give you the third (40kHz) or second (60 kHz). Square that up if you need digital.

Pere

Reply to
o pere o

60 -> 120k is low enough freq that with a sine you could use an audio OTA like the LM13700 to ring-modulate for the sum.

You can build a perfectly adequate filter out of the two NPN Darlington buffers that also come on the die - don't even need an external op-amp.

Reply to
bitrex

The second OTA in the package can function as a Schmitt if you need to square it up.

Reply to
bitrex

You've got to test the cheap xo's for phase noise. Apparently similar parts can be wildly different. Putting a little cover over them, to keep air drafts off, helps a lot.

Lately there are some phenomenal OCXOs for under $100. Cover them, too.

Me too. Never again, if I can help it.

formatting link

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Yeah, this is intended to be a low power design. Using a 120 kHz crystal would be better than using opamps to modulate a 60 kHz signal to

120 kHz. I also don't want 60 kHz on the board.
--

Rick C
Reply to
rickman

Overtone ones are fun, though--you get to design the LC Colpitts network so that it only looks capacitive at the right overtone. Just dicking with the C values in a normal Colpitts is asking for trouble.

(And yes, you do have to apply ALC and not use self-limiting to get the best results. Even in a self-limiting oscillator, it can be OK for many uses as long as it limits by BE cutoff and not CE saturation.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Or buy something and divide.

SiLabs and Fox make programmable XOs down to 100 KHz.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

WAY too much power consumption. Do you really think I can't read a data sheet?

I believe I can run the 120 kHz crystal at well below 1 uW. If I can't find a 120 kHz crystal maybe I can find a unit at 240 kHz and divide down and still be under 1 uW. lol

--

Rick C
Reply to
rickman

Why can't a LM13700 be low-power? It'll operate effectively down to a few uA bias current, and the total current draw IIRC is that + 2x that.

There's a circuit in the data sheet for a one shot/schmitt that burns 0 standby power.

Reply to
bitrex

Even so, that doesn't get around having 60 kHz on the board. 60 kHz is what I am measuring.

--

Rick C
Reply to
rickman

Looking at the tuning diode, I found some by Skyworks that look pretty good and are available at Digikey. But I can't tell what the reverse current is. It's not in the data sheet. They have parameters for the spice model, but I'm not confident in being able to create a model to then get the current. Can anyone divine the model data to glean that directly without having to do the simulation?

SMV1251 LS = 0.7 nH CJO = 52.48 VJ = 100 M = 76 CP = 2.00 RS = 1.4

DIODEM Diode_Model IS = 1.00e-14 RS = 0 N = 1 TT = 0 CJO = C JOM = M VJ = V JEG = 1.11 XTI = 3 KF =0 AF =1 FC = 0.5 BV = VB IBV = 1e-3 ISR = 0 NR = 2 IKF = 0 NBV = 1 IBVL = 0 NBVL = 1 TBV1 = 0 TNOM = 27 FFE = 1

There appears to be no shortage of devices available. I guess I should look for one that has a model available.

--

Rick C
Reply to
rickman

FYI, crystal overtones are just that -- overtones, not harmonics. They're usually a little bit above harmonic. There may be more, and stronger, spurious modes around the overtone as well (depends on cut and holder design), and the tempco will be a lot stronger (also depends on cut).

Normally, overtone crystals are cut for the purpose, to account for all of this.

If you're going to do it, please do report back how well it worked out -- if you can, try to plot the impedance and bandwidth of the overtone, and the spurs around it, and see what happens at different temperatures. I would expect a tuning-fork style crystal to have a strong overtone, but I don't know that it will be useable. Good luck!

On a related note, I have a 100kc crystal handy -- but it's not 120, and, as you might guess from the "kc", it's somewhat antique (dated 1959ish I think it was). Quite stately in its metal can. :-) Looks like a contemporary electrolytic capacitor, or octal audio transformer, but just two round pins.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Thanks for the offer, but I don't want crystals made in Kansas City. ;)

I'm avoiding the overtone thing. All I need is an unstable oscillator. I'm going to find a 120 kHz crystal (not made in Kansas City) or I can use a 240 kHz crystal. 120 kHz would be lower power, but if I can trust my simulations, the oscillator power is going to be true flea power, perhaps not even reaching 1 uW. So doubling that at 240 kHz won't be a problem. This won't run all the time anyway, so power isn't as crucial as the circuit that runs all the time, but power consumption can add up if you aren't careful about every little piece.

While I wait for crystal makers to get back to me I am looking at tuning the crystal. I've found a couple of varactors that might do the job although you wouldn't know by the simulation. Maybe my circuit isn't so good, but I can't get a change in load capacitors to change the frequency of the circuit in a measurable way.

--

Rick C
Reply to
rickman

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.