Crystal oscillators - Negative resistance and oscillation margin measurements

#134 In this video I continue looking at quartz crystal oscillator related measurements by taking a closer look at the negative resistance measurement. I look at what this measurement means, and how it can be turned into a more useful parameter which is the oscillation margin. Finally I look at what sort of possibilities there are when you run into problems with the measurement results.
More videos on crystal oscillators:
LTspice Crystal modeling: • Measuring and Simulati...
Oscillation frequency measurement: • Crystal oscillator - F...
Drive level measurement: • Crystal oscillator - D...
Negative resistance measurement: • Crystal oscillators - ...
Links and further reading:
www.maximintegrated.com/en/de...
www.cypress.com/file/88466/do...
www.murata.com/en-global/prod...
Special Thanks to all my supporters on Patreon! Especially @afiskon and Ralf B.!
If you liked this video be sure to check out my other videos and you can also subscribe to be up to date with all the new ones!
If you want to support the creation of more and better videos please consider checking out: / feszelectronics

[@alexengineering3754]

This is very interesting and most often completely ignored. Thank you for sharing!

[@acestudioscouk-Ace-G0ACE]

Good information, I need to watch this a few times to get it into my geriatric head.

[@randydireen3566]

Always love watching your videos. Thanks very much!

[@FesZElectronics]

Happy you are enjoying them! Thank you!

[@ivanm9674]

Great video! Nice explanation. Thank you!

[@wueric4076]

Good demonstration of the temperature dependent gm.

[@felixcat4346]

Another smart one ☺️.

[@stark9397]

Please come transistor, mosfets videos. Which important parameters, how can we choose and how can we control. Sorry my bad english, I hope you understand :)

[@toast2610]

To get the negative resistance higher, cant you attenuate the amplifier gain downward by adding a feedback resistor? That is a resistor across the input and output of the amplifier.

[@Nxb9ch]

Can you explain the transfer function of a crystal oscillator using mosfet?

[@SurvivalSquirrel]

I am searching for an answer, how the voltage of a microcontroller can fluctuate without any problematic effects. Especially with voltageripple and noise/spikes in the range of the maximum values of the microcontroller. I have not found any aspects yet, that would oppose the use of a supply with a voltage ripple of 0.5V, and some noise/spikes. Except, that of course it would affect ADC measurements. I could not find much to this, so i suspect its not problematic?

[@babhishek88]

Does the trimming pot add parasitics to the osc_in/out pins?

[@FesZElectronics]

It does. The recommended way to do a highly accurate measurement is to first use the trim pot to find the rough value, then use fixed value resistors to fine tune.

[@larcomj]

this video was released 27 seconds ago, how lucky am I

[@6c45pi]

How can you know if you're lucky until you've watched it?

[@Morganmcc6]

You have to watch out with trim pots when margining crystals. Usually trim pots add a lot of inductance which can non-negligibly affect your margin depending on the crystal

[@FesZElectronics]

May I ask what sort of differences did you encounter by using a trim pot as opposed to using fixed value resistors? Just to get an idea of the level of error.

[@Morganmcc6]

@@FesZElectronics I've seen as bad as 1.5K being the limit on a pot vs 600 with an 0201 fixed resistor.

[@FesZElectronics]

You must have very good eyesight to work with 0201 :D Anyway, I guess this is also frequency dependent - the impact of the extra inductance will be more severe with higher frequency crystals.

[@domtom128]

Great video as always! One phenomenon I'm trying to think an explanation of; from 12:27 there seems to be some kind of intermittent frequency mixing of crystal's fundamental and, I presume, its higher harmonic. Same thing was happening in the end with the hot air test (12:01), which is even weirder to me - seems like there is a higher margin for the higher harmonic, or something else?

[@FesZElectronics]

I was curious myself why that occurred, but I'm not quite sure yet. I guess it might have something to do with the averaging function used on the oscilloscope - to get eh waveform more static and uniform; but it might also have to do with the current probe and its impedance mismatch (between the probe inductor and the cabling)

[@domtom128]

@@FesZElectronics I should've looked more carefully, frequency of those oscillations seems to be ~100 MHz, so my money is on the averaging artifacts since your scope is 70 MHz BW.

[@FesZElectronics]

Just because the oscilloscope has a BW of 70, doesn't mean it can't display more than 70M, it will just be attenuated. But I still think its some issue with the oscilloscope :D

[@FesZElectronics]

I did some more investigations.. it seems my oscilloscope is on its last leg. I'm getting the oscillation on channel 1 even with the signal generator. Curiously, the oscillation does not appear on channel 2.

[@domtom128]

@@FesZElectronics that's interesting. Repair video maybe? :)

[@cahitskttaramal3152]

why dont you just pıt oscilkoscope probe on the xtal pins?

[@FesZElectronics]

The probe has 10-20pF of capacitance; adding the prove directly would completely change the oscillators behavior.

[@cahitskttaramal3152]

@@FesZElectronics What about active probes? They have like 0.2 pF caps at max.

[@FesZElectronics]

In what I observed from various application notes, measuring the oscillator frequency (the most sensitive test to external factors) should be done indirectly (either near-field probing, or using some other IC pins that output the OSC frequency, when available); but other measurements - like waveform shape or start-up time can be done with an active probe with direct probing.

[@cahitskttaramal3152]

@@FesZElectronics OK thanks! I am starting to measure my own crystals now :)

[@cahitskttaramal3152]

@@FesZElectronics Also I didn't catch how did you calculate the already existing 37 ohm ESR of the crystlal? Datasheet only states the maximum.

[@z3d12]

Barkhausen

[@LightningHelix101]

That guy really did a 180

[@xDR1TeK]

You have engineering hair! 🍻 You look tired man, have you been getting any sleep?