[EE] MEMS oscillator vs traditional quartz-crystal-inside integrated oscillator

[Electron]

Hi,
I must confess I didn't even know that MEMS oscillators existed. But the bugs in
many PIC32 (that prevent the use of crystals, see errata e.g. of PIC32MK, PIC32EC
and even PIC32EF chips) leaded me to select an alternative and I discovered these
MEMS oscillators that were previously unheard of, for me.

So I Googled and found some informations about them, that I'd like to share
because I think the technology is interesting, at least to know, if you don't:

As far as I can see, the advantages of Xtals over MEMS oscillators are:
# cheaper (albeit the price of MEMS *oscillators* is not prohibitive).
# consume less power (in certain MEMS oscillators models it is quite serious, ~30mA).
# have less jitter (but in some applications is irrelevant, we talk about ps anyway).
# Xtal oscillators startup more quickly.

And the advantages of MEMS oscillators over Xtals are:
# They are also quite stable devices, and they are actively temperature-compensated
(but when compared to a Xtal oscillator with temperature compensation, the latter wins)
# MEMS oscillators may be more robust in very high-vibration environments .
# They are often physically smaller than Xtal based technology, with the side-effect of
better signal integrity and reduced EMI. There exist small Xtal oscillators too, though.
# Oscillator vs oscillator, the MEMS are usually cheaper than the Xtal based ones.

MEMS oscillator technology is improving, so, someday, they may replace Xtal oscillators
even in RF applications, but certainly not today.

In communications equipment, Xtals are definitely the way to go. But since my interest
is in PIC32's, and in many models the primary oscillator is buggy and you just can't
use it (I know that latest (2017) IC revisions often have working primary oscillators,
but it doesn't matter when you've already bought hundreds, buggy, microcontrollers) if
the alternative is between a full blown external Xtal oscillator vs a MEMS one, the
latter may be cheaper, smaller and probably generate less EMI as well.

Generally, Xtal oscillators are of course superior, but in more and more specific cases,
MEMS oscillators are proving to be more and more a valid alternative.

I hope I didn't write anything wrong.. in case, there are a lot of knowledgeable people
here in the list that may correct it. I just wanted to write this as a pretext to talk
about the technology, and to bring the topic.

With kind regards,
Mario

[Sean Breheny]

Hi Mario,

Nice summary. I believe one more advantage that MEMS oscillators have over
crystal-based ones is temperature range. A couple of years ago I needed to
find an oscillator which was rated up to 120C operation and the only ones I
could find were MEMS-based. I don't know if there is some inherent reason
for this or if they simply tend to use the MEMS technology in making more
rugged oscillators, possibly for vibration reasons, and therefore put extra
effort into high-temperature operation of them.

Sean

Post by Electron
Hi,
I must confess I didn't even know that MEMS oscillators existed. But the bugs in
many PIC32 (that prevent the use of crystals, see errata e.g. of PIC32MK, PIC32EC
and even PIC32EF chips) leaded me to select an alternative and I discovered these
MEMS oscillators that were previously unheard of, for me.
So I Googled and found some informations about them, that I'd like to share
# cheaper (albeit the price of MEMS *oscillators* is not prohibitive).
# consume less power (in certain MEMS oscillators models it is quite serious, ~30mA).
# have less jitter (but in some applications is irrelevant, we talk about ps anyway).
# Xtal oscillators startup more quickly.
# They are also quite stable devices, and they are actively
temperature-compensated
(but when compared to a Xtal oscillator with temperature compensation, the latter wins)
# MEMS oscillators may be more robust in very high-vibration environments .
# They are often physically smaller than Xtal based technology, with the side-effect of
better signal integrity and reduced EMI. There exist small Xtal oscillators too, though.
# Oscillator vs oscillator, the MEMS are usually cheaper than the Xtal based ones.
MEMS oscillator technology is improving, so, someday, they may replace Xtal oscillators
even in RF applications, but certainly not today.
In communications equipment, Xtals are definitely the way to go. But since my interest
is in PIC32's, and in many models the primary oscillator is buggy and you just can't
use it (I know that latest (2017) IC revisions often have working primary oscillators,
but it doesn't matter when you've already bought hundreds, buggy, microcontrollers) if
the alternative is between a full blown external Xtal oscillator vs a MEMS one, the
latter may be cheaper, smaller and probably generate less EMI as well.
Generally, Xtal oscillators are of course superior, but in more and more specific cases,
MEMS oscillators are proving to be more and more a valid alternative.
I hope I didn't write anything wrong.. in case, there are a lot of knowledgeable people
here in the list that may correct it. I just wanted to write this as a pretext to talk
about the technology, and to bring the topic.
With kind regards,
Mario
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist