Hi, all,
I'm designing a mixed-signal board for a fire safety system intended to go on large pieces of agricultural equipment. It's doing DSP-type things, so it needs a decent clock.
Naturally I thought of using an external XO. However, these gizmos are liable to have a pretty rough life, with lots of vibration and impact and stuff.
Sooo, what do you use for a +-50 ppm, low jitter oscillator that will survive being dropped on the ground from a few metres up?
Thanks
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Survive damage, or bounce without a big phase hit?
The thing would drop out of phase-lock when someone unsnapped a front-panel SMB connector. The springs fixed that.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Fun. For this one, I'm mostly concerned about the resonator cracking due to shock.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
We do 40G hammer tests on products with crystals and resonators. I have never seen a problem with those components, but plenty with leaded caps and PCBs cracking
Cheers
Klaus
Rubber shock mounts?
Jon
Dropping a crystal on the floor can easily be 5000 g. It'll be a lot less with the thing on a board in a box, but I haven't used an XO in a ruggedized unit before.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
I'm thinking you won't have a problem, any more than with other parts. We don't get RMAs with failed Xo's but we get a lot of boxes back that are bent and bashed.
Drop a few and see.
If it's agricultural, maybe it will usually hit dirt.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Take a look at this App note from SiTime
I use their mems oscillators in one product. Not because of vibration but cost and power is a little better.
-- Chisolm Republic of Texas
Phil Hobbs wrote on 7/21/2017 3:00 PM:
They use crystals in lots of ruggedized equipment. Why do you think this will be a problem. I think if you have demanding specs for your time base it may be a concern, but if you are only using a typical crystal you likely won't have an issue.
Here is something from SiTime showing how "bad" quartz crystals are compared to their MEMS solutions.
Here is more info from TI.
They both test shock to 500 gs.
-- Rick C
urvive being dropped on the ground from a few metres up?
Not seen anything special in any of the military/aerospace/inflight refueliling stuff we built,including the latest 18Kw tank(battle) PSU/distribution system...
-- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
Might be a good fit for the LMK61E2. They are resistant to vibration, although they still jump around if you apply a lot of physical or thermal stress directly to the case.
-- john, KE5FX
Interesting, thanks.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Sure. What I'm trying to ascertain is how ruggedized it needs to be.
Because I've broken crystals by dropping them on the floor in the past. Not SMT ones, which is why I asked.
I think if you have demanding specs for your
Thanks. The floor tests to many times that. ;)
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Yikes, $16 in reels. Nice part though.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Don't drop your crystals before you put them in equipment. ;)
-- Rick C
They have a pretty decent AT-cut crystal inside, or at least I'm guessing it's AT-cut since it turns over at about 65C. I've been tinkering around with a proportional oven design for it.
DigiKey marked them up to $50/1 a couple of weeks ago, while Mouser was still selling them for about $20/1. I emailed DigiKey and asked them what they were smoking and if I could order some too. Turned out to be a listing mistake and they corrected the price, but even $50 wouldn't be too outrageous if it can "emulate" a $250 custom OCXO. I probably shouldn't say that in public where TI might see it...
-- john, KE5FX
** Vibration may be the bigger issue, most crystals hate it - standard HC49 ( tall)types in particular.
It is standard practice to wrap the radio receivers for RC models that use IC engines in soft foam to prevent crystal failures.
Vibration in the range of a few hundred Hz seems to be worst, the metal support clips holding the crystal inside the can simply sheer off if vibrated hard at their mechanical resonance.
A friend failed to follow this practice with his 1/8 scale gas car and suffered crystal failure in a few minutes one day. The max rpm of the 3.5cc engine was about 30,000 or 500Hz.
He gave me the dead 29MHz crystal and I opened the can with a saw for fun.
.... Phil
SiTime MEMS-stuff. AT-cut SMD crystals can take quite some shocks, but the 32.768/32.000 kHz RTC crystals are more prone to die on shocks.
The shock accelerations can usually be limited to reasonable numbers by thinking a bit the shock direction and PCB bending and installing methods.
ADXL001 is a nice part for measuring what kind of accelerations you can see on PCBs, even though they're limited to 500g. Used these in one wrist device and got bruises on my thigh while testing it by hitting myself..
-- mikko
SMT may be your answer. Some excerpts:
SMT military oscillator withstands 100,000-g shock
IQD's HGXO Series of surface mount oscillators is designed primarily for military applications such as smart munitions and projectile electronics although the devices can also be used in demanding industrial applications. Designed around a hermetically sealed high-shock crystal and a CMOS compatible integrated circuit, the new device is able to survive shock levels of up to 100,000 g and vibration levels of 20g at 10-2000 Hz swept sine in accordance with MIL-STD-202G. Housed in an industry standard 7-mm x 5-mm four-pad ceramic package with metal lid, five supply voltage versions are available at 5.0 V, 3.3 V, 3.0 V, 2.5 V and 1.8 V, thus providing compatibility with the latest generation of low power chipsets.
Output frequencies can be specified between 460 kHz and 5 MHz with a
15-pF HCMOS drive capability. Frequency tolerance is available down to +/-10 ppm whilst frequency stability can be specified at +/-40 ppm over the full military temperature range of -55 C to 125C down to +/-10 ppm over commercial temperature ranges coupled with low frequency ageing characteristics.High-shock quartz crystal oscillators
The classical quartz crystal oscillator is historically one of the most fragile components in an electronic system. This is not surprising since the quartz crystal resonator within the oscillator was composed of a large crystal such as a large round-blank AT-cut crystal mounted by metal clips inside of a metal housing. This construction could not survive shocks much beyond 50 to 100 g. While these crystal oscillators are superb for large benchtop instruments and similar devices, they are not well suited for applications where the device can expect high shocks such as handheld devices and munitions. In these cases, the accelerations can be on the order of thousands or even tens of thousands of g's. Clearly, the classical construction is not adequate for these applications.
The impetus to change the construction of quartz crystals and oscillators came from the continuing drive to miniaturize electronics. A key step in this miniaturization took place in 1970 with the development of the photolithographic and chemical milling processes for manufacturing quartz crystals. These processes, adopted from those used in the silicon industry, allow the precise milling of quartz crystals with dimensions under 1 mm and features as precise as a few microns. Another important step in this miniaturization was the development of the ceramic package for firmly mounting the crystal in a rugged housing. Together, this manufacturing and construction technique has become the de facto standard for miniature quartz crystals.
Miniaturization vs. benefits
Fortunately, the miniaturization of the quartz crystal has had the added benefit of greatly improving their shock and vibration survivability. Because of its small size, the resonator has low mass, and so the force on the resonator is low. Using strong mounting materials, the resonator is held firmly in place - the force due to acceleration is not sufficient to cause the crystal to dismount. Further, because of its small size (short blank size or short tuning-fork tines), the shear forces within the resonator are low and hence they can survive high shocks without breaking.
Another added benefit of the small size is that the frequency of the lowest flexure mode of the resonator can be on the order a few kilohertz or higher. This has at least two benefits.
First, for shocks that have a characteristic time of about 1 ms or longer, the shock can be treated as a quasi-static impulse - at any given time the shock can be approximated as a static acceleration. Because of this, the build-up in acceleration is sufficiently slow that it does not excite the flexure modes of the crystal.
Second, since these flexure modes are high in frequency, they will not be excited under vibration (which normally does not extend beyond 2 kHz in typical applications). This is important in both high-vibration applications and when manufacturing boards that are cut out using a router.
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.