Any MEMS people?

[rappyrex14]

Hi,

Curiously, anybody do any work on the mechanical engineering side of modeling the dynamics of MEMS gyroscopes? I know its a small niche, but figured I ask. Looking to model a 2DOF vibratory gyroscope, and could use some experimental data, to verify theoretically for a course project.

Thanks
Don

[uheli]

I do MEMS research and reply simply because I didn't expect to see this question. Are you an ME undergrad? If you are looking to validate your model with real data you'd need to match the geometrical considerations precisely. What geometry are you working with? If I needed this I'd model it in Comsol.

One route is to try searching the literature, e.g.
http://dx.doi.org/10.1016/j.sna.2006.06.010
http://dx.doi.org/10.1115/1.4001333


or start looking through the references of papers like this:
http%3A%2F%2Fdocs.lib.purdue.edu%2Fcgi%2Fviewconte nt.cgi%3Farticle%3D1161%26context%3Dnanopub&ei=_6q oT_X6Csix6AGJg8y5BA&usg=AFQjCNGUfyuxH4MSmE9XK5CjnZ 0QsMFRrA&cad=rja

[rappyrex14]

thanks for the reply! appreciate the links. currently i am a senior undergrad, and will returning for my masters next fall. Ive taken structural mechanics, composite materials, and this current MEMS course as a start to my graduate coursework. The university has licensing for ANSYS, but I am still in the process of learning the program and unfortunately wont be using it for this project. I have not taken a formal course in it yet either.

the course is on linear and non linear statics and dynamics of MEMS, my instructor wrote the text. I'm keeping things simple for the project - Coriolis vibratory gyroscope 1DOF drive and 1DOF sense. Have been using MATLAB to plot sense amplitudes, varying drive and sense frequencies (mode mismatch), stiffness's in the drive and sense directions, mass size. Basically varying parameters in the equation of motion. Again, pretty simple course project.

I have been reading many of the ieee publications, and alot of the works from Shkel and Acar as inspiration. Unfortunately, most of the works are very dense, and slightly above my knowledge base (masters/phd dissertations) - especially some of the MDOF gyroscope dynamics.

Can you tell me about the type of MEMS gyroscopes used in our heli gyros?! I have been searching high and low but information is very limited from futaba spartan beastx etcetc.

thanks again for the reply! its a small community, but such an interesting field. I feel like I have barely touched on the subject.

[desertstalker]

Originally Posted by rappyrex14

Can you tell me about the type of MEMS gyroscopes used in our heli gyros?! I have been searching high and low but information is very limited from futaba spartan beastx etcetc.

Most of them are Invensense systems, if you ask in a specific Gyro's forum you can probably get model numbers. You might try contacting Invensense, they may give you more detailed information than you can get from the data-sheets (usually they cover only what you need to use the chip).

[extrapilot]

Rappy

As for manufacturers, there are quite a few for affordable parts- including Bosch, Melexis, ST, Analog Devices, Silicon Sensing/PinPoint, and Invensense. You might try a patent search engine for these manufacturers- see if they have presented any details for a particular strategy or manufacturing method.

On your question, the published info I have seen says Futaba is running SiliconSensing (PinPoint), Spartan favors Analog Devices, and BeastX runs Invensense.

Just be aware when you ask about ‘which’ parts are in use, as there are a slew of less-than-obvious considerations in the part selection. For example, it may be that a particular design/component has superior performance in all aspects of interest, but its resonance is at some value that puts it at risk on a heli. Many you will find have resonance in the 7khz range, and that seems way out of the mechanical noise bandwidth. But, if you have a Rex450 with a motor turning 40,000 RPM with an 11-tooth pinion, that tooth noise is right in the resonance band.

Similarly, there are challenges with the physical implementation (i.e. analog vs digital interface, part voltage) which can affect the decision. You will find, for example, that Invensense is moving towards digital interface only for 2+ axis parts, and that may put it in disfavor in FBL. Likewise, some manufacturers (i.e. Analog Devices) tend towards 5v parts, where 5V is generally not available on an FBL, and will require a boost regulator (with attendant cost/footprint/noise).

Kind regards,

[*tc1]

I understood five words in this whole thread!





Not really, but you guys seem to know some stuff. It is nice to know that there are a few people in this hobby that actually know about the technology that goes into flying these little marvels.


Travis