Mems sensors

[JohnnyC]

I was curious. Has anyone ever instrumented thier bird with strain gauges, accelerometers, and the like? As cheap and light as they have gotten with MEMs sensors I would have thought so done somewhere would have tried it.

[wjb]

Originally Posted by JohnnyC

I was curious. Has anyone ever instrumented thier bird with strain gauges, accelerometers, and the like? As cheap and light as they have gotten with MEMs sensors I would have thought so done somewhere would have tried it.

All modern AHRS systems (Dynon, GRT, Advanced, Garmin) use MEMS gyros/accelerometers together with GPS measurements for their nav and attitude calculations calculations. Much better than the mass market cellphone-grade parts, these sensors are small, lightweight, and reliable.

[JohnnyC]

I am taking about strain rosettes to measure stress on the airframe. I know they are in avionics, I am talking about measuring vibration, flutter, stress, strain, displacements etc of the airframe. as you would have on test bed prototype. I have seen some as cheap as 3 dollars a sensor, wondered if anyone had done data acquisition on a vans type.

[wjb]

That now makes sense ... apologies! What you ask would be easy to do; we use a lot of Analog Devices ADIS 16xxx MEMS accelerometer/gyroscope sensors at work ... about 1 inch cubes with an SPI interface. A bunch of these and a laptop would make for easy data collection. There was a thread a while back about seemingly excessive tail shake while in a stall in a -10; that data would be interesting to analyze!

http://www.vansairforce.com/communit...ght=tail+shake

[David Paule]

You might be confusing motion sensing with strain sensing. Motion sensing wound better measure the damping leading to flutter onset, but strain measuement better measures the stress in a part.

Without a finite element model, though, amd a grount vibration test using accelerometers, strain measurements are only going to tell you what's happeninig at that specific location. You need the FEM to identify the critical locations for a given load case and correlate the sensor output to the bigger picture.

Another way strain sensors could help would be if their output were compared to a fatigue analysis, to give you an assessment of how much life you have left.

Either way, without some sort of structural model, there's nothing to compare it to, and therefore little point in adding strain sensors, at least in my opinion, unless you work for Van's.

Dave

[mill2978]

Originally Posted by wjb

That now makes sense ... apologies! What you ask would be easy to do; we use a lot of Analog Devices ADIS 16xxx MEMS accelerometer/gyroscope sensors at work ... about 1 inch cubes with an SPI interface. A bunch of these and a laptop would make for easy data collection. There was a thread a while back about seemingly excessive tail shake while in a stall in a -10; that data would be interesting to analyze!

http://www.vansairforce.com/communit...ght=tail+shake

I like it!

If they have an SPI interface you can use a Raspberry Pi(or similar device), it's tiny, light weight, cheap, and uses a lot less power.

[NTex]

I'd love to see a sensor on the nosegear for A models.

Lots of conjecture about why they flip over and shimmy, but no data. I don't really know enough of the hairy details to do the data logging and sensor selection - but it could be valuable information to have if you're considering the anti-splat modifications.

[JohnnyC]

I will look into this for my upcoming build. If I can find stress strain mems gauges for 3-4 dollars a pop or accelerometers for 20 bucks I could cover a lot of data points.