I think that for my reflow oven project, I will tack on another part to it since I am using a dsPIC for it (lol, overkill) and it has free quad encoder pins. I was thinking of also enabling it to control a spinning turntable and measure the speed of the table. Maybe the table could have a small vise whose that could travel along a locking rail at various distances from the center and maybe a temperature sensor (or some small heater of some kind). It's use would be to calibrate the accelerometer and gyroscope ICs to figure out their actual scale factors and biases at different temperatures by applying known linear accelerations and angular velocities to the ICs at known temperatures.

But I'm having trouble finding such a motor and anything resembling a small lightweight PCB vice to go on the table. THis is going to be a stationary project so It'd be nice to not have to use batteries. But AC motors tend to have fixed speed without complex electronics which I guess doesn't matter too much since I don't really need to test the devices at multiple speeds. But if I use a DC motor then I have to get batteries, or a high power AC-DC converter. I do have a 12V 20A bench supply, but let's just say I want this system to be self contained.

Can anyone think of any ideas for the motor? Like a motor that can run off of rectified, but unsmoothed DC current? or something like that? I'm also not sure how to apply a temperature to the test IC. I guess the temperature doesn't have to be accurate since the only thing that matters is how the scale factor and bias relate to the IC's internal temperature output. Just enough to get it above (and maybe below) room temperature by 30 degrees.

EDIT: I just thought of two problems- how am I actually supposed to get this IC onto the centrifuge? It's probably going to be a part of a much larger board and it's not like it's through-hole where I can just stick it into a calibration socket for calibration and then removing it and permanently installing it on the final board. And I'm not sure how to get the wires off of the spinning PCB either without them getting all wound up...

Why not just drop the accelerometer? If you have access to a decent scale you can measure the exact weight of the test board. Gravity is a very constant acceleration =) This will give you a highly acurate method of tracking linearity as well. Air resistance counts for a little, but so much as to alter your readings for short drops.

The accelerometer testing is just something extra. I will only ever use them as inclinometers where everything is relative, so scale factors (assuming they are for the same for all axis in the same IC) is irrelevent as long as biases are corrected for. I don't think I can drop it while applying heat to it though- not easily anyways.

The bigger problem is calibrating the gyroscopes where I actually need them
to be as accurate as possible.

I also realized that due to filter and amplifier offsets and gain errors and how ADC resolutions can vary that it's probably best to test the IC in the system. Which means I'll have to build the thing very generically and leave provisions so the system can connect to the centrifuge motor and control it to self-calibrate. Makes thing a lot easier. Might need a few generic circuits though like power supply but whatever. Still not sure how to route power onto a spinning platform without batteries.