IDG-500 vs ADXRS613

[Mitch]

I was also thinking a magnetometer would be a nice addition. It will allow a static calibration of the yaw axis and better error management at low velocities where the gps may not be adequate. I'm just starting to look at an HMC5843 breakout which is quite small, can be remotely mounted, and only requires a four wire connection - i2c plus power. Should be quite easy to integrate in the kalman. I don't think it would be reliable on board due to the stray fields. This configuration can be wing tip or boom mounted to distance it from the bulk of the electronics - same as full scale.

[estevanoveja]

Well, I'll be able to tell you very soon about the magnetometer behavior on a quadricopter as I ordered one for the project I'm working on.

It's a HMC6343, I know it's a bit foolish to spend so much on a sensor, but we have only 2 weeks left to finish the project so I wanted something very fast to integrate (and my school is paying). I will see how it behaves with the integrated tilt compensation and filtering dsp, especially concerning all this magnetic noise created by the motors.

I will keep you informed with the results as soon as I receive my pcb and start playing with it.

[marc]

Tom,
Concenring your concern about the ADXRS noise, when looking to the schematic, I see the ADC is fed via a resistor divider R2/R3...
According to me that R2/R3 impedance is much too high for the ADC input... it should be max 1K.... I will look to the dsPic datasheet later on...
The way to solve it is to add a cap of 0.1uF in parallel on the R3...
I will try it as soon as I have 5 min

CU

MArc

[marc]

Dear,

I confirm my previous post...
According the the datasheet, in the ADC electrical parameters section, the paramater "AD17" says that the "Recommended Impedance of Analog Voltage Source" should be max 200Ω..Therefore the impedance of the voltage divider made by R2/R3 is much too high and could explain the lack of accuracy. Let's try to put a small cap on R3.

Rgds

Marc

[Tom]

Thanks Marc,

I used these big impedances because I was looking at the datasheet of the gyroscope and it's maximum sourcing current (200uA).
Thus:
U = RI
3V3 = R * 200e-6
R = 16k

This was the impedance value of my first prototype. I thought that maybe I was pushing the gyroscope too far (noise), so I made it about a factor 10 smaller.

Your suggestion of 200Ohm would make the output current about 16mA, which is too much for this gyroscope? Maybe my modeling is too simplistic and I should include some internal resistor on the gyroscope and ADC side?

Why do you propose a capacitor (-> lag) and not a smaller resistor value?
However, your suggestion seems to make sense (because my formulas don't seem to make sense in the case without resistors). I will try it this weekend!

[marc]

Tom,

I think we should work on both side... reduce "R2+R3" to reach the "200uA Max" and also add a small cap on R3...
The idea of the small cap is to decrease the impedance that the ADC see towards the Gyro... but the disadvantage is that this cap will decrease the phase response... and should probably be compensated by a decrease of the cap between RATEOUT and SUMJ (C26).

Another question, when looking to the schematic, there is no dot on the connection C26-R2, are you sure they are connected? Is there +/- 180K when you measure C26?

Sorry for the theoretical approach... I hope to be able to practice soon...

MArc

[Tom]

I replaced R2 and R2 with 1k8 and 3k3, but didn't see any difference....

[pizza]

hi tom, should vref on IDG500 be connected to ADC pin? I don't quite understand the datasheet and the purpose of Vref..says it's a temperature independent reference... but how do we use it? is it useful? thanks!

[Tom]

@pizza: it's an output, you can use it to calibrate you ADC, or do nothing with it

[pizza]

thanks. do you have any advice on picking a yaw gyroscope? i.e what should the sensitivity/full range scale be? can't seem to find sites that carry ADXRS613 at the moment. i'm considering LY330ALH which is has a sensitivity of only 3.752 mV/deg/s and is ±300 degree/s

i noticed that you've used a separate IDO for the power supply of IDG500. will there be any major noise difference/problems if I just supply regulated 3.3V from cpu to IDG500? in the previous Sparkfun 5DOF( IDG300) they didn't actually put an IDO in the circuit.