Re: [PATCH RESEND v2]IIO driver for Analog Devices Digital Output Gyroscope ADXRS450

[Jonathan Cameron <jic23@cam.ac.uk>]

On 03/22/11 08:16, Hennerich, Michael wrote:
> Cai, Cliff wrote on 2011-03-22:
>>
>>
>>> -----Original Message-----
>>> From: Jonathan Cameron [mailto:jic23@cam.ac.uk]
>>> Sent: 2011年3月22日 1:40
>>> To: Cai, Cliff
>>> Cc: linux-iio@vger.kernel.org; linux-kernel@vger.kernel.org; Drivers;
>>> device-drivers-devel@blackfin.uclinux.org; Cliff Cai
>>> Subject: Re: [PATCH RESEND v2]IIO driver for Analog Devices Digital
>>> Output Gyroscope ADXRS450
>>>
>>> On 03/19/11 09:27, cliff.cai@analog.com wrote:
>>>> From: Cliff Cai <cliff.cai@analog.com>
>>>>
>>>> Change v2:v1:
>>>>
>>>> Make modification according to Michael Hennerich's comments,
>>>> correct the spi transfer way,use existing sysfs interfaces.
>>> Hi Cliff,
>>>
>>> As you are proposing a couple of new interfaces we need to have
>>> documentation for them. They are quad and dynamic_null_correction. We
>>> need to first establish whether they are of general utility and hence
>>> should be in the main abi doc. The quadrature one isn't something I've
>>> seen before. Is it common in gyros?
>>
>> I'm not sure about this.
>> Michael,do you have any ideas?
> 
> The ADXRS450 is a quite new part and features a new sensor design -
> So I don't think the quadrature error is to date very common.
> It might become in future...
> 
> From the datasheet:
> 
> "The quad sensor design rejects linear and angular
> acceleration, including external g-forces and vibration. This is
> achieved by mechanically coupling the four sensing structures
> such that external g-forces appear as common-mode signals
> that can be removed by the fully differential architecture implemented in the ADXRS450."
> 
> "The quad memory registers contain a value corresponding to the amount of
> quadrature error present in the device at a given time.
> Quadrature can be likened to a measurement of the error of the motion of the
> resonator structure, and can be caused by stresses and aging effects.
> The quadrature data is filtered to 80 Hz and can be read frequently to
> detect sudden shifts in the level of quadrature.
> The data is presented as a 16-bit, twos complement number."
> 
Cool. So what is a good general name for this?  I guess from this description
if it were in a multi axis device you would have this measure for each axis?

So perhaps gyro_z_quadrature_correction_raw?  This thing also looks
rather similar to a dynamically changing calibbias. Perhaps we need another
term for a general dynamic linear (I think this is linear) correction inside
a device?

Perhaps go with a gyro specific term for now and wait to see if we get
many more parts with this feature...
>>
>>> Dynamic null correction looks like it should be gyro_z_calibbias to
>>> me but I could be wrong.  The doc says "
>>> The user can make small adjustments to the rateout of the device by
>>> asserting these bits. This 10-bit register allows the user to adjust
>>> the static rateout of the device by up to ±6.4°/sec.
>>> "
>>>
>>> which makes me think it is an internally applied offset on the output
>>> signal and hence calibbias in our abi.
>>
>> Thanks
>>
>>> Other than that, various minor nitpicks inline.
>>>
>>> Jonathan
> 
> Greetings,
> Michael
> 
> --
> Analog Devices GmbH      Wilhelm-Wagenfeld-Str. 6      80807 Muenchen
> Sitz der Gesellschaft: Muenchen; Registergericht: Muenchen HRB 40368; Geschaeftsfuehrer:Dr.Carsten Suckrow, Thomas Wessel, William A. Martin, Margaret Seif
>