|
|
Post by jautor on Oct 29, 2021 16:20:44 GMT
Picking up a thread started by Bill altitech and building on it from some items I had gathered up looking at various IoT sensor products, here's a list of additional possible ReadingType values for a Sensor instance. I'd like to hear feedback on these - are they useful? Is the definition wide or narrow enough to avoid confusion? And probably most importantly - are the units correct given common usage. Note that in Redfish, we try to use SI units whenever possible, but there are cases where non-SI units were just too common to avoid. | ReadingType | Units | Description | | Vibration | [g]
| Dynamic vibration or force (g force)
| | Shock | [g]
| Static force (g force) | Velocity
| m/s
| Velocity (should this just be "Speed" since direction is a different measure?)
| | Acceleration | m/s2 | Acceleration
| | Radiation | Sv
| Radiation dose equivalent (sievert) | | Weight | kg | Weight, strain, or force | | Luminosity | lx | Luminosity or light measurement (lux)
| | Sound | dB | Sound pressure level C-weighted dB(C)
| | Noise | dB | Noise pressure level A-weighted dB(A) | Distance LinearPosition
| m | Linear position / offset
| Angle AngularPosition | deg | Angular position
|
* Units column lists the "case sensitive" normative values from the Unified Code for Units of Measure specification, as used in Redfish - see ucum.org/ucum.htmlAs feedback comes in, if there's agreement on any of these, I can bring them to the group to add to the schema. Adding enum values to Redfish is "easy" - we can add these over time as desired. I don't want to define enum values that won't be used - but I also don't want to hold up development that could use standard reading types for sensors.
EDIT: Updating the table to incorporate feedback as it comes...
Bring on that feedback!
Jeff
|
|
|
|
Post by altitech on Nov 2, 2021 0:06:58 GMT
"Distance" might be more intuitive than "LinearPosition" as the latter may imply a more static nature. Same applies to "Angle" vs "AngularPosition". Both are also shorter which helps with data scaling.
Weight, Mass and Force are somewhat different when it comes to sensing. The the first 2 are typically measured in kg but have slightly different meanings (gravitational weight vs atomic quantity). Force on the other hand is typically measured in Newtons. We actually have distinct sensors reserved for Mass, Force and Weight though in practice we have not had a real application for it (yet).
Redfish covers much more than just PDU's so these sensors may apply more to commercial or aerospace device applications.
Just my 2 cents ;-)
Bill
|
|
|
|
Post by jautor on Nov 2, 2021 3:47:16 GMT
"Distance" might be more intuitive than "LinearPosition" as the latter may imply a more static nature. Same applies to "Angle" vs "AngularPosition". Both are also shorter which helps with data scaling. Ah, those are good points - and the `position` aspect is really a use of the "Distance" measurement. Same for "Angle" - so I like both of those as improvements. Yeah, I had used the description from a weight sensor, and especially with your suggestion for shock/vibration types, I think we'd just call that "Weight" (kg) and keep that simple. That's our hope! There is a Redfish implementation on the ISS but I haven't been able to talk to it directly... (yet) Jeff |
|
|
|
Post by altitech on Nov 5, 2021 18:30:08 GMT
If the ISS is involved then Force (N) may be useful (thrust status, impact/contact sensor etc..). I'm sure these sensors are readily available to the average hobbyist on Earth ;-). Maybe its worth a ping to NASA though.
Bill |
|
|
|
Post by altitech on Nov 10, 2021 16:34:03 GMT
Jeff,
Do you know the next tentative Redfish release date that might include some of the new sensor changes discussed here or in other recent posts?
Bill
|
|
|
|
Post by jautor on Nov 10, 2021 22:42:22 GMT
Jeff, Do you know the next tentative Redfish release date that might include some of the new sensor changes discussed here or in other recent posts? Bill I can't provide any guarantees on release dates. But we generally do 3-4 releases per calendar year - the last one has always been "at the end of the year"... I've submitted the power/energy-related items from the other thread for the next release. The items in this thread I have not yet submitted - I'm hoping to get some more feedback, so don't expect to see those in the next release. Jeff |
|
|
|
Post by altitech on Nov 12, 2021 16:39:40 GMT
Thanks Jeff,
Wanted to ask because we still need to update our implementation to 2021.2 before next release slated for end of Q1. We will likely be tackling this in January just in case you have new additions by then.
Best Regards,
Bill
|
|
|
|
Post by altitech on Jan 7, 2022 17:32:12 GMT
Though I very much like Velocity, Speed is probably more generically appropriate.
Hmm...
Technically, Acceleration does have a directional vector component since it is the change in Velocity over time. Now that I think about it, Velocity would be more consistent.
Bill
|
|
