Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
SoftRF PowerFLARM compatible device?
1,062 views
Skip to first unread message
Moshe Braner
Sep 1, 2021, 9:58:45 AM9/1/21
to
How useful is this device in practice? Is the collision prediction
algorithm inferior to the "real" FLARM? No transponder/ADS-B reception?
Better than nothing, for gliders for which a real FLARM is not an
option, financially speaking?
https://www.ebay.com/itm/334100293758
algorithm inferior to the "real" FLARM? No transponder/ADS-B reception?
Better than nothing, for gliders for which a real FLARM is not an
option, financially speaking?
https://www.ebay.com/itm/334100293758
Eric Greenwell
Sep 1, 2021, 12:07:23 PM9/1/21
to
information. He's a real person, lives in Washington State, flies gliders.
--
Eric Greenwell - USA
- "A Guide to Self-launching Sailplane Operation"
https://sites.google.com/site/motorgliders/publications/download-the-guide-1
George Haeh
Sep 1, 2021, 12:50:12 PM9/1/21
to
Is the device FCC approved?
You might be able to buy a number of PowerFLARMs with the fine if not.
You might be able to buy a number of PowerFLARMs with the fine if not.
Davis Chappins
Sep 3, 2021, 1:37:21 PM9/3/21
to
Moshe,
The softRF device appears to output $PFLAA and $PFLAU sentences. If the device does set the alarm level flags in those sentences, I would expect the algorithms to be more simple than a real FLARM. However, since your softRF is broadcasting it's position to other FLARMs, their collision prediction algorithms should be unchanged.
In addition, you will be visible to other FLARM devices and you will have see other FLARMs on your connected display. And you'll be visible to ground stations and appear on sites like glidertracker.org
The device on ebay is https://www.amazon.com/gp/product/B097H11PJY/ref=ox_sc_saved_title_1?smid=A3GQP6749CQDZK&psc=1 with an 18650 battery and 3d printed case included. If you bought it off amazon, purchased an 18650 battery and charger, and found a 3d printed case you'd probably spend right around $120.
Hope this helps
John Galloway
Sep 3, 2021, 3:02:35 PM9/3/21
to
Andy Blackburn
Sep 3, 2021, 5:37:05 PM9/3/21
to
I realize the allure of this as a cost savings, but I have multiple concerns.
1) It's not FCC certified, so this may expose the user to regulatory action should they be reported and/or caught for using an unlicensed device. Is that likely? Maybe not, but it's a violation. These things are certified by the FAA for good reason.
2) It's been reverse-engineered. But the inner workings of Flarm units is likely impossible to ascertain simply from observing their function from the outside. That means it's not 100% certain whether this unit could adversely affect the function of other units under certain conditions - like a jammer. Flarms use a frequency-hopping technique to help keep the bandwidth uncongested. I am not at all certain whether that is a standard feature of the ISM-band chipset they use, or whether the chipset used in this device is the same - I expect it isn't. I doubt that large numbers of these devices have been tested in real-world conditions to see how they perform when there are a number of them operating since they haven't been certified. They may also leak RF like crazy since I wager the cases aren't metal. Flarm range is quite sensitive to the RF noise environment where they are operating. You will become very unpopular at your local glider field if you screw up everyone's Flarm range wherever you go. You'll be like the unvaccinated guy who shows up at a party. Even if it isn't you you'll get blamed for it.
3) I suspect it doesn't use any path-dependent collision prediction which in a thermal will make it either useless or debilitatingly annoying. Maybe it doesn't issue alarms at all which kind of defeats the purpose. I've been told that Flarm (as has been mentioned above) broadcasts path predictions. I am unclear how true this is as it seems that would eat up a lot of bandwidth, but the fact remains it's almost impossible to tell what a Flarm is doing on the inside. I worked with a developer to add NMEA compatibility and collision warnings to Stratux several years back so I could see UAT targets on my Flarm display - needless to say there was no way to replicate what Flarm did on collision alarms so we ended up with a simple proximity warning that was next to useless in the real world of glider flying - it went off constantly anytime someone was within a certain radius, so we turned it off.
4) My recommendation if you don't want to buy a Flarm and already have a Mode-S transponder is to add ADS-B 1090ES Out. At least it's certified. You won't get any traffic reports yourself, but you'll be visible to other aircraft - power and glider alike. If you don't have Mode-S (or want to see targets) I'd suck it up and buy a Flarm. There are used ones available, now that some folks are upgrading to the Flarm Fusion.
Hope that helps.
Andy Blackburn
9B
1) It's not FCC certified, so this may expose the user to regulatory action should they be reported and/or caught for using an unlicensed device. Is that likely? Maybe not, but it's a violation. These things are certified by the FAA for good reason.
2) It's been reverse-engineered. But the inner workings of Flarm units is likely impossible to ascertain simply from observing their function from the outside. That means it's not 100% certain whether this unit could adversely affect the function of other units under certain conditions - like a jammer. Flarms use a frequency-hopping technique to help keep the bandwidth uncongested. I am not at all certain whether that is a standard feature of the ISM-band chipset they use, or whether the chipset used in this device is the same - I expect it isn't. I doubt that large numbers of these devices have been tested in real-world conditions to see how they perform when there are a number of them operating since they haven't been certified. They may also leak RF like crazy since I wager the cases aren't metal. Flarm range is quite sensitive to the RF noise environment where they are operating. You will become very unpopular at your local glider field if you screw up everyone's Flarm range wherever you go. You'll be like the unvaccinated guy who shows up at a party. Even if it isn't you you'll get blamed for it.
3) I suspect it doesn't use any path-dependent collision prediction which in a thermal will make it either useless or debilitatingly annoying. Maybe it doesn't issue alarms at all which kind of defeats the purpose. I've been told that Flarm (as has been mentioned above) broadcasts path predictions. I am unclear how true this is as it seems that would eat up a lot of bandwidth, but the fact remains it's almost impossible to tell what a Flarm is doing on the inside. I worked with a developer to add NMEA compatibility and collision warnings to Stratux several years back so I could see UAT targets on my Flarm display - needless to say there was no way to replicate what Flarm did on collision alarms so we ended up with a simple proximity warning that was next to useless in the real world of glider flying - it went off constantly anytime someone was within a certain radius, so we turned it off.
4) My recommendation if you don't want to buy a Flarm and already have a Mode-S transponder is to add ADS-B 1090ES Out. At least it's certified. You won't get any traffic reports yourself, but you'll be visible to other aircraft - power and glider alike. If you don't have Mode-S (or want to see targets) I'd suck it up and buy a Flarm. There are used ones available, now that some folks are upgrading to the Flarm Fusion.
Hope that helps.
Andy Blackburn
9B
Andy Blackburn
Sep 3, 2021, 5:38:24 PM9/3/21
to
certified by FAA -> licensed by FCC - sorry.
paul.fer....gmail.com
Sep 3, 2021, 6:23:26 PM9/3/21
to
Matthew Scutter
Sep 4, 2021, 4:53:04 PM9/4/21
to
Any prediction happens in the receiver.
Eric Greenwell
Sep 4, 2021, 6:14:55 PM9/4/21
to
On 9/4/2021 1:53 PM, Matthew Scutter wrote:
> On Friday, September 3, 2021 at 9:02:35 PM UTC+2, jpg...@gmail.com wrote:
...
> On Friday, September 3, 2021 at 9:02:35 PM UTC+2, jpg...@gmail.com wrote:
>> Each Flarm and Power Flarm unit broadcasts its own (long developed) prediction of its future path with position and altitude error buffers based on the behaviour of gliders and not simply based on the instantaneous position or vector. A circling is more likely to continue for a few seconds than to stop. A very fast cruising glider is more likely to zoom upwards than suddenly descend etc. When a Flarm unit receives a predicted path that conflicts in time and space with its own then an alert is initiated. Does the unit in question definitely do all that? If not then it is a waste of money for gliders. From the data on that sale site "SoftRF is Power FLARM compatible device/software. It transmits and receives location information in 915mHz. It sends warning NMEA sentence when it detects near by glider traffic." Sounds like it just transmits and receives positions which could be helpful for general awareness but not for glider on glider collision warnings at the proximity and ways we may be flying together. Correct me if I am wrong but I can't find info to the contrary.
>
> I think 'broadcasting a future path' is a stretch. The protocol was reverse-engineered by Hiram Yaeger (that's how the OGN is able to exist), it appears to just broadcast position reports with horizontal/vertical speed, see http://www.dotmana.com/weblog/wp-content/uploads/FLARM-RADIO-PROTOCOL-VERSION-4-2008.txt
> Any prediction happens in the receiver.
The FLARM website claims each unit broadcasts a prediction of it's own path. Perhaps that
>
> I think 'broadcasting a future path' is a stretch. The protocol was reverse-engineered by Hiram Yaeger (that's how the OGN is able to exist), it appears to just broadcast position reports with horizontal/vertical speed, see http://www.dotmana.com/weblog/wp-content/uploads/FLARM-RADIO-PROTOCOL-VERSION-4-2008.txt
> Any prediction happens in the receiver.
is done because it is computationally easier than each unit having to predict the path of
several (or potentially ten or more in a thermal) gliders? A further advantage: each unit
has a complete record of all it's recent points, but may not have a complete record of the
recent points of other units, due to missed packets.
Mike Carris
Sep 4, 2021, 7:08:53 PM9/4/21
to
Eric Greenwell
Sep 4, 2021, 7:21:05 PM9/4/21
to
used? How have you upddated the SoftRF units to keep compatlbility with FLARM units?
Tango Eight
Sep 4, 2021, 8:05:23 PM9/4/21
to
T8
Mike Carris
Sep 4, 2021, 8:11:43 PM9/4/21
to
I have never used a FLARM unit and only know my experiences with the SoftRF Flarm. As a proximity device it worked well. Through XCSoar it gave audio alerts when other sailplanes got close on divergent paths. It was effective. I had it attached to the side of my LAK17a instrument paneI, which is nearly all carbon. I do not think it can be updated by the user. Also, I am not sure about the update capabilities of the Flarm devices used by RC pilots, and if the cores are the same. They are definately a lot smaller.
Mike Carris
Sep 4, 2021, 8:31:34 PM9/4/21
to
Martin Gregorie
Sep 5, 2021, 7:41:41 AM9/5/21
to
On Sat, 04 Sep 2021 17:05:21 -0700, Tango Eight wrote:
> A good guess would be that flarm broadcasts turn rate. In addition to
> speed, climb or descent rate, position, aircraft type, this is really
> all you would need to do "predictive" stuff on the receiving end.
>
The FLARM protocol document referenced by Matthew Scuttter (thanks for
> A good guess would be that flarm broadcasts turn rate. In addition to
> speed, climb or descent rate, position, aircraft type, this is really
> all you would need to do "predictive" stuff on the receiving end.
>
that!) shows that FLARM broadcasts a message containing just the glider's
ID and current velocity vector every 800-2000 mS. Presumably the
repetition rate varies depending on either the hardware in the particular
FLARM unit sending the message or on the rate at which it receives GNS
fixes.
This seems reasonable, since the receiver can deduce immediate threats
from a single message and, if it stores one or more of the preceeding
message(s), can then calculate whether the threat indicated by the latest
message from that glider is increasing or decreasing.
--
--
Martin | martin at
Gregorie | gregorie dot org
2G
Sep 5, 2021, 9:42:58 PM9/5/21
to
Tom
0 new messages