FlySight GPS

[Hellis 0]

You can calculate the distance traveld from the Flysight logs.

[crwper 7]

Quote

There may be people that want to mount on a flat surface, does the tape work well if the curve isn't a close match?

I've put together a couple of templates, which I've attached, and would appreciate any help you guys can offer with a bit of a test...

To use the template:

1. Choose the appropriate file depending on your normal paper size (letter or A4).


2. Print the PDF and cut out the template.


3. Using the template, check the curvature near the back of your helmet.

You may need to move the template around a bit to find a spot that fits well. The big question is: Can you find a spot on your helmet where the template is a good fit? Using 0.090" VHB tape to secure the mount gives a bit of leeway, but the mounting face (i.e. the faces marked by red arrows on the template) should fit within about 1/16" (1.5 mm).

If you're able to do the test, please PM me to let me know what kind of helmet you have, and whether or not the template fits. If it doesn't fit, is it too flat (i.e. the outside edge touches, but the inside is more than 1/16" from the template) or too curved (the opposite)?

Thanks so much for your help!

Michael

P.S. I'm working on the screw mount issue. With luck, I may be able to put a couple of holes on either side near the middle of the mount.

[DSE 3]

the template is too curved for my Tonfly 3X due to the size of the flat area.
It is just right on my Bonehead Mindwarp if I move it high up away from the bottom/collar. I'd prefer it to be as low on the back of the helmet as possible, which means I'd like the mount to be more flat (regardless of which helmet I was putting the mount on).

[LukeH 0]

Bonehead Mindwarp (quite an old one)
The bottom of the helmet at the back is quite flat, so template is too curved. There is about 2mm of clearance between the points either side of the Velcro slot and the helmet.

Bonehead Optik Illusion
The curves of the helmet are different in different directions and places, near the neck it's quite flat across the helmet, so it's similar to Mindwarp above, but up near the crown of the head it's fairly close to the template but still slightly flatter.

[crwper 7]

Here are a couple of new templates. These ones have two surfaces: The top one is the original, and the bottom one is a bit flatter. If I've understood correctly, the bottom surface should match your helmets much better.

Thanks for your help!

Michael

[LukeH 0]

Quote

Here are a couple of new templates. These ones have two surfaces: The top one is the original, and the bottom one is a bit flatter. If I've understood correctly, the bottom surface should match your helmets much better.

Thanks for your help!

Michael

Yes, it fits the Mindwarp quite well, it's a little too flat now if anything. Looks like you lowered the inner corners by about 3mm? I think 2mm would be enough and may work with a wider range of helmets.

For the Optik Illusion, is ok across the helmet the problem is that from bottom to top the helmet is more curved. Perhaps something in between the two templates will give a better compromise.

[crwper 7]

Quote

For the Optik Illusion, is ok across the helmet the problem is that from bottom to top the helmet is more curved. Perhaps something in between the two templates will give a better compromise.

How much difference do you see in the two curvatures?

Michael

[skydiver833 0]

I tried the most recent template on my Optic Illusion and for my preferred location (Just above the Bonehead Logo) the more curved template is a closer match than the flat one.
I believe a blend between the two that favors the more curved template would work well for my use.
Richard

[Yorick 0]

Quote

This is how I mount it...total cost is like $.03 and I trust it more than tape, velcros, or screws.

Very, very nice!
Thank you for inspiring me.
I received my device today and I did the same.

I also bought Phillips headphones for 6 euro's, cut them, and put them in the audible pocket. (They fit nicely)

And to avoid questions like: 'what is that thing on your helmet?'; I added the text 'GPS' next to the FlySight.

"The 'perfect' parachute jump was thought to be one where the opening shock and touchdown were simultaneous" -Lyle Cameron, ~1965
---
Falling-With-Style.com

[notsane 0]

I switched the tone setting in the config file to 0 (horizontal speed). We were doing some maximum flight speed jumps ... I topped out at 131mph (horizontal, not total). I noticed an annoyance .... when the tone is set for glide ratio, the tone doesn't sound before exit, for the most part ... presumably because the a/c is in level flight. But the horizontal speed tone continuously sounds, indicating the a/c speed. Without the freefall noise, it's pretty loud and annoying. Is there anyway to enable the tone after exit?

[LukeH 0]

I don't think there is a way to do this in mode 0 with the original firmware.

There is a threshold value that is applied in mode 2 and 3. Tones are silenced if you are below the threshold.

Edit: It looks like some changes were made in Jan to split the threshold into vertical and horizontal speeds and apply them to all modes.

Not sure if these changes are included in the stable or beta firmware here http://flysight.ca/wiki/index.php?title=Latest_firmware but the source is here: https://github.com/flysight/flysight

To check if you have this firmware version:
-take a backup of your Config file
-delete the file and turn the Flysight off and back on again
-a new config file will be automatically created with default values
-if there is a Thresholds section in the file you have this version.

[notsane 0]

No problem with flashing new firmware. I just did the Honeycomb hack to a nook color and added Marketplace with the Android SDK ... bitchin'!

I'm an engineer with Texas Instruments, so of course I'm wishing that you used the MSP430 or a Stellaris device ...
I'm looking through your C code ... you might add a "freefall detection" piece of code, like is used in audibles. Condition the tone generation based on that logical-and a new selection in the config file. Maybe a MUTE IN AIRPLANE 0,1 value.

Just an idea ... BTW, here's the Paralog interpreted data from the jump (attached). 131mph horizontal, 60 mph vertical sustained ... We can do better
Scott

[crwper 7]

Quote

I'm an engineer with Texas Instruments, so of course I'm wishing that you used the MSP430 or a Stellaris device ...

Quote

I'm looking through your C code ... you might add a "freefall detection" piece of code, like is used in audibles. Condition the tone generation based on that logical-and a new selection in the config file. Maybe a MUTE IN AIRPLANE 0,1 value.

An interesting idea... That might give a little more flexibility in the logic, while still allowing the FlySight to be used for non-freefall purposes (e.g. by swoopers).

If you haven't already joined the FlySight developers' mailing list, you should!

Michael

[notsane 0]

I'll join the list ... thanks, Michael! Nice job on packaging this thing. I opened it up to do a firmware update.

[Hellis 0]

Quote

I'm looking through your C code ... you might add a "freefall detection" piece of code, like is used in audibles.

My experience with creating a frefall detection is that its really hard to make it work with GPS devices.
I tried to make a code (to my other GPS, not the Flysight) that removed all the airplane and canopy data.

If i remember correct i made the code react on altitudes above 2000 meters, vertical speed of 100 km/h, looking at the other nearby vertical speeds and altitudes to make sure its correct numbers, and vertical speeds dropping below 50 km/h.

But because GPS devices sometimes get wrong data it could sense that i was in freefall when i was not.
And basicly if i changed the code to make it work on one of the jumpdata it would not work on the other because of that change conflicted the data on the other.

I have thought about trying to recreate it for the Flysight, but my previous attempts for my other GPS kind of holds me back.
The Flysight is more stable than my old one so its probably going to be easier, but


EDIT: one of the biggest challenges with GPS devices is when you turn them on and they start looking for satelites they are not accurate for S***.
And thats one of the areas where you get problems with freefall detection.
Other troubble areas is loosing connection with one and geting connection with another, this can create a "bump" in the data.
But ofcourse, this was with my old GPS, i have not tried it the Flysight.

[Costyn 0]

Quote

EDIT: one of the biggest challenges with GPS devices is when you turn them on and they start looking for satelites they are not accurate for S***.
And thats one of the areas where you get problems with freefall detection.
Other troubble areas is loosing connection with one and geting connection with another, this can create a "bump" in the data.
But ofcourse, this was with my old GPS, i have not tried it the Flysight.

Nearly all GPSes can give you a figure for their accuracy, the so called 'Dilution of Precision'. On my GPS I can ask it to log that too along with each trackpoint, which might've solved your problem by throwing away all trackpoints which weren't accurate. In the FlySight you could do something similar: only beep when you have a good fix and when you detect freefall.

Costyn van Dongen - http://www.flylikebrick.com/ - World Wide Wingsuit News

[LukeH 0]

Exit and canopy opening detection are quite difficult to do real in real time. Look at how the barometric loggers perform and they have the advantages of being on from takeoff and no loss of signal. Even Paralog doesn't always get it right and it's not trying to do it in real time.

When you start trying to cater for strange things happening in the aircraft (dropping altitude or holding) and different types of freefall (with and without wings) heavy flares in big wingsuits, and the transition from wingsuit flight to canopy flight it becomes a nightmare!

The threshold system of silencing the device is a reasonable compromise but could be tweaked a bit. I've also set mine to go quiet below 800m above the DZ elevation.

Maybe we could add a third threshold value so you have a min and max descent rate for tone. At the moment the threshold applies regardless of whether you are ascending or descending. If you are just turning the device on at the 2 min call during run-in the new version probably works very well.

[Hellis 0]

Hmm.. Good idea.
When you log the data that could work, but when/if you manipulate the already loged data it could be harder.

But i looked at one of my jumps with Flysight, and i think the horisontal-, vertical- and speed-accuracy could be used combined with the number of satelites.
The H and V accuracy was about 4 m during the jump, so if i would use a threshold of 7 or 8?

When i have a larger sample of loged jumps i will start with that.

Thanks for the idea!

[crwper 7]

Quote

Maybe we could add a third threshold value so you have a min and max descent rate for tone. At the moment the threshold applies regardless of whether you are ascending or descending. If you are just turning the device on at the 2 min call during run-in the new version probably works very well.

Looking at the configuration file the other day, I was thinking it might be best to use just the vertical speed threshold for all modes, and perhaps eliminate the horizontal threshold. Really, the horizontal threshold is only useful in the "inverse glide ratio" mode--not too important for the places FlySight is actually being used.

Applying the vertical threshold to all modes would make FlySight's behaviour very predictable, which I like, and also simplifies the interface a bit. For exceptional uses--e.g., flying a wingsuit with jets--there's likely no way to reliably detect the difference between "in flight" and "not in flight". I suspect the user would just have to put up with a few beeps in the plane.

On the topic of freefall detection... One advantage we have here, perhaps, is that the FlySight will always start a new log when you turn it on, and end the log when you turn it off. In contrast, most audible altimeters, for example, use freefall detection to delineate jumps. The nice thing about FlySight's way of doing it is that, if your detection algorithm is not quite perfect, it wouldn't start a new jump--just miss one or two data points on a single logged jump (or include a few points it shouldn't have).

Michael

[Hellis 0]

But an audible has the advantage that it does not need to know that its in freefall for the first 5 seconds.
The audible has one job and that is to beep when altitude is low, but the GPS needs to beep when its losing altitude but not in the airplane.
And on the jumprun you can easily lose altitude, and combine that with a wingsuit and the decentrate of airplane on jumprun can be very similar to a wingsuit in freefall.
Nightmare.

Does the GPS have any way of calculating stuff?
Can a freefalldetection algorithm be implemented in the software so that it will delete or not log when you are in the airplane?

I just had another idea. You are "supposed" to turn on the GPS at the 2 min call.
But what if the GPS is turned on at the 2 min call but does not start logging untill 2 min later.
Think of it as a timer, you turn on the GPS and 2 min later you hear a beep that means it starts logging?
On the next version of Flysight we might even see a button to start the logging?

[LukeH 0]

How about a 3 position switch.

OFF - ON (No logging) - ON (Logging)

The middle position could be used when acquiring Sats on the ground and in the plane, being flicked to Logging shortly before exit.
The middle position could also be more or less silent, perhaps a couple of quiet beeps to indicate that it's working and got a fix.

Perhaps an option to auto adjust the volume depending on descent rate. I.E. quieter in the aircraft and under canopy and louder in freefall, similar to the way a car sat-nav increases volume with speed.

Maybe we are just trying to fix problems that don't need fixing ;-) I like the simplicity of it as is.

[crwper 7]

Quote

Maybe we are just trying to fix problems that don't need fixing ;-) I like the simplicity of it as is.

I'm in the same boat... One of the things I like about FlySight is that I know it will start logging when I turn it on. If something unusual happens on the jump, or if I fly particularly well, I'm not worried about whether or not the FlySight will decide I've ended the jump.

I'm always wondering if there might be a way to make things a bit "smarter", but at the same time, I feel like a lot of that should be deferred to the analysis software (e.g., Paralog). If the analysis software fails, you can always crunch the data manually, or try a different program. However, if the FlySight fails to log a jump, there's nothing you can do.

Michael

[mariobat 0]

Hi all!

I ear the glide tone from my FS only under canopy and not during the WS jump (the volume is set at the top).
Could you help me?

The glide is set at 2 and downloading data from FS I usually fly beetween 1,8 and 2,2.

Blue Skyes
Mario

[antonija 0]

Try louder earphones and/or quieter helmet.

You will hear it 100% if you get one of those in-ear phones (they isolate outside noise and produce sound inside your ear canal), but then you will not hear anything else around you under canopy.

Or you could try one of those amplifiers that motorcycle riders use sometimes, but that's more weight on you and another battery to think about.

I understand the need for conformity. Without a concise set of rules to follow we would probably all have to resort to common sense. -David Thorne

[LukeH 0]

Can you attach the config file from your flysight?