Basejumper.com - archive

I've been working on a project which is now in its second iteration. The idea is to develop a device which will dictate my glide angle every second or two during freefall. This will allow me to make adjustments to my body position and determine immediately what effect those changes have. I've seen a bunch of GPS data indicating that people are wearing GPS on skydives and BASE jumps, and later viewing the data to see what works and what doesn't. It seems to me that having this information in-flight would be a considerable improvement.

Since I am a programmer, I figured the best chance for this project to get done would be if I turned it into a programming problem. My first attempt used a Garmin Geko GPS attached by cable to a Palm Tungsten. The headphone output of the Palm Tungsten went to a home-built amp, which went to a pair of earphones. On the software end of things, the program took position data from the GPS and calculated glide angle. To dictate the glide angle, it played one of thirty sound files (me saying, "0.0", "0.1", etc.) depending on that value.

Problems noted: First, the cable was fussy as hell. As often as not it would come undone on jumprun or something, and I'd have to abandon the test for that skydive. Second, the audio was too quiet to be heard in freefall. Since the amp was really just acting as a voltage follower (a 1:1 amp), there is plenty of room to improve on volume. I did not get a chance to try the device using my wingsuit, which I thought would be a bit quieter than regular freefall.

Successes: Even though I couldn't make out the numbers in freefall, I could hear that it was saying something, so I know that the GPS is working properly very soon out the door. The Garmin GPS has a time to first fix of something less than 10 seconds when hot, i.e. if signal is lost for 15 minutes.

So, on to the second version. This one is a Garmin GPS 10 (a wireless GPS with accurate velocity output), plus the Palm Tungsten again, and the amp and earphones. I've also re-written the software to make it easier to tweak. The new version has different behaviours for freefall (>13 m/s down), under canopy (0-13 m/s down), and in aircraft (climbing). So far I only really know what I want in the freefall stage, which will be similar to the previous version. I've tested the wireless, and that works really well. I also tested the velocity walking to work this morning. The specs say it's accurate to 0.1 knots, and in fact that seems true. It was stable when I was walking a steady pace, and if I slowed down just a bit it would drop instantly by 0.1 or 0.2 m/s. The amp will be easy to change so the output is louder.

Currently the interface on the software is pretty simple. It's just echoing what it gets from the GPS, and it has a button to connect/disconnect. I will likely be adding to this a page with a position/velocity readout, as well as a "satellites in view" page for checking things out on the ground. I will also be adding a position/velocity log, so you can go back later and check things out.

So, why am I posting this? I think something like this would be very useful for anyone interested in optimizing their track or wingsuit flight. I will definitely be releasing the software, along with source code, for anyone who is interested. You'll need a Palm Pilot and a GPS. The amp is easy to build, but this is something I could probably build for cost if someone is not able to build it themselves. Does anyone know of a commercial alternative (i.e. a small battery-powered audio amp)?

I am curious to know 1) if anyone else is interested in this kind of thing, and 2) if you are interested, what features would you like to see?

Michael

http://www.ballvarios.com/

The product exists for paraglinding, it would take some treaking to deal with freefall speeds, but for under canopy it works perfect.

This is fantastic! I'm very interested, and maybe can help you out with coding, too.

I'm working on my tracking skills using Etrex Vista C. For now, it's just postflight analysis, but I'm going to mount it on my cheststrap and view the altitude/distance plot in real time.

The audible output is even better.

It might be easier to "hear" the glide angle by playing a tone where the pitch gets higher as the glide angle increases. That way you don't have to hear and understand actual words, you just listen for a tone.

Kind of like how analog gauges work better in a race car. The driver just needs to check and make sure the needle is pointing straight up rather than trying to read and comprehend a digital readout.

Lou

The usual argument with analog gauges is that they are the best way of communicating both absolute value (and moreso position within a range of values) and change in value using one instrument.

My concern with tones is that it might make it difficult to compare what you are doing now with what you did earlier, unless you're very good at distinguishing frequency. However, this may not be an issue, and at any rate tones are very good for communicating change, so it's definitely worth trying it out in freefall.

Michael

Is there an instrument there that indicates glide angle? I can only find varios on that website, which indicate vertical speed but do not take horizontal speed into account.

Michael

Yes, variometers that cost $300 and up show the glide angle. (although I think they won't work at freefall speed). There are also vario+GPS combos. I use Mini-Vario (descent version) under canopy. It's a small device the size of protrack that beeps faster and in higher tone as your vertical speed increases.

my vario samples every 10 seconds I think not a real good freefall stat, but great for glide ratio under canopy

Cool idea Michael.
Ive been a hang glider pilot since 1987 so I have a lot a flight time with a vario. It sounds like there are some similarities between a vario and the instrument that youre working on but I think there are some differences. Im not that familiar with the latest and greatest varios on the market which have additional instruments such as glide angle and GPS, but the vario I have has a visual display (like in an airplane) and an audible beeping indicating the climb rate (with a happy tone) or descent rate (with a different bad tone). Like veter_ said, the tone beeps faster as the vertical speed increases either up or down. For it to mean something accurate and useful to the pilot it has a starting point of 0 which is a vertical speed up or down of 0 (maintaining the same altitude). You calibrate it to the current barometric pressure on the ground before you launch to obtain the starting point. There is an error factor, which is any barometric pressure change as you move through the airspace or as the barometric pressure changes throughout the day.

For your device to work like a vario (not the glide angle or gps report), I would think it would need a similar starting point such as a certain glide angle and then give the pilot different tones and beep rates as your glide angle becomes shallower or steeper from the starting glide angle. The problem I see with that is how do you determine the starting glide angle and how would that apply from pilot to pilot?

Keep in mind Im just brainstorming here.

If I understand what your aiming at correctly.......Are you simply calculating the altitude lost by distance gained during a time cycle and reporting the glide angle with one of a series of numbers; each number indicating a different glide angle?

Maybe it could have the beep rate increase as the glide angle became shallower.
Falling straight down would report no beep and a shallow glide angle beeping like crazy with a happy tone - in hang gliding using a vario, this kind of audible report means you have a killer climb rate, which is a good thing (unless your standing on your wingtip and its off the chart and youre approaching the cloud base ).

One thought comes to mind. If you turn during your flight, maybe your device could detect that right away with the gps data so that it would still give you a good glide angle report immediately following (or during) the turn. You probably already have that figured out.

That's the general idea.


Right. The first version calculated the altitude lost and the horizontal speed from position information, and then used this to calculate glide angle. If the glide angle was 2:1, it would say, "two point one".

The new version uses velocity data coming straight from the GPS, rather than calculating velocity from position. As I understand it, this should be considerably more accurate. Calculating from position, you have (best case) about 5 metres error in position, which means your velocities are only accurate within something like 5 metres per second. Taking the velocity data from the GPS, I believe it uses something like carrier phase to measure the velocity, rather than taking differences. As a result, velocity is accurate (according to Garmin's specs) to about 0.1 knots (about 0.05 m/s, or about 100 times more accurate).

My experience with the previous version was that, although the dictated data tells you exactly what glide angle you achieved, it is difficult to hear over wind noise. I can always make the output louder, but it seems to me there is a basic problem here that wind noise is similar in some important ways to the human voice. A tone, on the other hand, is quite distinct from wind noise. So, perhaps a tone would be better.


Yeah. Another advantage with tones is that we have two dimensions to work with. One is the rate, and the other is the pitch. We could do something like have the rate indicate your glide angle, so as you say it is beeping like crazy when you're doing well, and the pitch could indicate change in glide angle (high pitch if your glide angle is improving, low pitch if it's getting worse). That would correspond with the "happy" and "sad" tones, right? A couple of options here would be (a) to have a continuously changing pitch, so you know just how fast you're improving or (b) just a couple of user-defined tones which indicate + or -. I have a feeling that from a human interface standpoint option (b) will be better, since (a) might just be too much information. Any thoughts?


I think this happens naturally, if I follow what you're saying. I am getting east, north, and up velocities from the GPS. I use east and north to calculate a horizontal velocity which is independent of your direction. So, even in the middle of a turn you should be getting a good indication of glide angle.

Michael

 

I think the problem with re-mounting the display is mainly that it is very difficult to focus your eyes on a display that's that close. The heads-up systems (like on L&B's Titan) usually move the focus to infinity, if I'm not mistaken, so it's easier on the eyes.

As for speed, that's one of the things I was thinking. It will be easy to have the thing indicate speed instead of glide angle. For example, a jumper might be more interested in optimizing horizontal speed, or in the case of speed skydiving, vertical or total speed.

I was inspired to start this project after watching a video with some guys using an Atmonauti-style track off a big wall. What I was thinking is, how much have we actually played with tracking positions in freefall? If we had a device which gave instant feedback on the effects of a change in body position, how much fun would it be to try all kinds of different things in freefall? Surely this will at least give a more intuitive feel for what works and what doesn't.

Michael

That, in my opinion, would be the weakness of beeps. Beeps are great if you're trying to optimize your glide angle on this dive. But if you land, repack, and head up again, are you really going to remember how fast the thing was beeping last time?

A glide ratio of 2.2:1 will always be better than a glide ratio of 2.1:1. But is "beeping like crazy" better or worse than "beeping super-fast"? Increasing the volume, I think, would be worth a shot, and ultimately the more versatile solution.

I think option (a) would be redundant info since you will hear and understand the beet rate changing and how fast it changes. For option (b), you might want three tones (+, 0, -). That way if youre maintaining a glide angle (if just for a few seconds) you still hear a tone with the beep rate report to indicate how well youre doing.

---

base736,
You do have a point in that you wont know exactly what the glide angle is. But I think after using the device enough times you will become acclimated to it enough to understand and remember what the different tones and beep rates mean to you dive after dive. I say this because with my experience using a vario in a hang gliding environment - at first with a new vario I would find myself looking at the visual report to see what the climb/descent rate is with what I was hearing. But it doesnt take long to know your climb rate by listening to the beeping without having to look.

Your post gave me an idea, maybe the device could store the glide angles throughout the dive (and maybe playback the tones it produced with the glides angles it stored) so you can study it after you land. I think after doing that enough times you would start to get a feel for approximately what the glide angles are for the different tones and beeps rates. Still, I agree that knowing the exact glide angle in freefall would be ideal. Also I think maybe the tones could produce more of an instant feedback since a beep takes less time then saying the glide angle; maybe it wouldnt matter that much anyway though.

---

One more thing I have been pondering is the wind factor. I dont even know what to say about it or if it should even be considered. I can see you having a great tracking body position, but with a steep glide angle due to a headwind or vise-versa having a poor tracking position with a shallower glide angle than normal due to a tailwind. Maybe you would just have to know what the winds are doing and take that into account.

michael

for what its worth.....re your question about a small battery powered amp. I travel a lot for work, and have a set Philips earplugs with a noise cancelling AAA battery powered amplifier (about chewing gum pack size). On max volume with aircraft audio or an mp3 player its fairly loud (enough to want to turn it down). The noise cancelling feature makes a definite difference in flight should help with bringing the signal up against the background freefall wind noise. They seem reasonably strong and will take a far bit of handling stress. Last time I looked they cost about A$70..USD55 ish....in most well stocked airport/travel shops.

good luck with making the whole package work...cool idea

regards

Steve

www.frwd.com, Though I offered them a way to track aerials, They dissed it. My firm works with positioning and are quite accurate with the technology. Check out these guys. Though they are shortsighted, they could maybe help or steal your idea.
take care,
space

I was thinking about this. One thought is... I am planning to have the device log jumps as well, so you can look back later. In freefall, it seems to me that the thing you are most interested in is "does this improve things, or make them worse?" If you were optimizing your body position with a wingsuit, for example, the information which is most immediately relevant to you is relative improvement. Later on you might look back and say, "Geez, I only got 2.0 on that flight." But during the jump, the numbers may not be as helpful as I originally thought.

Another advantage of beeps over words is that with beeps we can update the data every second. With words, I think it's going to have to be dictated every two seconds. One possibility is a hybrid design. What if it beeped most of the time, and every ten seconds it played an asynchronous (i.e. the beeps keep going) sound indicating your actual glide angle? It would be like you were listening to a vario and taking a peak at the dial every ten seconds.

Johnny, I've thought about the wind as well. I don't think there's any way we could factor it out even if we wanted to. But the more I think about it, the more I think we don't want to. Wind is part of the environment, and it's a variable that should be experimented with. I think perhaps one advantage of using beeps instead of an actual number is that it puts you in the domain of relative change. So, while you may notice that you can't get it beeping as fast as you could previously, you will still be able to optimize your flight for the current conditions. Under canopy, this could be helpful for understanding the relationship between winds and optimal flying style.

For my Geko301 I've built a chest mount with a clear window on it. It's basically an old altimeter pillow with a pouch attached to it. This way I can see my altitude, forward speed and vertical speed while in freefall. Unfortunately the software can't really handle the freefall speeds; the Glide Ratio indicator stays blank.

Surprisingly, I've had no problems with GPS reception even though my body is blocking the view of the satellites.

What's that and where can I find more information? There doesn't seem to be anything on the L&B website.

 

Michael,


The Freefly-team Alchemy has developed a noise-cancelling system with which you can communicate in freefall. They are using it for their freefly- and tunnel-training. I did some training and tunnel time with Alaska Jon in March in Perris, and this system works just perfect.

So maybe this is another solution to think about.

Check out
http://www.raredynamic.com

Otherwise your project sounds super fascinating and I am looking forward to hear about your progress.


All the best.

http://www.alti-2.com/Titan/titan.htm

Of course, it's Alti-2, not L&B. Thanks.

MIchael

I´m using FRWD and its working really well. i recomend.
If someone wants to have one i can help !!
they are making those devices in city of Oulu, near my home.

check www.frwd.com

Also, if you know any event where they can demo this device, they might be interested...they just asked me about this couple of days ago.

The Altimaster Titan is a system that will have several different peripheral options; the Head Up Display, SportVue, will be the first introduced. We'll also bring in GPS, audible, visual, and more.

We'll be releasing a full SDK for anyone who wants to write their own programs for it. Let me know if any of you want to be notified when this happens.

~ Lara

you may want to check out the hmd for palmtops from icuiti: http://icuiti.com/work.html

we have used these in freefall.

-dan<><>

Different people absorb data in different ways.
For example, I absorb data rapidly if its presented as angles. Have you ever noticed that the better automobile speedometers display city speed limits with the needle near horizontal and highway speed limits near vertical?
I absorb data slower if it is presented as "beeps" per second.
I absorb data slower if it is presented as "changes in pitch" (musical tones).

Pure numerical data is almost gibberish to me.

Would it be fair to say that you absorb data slower generally when it's auditory, or can you think of an audible indicator that is not rate or pitch change in a tone? One thought is the method where the device dictates the glide angle to you as a number, but I assume this would fall under the category of "pure numerical data"?

A visual signal would be interesting to experiment with, but I don't think is practical for me at this point.

I was looking at the techincal data on the SportVue used in the Altimaster Titan, and it looks like the LCD screen which is being projected in front of you is of the "custom-tooled" variety and not just an array of pixels. This has the advantage of higher resolution, but comes as a bit of a disappointment to me because it means less flexibility in the display.

I've ordered a relatively cheap pair of Etymotic Research ER-6i earphones, which are the in-ear type. This should reduce outside noise by about 30 dB. The earphones themselves are very sensitive to voltage, which means I might be able to eliminate the amplifier from my system. I'm pretty stoked to see what these things sound like.

Michael

Edit to add: With the in-ear earphones, I've also considered that this will make an audible altimeter almost inaudible. I'm thinking I will have the device dictate altitude every thousand feet or so, as well as a distinctive tone at pull time. Obviously this is not the only altitude reference I use, but should help keep me from fixating on the other data.