df8m1

I had a moment and some good news (well I think it is good news anyway lol) so I thought I would post an update. As previously mentioned, I have been tweaking the electronic design to try and reduce the power consumption of the Sport model. The good news is preliminary testing indicates that the hardware changes to the Sport model will reduce peak power consumption by over 70% compared to the Military model. That "should" double the number of jumps on a set of batteries to get us in the 1000 jump range. Actual field testing will have the final say, but I am very pleased with the results of the changes. The bad news is the new “lower power” components are super expensive!!!! Noting is free I guess lol. Really looking forward to testing this year!

Good questions! First I would like to clarify that in order for the AAD to "disarm" it needs to see a good canopy over head. How I determines that I can not say as there are "others" reading this too lol.. So until it sees that you have a good main canopy, it remains "armed". However, if you have a malfunction this AAD will calculate at what altitude it will need to fire at for a reserve canopy over head by 300 ft AGL, (as apposed to the primary reserve over head altitude of 500 ft AGL). This delay allows the jumper to use any available time to deal with the malfunction and clear the air above the reserve before reserve deployment. Regardless of whether the jumper has been able to clear the malfunction or not, it will fire if the jumper is still in danger at the point where the calculated altitude is reached. This delay is an attempt to allow a higher activation altitude while reducing the chances of a two out. Back to the cutaway: Where the Cutaway awareness and the 4 seconds comes into play is after you have a good main, you experience a problem during the canopy flight, such as a mid air collision, or something that you did not notice during your canopy check. An AAD is not intended to activate the reserve on behalf of a perfectly capable jumper or a properly equipped rig, meaning that it is not an RSL, nor is it meant to "pull" for a jumper that is perfectly capable of doing so. Based on your scenarios and concerns with them, the best back up device would be an RSL. (Something else for everyone reading as some are new jumpers... Review your emergency procedures, and you decision and action altitudes. It is "not recommended" that one cuts away bellow 1000ft. RSLs, and MARDS might have people thinking that recommendation does not apply to them and that is not so.) However, there are times when a jumper may not want the reserve deploying immediately, and therefor do not use an RSL, those considerations are some of the reasons for the 4 second delay. There are some that believe that an AAD should act for the jumper and not allow the jumper the option of delaying for what ever reason. Personally, I do not like the idea that a device will subvert my intention. I believe that an AAD is there to act on behalf of a jumper who can not, for what ever reason, take action in time, and up until "the point of no return" is reached, believe that the jumper should be the "pilot in command". When I started jumping AADs were not trusted, now new jumpers believe that their AAD will save them regardless of the situation they put themselves in. I am not sure how to change that mindset. Very good questions! Apposing viewpoints welcome:)

Exactly. One of my requirements when I started to work on this AAD is that it would "resist" firing in an aircraft regardless of the aircraft's altitude and speed, or firing speed setting. Other requirements were that it would be able to identify Wingsuit flight, RW/VRW flight, and canopy flight on it's own, in real time, and be confident enough in it to not need an audible. I think that raising the firing speed for a HP canopy only reduces it's effectiveness when it counts (when there is not a canopy over head). I prefer being able to identify when a good canopy is over head and disarm; except for when a cutaway has been detected, at which point it will rearm, and if a reserve deployment is not detected in 4 seconds, if the conditions confirm the jumper is in danger and within the firing altitudes, it will fire. Right now my Wingsuit firing speed is 27mph (for reference), and you will be able to swoop the pond without fear of a two out or compromising your protection at any point during the jump.

That is an interesting request... The problem with a "wild west" approach to AAD activation altitudes and speeds is it increases the chances of an in aircraft fire if the "custom setting" is outside of the pilots knowledge when descending with jumpers, and possibly doing so in stressful conditions. AAD manufacturer's are concerned about liability as it is, now factor in the jumper's ability to mess with the activation speeds and altitudes as a matter of "normal operation". However, I can see there being some extreme situations where being able to alter the activation settings would increase the effectiveness of the AAD in a unique situation. If you have a unique application, perhaps one of the three current AAD manufacturers will sponsor you and make you a custom settable AAD?

As I am tightening up / fine tuning the settings for the Sport AAD, I keep having the same thought go though my head and I am interested to know what others think about it. Would it be better if there was a standard AAD X Ft AGL "Arming" threshold, a standard "Expert" firing speed, a Standard "Disarming" altitude, etc.. ? Now I know perfectly well that each AAD manufacturer chose their settings for some reason, if not only to be different, but in regards to jumper training it just muddies the water. I recall several discussions at the picknick table at the end of the day about what the arming altitude is for X AAD, and the disagreements that quickly popped up lol.. Yes, it is a matter of "reading the manual", but wouldn't it be nice if at least the arming and activation altitudes were standardized?

Good questions; and ones that I have been asking myself for a while now lol. Because the sport AAD hardware and software will piggyback on the military free fall AAD it’s evolvement time period should be accelerated and I want to be jumping Beta sport AADs this spring. Now I know I have said that in the past, however the opportunities to apply our technology to the military applications swooped in and we pushed the sport AAD to the side. Now that both military AADs are at a point where we are tidying up things, the sport AAD is front and center. I made the decision some time ago to self fund instead of taking on additional investors or outside money because I wanted to be able to take the time I feel/felt is/was necessary to achieve the level of performance and quality that I want before I would trust it on the backs of other jumpers and I put my name behind it. I decided that it makes the most sense to get all three AADs to a demonstration level to prove they can do what I say they can, and see what opportunities, if any, can be identified that would be beneficial to my team and investors, which is where my looking at different business models comes into play. How we sell them may very well be dependent on a strategic partnership yet to be identified; for example, Bill Booth has Vigil America, SSK has Airtec, and Alti-2 jumped on Mars. Is there someone else that wants to get in on the next generation of advanced AADs and snatch us up? Time will tell, but for now I am focused on getting an airworthy sport AAD in the air. The right partner / backer could greatly accelerate the path to market for the sport AAD once it is ready. The exciting part is we are a few months away from getting in the air. I know that doesn’t answer your questions, and I don’t mean to make excuses. If you are asking about a release date because you are looking for a new AAD, my advice is, if you want an AAD to not wait for mine to come out and get one of the three currently available, which you can sell once mine is available : ) Although this AAD will work fine for “Traditional” skydiving disciplines, the target markets will be Wingsuiters and High Performance Canopy Pilots as it will operate in all phases of those jumps without compromising performance at any point between the plane and the ground, and will not need/use any peripheral devices, such as an audible. Keep an eye on this thread as you never know what tomorrow may bring!

I am always surprised how fast time goes by when one is having fun lol.. I didn’t mean to go this long without an update, so I will bring everyone up to speed. My team and I have been focused on the military AADs (Static Line and Manned Free Fall) fine tuning the hardware and firing algorithms, and I am pleased with how each AAD is turning out. I am trying to be mindful of the workload, as I do not want to burn my team out, so the evolvement has been slower but steady. Now that my confidence in the military AADs is high, I will be turning my attention to the Sport AAD hardware and firing algorithms. It was my initial intention that the military and sport hardware would be the same, but the features that the military is asking for require hardware and software that is not of value to the sport market, and in an effort to keep the cost of the sport AAD down and reduce power consumption, dedicated sport AAD hardware is needed. Fortunately, the changes are mostly elimination of components, such as the BLE, and changing out the 32MB Micro SD-card with smaller but less power hunger onboard memory, so the revision design effort will not be bad. It just takes time to make the changes and have them reviewed, update based on the results of the review, resubmit for design review, yada yada.. Very important process, but can take some time. I also have been making some changes to little things from a “fit and finish/feel/perception” perspective. So far the changes have increased quality feel and performance while reducing the cost. Details are important to me and I am glad we are taking our time to get it right. After taking with many jumpers of varying backgrounds, it is very apparent that the skydiving industry is not ready to embrace some of the things that I wanted to carry over from the automotive industry, mainly the remote monitoring of the AAD self test results. Infrastructure challenges aside, jumpers are just not ready for that, so I am shelving that feature for the sport AAD. One positive side to doing that is I can eliminate the BLE circuitry and code; however there will still be a Micro USB connector in the interface. As if all that is not enough, I am also looking at different business models in an effort to identify which ones make the most business sense as we come closer to going to market. There are a lot of moving parts and the right strategic partner can increase the success dramatically. I keep telling myself “if this was easy everyone would be doing it” lol..

I have been getting a lot of questions such as "what makes this AAD any different?" and "Can a jumper change the activation altitude?".. I am finding that I am too close to this thing to keep it simple when I try to answer those questions lol, so I thought I would post some graphics from one of the PIA seminars I put on in Dallas. Question: Can a jumper adjust the activation altitude? Answer: No.. This AAD actually adjusts the activation altitude, within a window, based on the jumpers descent rate. At a descent rate of 170ft/sec or slower, the activation altitude is 998 ft AGL. Descent rates above 170ft/sec but bellow 250ft/sec will result in an activation altitude between 1200 ft AGL and 998 ft AGL. Speeds above 250 ft/sec will activate at 1200 ft AGL. The goal being to have an open reserve over head by 500ft AGL regardless of the jumpers descent rate through reserve deployment. The current AADs use a fixed activation altitude regardless of the jumpers descent rate which puts all the variable tolerance (how far a jumper will travel during the reserve opening sequence) on the bottom end (between the fixed activation altitude and the ground). In contrast to that, by automatically increasing the activation altitude as the jumper's descent rate increases, I are attempting to put the variable tolerance on the upper end and provide a reasonable cushion between jumper and the ground by the time the reserve is open and over head. There is an exception to that where the AAD will delay activation, and that is if a main deployment was detected prior to reaching the activation altitude. This delay is based on the jumper's real time descent rate and a reserve canopy over head altitude of 300ft AGL. This delay allows any usable time for either the main to open (if the jumper pitched low and the canopy sniveled) preventing a 2-out, or allow the jumper to use any valuable time to try and deal with a main malfunction and clear the air above the reserve prior to automatic activation in an attempt to prevent a duel entanglement malfunction. This ability eliminates the need for the jumper to manually adjust the activation altitude. I would like to note that the jumper will be able to adjust the DZ elevation for a remote DZ which is one way some jumpers are raising the activation altitudes on their AADs now, but it should be noted that doing this also raises the altitude where the AAD will no longer fire, usually around 300ft ish AGL. Question: "What makes this AAD any different?". One thing is that this AAD is able to identify where it is during a flight. The Situational Awareness graphic shows the flight mode changes that this AAD actually identifies every jump. This ability allows this AAD to resist firing in an aircraft regardless of the altitude and descent rate. It also allows for the detection of a main deployment, if it has malfunctioned or has opened and is flying, and if so, locks out the ability to fire regardless of the jumper's descent rate while under canopy (preventing a two out due to a high performance landing). It can also detect a cutaway, and if a reserve deployment is not detected in 4 seconds, it will determine an activation altitude based on the jumper's real time descent rate and a 300ft reserve over head altitude, and if a reserve deployment is not detected by that altitude, the AAD will activate. This allows for a delayed reserve activation provided the altitude is available, as there are times when a delay between a cutaway and reserve activation is beneficial, and I do not want to take that "pilot in command" decision making power away from the jumper. I am however comfortable in saying "times up"..lol. The Dynamic Activation graphic below shows the difference in activation altitudes based on speed and if a main deployment has been detected or not.

I jumped an Alpha all last summer and there will be some Military Betas in the air this summer. I pivoted our focus from the Sport AAD to the Military one as the U.S. Military is looking for a replacement for the Cypres-2 and Vigil AADs. (The Sport and Military AADs are different in regards to hardware and software). I would recommend that you go ahead an get an AAD, and when this one is ready, you will always be able to sell the one you got as a stopgap. I am excited about live testing the Military AAD this summer.. If the U.S Military does not give us the nod then we will pivot back to the Sport version. I am very pleased with how both AADs are evolving as we improve / refine the algorithms, and address issues that could arise from potential scenarios; of which there are an incredible amount lol.. The algorithms are performing at a level that I never thought possible when we started these projects.. It is like watching a child grow up lol.. I will try to keep everyone updated this summer.

And now a jumper who is whistling in can think to themselves, "Boy... I'm glad my AAD battery is only 13 years old" lol.. Batteries are cheep IMO

You are not a race car guy are you? lol.. the seat belts in "daily drivers" are pretty much the same as you pointed out, but a 5 point race car seat belt is very different, as it is designed for a wide variety of extreme conditions.. I like you seat belt comparison.. the current AADs do just fine for the type of jumping that they were designed for (and I have consistently said that ), just like standard seat belts do just fine for every day driving, which they were designed for.. But like a race car 5 point harness that is designed for wider range of extreme conditions, this AAD is designed for wingsuit and HP Canopy flight conditions, and can do both without specific modes that only let you chose one or the other. Is it overkill for your type of jumping?... sounds like it.. This is a good conversation as I can see that I am going to have to figure out how to better covey that this AAD will handle "pro, wingsuit, and CP" conditions without needing separate "modes".. they are all handled in one "mode' (to keep the terminology consistent) I am in the camp that "Student Mode" AADs are not a good thing, but that is a different topic for a different thread.. I do not recall ever hearing of a Tandem AAD not doing the job, so I see no need, at this point, to have a Sport Tandem SKU on the shelf.. There will be military tandem AADs for sure, but until the sport market indicates that the demand is there for an advanced tandem AAD, I think the current AADs have it covered. This is a competition race car level product, and you are correct that most of the jumpers are covered by the current AADs. Jumpers that are strictly looking at price have three other options, the ones that want more performance will have an option soon.

I actually think that so much flexibility in an AAD is a horrible idea. If that was possible you'll have all sorts of misconfigured AADs and a spike in misfires. People don't know how to turn off their VISOs, why should we play with AADs settings that freely and dangerously? Multimode AADs and user selectable altitude are already a possibility. And despite that simplicity there were misfires. Well if they offer a wingsuit AAD, then most of the demands for AAD modes will be met. But to date, aside from the really overpriced and quite strange Cypress WS AAD, there are no WS AAD modes. Set your AAD to pro and you risk no fire on a WS. Set it to student and you risk a two-out if you swoop. Why not have an AAD mode that activates at 60 MPH instead? It will allow canopy work that is within the realm of what most WS fliers would be doing, but not so low that it would risk a two out. If I think of the foundation of design and operating philosophy of the current AADs like a family tree, even though the three AADs that are available today have their own branch, they are all running parallel in that at their foundation, they are using the same information to base decisions on. This approach was sufficient 20+ years ago.. but today with High performance canopy piloting and Wingsuits as individual disciplines, as well as potentially blended, the limited functionality leaves little recourse but to try to "split the difference" in the firing speed, which has the propensity to make it inert to in one phase of the jump or another. My approach is vastly different, and my branch will be 90 degrees off of the trunk of the AAD family tree. My target markets are the wingsuits and HP canopy pilots. You mentioned price and I see it this way.. It comes down to the value proposition.. This AAD will not be cheep, but it will be within range of the current WS AAD, only it will offer a lot of advanced functionality for that price. What are you buying for the money?? is it polished 20 + year old technology (which still works fine for the types of jumps it was developed for), or is it up to date with increased capability that is compatible with today's disciplines? This AAD will be able to handle any type of jump from an aircraft without modes and without operational compromise or peripheral devices that indicate if the AAD is doing what it is supposed to do, (not that their is anything wrong with that, I just see it as unnecessary). This AAD collects enough data, that if properly processed, you can generate a 3-D image of the jumper throughout the jump. I look at it like the current AADs are equivalent to flip phones, and this AAD is a smart phone. Each will have their place in the market. Technology has changed a lot in 20 years. I am just applying it to an AAD and utilizing its capability.

I actually think that so much flexibility in an AAD is a horrible idea. If that was possible you'll have all sorts of misconfigured AADs and a spike in misfires. People don't know how to turn off their VISOs, why should we play with AADs settings that freely and dangerously? Multimode AADs and user selectable altitude are already a possibility. And despite that simplicity there were misfires. First I would like to say that I really do appreciate every thought and comment and encourage everyone to post them. I am in the camp of most people should not be allowed to mess with AADs... What could possibly go wrong lol.. Predictable performance is a big deal. If someone is jumping a reserve that takes longer than 3 seconds to open, I'd say they have other problems that need to be delt with... Needing different firing speeds during a jump is a different, and very real problem all together.. Confidence in the AADs ability to correctly identify the flight mode that the jumper is in, is paramount. I think that a wingsuit pilot should be able to swoop the pond without sacrificeing activation performance during any phase of the jump... And I am doing it without an audible...

How about a touch screen?!

That is a work around for the current AADs. This one is very different.. "dynamic" comes to mind when describing it, but I am not sure that is the right term.. If you read back though the thread, I have described how the firing altitude is arrived at and self adjusts within a window of speed and altitude in the effort to get a canopy over head by around 500ft. If it detects a main deployment just before the primary firing hard deck, then it will delay firing until the point of no return hard deck, which is dependent on the descent rate of the jumper. The point of no return hard deck is based on a lower canopy over head altitude. This delay gives the main as much time to open without automatically getting two out, or gives the jumper as much time as possible to deal with a malfunction in an effort to avoid firing a reserve into a main malfunction unless that is the only option left. It can also determine if the jumper is under canopy or is flying a wingsuit so there is no need to sacrifice performance in one phase of a jump in order not to fire when it shouldn’t in another. There is no need for Pro, Expert, Wingsuit, etc settings.. I also don’t see a need for a Student Mode or model. The only Mode it would need to be put in would be Tandem, if I decide that I want to offer it for tandems; the only difference being the firing altitudes for tandems. Now I can only imagine the dressing down I am going to get for not wanting (at the present) to offer student or tandem AADs lol… Thank you for your suggestions! Do you have any questions regarding how I have previously addressed the reasons for wanting manually adjustable settings?

Thought I would take a moment and post an update... I have been jumping a rev-1 alpha military EEAAD all summer collecting jump data and evaluating battery performance. Last April all of the AAD manufacturers were invited to come and meet with Army Research to discuss the performance requirements for the EEAAD, and to “pitch” each of our AADs.. It sounds like the Army has started the RFP process over and is adding the features that they liked from each AAD to the requirement, so we are in a holding pattern until the new requirement comes out. Although the military EEAAD is our primary focus at present, for obvious reasons, the sport version is not to far behind. Based on what we have learned testing this summer, some changes will be made to the sport version, specifically to enhance battery life, as the military and sport operational requirements are different. This summer has been fun, and I have had a lot of great conversations with many people about this AAD and AADs in general. On several occasions it was brought up that it would be great if an AAD was capable of cutting away a main if it spun up (regarding super tiny canopy test jumpers).. When I tell them that this AAD platform as two separate firing circuits, and with the right program, could handle that no problem, they get real quiet lol… It is like they were asking for the impossible and I say “our AAD platform could do that”, and they don’t know how to react lol.. Algorithm enhancement is down to fine tuning, splitting hares almost in some ways.. My focus is on reliability, performance, and simplicity; it’s nice when all three can be increased with the same change. Now that winter is almost here and jumping is reduced to only “as required”, the focus goes back to hardware and data analysis. Data from every jump is run through the simulation so I know on a weekly basis if a problem has been identified, so the algorithms are always up to date. I also will continue to generate custom data sets to create specific situations in order to see how the algorithms handle them. It won’t be much longer now

Please clarify the distinction between deployment and deployment detection, and how canopy detection differs from

Full Disclosure: I am working on a new AAD.. This discussion made me take a look at the AAD altimeter readings from the point where the deployment is detected and when the vertical descent rate is

Thanks for the input. I was not sure if Vigil retained the "off set" for a different DZ landing altitude. This AAD will have a preset reserve activation of 1000 to 900 ft, and will automatically increase the activation altitude proportionate to speed above XXX ft/sec to compensate for the additional distance traveled during reserve deployment due to the higher speed. The goal is to have an open reserve at a consistent altitude regardless of the jumpers descent rate at the time of reserve deployment. The Max increase will be 250Ft which would be at the Max TSO Placarded speed. This automatic increase in altitude above XXX Ft/Sec speed is to help prevent jumpers going in at reserve line stretch after an AAD fire at extreme free fall speeds. Now I know there are people probibly screaming at their monitors that automaticly increaseing the Resereve activation altitude will increase the chance of a two out situation. To address that, this AAD is able to detect a main deployment, and if it has detected a main deploument it automaticly adjusts the Reserve fire altitude to a lower "point of no return" altitude which again is dependent on the jumpers descent rate. If they have a high speed bag lock, they will not have as much time to deal with it as they would with a lower speed malfunction. The Smart Reserve Activation altitude in intended to allow as much time for a normal AAD fired reserve activation to occur, prevent a reserve fire from low pitch and snively canopy, and allow the jumper as much time as allowable to clear a malfunction before the AAD fires the Reserve should it be necessary. Now that I have gone off into the weeds.. My original thinking was to maintain the DZ Offset until manually cleared or the AAD is turned off. The key aspect with this AAD, as with any piece of equipment, is to not only know how to use it, but to know what it is going to do. It is starting to get exciting now that I am at the point where I have to nail down the little details..

Quick question I hope... In regards to offsetting DZ altitude... As far as I am aware, every AAD currently available requires the user to set the off set before every jump, or in other words, the offset clears upon landing and will need to be reset before the next jump if the jumping back into the same DZ from the same airport. I do not recall hearing people grumbling about having to off set the DZ altitude for every jump, and there has to be a few DZs that take off from one location and the DZ is at a different altitude. Would it be preferable to have the offset remain until either the jumper clears the off set or the AAD is turned off? Or is requiring it to be set on every jump not a big deal? Thanks in advance :)

What you said lol... I built some parachute flight data recorders for the Forest Service Smoke jumpers several years ago now. It had a pitot tube that measured indicated air speed. It was calibrated in a wind tunnel. Depending on what you are looking at doing, you may want to look at model rocket / airplane airspeed measurement systems.

Sorry for the delay.. I believe C, or some form of it anyway... The cool part is the firing software is not part of the operating system software, so we can put the firing software on any hardware platform that has the instrumentation we need and can supply data at the required rate. To demonstrate this they are going to install it on a smart phone.. For jumpers this means nothing, but to a competitor who would be interested in buying the firing software, it makes it really easy to integrate into their AAD platform.

It all depends on the operational range of the AAD. If the goal is to mimic a mechanical AAD, then the current AADs do that very well, there is one for every individual discipline, but when used outside that discipline the performance may not be a good match. For example a wingsuit and a speed AAD. But if you want one that will handle multiple disciplines on the same jump, guard against firing in an aircraft, reduce the chances of a two out because you set the firing altitude higher, etc, with out sacrificing performance at any point during the jump, then it gets a little more complicated lol.. The current AADs work well when used within their specific performance windows, and I would say that covers probably 80% of the sport jumpers, but does not meet the new military requirements. The military is requiring higher performance AADs that will not fire in an aircraft. The trick is to make the complex as simple as possible. Ten years ago we covered a 4X8 white board with logic written in small print. Today, we are exceeding the functionality of all that logic with refined logic that takes up less than 1/4 of that space. Our process is to identify function, for example, to delay reserve deployment if a main deployment was detected in an effort to prevent two out at 900ft because the jumper was low and sniveled big time... Then come up with a process that will achieve that, (not worrying about complexity at that point). Test the logic using test data in a simulation, and over time refine the logic over several iterations. We are still in the testing the logic phase for a lot of this system, so things may seem overly complex and awkward, (for lack of a better term). Two months from now I will probably look back at the logic I am playing with now and say to my self why the hell did I do it that way??! lol The process can seem messy and can make some people uncomfortable lol.. I guess that is why manufactures keep things behind the curtain.. I think there is value in the questions and comments.. Monosa's post has me thinking again about some things that I set aside until we were ready for them.. I think it is better to have a lot of options and choose wisely, then to not have any choice. Fixed speed and altitude works for the majority, I think everyone else shouldn't have to compromise.

Good questions!! I don't really want to get into too much detail as to our processes and such, but I will attempt to walk the line so to speak lol.. I am not a programmer so I am out of my area, but in regards to validating the algorithms, making sure they respond predictably and consistently, for example after a change is made, we have "gold standard" data files that we rerun in place of sensor data which allow us to identify any differences in the performance of the code. The processor thinks it is reading data real time, as if it were in the air. If there is the slightest difference in the out put we will see it. The programmers also asked me to make special test data sub sets and that can be injected at any point during a test, even garbage data to simulate a bad sensor reading. This is a very time consuming process that has to take place every time a change is made. Nothing really special about the process.. This is of course in addition to use of best coding practices and external debugging and validation software. As for redundant hardware, and how it would react to a malfunction.. The only actively redundant component (at this point) is the baro pressure transducer, and that is mainly due to the hardware being used for multiple applications. (One application uses one transducer, another uses the other)... One benefit to having two is we can compare the readings during the self test.. If they differ too much, then a fault code will be set and the AAD will shut down. If the baro pressure transducer that is being used drops out during a jump, the other one will be used until the jumper has landed, at which point it will shut down. The system monitors every communication action on every buss, and records faults if there is a momentary problem detected. With the Sport AAD, the baro transducer is on a different buss than the other sensors, so if, lets say, the accelerometer froze and did not release the buss mid jump, then the AAD would go into limp mode, and become a good old fashioned speed and altitude AAD, relying on the baro sensor for the remainder of the jump. Same would go if the baro sensor froze, then the back up one would be used until landing. The alternative is to just shut down, which we may end up doing for simplicity sake. Just because something can be done, doesn't mean it is the best course of action for every situation. We are testing different ways to deal with different scenarios, and evaluating them as part of the development process. If the processor were to crash and reboot there are a lot of different things that "could theoretically" happen, the most likely would be that the power would turn off and that is that... The worse case is if it "hiccups", where it stops looking at the data for a period of time, then comes back on line and sees a dramatic difference in readings and reacts unpredictability. That is why we do the testing mentioned above, to see how it reacts and put in place checks to detect the hiccup. The unsaid concern is "will/can/would it fire when it should not?". The firing circuits are designed with several layers, requiring a sequence of events to take place in the correct order, in order for the cutter to fire. In my mind that is the best defense if the processor should really shit the bed. That being said, there are other ICs on the the boards that have processing capability that could be used as a check for the main, which is something that we have talked about. That however adds more complexity, but it may be worth it.. Aircraft have three computers for this very reason. It is the "what ifs" that keep me up at night.. The systems programmer has worked on guided missile systems, military radar systems, and autonomous vehicle systems. He has said that this AAD is more complex than anything he has worked on before lol.. It has to almost think!

After reading your post I am having second thoughts!! I guess if it were easy everyone would be doing it lol... And you are correct about the market being small, especially when compared to the development costs. When I talk to potential investors about the market size the standard answer is "it's too small". A comment was made pointing out the military market... Currently the USA pays around $5,000 for a military Cypres, and the military are tired of that. At that price point, yes, it is worth doing.. but the military is looking to drastically lower the cost of AADs, and they order sparatickly in little batches which is not ideal.. I won't speak for the other AADs, but our military and sport AADs are quite different. The military has requirements that the sport does not need, and the sport has requirements that the military does not need. Each system is designed and validated for the end user market so it is like developing two separate AADs, with two different revenue streams to pay for their creation.There is a reason why the current AAD technology has not been updated in 25 years lol... Fortunately we are at the fun part now. I have been test jumping an Alpha unit and we are building drop test dummies that are brave enough to get out at 500 ft to test the Static Line AAD, and 1,500 ft for the sport AAD. The cutter is coming along nicely and summer is right around the corner.. or so they say.. But seriously, we do appresheate everyone's words of encouragement. Sometimes I stop for a moment and ask myself, what the hell am I doing?!