Hawk Vario Optimism

[Britton Bluedorn]

Like others I’ve had the hawk vario for a bit now and have learned a few things. I recommend that everyone perform a test and display both the hawk AoA and AoS navboxes on your screen and observe the readouts while in flight. It’s pretty telling.

For my glider when flying wings level with the yaw string straight, the AoS readout shows about 4deg. It really takes a good bit of rudder to make the hawk beta value go to zero, and then the yaw string deflection is obnoxious.

I believe the yaw string as truth for indicated beta angle when flying straight, and the readout of hawk beta is incorrect in my glider (more on why comes later in this message).

My hawk Vario optimism is definitely skewed when circling to the right and better matched circling to the left when beta is reported to be smaller.

Digesting Professor Heinrich papers and talks, he’s got the math and theory right, but he also assumes a perfect installation too. This is where LXNAV needs to step in.

We have the leveling procedure to get the alpha correct, but there is no adjustment for installed beta that I’m aware of.

The pitch axis is easy because it can be read easily on the ground as installed theta angle relative to a ground plane, hence the leveling procedure. Beta on the other hand needs a bit of understanding and probably flight trials to dial in.

Heinrich’s argument when thermalling is that the blue needle is if you flew more accurately (reducing the beta) would get the TEK needle to become closer to the theoretical best climb of the hawk needle. So beta *is* being taken into account with the blue needle display (via nulling beta).

In the real world, our machines are hand crafted and have some slop built into them so everything fits together. My glider specifically was involved in an accident prior to my ownership that tweaked the canopy in its open position and it’s never been like factory fit since. I’m the 3rd or 4th owner and indeed you can almost see the skew in the instrument panel with the canopy off. Whether or not this is from the accident or from the prior owner’s adjustments I can’t tell. I need to measure with a tailors measuring tape from the nose and send to LXNAV to prove to them not every glider is perfect. (You’ll remember we had to do the same for the pitch leveling argument too before they did something about it).

I believe that if we had additional configurable alignment settings for beta (just like we do for alpha), then we could dial in our ‘installed beta skew angle’ and the in flight optimism would be much reduced and useful.

Until that happens, I fly with Hawk in cruise and TEK when circling.

Britton


P.S. I’ve been meaning to send my thoughts to LXNAV, but need to take some measurements first before I inundate them with my engineering logic. If anyone else has measurements of their panels (nose to panel left corner, nose to panel right corner; yes I’m assuming the panel is flat which is another headache) feel free to send my way or beat me to the punch when emailing LXNAV.

You’ll notice in the leveling screen small numbers in the upper left corner. R, P, B which I assume is roll, pitch and beta (though I struggle to understand how they can compute a beta while stationary upon system initialization). I also wonder how much the roll value is influencing our readout too, as mine is definitely not zero but 2.0deg after I’ve leveled in roll via an external level. But let’s argue for beta adjustment first before roll adjustments.

I also might experiment with the 3D printed shims just for fun too until LXNAV adds these alignment features into the software.

[John Mclaughlin]

I think the Hawk optimism only happens in some thermals and Hawk is fine in other thermals. I guess sometimes, there's something about a vertical gust which can fool Hawk into believing this is a climb, when it was only a gust. The initial vertical acceleration must be of great importance to Hawk in determining the presence of a thermal and if the vertical acceleration is not sustained, it should detect this (like we feel an elevator start and stop, but nothing in the middle). Something seems to go wrong with this process and once Hawk has decided that you're climbing at 6kt in a 1kt thermal, it will continue to do so until it eventually sorts itself out. I like Briton's theory about beta angle, but in that case, the error would be there for every thermal. For me, for most thermals, the error is not present.

John

--
You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/519479B9-3449-4EE9-886B-D1A102B65E2C%40gmail.com.

[Britton Bluedorn]

Hi John, yes for my case it’s so consistently wrong when circling right (and not left) I think my issue is the initial beta offset when flying straight. I still recommend everyone show the AoA and AoS value to understand the trends. Too bad we don’t get this info written into the log file. 

I do like your idea too. Makes me want to tear into the unit and find the part number of the MEMS unit and look up the specification data sheet for their accelerometer feature. 

There’s typically a range that they can measure, and if the accel is too large it pegs to the max for the duration. I’m not sure about any hysteresis once it’s out of limits, or if it can successfully read the other axis if one is pegged. If I see my instrumentation friends at work I’ll ping them. 

I’m used to looking at time history strip charts from flight test data. Would love full access to Heinrich’s full rate data to back out what’s going on under the hood. 

Britton 

On Jun 19, 2023, at 1:05 AM, John Mclaughlin <9whit...@gmail.com> wrote:



[John Mclaughlin]

Hi Britton,

I did try flying with Hawk sideslip angle displayed and there was no discrepancy between it and the string - I stopped looking at it because the numbers seemed reasonable. I did this mainly to eliminate the possibility I was applying too much sideslip, as has been suggested in the video a few months ago. I think you've possibly identified two separate problems with Hawk. I do like it though and I'm glad I've got it - I just hope they continue to develop it and sort out these glitches.

John

[Jim Staniforth]

I installed the HAWK demo today in the S-10.

Found it mostly to be useless information. Is it optimism or bullshit?

The Air-Glide S was a better vario, and I trusted the winds more.

Jim

[Francois HERSEN]

The most important thing to have a good accuracy with Hawk is the levelling (see 8.7.4). To do that I spent half a day!

François

[Ludovic Launer]

@Jim Staniforth

I don't think your comment is fair, as you said you've only tried it once.
I initially had a mixed feeling, but you have to understand what id does and what it's showing to appreciate it. This is a good source to start:

https://www.youtube.com/watch?v=A82x2xlIMM0&ab_channel=InfoLxnav

And then, as @Francois Hersen said, you have to make sure the levelling is done correctly, and then set the Hawk parameters to your liking.

I have for instance Wind Variance = 0.05 which is far from the default value. But that's the way I like my vario needle to move.

Having flown with the Hawk for 2 seasons now, it is for me a must have. It gives valuable information both for wind and vario.

I love the ability to have 2 needles on the vario as they're not indicating the same thing it's very useful to detect gusts and center weak or changing thermals.

Regards,

Ludovic

[Ian Molesworth]

Having installed Hawk more than a year ago and having flown in  Ridge, Thermal, Wave and convergence. With the instrument in my LAK17A I still prefer it to the old way of doing things. However, when everyone was complaining about not being able to zero out the panel tilt. I seemed yo have little problem in spite of the LAK panel tilting away at the top massively. ( 20 degrees or more ) I zeroed the system in the attitutde I could get to. With probably 10 degrees of tilt on the sensor. Hawk worked pretty well and wasn't optimistic at all. 

Then I got a 3d printed shim, took out most of the tilt and did the relevelling properly.  Now the Hawk is drastically different. I too have instances where I have a significant positive climb indicator and an ACD-57 altimeter slowly loosing height in very weak thermals. 

Im tempted to relevel in a nose up attitude to see what the real effect is.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/38b33f12-29f1-4125-809c-8343c9255101n%40googlegroups.com.

[Morgan Hall]

One item of feedback I had for LX that I sent yesterday was the notion of test flight following installation and leveling.  It's a good starting point to put the glider in a pitch attitude for weighing, but I would think that what you really want is "level" to be at some inflection point of airspeed on the polar. Obviously this will change with wing loading, but since you can put your loading and W&B into the 90X0, that should be a manageable calculation.  My thought is that you install the V8, level to factory spec and then go fly it with the ahrs indicating level and see what airspeed you are at.  If it is not "60kts" or whatever the right value is, you know you need to adjust the offset +- to achieve that airspeed. 

It seems to me that without going through a calibration and verification that we are operating in unknown regions of the polar and hawk is making serious assumptions based off level meaning something.

I absolutely see the potential of Hawk, but I also have seen massive errors when conditions don't match the ideal that the designer used for his algorithm.  So it needs work.  I already bought it for my Duo, so I'm just going to keep pressing for improvement.  What LX and the professor that designed it need is a bit of humility about the feedback they are getting.  The initial response tends to be some variation of "RTFM" and "you are flying wrong" and while it is important to read the manual and understand the complexity, it should also pretty much work with defaults and basic setup of leveling the glider.  Flying with too much sideslip should be expected to degrade your climb rate by .5 or 1kt, maybe more, but not 8kts.  So reports of massive discrepancies need to be looked at closely since that's not reasonable for piloting issues.

I suspect there are more parameters that need to be added to move Hawk in the right direction.  Maybe Kalman filters are truly magic, but it seems like having the ability to compensate for pitch, roll and yaw errors along with position of the sensor box relative to the CG would all improve the baseline.  A 2 seat glider installation has the sensor box about twice as far from the CG as a single seater.  That has to impact the results as pitch and yaw see nearly twice the movement for a given pitch or yaw change.  It seems like a really hard problem to solve, but Air-Glide managed to do it with their vario for the most part, so I think LX can get there as well and with the sales of Hawk license and new hardware I think the funding should be there.

Morgan   

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAMJg6AFwOeusnnZsCayG9NwokeA8Rh9HhE%3DCkSGvHuKi923g-Q%40mail.gmail.com.

[Jim Staniforth]

Levelling was done with the instrument panel in "weight and balance" attitude (reference angle from the tail boom, calculate offset, prop up the canopy) and level checked in roll.

The AHRS behaved properly when checked in flight.

I used close to Morgan's suggested Wind Variance value 0.1, as 0.11 is not an option in S10. The wind seemed OK, but at one point believed it too much and used the downwind side of a ridge. It did wake up and state the obvious at that point. Work needed! Still, if it was available I'd buy the wind only licence. It isn't.

The convergence was low above the rocks, so it was not easy to evaluate HAWK winds in that respect.

As far as jumping to conlusions too soon, with 30+ hours flying a HAWK 9070 I never got comfortable. Imaginary thermals even on the last day of messing with it. For example, HAWK: 4+ Knots UP. Digital Altimeter: DESCENDING. +1 for digital altimeters!

I update instrument panels all the time, am all for trying new technology, but looking at conflicting information in the panel and trying to fix it during flight seems something beyond unnecessary.

At one point in yesterday's flight wanted to get rid of the HAWK vario needle, but it seemed best to keep evaluating. At least I didn't let it drive audio.

I preferred the Air-Glide S simple wind arrow and number, and the way it updates. Nothing wrong with the vario, after installation difficulty - and now support.

At this point, my S-10's own electronic TE vario reads the air better than HAWK. Will fly first, adjust second (don't hit the rocks) and reevaluate.

Jim

[Ian Molesworth]

I re-leveled my LX/Hawk a couple of weekends ago in the weight and balance attitude again.

My Hawk overoptimism is now wildly out. I am getting 6 knot indicated on the blue line and as little as 1.5 achieved. Now I know I'm not the best thermal centering pilot but this is making it pretty useless.I frequently find myself making a turn into a solid 3 knots indicated only to find the ADC 57 altimeter winding down slowly before I scrape into the 1 knot or so that's really there!

I am going to try re-leveling at a 'normal flying attitude' to see how that affects it. It can barely get any worse!

Is there a way to record the parameters of the system alongside the actual ( barometric) achieved rates and let the developers have a look at it to try and find out what our problem is?

[Richard Pfiffner]

Have any of you that have issues with the Hawk checked for leaks in your Pitot system or mulitprobes.

Leaks will have an effect on Hawk.

From  LXNAV

Maybe multiprobe tube?

These are common root of problems

Richard

www.craggyaero.com

[Morgan Hall]

Richard,

My Duo uses electronic TE and I thoroughly tested my static and pitot system when I installed the panel.  For leaks prior to installation of instruments and then post install checks for pitot accuracy and static. I have access to a certified pitot/static test box, so this was more than just applying pressure from a syringe, I did full leakdown checks using the box.  In my Duo, I'm confident it isn't a leak problem.  In the single seat discus, it does use pneumatic TE and I've leak checked that as well as performed the same pitot static tests. 

I don't think we are seeing these errors because of minor leaks in pitot or static systems.  Just like I don't think thermalling with too much sideslip is an acceptable justification.  I'm not complaining and reporting a bug that I see 3kts relative on the hawk needle and when I turn it's only 1.8kts average.  I'm seeing 10kts, maybe even 10kts on the 10 second Hawk Netto Average and my digital altimeter is winding down on both the ACD-57 and the LX9070. 

Unfortunately the crazy optimism is reproducible, but rather randomly.  Maybe 10% of the time it's completely unreliable.  By using the TEK and Hawk needles and waiting a bit you can reduce the chances of it misreporting, but that means delaying slowing and turning by a couple of seconds.  There are ways to compensate for the behaviors, but that sure is a frustrating accomodation to make for a fairly expensive upgrade.  

The logging mechanism in the V8 is present to capture whacky data, but turning logging on or off is a lengthy distraction of navigating to the menu and password function and then entering a 5 digit code 43001 to start logging.  Not ideal when you may already be in a ratty thermal trying to make it work.  And if you forget to the turn off the logging and let it run for too long, it can essentially disable your vario for 10+ minutes while it writes the log to the SD card.  That's fun. 

I'm sure that LX can eventually fix this. But they do need to take the reports seriously and probably produce a beta software that can capture logs more efficiently because the transient and occasional nature of the problem doesn't lend itself to a long logging entry process.

Morgan  

--
You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/6691d531-77a4-40f8-9ada-fa64d4fb6c7cn%40googlegroups.com.

[John Mclaughlin]

My Hawk (S100) was working perfectly well. Then I updated the firmware to the latest version and it's gone bad - showing lift where there is none. I'll try a sensor reset if the bloody rain stops!  

--
You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/9be98d9b-3dad-472c-bc71-bd279dfbee36n%40googlegroups.com.

[Steve Koerner]

My experience matches what Morgan describes.  Sometimes it seems to work correctly except for a scaling error.  Other times it just goes to hell. I wish I hadn't wasted so much money on Hawk; alternatively, I wish it worked reliably.  I've leak tested as well.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAPaUW1TYmZzy2PNRErXfN9b4rduKguLL-7oYCuUFdRCJJPEm%3DA%40mail.gmail.com.

[Ian Molesworth]

Flew today in interesting conditions in North Wales. 

Relevelled the ship with just the tail dolly on, tail low about 300mm below weighing attitude. Hawk far less optimistic and much more believable.

Thats the way its staying until something comes up from LX. 

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAJGEj%3DC-afZByepK8nLqmhfFY2h%3DPGAwV7Ny8MdxBYe_TiTyLA%40mail.gmail.com.

[Richard Pfiffner]

Using a certified leak check pitot/static test box does not necessarily  mean you did a proper leak check.  We found that looking at the IGC file from the flight on SeeYou that the TAS graph many times went to zero.

Here is the way I checked the customers pitot system.    Inserted the Multiprobe or Pitot probe and taped the fitting plugged the pitot hole in the pitot or pitot hole in the multiprobe with tape,  remove the tubing from the PTOT on the V8 and pressured with an airspeed indicator and syringe to 80 kts.  This checks the entire system which included a Clear Nav Vario, Airspeed indicator, and airspeed pressure switch for the TT22 Transponder (we found this leaked).     The Airspeed pressure switch had tubes that were tool large for the barbed fittings.    One we replace the tubing with smaller tubing the 80 kts pressure it held for 10 minutes.

Richard

[Morgan Hall]

Come on Richard, this is like Uros telling me I'm thermalling with too much slip and that is where the 10kts of difference is coming from. 

I've performed leak checks using your method, it usually works fine, although you are changing one item in the system and have to test for that as well.  Using the system that I have access to, the attachments are all external to the gliders tubing system.  So short of the fitting to the pitot or the static port leaking, you are actually testing the entire system as flown.  You are only subject to a couple of locations that might leak and the computer in the test box gives you a leak down rate for both pitot and static systems. In my case, my initial leak check found a leak on the pitot.  The glass of the front cockpit ASI was leaking slightly.  Snugged down the bezel and everything sealed up.

Side note of caution for anyone checking their pitot/static system with an LX or other electronic pitot/static device attached.  The pitot/static sensors on the LX are differential sensors. You should not pull a vacuum on a static line to test the altimeter for example.  That is one way that you traditionally could check for a static leak.  Pull a vacuum to get the altimeter to indicate 1000ft higher than field elevation and clamp it off.   If it returns to field elevation, you've got a leak somewhere.  100fpm of "leak" is the max allowed for an IFR aircraft, so if you are less than that, you're probably fine.

With the LX and some other digital equipment, pulling a vacuum on just the static line will cause the pitot to register the differential and airspeed will shoot up.   1000ft difference isn't likely to cause any damage, but you need to be aware that you can harm the sensors or cause issues with their zero point according to LX. They never said how much was too much, just that it could be a problem.  Probably worth checking with LX if you need to check your static system and verifying how they recommend you check it. 

Morgan

--
You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/ceb3e60e-b037-49b7-89cc-f425149c2933n%40googlegroups.com.

[jpg...@gmail.com]

I have previously had excellent Hawk performance when circling in thermals but on my last XC flight, a week ago, in very weak broken thermals, it was hopelessly optimistic and unstable.   Hawk cruise netto and winds were fine.   I was on beta 9.06f.  I restarted Hawk several times during the flight but it made no difference.  I wondered if a software bug had been introduced(??).   My pneumatics are good.   I haven't had a chance to fly with 9.06g yet.

[Andy Blackburn]

I did a P-S check on my glider after installing the Transponder Squat Switch (if you have ADS-B Out this is an essential piece of equipment if you don’t want the FAA giving you a call). I put some pressure and vacuum on the unit - mostly the latter for a leak-down test. Turns out my brand new glider had a break in a t-connector in the tailboom, just aft of the engine bay. Fortunately that was accessible without cutting holes in the fuselage. I’ve been flying the past couple of years with a static system that was at whatever pressure the inside of the tailboom sees.

Worth checking.

Andy

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAPaUW1SVrnxmRFi_tx7qy3v5waUh5oUveJvLPvRu%3DXWER9og-w%40mail.gmail.com.

[Matthew Scutter]

I also found the best performance on my Diana 2 more or less inadvertently, ignoring the instructions to set it at W&B position (very wrong), and also not setting the horizon so it was visually 'right' (terrible stick lift), but rather in an approximately tail dolly position.

I would be interested to know the pitch angle/offsets anyone flying with a JS3 has, because I cannot get good performance at the moment from the one I'm flying at the moment, having tried three different positions. 0 degrees/uncalibrated was actually quite ok but not as good as my Diana, and I've only succeeded in making it worse so far.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAMJg6AGapkhL5%2BRrUBqXnu-zproJR9aJ9vfONOWsy77VG_hCwQ%40mail.gmail.com.

[Nelson Howe]

I've flown with Hawk for the last year in my JS3.  I tried to use Hawk audio in the climb as it seemed like a good idea.  I found it frustrating because of the optimism:  crank into a thermal work it hard, hear the vario sounding very excited, and then listen to the voice tell me "1.2" ( and this is knots).  I shrunk the range on the audio and found it helped, but the lying continued.  It became something I was having to expend extra mental energy on to understand and work with.  Finally, this spring I changed my setup so I use Hawk audio in SC mode in Relative and TEK vario in climb.  It was a nearly instantaneous "aha" moment for me.  Suddenly the vario acted like a vario and I could make adjustments in the thermal based on what I heard, slide over to the core or know when I moved the wrong way.  It was a vario working like a vario, and I didn't have to expend extra energy interpreting it.  I still use Hawk in cruise and it is very good at keeping me from turning in horizontal gusts.  When I think I have something, I listen to Hawk and glance at both needles, as well as the needle on my Winter mechanical and (yes, I know) my Butterfly.  I have some experimenting deciding when to push the speed/climb button to change over to TEK, but for the most part I am much happier using the instrument this way.

I did calibrate Hawk once with a TLAR approach, and again with the W&B position, and found no difference between the two.

It seems like Hawk should work better as a vario, but for me it doesn't.  It is still valuable, and I use it for winds and, mostly, for deciding when to stop to climb.  I find the Professor's explanation of why it does work and why we are wrong to be an unconvincing argument.  One thing I noticed the other day was that I was thrashing around trying to work a small core, trying for a steep bank angle to see if I could get something more, and the blue needle was reading about 5-6kts while the TEK was barely calling out 1.5.  Sideslip was 10 much of the time as I was thrashing a bit looking for a way to make something of the climb.  The thing is, this was never going to be a 6 kt climb.  The professor explains that if I could fly that with 0 sideslip, I should be able to achieve 6kts, but as soon as I reduced my sideslip and my bank, the blue needle came down to a more realistic reading of 1.5 knots.  Understand, this was not a good day.  But it got me to thinking that the problem is not that we are not flying smoothly enough to achieve the ideal climb rate the Hawk advertises, but that the Hawk is disturbed by any rough or uncoordinated flying, and tells fairly large lies about climb rates as a result.  TEK doesn't do that.

[Ramy Yanetz]

Sorry Andy, every time I hear a mention that the transponder squat switch is a must I’ll point out it is not. It is something that Darryl insisted which I found to be incorrect, I m using the software switch between ground and air for years and it is working fine according to the reports I pull and no calls from FAA. The issue you guys had at Nephi few years ago was due to manually switching (or forgetting to switch). Perhaps someone can explain why a software can’t easily tell if you on the ground (moving slowly, stable altitude) vs flying (moving faster, altitude fluctuates even in wave). 

Those without switch, just turn on the software  option (I don’t recall how it is called). 

Ramy

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/BE4D307D-13EC-4D79-8995-19296B0A23F7%40gmail.com.

[Ramy Yanetz]

Reading all the issues with Hawk I have to say I am very pleased with my decision not to upgrade to Hawk. What’s wrong with TEC varios? They worked reliably for decades. Horizontal gusts don’t feel like lift. Sure I am making many unnecessary turns in sink, more than I am willing to admit, but those are simply unworkable thermals or turning the wrong direction, not horizontal gust. If someone will invent a vario which will help you always turning in the correct direction I’ll be very interested. 

The instantaneous wind is valuable, although I find the instantaneous HW/TW almost as valuable, and it is available for free as a nav box without hawk. How many are using it? I can’t imagine flying without it. 

I completely agree with Morgan, blaming 5-10 knots error on slip is ridiculous. If anything, slips improve climb, but even if someone over slip, it my increase sink rate by maybe  1 knot at most. 

Bottom line, I’ll take a reliable vario which may occasionally register gusts as lift anytime over complex vario which requires endless tinkering. I wonder how many flights were ruined by hawk so far? 

Ramy

https://groups.google.com/d/msgid/lxnav-user-group/4a322661-823c-49a7-b18e-bf892fac3bf3n%40googlegroups.com

.

[lonkelly]

I sm happy enough with Hawk, and find the wind and AI quite helpful. I found this YouTube lecture very informative about modern attempts to build better varios. The initial discussion of vario indications vs seat of the pants squares with my limited experience. As I get more time in the air I rely less on the vario and more on my eyes and bum when deciding when to turn in which direction. Still, I will take all the help I can get and I wonder how the RTK Larus will perform. Looking at the Larus GitHub repos it's interesting they are thinking about temperature and humidity sensors to add to the model.

[jpg...@gmail.com]

I don't regret paying for Hawk at all.  The winds are a lot better.  In the cruise it reliably indicates thermal lift sooner and more reliably than TEK varios.   In the climb your TEK vario hasn't disappeared - both needles can be displayed and you can choose to hear TEK or HAWK audio.   Except for my last flight that I posted about, I normally have used the HAWK because mine usually is better for centring and usually agrees closely with TEK for the overall climb rate.  On my last flight, when it didn't, I simply switched to TEK climb audio for the duration.

[John Ferguson]

My experience with Hawk is the same, 1st got Hawk April 2022, thought this is actually pretty good, glider and I felt in tune with Hawk responses, was impressed. But ... as time has gone on and software updates have rolled in, Hawk as a vario is useless. Yes, done my weight and balance, done the AHRS levelling, haven't flown 9.06 yet as weather is rubbish.

I did change my glider March 2023 and bought Hawk as much for the AHRS as anything else, still not impressed by Hawk as a vario, have gone back to TEK for vario and cruise.

My day job is software, if any of my software projects behaved as Hawk does I would ask devs to roll back the code to a working set and to figure out what changed that altered the desired behaviour of Hawk.

John

[Neil McLaughlin]

Nelson,

Your comment here- "Finally, this spring I changed my setup so I use Hawk audio in SC mode in Relative and TEK vario in climb." 

Can you or anyone else confirm, when you use HAWK audio in SC mode, you are getting a HAWK speed control sound, not the HAWK Vario audio sound, right? As thats what I thought when I tried using it this way, and I've never been a Cruise/Climb switch kind of guy. so didnt like it. I asked LX about allowing HAWK Vario Audio instead of SC on the SC button, but they told me it already does this.. Who is wrong here, me or LX?

Thanks,

Neil

[David Anisman]

John,

Like you, I got Hawk in the Spring of 2022. Had the same experience, very satisfied with it. But, I have not "upgraded" firmware since then, so I still have the same version of Hawk running. I believe serious bugs were introduced since then. I agree that providing the ability to go back to a known good working version should be considered essential so that owners can still have at least a usable form of it.

I see none of the reported issues (Hawk singing 8 knots while glider climbing at 1 knot).  After communicating with Morgan in detail about it, I do believe the issues are real and attributable to recent and  current versions Hawk (and not something else) at least on the 90xx platform. I

Not sure what device are you running Hawk on. I am running it on an S100. If I had the version problem I would try to get LXNAV to provide a way to install a previous version. On the 90xx platform not sure is that's feasible.

David

[Ramy Yanetz]

On the LX9000 about page, there is an option to go back to latest stable version (not beta) at least for those using Wi-Fi. 

Ramy

https://groups.google.com/d/msgid/lxnav-user-group/72054317-dda1-4b2f-97de-85ef5735787en%40googlegroups.com

.

[Morgan Hall]

Hi Neil,

If you want the hawk vario all the time, simply stay in thermal mode. If you set the audio source for the cruise mode to Hawk Relative and the vario audio source to Hawk, aside from volume(which are independent), you won’t hear much of a difference when in cruise mode. If you want more differentiation in sound choose another audio driver, but Relative sounds pretty much the same as vario when either Hawk or TEK is used. 

However I think there are still bugs in the SC audio in these latest betas.  I had some weird audio issues in a week of flying from Ely.  Couldn’t reproduce them reliably so I did not send a bug report, but I still get a constant tone from the system when it should be silent. 

One behavior I noticed was that the speed command chevrons seem to be fixed to the TEK vario.  In cruise you will get conflicting visual and audio cues when Hawk and TEK are at odds with each other. 

Morgan

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/060a6969-f85c-415a-8a74-474a43cf9b50n%40googlegroups.com.

[Nicholas Thomas]

Hi everybody. I have been lurking waiting for the following: “One behavior I noticed was that the speed command chevrons seem to be fixed to the TEK vario.  In cruise you will get conflicting visual and audio cues when Hawk and TEK are at odds with each other.”

I too would like the chevrons to match the audio.

The best part about Hawk are these discussions. Did you screw up the installation, do you suck as a pilot? I have never been so prepared to defend user error.

I flew my whole life on steam gauges, no TE. The clearnav vario was my first and we have been having arguments since its installation. I just bought the S100 with hawk and now the three of them have an argument among themselves. Oddly, this is very calming.

The pneumatic vario is just wrong, so the reply is: “Jane, you ignorant slut.” However the other three lie to me and I have yet to come up with a good reply.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAPaUW1Qg4cRwxze2TLzR98NxeZF2PMotGbRY%2BJARWJ-c4tJShg%40mail.gmail.com.

[Nelson Howe]

Hi Neil,

 

I’ve never found much use for “push/pull” speed command audio.  I used to use SC Mixed when I used Hawk for all phases of flight.  I have since decided the speed command sounds become background noise, and get I more useful information out of hearing the vario sound at all times (no deadband).  In cruise, I use Hawk Relative vario, the notion being that it is reporting at what rate I could expect to climb if I stop to turn.  I choose the speed I want to fly more by using the block STF based on the current MC setting.  That block speed is displayed in a data block on my map page.  If the air is going up, I may be pulling up and flying slower, but I do that as a judgement call based on what I see and feel, not just on what the instrument says.  Likewise, if I find the vario making unhappy sounds in sink, I might glance at the chevrons and see how far off speed I am at the moment.

 

The manual:

 

SC audio mode has four modes:

• SC positive: sound is interrupted with silence every few milliseconds when the needle is positive; on negative side sound is linear (not interrupted).

• SC negative: inverse function to SC positive.

• SC: sound is linear and non-interrupted in full scale range.

• SC Mixed: for positive relative values the sound represents relative; for negative relative values the sound represents SC (for that setting it is recommended to set SC needle to relative).

• Relative: it will beep with same tone as in vario mode for relative value. [I USE THIS]

• Netto: it will beep with same tone as in vario mode for netto value.

• Vario: it will beep with same tone as in vario mode

 

Vario audio source is shown when HAWK option is activated. You can choose between HAWK or TE vario audio source for variometer sound. SC audio source is shown when HAWK option is activated. You can choose between HAWK or TE vario SC audio source for speed to fly sound.

 

Here is hint for one audio setup option, if you are not using SC mode and HAWK is enabled. Set SC audio to Vario and set SC audio source to HAWK. Now you will have audio based on TE once in vario mode and audio based on HAWK once in SC mode. [I DO THIS]

 

This is the setup I’m using.  Didn’t actually know the hint was in the manual until I found it just now.  The only drawback I have in this arrangement is the autoscaling on the map when I push the SC button.  If I am climbing in climb mode listening to the TEK, my map is automatically zoomed in so I can see the snail trail of my circles as I work the thermal.  As I get higher, I like to be able to scale out on the map to think about where I am and where I need to be going next in the task.  The fastest way to do that is to click the SC button into cruise mode and return to the cruise page, which is already scaled out.  But when I do that now, I put myself back into Hawk audio source.  It’s almost as easy to push the HAT button sideways and adjust the zoom.  Almost, but not quite.  Right now, this is a minor quibble I can ignore.

 

Nelson

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/060a6969-f85c-415a-8a74-474a43cf9b50n%40googlegroups.com.

 

[David Anisman]

Nice post.

I think that if the needles and audio in cruise ("SC mode") are set to Relative, a discrepancy with the chevrons - which would continue to indicate push-pull classical speed command - would be expected. Or not? If they are set to SC, then they should match unless the time constant is set to a different value (don't recall if it can). I take advantage of the chevrons rather loosely.

David

You received this message because you are subscribed to a topic in the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lxnav-user-group/nz4npqP4cyw/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lxnav-user-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CABD915C-71F1-480F-92E4-67463996A77F%40hxcore.ol.

[Ramy Yanetz]

Hi Nelson

I am also flying block STF, but only showing a single STF for the MC setting. 

How do you display block speed in a data block? 

Ramy

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CABD915C-71F1-480F-92E4-67463996A77F%40hxcore.ol.

[Nelson Howe]

Hi Ramy,

 

I may have misstated what I do.  I suppose a true block speed would be a range, which I don’t do.  I mean a set speed which changes as I change the MC setting.  This speed is an available data block in “Aircraft Parameters” which I have up on most of my pages.  I use it as a general target speed on a glide.  I suspect that is also what you do.

 

Nelson

To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro..@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAMJg6AGapkhL5%2BRrUBqXnu-zproJR9aJ9vfONOWsy77VG_hCwQ%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/060a6969-f85c-415a-8a74-474a43cf9b50n%40googlegroups.com.

[Mike Clarke]

I also feel that Hawk worked much better last year on my 9070, I don't recall ever seeing 6.5kts up on the vario and the altimeter going down then. I do updates via a USB rather than WiFi, I have kept a lot of old versions on the USB and yesterday I reverted to 9.03a which is before the AHRS system pitch offset was increased to +/- 20 degrees. I only changed the 9070, not the V8. I just need some summer weather to test it.

[John Ferguson]

Do any of the LXNav team monitor this group, it seems many of us hawk users have similar experience, we can't all be useless at thermalling ... it suggests a look at the code would be a good idea.

Mike - I have found in the past that if the nav unit and the vario are not at the same software level I get errors in the IGC files for example.

John 

[jpg...@gmail.com]

That's a thought - the AHRS pitch offset increase was introduced in Beta 9.03c and then in 9.05 stable in April 2023.   I wonder if it is worth re-doing the AHRS calibration if it was last done prior to that change (??)

[Andy Blackburn]

Hi Ramy,

Yes, the GPS ground speed software switch ought to be fine so long as you don’t park your glider in the wave at low or zero groundspeed. I believe in that case you show up to ATC as on the ground because the switch activates. I guess you could re-program your transponder to do manual switching when you anticipate potential for wave soaring. I don’t fly much wave and don’t ever park in one spot, but Darryl suggested I do it after my ADSB-Out PAPR got so screwed up that the FAA called me.

I’m happy with the airspeed rather than groundspeed switch. It has been working great and I just turn my transponder on as part of my pane power-up sequence. And of course had I not pout it in and tested the pneumatics I would not have discovered that my static system pneumatics were AFU.

Now back to Hawk…

Andy

[Morgan Hall]

If it was just a pitch calibration issue across versions then those of us with recent calibrations wouldn’t be experiencing the issue, but unfortunately even with fresh calibrations to W&B specification I have had the issue across three separate gliders. 

I still wish there was a verification step to the leveling process.  My assumption is that the Hawk algorithm expects “level” to be at some general point on the polar and all circling polar calculations are then based off that baseline. 

Oh the joys of testing black box software. 

Morgan

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/7e18361f-e885-4fad-b630-716ab821cee3n%40googlegroups.com.

[John Ferguson]

Flew today in weak english thermals with 9.06, Hawk was very optimistic. TEK showing 1 or 1.5 and Hawk showing 4 and 5 - in knots. Found one good thermal that gave 5 knots on TEK, Hawk was almost in agreement although that might just be because Hawk wants to sit at 5 knots. W and B done, Levelling done, ASG 29 with LX 9000 gen 4, Hawk sideslip varied from 2 to 10 degrees, no noticeable difference in Hawk response.

It does seem something in the circling polar model or calc is off. I presume that circling polar is a generic model rather than being a specific model for each individual glider type. Or could it be the wind calc that is off ?

John

[Jim Staniforth]

Update!

After several flights with this beast, I found that these settings made me happier with it as a vario. Not happy, just closer to being content than originally.

Wind variance 0.05

Wind avg horiz. 20.0.

Wind avg. vert. 10.0

Tried all sorts of variations. Constantly adjusting it in flight is stupid and dangerous.

Tried different setups including allowing HAWK to run the audio in cruise... Just a let down when turning into the incredible lift predicted that was a Knot.

Do people who like HAWK fly in places with weak lift and sink?

I found it best to use electronic TE and delete the idiotic blue needle.

Can't imagine paying for this thing unless only wind is available for a reasonable cost.

Bottom line for me: A (properly installed) Air-Glide S is amazing, HAWK is not.

Jim

[Morgan Hall]

LX sent me an updated firmware on Monday for the V8 that is using an older revision of the Hawk algorithm.  I have not had a chance to install it or fly with it yet, but I am hoping that this is a step in the right direction even if that is backwards to a version that was more stable. 

We shall see. 

Morgan

--
You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/6e7a76c7-ef89-4175-a6be-cbfcded3ee60n%40googlegroups.com.

[Richard Pfiffner]

Morgan,

What is the version?  Please

Richard,

[Richard Pfiffner]

I looked thru the posts and found only 3 times the firmware versions were mentioned.

It appears to me that we are comparing apples to pees.  It would be appropriate to mention the instrument and firmware version in the post.

Richard

[Mike Clarke]

Advice from Uros today was that if reverting to older firmware the peripherals need to revert also. I am now going back to 9.01 (LX9070)  from June 2022 (which seemed good running netto for cruise and climb) and if that works I will try 9.02. And so on…

[Andy Blackburn]

Is there a location where you can download earlier versions.

My problem with going back to last year's firmware is that the instrument quite often cannot get a GPS lock, so that’s worse than not having Hawk. When was 9.01 released? Where do I get it if I don’t have a copy?

Andy Blackburn

9B

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/2e96ec53-b3bc-4106-90c3-daf07a67eb61n%40googlegroups.com.

[Richard Pfiffner]

in...@lxnav.com 

Richard

[Mike Clarke]

I reverted to beta version 9.01a today on my LX9070 (from March 2022) with v9.02 for the sensor box and v9.02 for the V8 vario. I also set it to netto in cruise and climb. The useful performance was excellent, I am not sure if this was because of the version change, moving from vario / relative to netto / netto or a combination of the two. The Hawk needle on the V8 acted like a very fast gust-free vario with a readiong a littl higher than the TEK needle.My next experiment will be the stable version of 9.01 which came out in June 2022.

I will be leaving it on netto / netto for now; this is one of the alternatives suggested in the "Lessons Learned" video. I have a sense that things are not quite right with the algorithm when Hawk makes adjustments to get from pure netto to vario.

[jpg...@gmail.com]

Presumably if netto is operating perfectly it must always over-read the actual circling climb rate.   Id rather stick to "vario" in the climb because even if the Hawk is over-reading I still have the red TEK needle operating entirely traditionally.

[Mike Clarke]

John,

I believe if the vario needle setting on the LX9xxx and v8 is Netto then both the TEK and Hawk are showing netto. I suspect the slight difference I saw with the Hawk reading higher was related to the speed of response on the Hawk. I tend to look at the vario when I feel an upwards surge. It might be nice if we could select vario for the TEK and netto for the Hawk but that does not appear to be curretly the case. 

[Nelson Howe]

Mike,

 

It is possible to select vario for TEK and netto for HAWK.  Or relative netto for HAWK, which is what I do.

 

Nelson

 

From: Mike Clarke
Sent: Friday, August 4, 2023 6:19 AM
To: LXNav Soaring Glider Equipment User Discussion Group
Subject: Re: [LXNAV-user-group] Hawk Vario Optimism

 

John,

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/3e78373c-f4e8-49e2-8976-20fda967fca4n%40googlegroups.com.

 

[Mike Clarke]

Nelson,

Page 95 of the LX manual says “The HAWK needle will show same parameter as Vario needle but calculated with HAWK system.” S,o in vario mode, and cruise mode, both needles show the same thing. That is my experience.

Mike

[Nelson Howe]

Mike,

 

Is it possible your manual is not up to date?  Page 95 in my manual (May 2023 version 9) refers to password resets and brightness controls, not HAWK needles.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/d589dcb7-6133-4383-a2ea-a706b3b817ecn%40googlegroups.com.

 

[Mike Clarke]

Nelson,

I was using #51, apologies. The latest version of the manual is June 2023 #53 and it is on page 96 under the heading "Numerical (navbox) Mode", second bullet point. The words are the same, the Hawk needle will show the same as the  Vario [TEK]   needle.

Mike 

[Nelson Howe]

Mike,

 

Upon reading more closely, I see your point, and agree that the blue needle will read the same as the red needle, though derived by HAWK not TEK.  I hadn’t picked up on that.  That doesn’t strike me as much of problem for how I am using the instrument.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/c7d00d0b-cf18-493a-94c9-46990917f427n%40googlegroups.com.

 

[Morgan Hall]

I finally had a chance to fly with 9.18 custom build for the V8. This has the older Hawk algorithm, but all the other features of the newest builds. 

Night and day difference in Hawk behavior.  All of the sudden it is very reliable and reasonably accurate.  Still a little bit of optimism, but totally acceptable to me if Hawk indicates 6 on average and TEK is 5.  

I didn’t have a single instance of Hawk on crack today and the lift was varied from snotty convergence to smooth wave at one point.  

I’ve let Uros know it seems to work. Hopefully they can figure out what code changes caused the issues in the current beta and stable releases. 

I will try again tomorrow.  

Settings wise, I am still using the same settings as before with the audio band shifted down from 1500hz to around 1100hz and the variance around .10.  Winds seemed reasonable as well, but they were all over the place depending on terrain and altitude.  

Nice improvement though. 

Morgan

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/B5C0A02B-BCF6-4C48-ABD2-CDA1B3382965%40hxcore.ol.

[Jim Staniforth]

Morgan and I spoke during flight. Wondering if the update will follow for S10 / 100 varios.

Jim

[Hal Woodruff]

Morgan,  

I've been following this discussion for some time now and sounds as if you've persevered until a solution came forth. I've for a long time have had very optimistic Hawk various readings to the extent that if I was centered in a 4.0 knot thermal the Hawk would be screaming at me that I should be climbing at 10-12 knots. No amount of my "poor thermaling" technique (as suggested in the Lessons Learned video) could possibly produce such a drastic difference. 

So, how does one go about getting the v9.18 build for the V8? Is this a build for the 90xx series or just for the V8? At one point someone suggested to re-calibrate the AHRS while in the tail dolly as opposed to a normal W&B calibration. What suggestions do you have for calibrating the AHRS?

Thanks again,

Hal Woodruff  (HW)

[Morgan Hall]

Hal,

I will send you the firmware.  Uros said I could share it if I was happy with it and it took until yesterday to get a flight. 

In a comical bit of irony, yesterday was the last day of my demo license. So today I had no Hawk and didn’t get to test it further.  So 99 of 100 days had the bogus firmware.  

I already purchased the license for my Duo, so I will do some more testing in that, but this “older” algorithm seems to work a lot better.  

I didn’t change anything about my level settings from the last time I flew with it. I leveled it per the W&B instructions from Schemmp. 

I did double check my speed with the AHRS indicating level and unballasted I was at about 60kts.  This is a sanity check I think LX should develop. I don’t know what speed it should be, but my guess is that it should be somewhere near best glide.  So that could be a verification step post install in order to confirm “level” is reasonable.  Just a quick gut check. 

Morgan

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/700fc046-a415-4bcb-aff6-dd7af15764fbn%40googlegroups.com.

[Morgan Hall]

Oh and to be clear, the firmware is the “sensor box” firmware, not the 90x0.  Athough you install it from the 90x0. 

[Neil McLaughlin]

Could you send to me also or drop a link we can use to download? 

Thanks!

Sent from my iPhone

On 15 Aug 2023, at 05:21, Morgan Hall <mor...@gmail.com> wrote:



You received this message because you are subscribed to a topic in the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lxnav-user-group/nz4npqP4cyw/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lxnav-user-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAPaUW1S3sGRhj6zJxf4JoxMDt0DPiSDOjwBTpjODx_7FUvtZww%40mail.gmail.com.

[Francois HERSEN]

I fly with LX9000 ver 9.06g, Vario ver 9.17 and  Indicator ver 9.13

For Sigwing I use 0.7, 0.01, 0.5

I spent half a day for leveling (AHRS calibration correction for my glider -9,7°).

I am very happy with the Hawk, and now I fly with only the bleu needle with Hawk vario sound, and pneumatic TEK vario. On the moving map the trace is Hawk netto.

Hawk is a great improuvement!

I check if an update 9.18 is available with internal the wifi, not yet.

[John Ferguson]

If I could have a copy that'd be great, actually got some weather to fly in, in the UK

John

[John Ferguson]

I loaded the "old" Hawk sensor box version and after having flown in some weak and strong conditions in the UK I would say the extremely optimistic reading of HAWK has been tamed. Still a little optimistic in thermals at times but very much improved response.

John

[Mike Clarke]

LX published stable V9.07 for the LX90xx and LX80xx today. Uros confirmed to me that this incorporates the sensor box reverting to the earlier HAWK library. I ran the update on my 9070 and it “updated” the sensor box to V9.19 and the V8 to V9.15. I have not yet flown with it, but I had two flights last week with the beta version of this and HAWK was working very well.

[David Anisman]

This is very good news. Seems like the firmware regression hypothesis has good support so far.

My last 3 flights have been with only Hawk, no TEK at all. This is running an S100 with Spring 2022 firmware, before the regression.

We can now ask for a similar version for the S1X series.

By the way, I started sensing less discrepancies between Hawk and realized thermal climb after I switched the needle to Vario (from Relative) in climb mode. Now typically those two don't depict the same parameter. One is a time window average (typically 20 sec) and the other averages the full climb. So they are not that easy to compare. But one can get an approximate sense by observing them whether they diverge more than they should.

David

C6

[Paul Ruskin]

I've been flying with the latest beta on a 9070 this week - and it's working fine.

I discovered that you can program the function key on the stick to switch between Hawk or TEK audio for either climb or cruise.  The symbol on the vario changes colour when you do so.  So if I do feel uncomfortable with Hawk I can easily switch back to TEK - but actually that's meant that I've stuck with Hawk almost all of the time.

Paul

Paul Ruskin

+44 (0) 7768 615278

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/f47ca948-154b-4ca6-98d6-6431574633a5n%40googlegroups.com.

[John Mclaughlin]

I was advised to use the latest beta version for S100, available here: https://gliding.lxnav.com/beta/

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/75a83456-f101-4b96-87a1-714ed9b60dban%40googlegroups.com.

[Ian Molesworth]

After updating my 9050 to 9.07 ( which degrades the V8 to an older release I understand ) last week and finally getting to fly in very weak and broken conditions yesterday I can confirm that the vario function is back to normal.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CA%2BVVBinuoxyjH47%3D%2BXk%2B%3Dh6PoeOfHcnRwtzP1LFH9b2ahwuzcg%40mail.gmail.com.

[211fl...@gmail.com]

Only if normal is bat-shit crazy

 

Scott Fletcher

211fl...@gmail.com

864-420-1749

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/CAMJg6AHyKWm6u8E9iGQZ1K3dLoiHm4%3DtBb2Y0r39Ubs12sBjyQ%40mail.gmail.com.

[jpg...@gmail.com]

I flew with 9.07 last week and for me Hawk was back to performing really well, and not at all bat-shit crazy.

[John Ferguson]

Has anyone updated to the most recent LX 90x0 firmware. I did so by mistake earlier this year overwriting 9.07 and found the Hawk craziness came back. Unfortunately I haven't flown much this year and am wondering if it was just one of those things or has the software problem recurred. I still have 9.07 on USB and can apply it again.

Thanks for info

John

[Nelson Howe]

Actually, it's been working well for me all year.

Nelson

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/681843c1-e09e-49b5-9427-74250dfa5eb5n%40googlegroups.com.

[Bif Huss]

All,

One thing I've noticed that is odd with Hawk is that often the TEK needle indicates a climb and the altimeter shows increasing altitude while the Hawk indicates sink and then finally catches up and indicates lift. This seems to happen most often in weak 1-3 kt lift. I know that this will be explained as the TEK reacting to a wind gust,  but I sure gain a fair amount of altitude for it to be a gust. Anyone seen the same thing? For this reason, I've switched the audio in climb mode back to vario vs Hawk. Don't get me wrong, I'm an avid user of Hawk, but I find I need to use it in conjunction with the TEK, not by itself.

Bif Huss

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/681843c1-e09e-49b5-9427-74250dfa5eb5n%40googlegroups.com.

[Marc Teugels]

Hey .

I tend to select the audio to Hawk in order to find and center the thermal , once established I tend to select TEK in the stable thermal . In cruise always on Hawk relative netto. 

It is also my feeling that Hawk works less well in turbulent conditions , it varies a bit . But to get into the thermal and avoid missed entries it is a dream .

[Jan Waumans]

How do you switch the audio in flight ?

I have only 10 hrs of so with HAWK but I switched the audio back to TEK (vario) in the first place since after the tow HAWK is completely inaccurate (shows lift while altitude is going down).

I've been told you should reset HAWK (rHAWK) after tow ...

Watching the HAWK needle to turn in works fine though.

Wind speed last weekend was double that of XCSoar (with LXNAV data), maybe high but more correct.

Op dinsdag 11 juni 2024 om 09:08:56 UTC+2 schreef Marc Teugels:

[Marc Teugels]

You need to wire a switch to the v8 and go somewhere in the menus to give a meaning to the switch , something like  'change audio source'

Cannot remember by hart. Play with the menu's

[cdeerinck]

>Setup>Hardware>Remote Stick>Front SC Button and then select "Toggle Audio Source".

[Marc Arnold]

Good tip. Thanks.

[Marc Arnold]

I tried this without success today. “Toggle Audio Source” produced no discernible change.

On Jun 14, 2024, at 9:07 AM, Marc Arnold <ma...@rockymountainsoaring.com> wrote:

Good tip. Thanks.

[Chuck Deerinck]

Do you have the stick with a digital interface? The older sticks had wires, and that would require you hooking it up.

~Chuck

On Jun 15, 2024, at 3:23 PM, Marc Arnold <ma...@rockymountainsoaring.com> wrote:



I tried this without success today. “Toggle Audio Source” produced no discernible change.

[Marc Arnold]

That could be the problem. Thanks.

I can (easily) live without it.

I have SC set to Relative, and Vario set to Vario audio. It seems to be working adequately for me.

Cheers,

Marc

[Marc Teugels]

When it is correctly wired etc , and when actuating the correct button the small "circling" symbol on the V8 should change blue when on Hawk vario and to white when on TEK.

(i see you put it at rear SC button - no idea which stick you use - but i don't think that is the correct solution ) 

Actually when you have a FN switch I think you should use it for that purpose

I personally wired a switch to the V8 , and programmed in the V8 that particular wire to "change audio source" . 

I don't have a FN on my stick , that is why I use the separate switch , I have to change position , and then once again. Works in a funny way but it works 

[Morgan Hall]

The speed command audio switching from the back seat stick may not even work. I feel like I tried it on my Duo back seat on a recent flight and it wasn’t working.  It does work from the front seat. 

But you can only get to the config page on the front 9070, so it may have just been that the different profile in use had an old setting and I couldn’t verify the setting during the flight. 

Morgan

--

You received this message because you are subscribed to the Google Groups "LXNav Soaring Glider Equipment User Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lxnav-user-gro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lxnav-user-group/a82f8a7e-f2e5-448a-81d5-2397b0ef7c52n%40googlegroups.com.

[Craig]

BAck to the original poster. I find my Hawk is also way more optimistic turning right compared to left. Hawk will have av climb at 6-7 kts, Oudie 3 kts,  It has been perfectly balanced in pitch and roll. Interesting that it may be a yaw problem. Any updates on how to correct?

Overall i still find it quite a large advantage in removing much of the gust noise and the instant changes in lift, even though the constant is optimistic when going right. And possibly pessimistic when circling left.

On Monday, 19 June 2023 at 10:04:39 UTC+10 Britton Bluedorn wrote:

Like others I’ve had the hawk vario for a bit now and have learned a few things. I recommend that everyone perform a test and display both the hawk AoA and AoS navboxes on your screen and observe the readouts while in flight. It’s pretty telling.

For my glider when flying wings level with the yaw string straight, the AoS readout shows about 4deg. It really takes a good bit of rudder to make the hawk beta value go to zero, and then the yaw string deflection is obnoxious.

I believe the yaw string as truth for indicated beta angle when flying straight, and the readout of hawk beta is incorrect in my glider (more on why comes later in this message).

My hawk Vario optimism is definitely skewed when circling to the right and better matched circling to the left when beta is reported to be smaller.

Digesting Professor Heinrich papers and talks, he’s got the math and theory right, but he also assumes a perfect installation too. This is where LXNAV needs to step in.

We have the leveling procedure to get the alpha correct, but there is no adjustment for installed beta that I’m aware of.

The pitch axis is easy because it can be read easily on the ground as installed theta angle relative to a ground plane, hence the leveling procedure. Beta on the other hand needs a bit of understanding and probably flight trials to dial in.

Heinrich’s argument when thermalling is that the blue needle is if you flew more accurately (reducing the beta) would get the TEK needle to become closer to the theoretical best climb of the hawk needle. So beta *is* being taken into account with the blue needle display (via nulling beta).

In the real world, our machines are hand crafted and have some slop built into them so everything fits together. My glider specifically was involved in an accident prior to my ownership that tweaked the canopy in its open position and it’s never been like factory fit since. I’m the 3rd or 4th owner and indeed you can almost see the skew in the instrument panel with the canopy off. Whether or not this is from the accident or from the prior owner’s adjustments I can’t tell. I need to measure with a tailors measuring tape from the nose and send to LXNAV to prove to them not every glider is perfect. (You’ll remember we had to do the same for the pitch leveling argument too before they did something about it).

I believe that if we had additional configurable alignment settings for beta (just like we do for alpha), then we could dial in our ‘installed beta skew angle’ and the in flight optimism would be much reduced and useful.

Until that happens, I fly with Hawk in cruise and TEK when circling.

Britton


P.S. I’ve been meaning to send my thoughts to LXNAV, but need to take some measurements first before I inundate them with my engineering logic. If anyone else has measurements of their panels (nose to panel left corner, nose to panel right corner; yes I’m assuming the panel is flat which is another headache) feel free to send my way or beat me to the punch when emailing LXNAV.

You’ll notice in the leveling screen small numbers in the upper left corner. R, P, B which I assume is roll, pitch and beta (though I struggle to understand how they can compute a beta while stationary upon system initialization). I also wonder how much the roll value is influencing our readout too, as mine is definitely not zero but 2.0deg after I’ve leveled in roll via an external level. But let’s argue for beta adjustment first before roll adjustments.

I also might experiment with the 3D printed shims just for fun too until LXNAV adds these alignment features into the software.

[Britton Bluedorn]

Hello, when you say perfectly aligned in pitch and roll, what are the R, P, B values in the upper left corner of the AHRS alignment screen? Have a picture?

I completed an extensive write up on my system and posted it to this group somewhere mid-2024 timeframe I believe.  My first impressions (mid-2023) was that yaw was the culprit turned out to be a roll error that couples into the yaw response for my system.  The enlightening item was the readout of the R, P, B values that really inform the user if their unit is really multi-axis level which is what the HAWK algorithm assumes.

I'd start by reviewing your R, P, B values with wings level and check if R=0.  Beyond that LXNAV has a new feature to compensate for yaw alignment in the AHRS alignment page, though I haven't tried it yet.

I ended up making a shim to roll my unit back into aircraft roll level (as my instrument panel is not perfectly aligned in roll when the canopy id closed).  Once I did that HAWK behaves as advertised, but it was understanding the R, P, B output that really dialed it in.