GPS DriftCast (GPS Drift 2.0) Vastly improved landing location prediction based on winds aloft forecasts

[kjhambrick]

@dvdsnyd --

Another week and a new Q about weathercocking -vs- wind speed and GPS DC interpolation / extrapolation ...

Note: The Saturday, Oct 19 NASA Houston Rocket Club launch at the JSC Site was canceled last nite due to a burn ban,

It's sad because this might have been our last chance to fly at JSC.

Forever.

Because ground-breaking begins in November for new TAMU / ACMI facilities at JSC.

I have been reading about the NHRC launches since the 1990s and my brother Mike is a regular attendee.

I have always wanted to attend a NHRC launch.

And I was hoping ot make a weekend of it with the granddaughters -- definitely doing a tour while we were there.

Anyhow ... the Oct 19 launch has been canceled.

Sad ...

But as usual for our local TX launches, I've been running GPS DC for the site day-by-day.

As for GPS DC ...

My early morning run today was interesting in that each of the hourly wind forcasts for Saturday exceeded 20 mph.

Which reminded me of a question.

When setting up GPS DC, one can enter weathercock distances and altitudes from 0 -to- 20 mph.

It looks like GPS DC factors the actual wind speed into the "Est Dist between Launch / Land" when the wind exceeds 20 mph ...

But what does GPS DC do with weathercocking and altitude when the wind speed exceeds 20 mph ?

This is HEDLEY on an F67W to about 1000 ft in dual deploy mode ( not exactly sure of the Pad pin -- it came from the NHRC web site ).


I am not sure what GPS DC should do with such site conditions.

And, I know that the safety code says we should not fly when the wind exceeds 20 mph but it looks like GPS DC simply reports the weathercock distance and altitude for 20 mph if the wind exceeds 20 mph.

Anyhow, something i ran into a few months ago for a TX launch and forgot to ask.

For the record, these are the Google Earth landing scatter pins for Saturday, Oct 19:



-- kjh

 

[311stylie]

It's sad because this might have been our last chance to fly at JSC.

I know that arrow! Hopefully they find another spot. I only flew there a few times before my flights were all over the waiver and switched to flying with Tripoli Houston solely.

 

[kjhambrick]

I know that arrow! Hopefully they find another spot. I only flew there a few times before my flights were all over the waiver and switched to flying with Tripoli Houston solely.

@311style --

Seabrook is 'right there' ...

We lived in Dickinson back in the Moon Launch days while Dad worked in Texas City.

That was a such a great area for a kid way back then !

-- kjh

 

[dvdsnyd]

@dvdsnyd --

Another week and a new Q about weathercocking -vs- wind speed and GPS DC interpolation / extrapolation ...

Note: The Saturday, Oct 19 NASA Houston Rocket Club launch at the JSC Site was canceled last nite due to a burn ban,

It's sad because this might have been our last chance to fly at JSC.

Forever.

Because ground-breaking begins in November for new TAMU / ACMI facilities at JSC.

I have been reading about the NHRC launches since the 1990s and my brother Mike is a regular attendee.

I have always wanted to attend a NHRC launch.

And I was hoping ot make a weekend of it with the granddaughters -- definitely doing a tour while we were there.

Anyhow ... the Oct 19 launch has been canceled.

Sad ...

But as usual for our local TX launches, I've been running GPS DC for the site day-by-day.

As for GPS DC ...

My early morning run today was interesting in that each of the hourly wind forcasts for Saturday exceeded 20 mph.

Which reminded me of a question.

When setting up GPS DC, one can enter weathercock distances and altitudes from 0 -to- 20 mph.

It looks like GPS DC factors the actual wind speed into the "Est Dist between Launch / Land" when the wind exceeds 20 mph ...

But what does GPS DC do with weathercocking and altitude when the wind speed exceeds 20 mph ?

This is HEDLEY on an F67W to about 1000 ft in dual deploy mode ( not exactly sure of the Pad pin -- it came from the NHRC web site ).
View attachment 672327

I am not sure what GPS DC should do with such site conditions.

And, I know that the safety code says we should not fly when the wind exceeds 20 mph but it looks like GPS DC simply reports the weathercock distance and altitude for 20 mph if the wind exceeds 20 mph.

Anyhow, something i ran into a few months ago for a TX launch and forgot to ask.

For the record, these are the Google Earth landing scatter pins for Saturday, Oct 19:
View attachment 672329


-- kjh

Hi Konrad,
For winds in excess of 20mph, the weathercock simulation just uses the 20mph data point in the table.
The intent at the higher wind speeds is more of a deterrent than anything to get the user thinking about whether they really want to fly at that ground wind speed. 20mph is usually the cut off for a safe launch anyways, so having simulations above that is moot.
Hope this makes sense and helps.
Dave

 

[kjhambrick]

That makes perfect sense, Dave.

I usually scrub my launches if the wind exceeds 15 mph or so which renders my Q as moot.

-- kjh

 

[Rocketclar]

Ran DC 2.30 Friday for yesterday's launch (our last of the season). Attached are the predictions. It was windy and I, along with several others, had several smaller rockets land near the predictions. The yellow line shows the adjusted direction and distance to the landing spot of my Prometheus 3 on a J460 launched about noon. Boost was good and straight to 3186' (vs RockSim est of 3373') , but it does have a lot of fin surface area. We've observed that our site has many wind changes vs the models throughout the day. I'm still encouraged based on many of the other comparable flights and their landings Saturday. This was especially important for this launch with the wind directions predicted as we must not drape a rocket across the high voltage lines that run on the W part of the site. If we do, we have been told we WILL lose the site. At the beginning of the launch Saturday we were limiting HP launches due to the winds at various levels blowing toward the power lines. The model pictures helped us express our concern and explain our reason for limiting impulse.

 

[DavidBellhorn]

A web version of GPS DriftCast is now available.
https://www.gpsdriftcast.com/

This initial release lacks the ability to save forecasts. Everything should function identically to the Excel spreadsheet.

One new feature is a landing scatter map embedded in the webpage. It is using Google Maps' static view in an effort to stay within their free user tier.


The visual design was originally intended to work equally well on desktop and smart phone screens. Ideally it can be run on cellphones at a launch site. Unfortunately the addition of the results data table causes odd scaling with vertical orientation, but hopefully that can be optimized soon.

Please share any other suggestions and issues. @dvdsnyd and I have discussed some potential improvements for both versions. @kjhambrick helped identify some problems which have since been fixed, but there are likely more.

 

[kjhambrick]

Thanks David and Dave !

It will be nice to have GPS DB ( GPS DB == David Bellhorn's GPS DC WebApp ) out at the launch site on my phone from time to time !

-- kjh

 

[NTP2]

A web version of GPS DriftCast is now available.
https://www.gpsdriftcast.com/

This initial release lacks the ability to save forecasts. Everything should function identically to the Excel spreadsheet.

One new feature is a landing scatter map embedded in the webpage. It is using Google Maps' static view in an effort to stay within their free user tier.
View attachment 678820

The visual design was originally intended to work equally well on desktop and smart phone screens. Ideally it can be run on cellphones at a launch site. Unfortunately the addition of the results data table causes odd scaling with vertical orientation, but hopefully that can be optimized soon.

Please share any other suggestions and issues. @dvdsnyd and I have discussed some potential improvements for both versions. @kjhambrick helped identify some problems which have since been fixed, but there are likely more.

so cool.

 

[Buckeye]

Nice work on the web version!

 

[kjhambrick]

Very Nice Job, David !

I especially like the color-coded pins

Below is an imaginary launch at the AARG Hutto site on Sat, Nov 23.

Sorry about the size ... I need to learn how to take a screen shot of a web page

Eeek ! That's a lot of RED in that data table !

This is the LandingScatter.kml in google earth ...


Interesting ... that's just about where my 4000 ft Level 2 flight landed on Nov 4, 2023

And the FlightScatter plot rotated about 90-deg to show a little detail.

I DO like the ground tracks showing the weathercocking distances -- you can even see the little draw beneath the Impenetrable Thicket !


Sorry about my crappy point-of-view ... I am not very good with the 3D GE displays.

I especially like the [Help] button !

Thanks again guys !!

-- kjh

 

[FlyAwaySam]

@DavidBellhorn Notice a bug. On the IOS web page for it, the latitude and longitude fills in only does the 0-9 numeric pad without the decimal point. Kind of makes it hard to fine tune it.

 

[dvdsnyd]

@DavidBellhorn

Maybe just me, but I'm getting a webpage can't be reached when attempting to access https://www.gpsdriftcast.com/ this morning.

 

[DavidBellhorn]

It works on my machine. ¯\_(ツ)_/¯

Are you still having trouble connecting @dvdsnyd? The website comes up for me today, and my hosting company reports everything running normally.

@FlyAwaySam sadly your problem is not entirely unexpected. I previously had to modify the latitude and longitude fields in order to get a negative sign option on my Samsung phone. I will investigate how to better support the IOS virtual keyboard.

Glad you like the color coded pins @kjhambrick! I plan to show the color with hours in the results table too. I'm hoping that helps when the pins mostly cover each other.

Google Earth draws flight path lines okay, but seems confused by custom pin colors. Most of my time up to now was spent researching Google Maps customization options. Reading more Google Earth documentation is on my to-do list.

 

[dvdsnyd]

It works on my machine. ¯\_(ツ)_/¯

Are you still having trouble connecting @dvdsnyd? The website comes up for me today, and my hosting company reports everything running normally.

It's working at home. I tried one more time at work, and it was all out blocked there now. Went from not reachable, to blocked in an hour or 2. Must not like something about it.
Guess no GPS DC at work for me
Mystery solved.
Dave

 

[ksaves2]

Couldn't find a version I could download to my Android phone. Kurt

 

[kjhambrick]

Couldn't find a version I could download to my Android phone. Kurt

On my up-to-date Samsung Galaxy Note 20 Ultra, I browsed to https://www.gpsdriftcast.com/ and it prompted me to install the app.

Now I have an app:


And it works !

HTH ...

-- kjh

p.s. I am running Android version 13 ...

p.p.s. Nice Icon, David !

 

[DavidBellhorn]

@FlyAwaySam Latitude and longitude fields should now be using a standard, full keyboard layout. Best I can tell, none of the IOS numeric keypads include a negative sign.

@ksaves2 @kjhambrick When are you receiving prompts to install an app? It is just a web page that should open inside a browser. That is what happens on my Android 14 phone.



Though I am glad to see the icon scales properly for your icon! I spent way too long drawing it instead of writing code.

 

[kjhambrick]

David --

When I used chrome to browse the site yesterday nothing happened on my phone -- the site opened in chrome as it does on my laptop.

But today when I opened www.gpsdriftcast.com in chrome on my phone, there was a prompt to install the app before the site opened.

I said yes and an app installed, including the beautiful icon.

When I tap the new icon, the site opens as expected:


-- kjh

 

[DavidBellhorn]

Weird! That sure looks just like the website. I've been running tests on my phone for a couple months without experiencing what you described.

I purposely wrote everything with only normal website functionality to keep it as universally accessible as possible. I have no idea what would trigger a prompt to install. Sounds like there is more researching documentation in my future.

 

[kjhambrick]

Weird! That sure looks just like the website. I've been running tests on my phone for a couple months without experiencing what you described.

I purposely wrote everything with only normal website functionality to keep it as universally accessible as possible. I have no idea what would trigger a prompt to install. Sounds like there is more researching documentation in my future.

Yes it does look just like the website.

I am baffled, as usual

One thing I am sure I did yesterday when I got the install prompt is I browsed to www.gpsdriftcast.com where the day before I browsed to gpsdriftcast.com without the www. prefix ...

Maybe ?

-- kjh

 

[NTP2]

I am baffled, as usual

One thing I am sure I did yesterday when I got the install prompt is I browsed to www.gpsdriftcast.com where the day before I browsed to gpsdriftcast.com without the www. prefix ...

Maybe ?

-- kjh

I don’t think so, chrome will only show the prefixes if it thinks you’d like to see them, EG I don’t see www.rocketry forum.com, I see rocketry forum. But if I had a problem or anything odd is happening I’ll usually see the full URL.

 

[DavidBellhorn]

Couldn't find a version I could download to my Android phone. Kurt

I think I figured out why you received a prompt to install. It should just open as a normal website now.

Long answer: This behavior appears to have been caused by some of the icon settings. Getting an icon to appear on all the various web browsers available on different operating systems requires jumping through a lot of hoops. One of the values I set is associated with 'Progressive Web Apps'. Hopefully removing that avoids the issue.

FYI @kjhambrick, I'm not sure what behavior will occur for existing installations. My guess is it will still work, but opens inside a browser UI.

 

[kjhambrick]

My GPS DC Icon no longer works but no worries, I added a mobile bookmark which is about the same for me.

I've been running GPS DB ( what I call the online version ) and comparing it to GLS DC ( the Excel version ) just about daily for the past few weeks.

The results have been virtually identical for me.

This week there is a real launch at the AARG Hutto site but the weather is looking a tad moist and the cloud cover is forecast pretty low.


What the heck, GPS DC / GPS DB don't know no steenkin' couds nor rain !

But windy predicts cloud cover between 3300 and 3700 ft ( when it isn't actually raining )

This is TP with an AT RMS 29/180 H128W to about 3100 ft with the Landing Scatter from GPS DB imported to googleearth:


Looks safe enough relative to the high voltage power lines and trees and ponds and stuff and I've never landed up in that section of farm land !

One more suggestion ...

My rockets weathercock the same way with the same motors at the same site ...

Would it be hard to add a [Save As] for my rocket+motor+site weathercock data so I could load the data for different combinations of rockets and motors and sites for subsequent sims ?

Thanks guys !

GPS DC and GPS DB are awesome !!

-- kjh

 

[kjhambrick]

@DavidBellhorn --

I forgot to mention that after I added a shortcut for GPS DB to the Home Screen on my Samsung Phone, my Icon started working again !

-- kjh

LMGTFY: http://lmgtfy2.com/?q=samsung+galaxy+note+20+ultra+add+website+to+home+screen

 

[DavidBellhorn]

@kjhambrick do you mean the original Icon started working after you added the new shortcut?

Also, are you using the download version of Google Earth? Your image reminded me to investigate the weirdly colored placemark icons. Except I just loaded examples in the Google Earth website, and now they all look like Google Map icons with colors correctly matching my flight paths.

 

[kjhambrick]

@kjhambrick do you mean the original Icon started working after you added the new shortcut?

Yes, I believe so.

Your Icon was so pretty I didn't have the heart to delete it from my phone

I tried it again shortly before I found the instructions to add a web shortcut to my Home Screen as I linked above and I still got the 'must run in Chrome' error ( sorry ... I don't think I ever took a screen shot because you expected that message ) ...

Then I added the shortcut to my home scteen and the icon worked after that ...

This is your icon on my phone home screen:


I don't know if the old icon was replaced by the shortcut ops I linked above ...

And this is the best I could do to capture DPS DB as it started up on my phone:


What I actually saw was a large version of your icon which then faded out and was quickly was replaced by the usual entry forms.

And this is TP on an H128W at Hutto today flying dual deploy to about 3100 ft:


Sorry I don't know how to make the GE image bigger ...( is there a way for me to do that ? )...

Also, are you using the download version of Google Earth? Your image reminded me to investigate the weirdly colored placemark icons. Except I just loaded examples in the Google Earth website, and now they all look like Google Map icons with colors correctly matching my flight paths.
View attachment 682106

Yes, I have been running the Linux Desktop version of googleearth.

I did not even know there was a WebApp version until you asked about it just now !

Thanks for the CLUE !

I just now reran TP on an H128W to about 3100 ft for today's launch at Hutto in chromium-ungoogled on my desktop.

This is the juicy stuff from the GPS DB session in chromium-ungoogled on my Linux Desktop KDE / QT Session:

And this is the Linux googleearth app on my Linux Desktop ( the only way I have ever used GE until just now )


After learning there was a GE WebApp, I fired that up in chromium-ungoogled and this os the same sim in the GE WebApp:



The pin-colors are definitely different in the fat Desptop App !

HTH and thank you both for GPS DC !

I find it VERY useful no matter which-a-way I fire up googleearth

-- kjh

 

[DavidBellhorn]

Interesting! Sounds like there will be more research into phone installation options in my future.

The web version of Google Earth popped up the first time I went to their site. I was already used to that by the time I eventually found their download page for desktops.

Sorry I don't know how to make the GE image bigger ...( is there a way for me to do that ? )

Assuming you mean the image embedded in the GPS DriftCast website, no there is not a way for you to increase the size. It is currently at the maximum resolution for Google's static map format.

Ideally I can switch to a more dynamic map format. The current version was selected for financial reasons. Google tracks every time the website generates a landing map. They provide a certain amount per month for free. That limit allows 4x more static maps versus dynamic.

David

 

[kjhambrick]

Interesting! Sounds like there will be more research into phone installation options in my future.

Whee ! It gets better-n-better

The web version of Google Earth popped up the first time I went to their site. I was already used to that by the time I eventually found their download page for desktops.

Assuming you mean the image embedded in the GPS DriftCast website, no there is not a way for you to increase the size. It is currently at the maximum resolution for Google's static map format.

Ideally I can switch to a more dynamic map format. The current version was selected for financial reasons. Google tracks every time the website generates a landing map. They provide a certain amount per month for free. That limit allows 4x more static maps versus dynamic.

No worries, David.

It is REALLY easy to [File]->[Open] the .kml files.

I just wondered if I was missing another 'fone-feature'

-- kjh

 

[DavidBellhorn]

Would it be hard to add a [Save As] for my rocket+motor+site weathercock data so I could load the data for different combinations of rockets and motors and sites for subsequent sims ?

Huh, it seems my original response to your question never posted for some reason.

My instinct is to save rocket+motor combinations separately from sites. That avoids a need to re-enter identical data when circumstances require launching at a different event than originally planned.

There are some practical questions to consider. Do we allow unlimited entries to be saved? Can someone easily load one combination from among 100s or 1000s? Is older data automatically deleted when maximum capacity is reached?

Future updates to weathercock functionality may even render this save/load feature moot. Of course if/when that becomes a reality is unknown. Implementing your suggestion in the mean time may be a better short term solution. Definitely worth discussion.

This applies to both Excel and web versions by the way. Obviously they require different efforts, but these underlying questions remain.