Today it was overcast (as usual for this year) and I decided to take a short flight between rainfalls. The rain had stopped and I went up on the roof to my usual launch location.

Started up with 100% flight battery. Waited about 55 seconds before status went green (GPS). Had 10-11 satellites. But no map and no home point. After waiting 3 minutes, I decided to start the motors and lift off about 20' altitude and see if I get home point then, as that's often been where I get "home point recorded". But this time nothing. I hovered until the battery level got to 80% and was thinking of aborting my "mission" when finally "the home point has been updated" with an altitude value of about 40'.

Often as is the case, the thing that holds me up from flying is waiting for that home point to update so I can see a flight path.

If the craft has GPS, then why doesn't the home point update then?

I would say 8-10. But in my area, I regularly get 18 or more all the time...

It’s good to be well connected. Haha


RedHotPoker

Hi, Sorry for the troubles that it caused. Did you try to change the location where we can get higher GPS? Did you calibrate the IMU and compass of your drone? Please keep us posted. Thank you for the support.

Your P4 pro receives signals from US GPS satellites and Russian Glonass satellites.
To get a good GPS location fix, it needs to have 6 satellites of one kind.
The display in the app does not show the split between US and Russian satellites.
If you had 5+5 = 10 sats you would not have had an acceptable GPS position fix.
It would be an uncommon, but possible scenario, particularly if you don't have an unobstructed skyview for the Phantom.
There is also the possibiity that one or two sats may have been too low on the horizon and rejected because of poor geometry for a good fix.

If you have buildings, trees, terrain etc mazking a significant part of the sky, it may be better to go higher in atti till the Phantom does get a clear skyview and picks up more satellites.

If you could only receive 10-11 sats, it seems that your launch spot does not have a good skyview.

If launching from an open area with a clear horizon, it would be normal to have 16-20+ sats and GPS lock in less than a minute.
The more sky your GPS antenna can see, the more sats you acquire.

DJI Tony Posted at 2018-10-2 13:54
Hi, Sorry for the troubles that it caused. Did you try to change the location where we can get higher GPS? Did you calibrate the IMU and compass of your drone? Please keep us posted. Thank you for the support.

Did you calibrate the IMU and compass of your drone?
Why do you ask that?
It's completely irrelevant to the situation he described?

That’s actually pretty funny.

Haha, oh well, at least they keep abreast of the situation, being present.  ;-)


RedHotPoker

There really isn't a magic number of satellites. The home point will be set when the GPS health reaches 4. That means the GPS icon at the top of DJI GO will have at least 4 full bars.

I am not sure what the number is.  At least one member has posted a number of times with great confidence that the number is six on one or the other satellite system, which makes sense considering the older generation required 6 GPS satellites.  But when I see GPS/GLONASS logs with 7-8 satellites acquired and yet it includes a satellite fix, I have to wonder to myself if their isn't some other kind of combination that works too, in the DJI system, such as 5&5 or even 4&5.  How about 4&4?  5&3?  6&1 or 6&2 is possible of course, but it seems unlikely.  After all, the GPS used in motor vehicles only need four GPS satellites.  Anyway, in my simple mind, it is an unanswered question.  

MHO

Mark The Droner Posted at 2018-10-2 14:54
I am not sure what the number is.  At least one member has posted a number of times with great confidence that the number is six on one or the other satellite system, which makes sense considering the older generation required 6 GPS satellites.  But when I see GPS/GLONASS logs with 7-8 satellites acquired and yet it includes a satellite fix, I have to wonder to myself if their isn't some other kind of combination that works too, in the DJI system, such as 5&5 or even 4&5.  How about 4&4?  5&3?  6&1 or 6&2 is possible of course, but it seems unlikely.  After all, the GPS used in motor vehicles only need four GPS satellites.  Anyway, in my simple mind, it is an unanswered question.  

MHO

Where I am, there are usually less Russian satellites visible than GPS, so it would be possible to have 7-8 sats total and have 6 of them being GPS sats.
Are you sure you are talking about a P4 flight log though?  7-8 stas is very low for a P4 pro to have with a good sky.
If you were looking at a P3 standard log, it only received US GPS sats so having 7-8 sats would be plenty to have a good fix.
Or a third possibility I have seen is where the Phantom goes to a position where some sats are obscured by trees or other obstacles.
The Phantom will carry on for a short period without dropping into atti like it is giving the sat numbers a chance to come back.

Labroides Posted at 2018-10-2 15:38
Where I am, there are usually less Russian satellites visible than GPS, so it would be possible to have 7-8 sats total and have 6 of them being GPS sats.
Are you sure you are talking about a P4 flight log though?  7-8 stas is very low for a P4 pro to have with a good sky.
If you were looking at a P3 standard log, it only received US GPS sats so having 7-8 sats would be plenty to have a good fix.

It was a GLONASS log posted here.  The log showed 8 sats and then 7 for a brief second or less and then back to 8 but still had the fix.  I even mentioned the low sats w/ the fix in the thread and wondered about it  This was a while back maybe 3-4 months ago?  I wouldn't want to try to find it haha

Labroides Posted at 2018-10-2 14:03
Did you calibrate the IMU and compass of your drone?
Why do you ask that?
It's completely irrelevant to the situation he described?

I think this is their standard reply to everything. That and "May I know what aircraft and device you are using"

Labroides Posted at 2018-10-2 14:03
Did you calibrate the IMU and compass of your drone?
Why do you ask that?
It's completely irrelevant to the situation he described?

Hi, Thanks for noticing our comment. The steps that I provided is for troubleshooting and for the drones to read accurately as the customer has mentioned that the home point cannot be recorded right away.

DJI Tony Posted at 2018-10-2 16:00
Hi, Thanks for noticing our comment. The steps that I provided is for troubleshooting and for the drones to read accurately as the customer has mentioned that the home point cannot be recorded right away.

Do any of the current moderators actually fly Phantoms?

Labroides Posted at 2018-10-2 17:13
Do any of the current moderators actually fly Phantoms?

Haha. Here is the crunch...



RedHotPoker

Labroides Posted at 2018-10-2 13:57
Your P4 pro receives signals from US GPS satellites and Russian Glonass satellites.
To get a good GPS location fix, it needs to have 6 satellites of one kind.
The display in the app does not show the split between US and Russian satellites.

Very good point about the split between GPS/GLONASS satellites. I am in a forested area, but that's why I launch from the roof. Typically, I get GPS (green) in about 50-60 seconds and usually a few seconds after that, "the home point has been updated" when launching from this spot.

It may be possible that the density of the overcast may have added enough additional attenuation that my bird didn't acquire at least 6 of either the American or the Russian satellites.

Normally out on a soccer field, I can acquire 17 satellites. 10-11 is about as good as it gets up on the roof with the trees surrounding us. Down at ground level, it's 5-6 and oscillates from ATTI to GPS and back. That's why I figured raising the Phantom's altitude would have helped acquire more satellites, but once I saw the GPS go condition green, I expected the home point update to happen very shortly thereafter, as is usually the case. Chalk it up to bad weather. More tornadoes blew through here just an hour after I landed.

Mark Weiss Posted at 2018-10-2 18:18
Very good point about the split between GPS/GLONASS satellites. I am in a forested area, but that's why I launch from the roof. Typically, I get GPS (green) in about 50-60 seconds and usually a few seconds after that, "the home point has been updated" when launching from this spot.

It may be possible that the density of the overcast may have added enough additional attenuation that my bird didn't acquire at least 6 of either the American or the Russian satellites.

It may be possible that the density of the overcast may have added enough additional attenuation that my bird didn't acquire at least 6 of either the American or the Russian satellites.
That's one thing you don't have to worry about.
GPS works perfectly well in all weather.
Cloud cover wouldn't make any difference.

Unless the roof is higer than the forest then it will impede on the gps signal depeding on were the sats are.

My bug bear it the fact that the P4P takes a day and an age to aquire a lock yet mateys MP is locked on about 10x faster, hes up and in the air, taken 3 hrs of video and 2132 photos before i even get a lock

If you are out in the open and it takes more than 1.5 mins, something isn't right.

The only time weather makes any difference to Sat acquisition can be but not always in times of severe electrical storm or Sun Spot activity.

Although the actual signal strength of a sat is weak - the frequency used is such that it is a 'get through all' system.

Going back to OP's location and using the roof. I sometimes use my balcony - which is near all wood construction. Its large and easy to take-off / land from. I get sat lock literally as quick as when out in the field along with P-GPS. But I have noticed odd times that Home Point Recorded can take significantly longer ... why ? No idea ... but yes - it does happen.  Note I am using a P3P ... and my old P3S was similar.

Nigel

I go by the bar and not the numeric count - I've lost GPS signal before with the app displaying 20 sats... I wait for at least three bars before take off

Labroides Posted at 2018-10-2 15:38
Where I am, there are usually less Russian satellites visible than GPS, so it would be possible to have 7-8 sats total and have 6 of them being GPS sats.
Are you sure you are talking about a P4 flight log though?  7-8 stas is very low for a P4 pro to have with a good sky.
If you were looking at a P3 standard log, it only received US GPS sats so having 7-8 sats would be plenty to have a good fix.

Wow I found the thread and it didn't take too long either... please see post #213 here:  

https://forum.dji.com/thread-38269-6-1.html

Unfortunately the log is no longer visible...

Mark The Droner Posted at 2018-10-3 02:39
Wow I found the thread and it didn't take too long either... please see post #213 here:  

https://forum.dji.com/thread-38269-6-1.html

Some guy in post #214 made a comment much like the 3rd point in post #9 of this thread.

I've long suspected cloud cover affecting GPS because on dark overcast days, the GPS navigation system in my wife's SUV takes up to 45 minutes on the open highway to acquire enough satellites and show the map and our current location. On clear days, the satellites acquire before I even leave the driveway. The high frequencies that GPS operate at are subject to attenuation from moisture, fog and cloud cover. Ever notice that satellite TV doesn't work during a heavy downpour?

From the rooftop, I can see about 35° of unobstructed sky. The rest is obscured by trees. That improves dramatically when the drone is 40' above the roof and almost level with the treetops.

I've never paid much attention to the signal bar readouts, but I will observe them in the future and look for any correlation with home point recording and number of bars shown. Apparently the AC has a lower threshold for P-GPS mode than for recording home point, in terms of satellite signal quality.

Labroides Posted at 2018-10-2 20:23
If you are out in the open and it takes more than 1.5 mins, something isn't right.

I will have to time it, but if its shorter than that, i would be very surpised, i am referring to the point where she says home point locked..... i think thats about 10 sats, i normally see around 17/18 sats once fully acquired, just seems to take an again befroe she says i can fly....

Mark Weiss Posted at 2018-10-3 08:53
I've long suspected cloud cover affecting GPS because on dark overcast days, the GPS navigation system in my wife's SUV takes up to 45 minutes on the open highway to acquire enough satellites and show the map and our current location. On clear days, the satellites acquire before I even leave the driveway. The high frequencies that GPS operate at are subject to attenuation from moisture, fog and cloud cover. Ever notice that satellite TV doesn't work during a heavy downpour?

From the rooftop, I can see about 35° of unobstructed sky. The rest is obscured by trees. That improves dramatically when the drone is 40' above the roof and almost level with the treetops.

I've long suspected cloud cover affecting GPS because on dark overcast days, the GPS navigation system in my wife's SUV takes up to 45 minutes on the open highway to acquire enough satellites and show the map and our current location. On clear days, the satellites acquire before I even leave the driveway. The high frequencies that GPS operate at are subject to attenuation from moisture, fog and cloud cover.

It's not cloud cover that's affecting things.
GPS was designed from scratch to be an all-weather navigation system

Labroides Posted at 2018-10-3 21:54
I've long suspected cloud cover affecting GPS because on dark overcast days, the GPS navigation system in my wife's SUV takes up to 45 minutes on the open highway to acquire enough satellites and show the map and our current location. On clear days, the satellites acquire before I even leave the driveway. The high frequencies that GPS operate at are subject to attenuation from moisture, fog and cloud cover.

It's not cloud cover that's affecting things.

+1

GPS was a vast improvement on Transit - which suffered bunching of sats in passing over ... so a lot of blank periods of no sats ... reduced signal during bad weather etc.

GPS only suffers any reduced - IF ANY as its rare - in severe electrical storm or Sun Spot activity. For practical purposes - its literally 100% capable.

There must be other reason for any problems.

Nigel

solentlife Posted at 2018-10-4 05:22
+1

GPS was a vast improvement on Transit - which suffered bunching of sats in passing over ... so a lot of blank periods of no sats ... reduced signal during bad weather etc.

GPS was a vast improvement on Transit - which suffered bunching of sats in passing over
Or if you were close to the equator, the sats were too spread out.
My longest between (acceptable) sats when using Transit was 8.5 hrs.
But we still thought it was a huge improvement over using the sextant and drawing triangles.
Being able to get a fix a few times a day even when the horizon was obscured was fantastic.

Labroides Posted at 2018-10-3 21:54
I've long suspected cloud cover affecting GPS because on dark overcast days, the GPS navigation system in my wife's SUV takes up to 45 minutes on the open highway to acquire enough satellites and show the map and our current location. On clear days, the satellites acquire before I even leave the driveway. The high frequencies that GPS operate at are subject to attenuation from moisture, fog and cloud cover.

It's not cloud cover that's affecting things.

How would you explain the long satellite acquire times for both my Phantom and the navigation system in my SUV on cloudy days then?

A fact of physics and radio propagation is that the higher the frequency, the more susceptible to attenuation by atmospheric factors. It's 1575MHz roughly, which is not nearly as bad as C-band satellite, but it's not infallible.

Labroides Posted at 2018-10-4 05:58
GPS was a vast improvement on Transit - which suffered bunching of sats in passing over
Or if you were close to the equator, the sats were too spread out.
My longest between (acceptable) sats when using Transit was 8.5 hrs.

Tell that to navigators in Palawan Passage ..............

Sats would bunch up and then you would wait 12 or more hours for another 'fix' ...

We were contracted to test the Magnavox machines on the Gas ships running Brunei to Japan ... bl**y awful machines !!
In fact we reverted to tried and tested old methods ... we actually knew the reefs and wrecks and had turn points marked on charts using echo sounder results .... because Transit was so bad we could never rely on it.

As to Sextant ... until my brother 'pinched' it - I had a beautiful sextant from my old days. I still have my old sight reduction works books and what was the best 'nav calculator' - Ti59 Module based programmable ... still works today !!

Nigel

They're putting down process stone on our private road today, so I decided to take the Phantom for a local flight to see how it looks from above. I waited til I had 10 satellites and full five bars of GPS signal, but the home point never updated. Since I was staying within 800', after waiting five minutes with the bird hovering at above treetop level, I decided to fly my route up and down the road and come back. Even after that entire flight, the home point never updated and I had no map display. It was starting to sprinkle a tad as I was returning. Above treetop level, there should be no obstructions and satellite signal quality was maxxed out at five bars. There must be another factor to setting the home point besides GPS. Within the first minute, I get GPS lock and seconds later it tells me what class of airspace I'm in, but no home point recorded. Seems to only happen on cloudy days.

Mark Weiss Posted at 2018-10-4 06:32
How would you explain the long satellite acquire times for both my Phantom and the navigation system in my SUV on cloudy days then?

A fact of physics and radio propagation is that the higher the frequency, the more susceptible to attenuation by atmospheric factors. It's 1575MHz roughly, which is not nearly as bad as C-band satellite, but it's not infallible.

How would you explain the long satellite acquire times for both my Phantom and the navigation system in my SUV on cloudy days then?
It's not due to cloud so it must be something else like a poor antenna, bad connections, terrain, tree cover, interference from other devices, coincidence, operator confusion etc, etc.
It really is not clouds.

A fact of physics and radio propagation is that the higher the frequency, the more susceptible to attenuation by atmospheric factors. It's 1575MHz roughly, which is not nearly as bad as C-band satellite, but it's not infallible.
GPS was developed specifically to provide an all-weather navigation system.
Not something which only works in clear weather.
http://gpsinformation.net/gpsclouds.htm

Mark Weiss Posted at 2018-10-4 12:56
They're putting down process stone on our private road today, so I decided to take the Phantom for a local flight to see how it looks from above. I waited til I had 10 satellites and full five bars of GPS signal, but the home point never updated. Since I was staying within 800', after waiting five minutes with the bird hovering at above treetop level, I decided to fly my route up and down the road and come back. Even after that entire flight, the home point never updated and I had no map display. It was starting to sprinkle a tad as I was returning. Above treetop level, there should be no obstructions and satellite signal quality was maxxed out at five bars. There must be another factor to setting the home point besides GPS. Within the first minute, I get GPS lock and seconds later it tells me what class of airspace I'm in, but no home point recorded. Seems to only happen on cloudy days.

Can you upload todays flight and post result here, there should be information there confirming home point set or not:-

http://www.phantomhelp.com/LogViewer/Upload/

Aardvark Posted at 2018-10-4 14:30
Can you upload todays flight and post result here, there should be information there confirming home point set or not:-

http://www.phantomhelp.com/LogViewer/Upload/

Quoted for emphasis.

I'm going to take a wild guess that the log as described in post #30, if we ever get to see it, is going to fall under the "operator confusion" category mentioned in post #31...

Mark Weiss Posted at 2018-10-4 12:56
They're putting down process stone on our private road today, so I decided to take the Phantom for a local flight to see how it looks from above. I waited til I had 10 satellites and full five bars of GPS signal, but the home point never updated. Since I was staying within 800', after waiting five minutes with the bird hovering at above treetop level, I decided to fly my route up and down the road and come back. Even after that entire flight, the home point never updated and I had no map display. It was starting to sprinkle a tad as I was returning. Above treetop level, there should be no obstructions and satellite signal quality was maxxed out at five bars. There must be another factor to setting the home point besides GPS. Within the first minute, I get GPS lock and seconds later it tells me what class of airspace I'm in, but no home point recorded. Seems to only happen on cloudy days.

I waited til I had 10 satellites and full five bars of GPS signal, but the home point never updated. Since I was staying within 800', after waiting five minutes with the bird hovering at above treetop level, I decided to fly my route up and down the road and come back. Even after that entire flight, the home point never updated and I had no map display.

Are you sure that the homepoint never recorded?
Or was it just that you did not hear the confirming message?
No map doesn't mean no homepoint. - The map comes from your wifi or cached map, not from GPS.
Were you flying in atti mode or P-GPS mode?
If you have GPS mode, you have a home point.
Did you have a distance showing in the app?
That distance is the distance from  .... home point and until your Phantom records a home point, it won't show any distance.

Post your flight log and it will clearly show whether you had a home point or not.
I'm guessing it will show one.

The GPS system is supposed to be transparent, but that doesn't mean signals are not attenuated by water vapor. 1575MHz is conveniently between the GSM cellular and PCS cellular bands. There is nothing magical about that frequency.

There is no user confusion about using the GPS. Either it locks on or it doesn't. I can only state that our SUV's GPS takes a lot longer to lock on in cloudy weather than it does on clear days.

I was flying in P-GPS mode here.

Here's the log from today's flight:
http://www.phantomhelp.com/LogViewer/SN26OHASNZWWN6UJ8WJA/
When the home point gets recorded, I get an audible and a visual indication. The map appears with an "H" in a blue circle. That didn't happen today.

I do see that the log shows home point recorded 3.3 seconds after motor start. The question is why it didn't show up on the tablet screen as it normally does.

Another concern is my battery is showing a lot of cell deviation before takeoff. Then it normalize. Impending battery failure? Should I send this battery in for replacement, as it's coming up on 6 months. I have about 30 cycles/ flights on it.

I think you are worrying more than you should be, it does show that it was updated pretty darn quickly, wow

Even after that entire flight, the home point never updated and I had no map display

Why did you think it hadnt updated the home point? if you had no map display then you wouldnt have seen it on the map.
I think its possible you just didnt hear it because you wasnt expecting it to update 3 seconds after turning it on, sound feasible?

Mark Weiss Posted at 2018-10-4 18:44
The GPS system is supposed to be transparent, but that doesn't mean signals are not attenuated by water vapor. 1575MHz is conveniently between the GSM cellular and PCS cellular bands. There is nothing magical about that frequency.

There is no user confusion about using the GPS. Either it locks on or it doesn't. I can only state that our SUV's GPS takes a lot longer to lock on in cloudy weather than it does on clear days.

The GPS system is supposed to be transparent, but that doesn't mean signals are not attenuated by water vapor. 1575MHz is conveniently between the GSM cellular and PCS cellular bands. There is nothing magical about that frequency.
The US government spent about $12 billion to build the GPS system and about $750 million per year to maintain it.
The reason they spent that money was to create an all-weather navigation ststem that workd anywhere in the world.
A critical navigation system that failed because of cloudy weather would be an epic failure.
They have done an excellent job with it and it truly works in all weather - even on cloudy days.
A critical navigation system that was unreliable because of cloudy weather would be an epic failure.
Imagine the maritime and aviation users relying on GPS for critical navigation functions.  

Do you think they would accept a system that doesn't work on cloudy days?

Here are a couple of general references to GPS that all confirm within their first few sentences that weather has no effect on GPS:
https://www.navcen.uscg.gov/?pageName=gpsmain
https://www.jurovichsurveying.com.au/faq/what-is-gps
https://www.civilsimplified.com/resources/what-is-gps
https://www.gtav.asn.au/documents/item/422

I can only state that our SUV's GPS takes a lot longer to lock on in cloudy weather than it does on clear days.
There must be other factors involved.
Cloudy weather has absolutely zero effect on GPS reception.

Here's the log from today's flight:
http://www.phantomhelp.com/LogViewer/SN26OHASNZWWN6UJ8WJA/
When the home point gets recorded, I get an audible and a visual indication. The map appears with an "H" in a blue circle. That didn't happen today.
I do see that the log shows home point recorded 3.3 seconds after motor  start. The question is why it didn't show up on the tablet screen as it  normally does.
The app may be busy with multiple operations and not get to display the homepoint message or you may have the volume down but that doesn't matter.
As soon as the Phantom gets a good GPS position, it records that as the home point.
Watch next time and you'l see the distance shows as N/A until you get GPS and then it shows you the distance from the home point which has been recorded.

After launching, your GPS health went down and did not get back to 4 bars until you were at 50 feet and 5 bars at nearly 100 ft with 15 sats.
If you are interested, you can see the split of Russian and US satellites with an Android app called GPS Status.
It shows the different sats quite clearly.

Another concern is my battery is showing a lot of cell deviation before takeoff. Then it normalize. Impending battery failure? Should I send this battery in for replacement, as it's coming up on 6 months. I have about 30 cycles/ flights on it.
That doesn't look like any cause for concern.
The cell differences are only 0.1 volts and only for the first few seconds.
After that they level out to be in the order of 0.01 volts.

Bashy Posted at 2018-10-4 20:11
I think you are worrying more than you should be, it does show that it was updated pretty darn quickly, wow  

Even after that entire flight, the home point never updated and I had no map display

Nope. Two reasons, as I mentioned earlier:

1. Lacking the usual audible announcement
2. Lacking the "H" in the blue circle and flight path track on the map window.

So in essence, other than seeing my drone by line of sight, I have no way to tell where the drone is via the display.

The fact that the home point is apparently updated (as the log shows) but not doing so on the screen is a relatively new development. We haven't had a sunny day in quite some time, so my last flight on a sunny day, I had a home point and a flight path, even without a cell signal to download a map. I simply get a flight path over a black background instead of terrain map. None of this is appearing the last few times I flew on overcast days.

Mark Weiss Posted at 2018-10-4 20:35
Nope. Two reasons, as I mentioned earlier:

1. Lacking the usual audible announcement

Forget overcast.
It really, really, really has nothing to do with your observations.
Your map doesn't come from GPS, it comes from the internet.
GPS only supplies position information (Lat/Long and time)
The app displays your Phantom's location (from GPS) on the map it downloaded from the internet.
Take your tablet inside, check your wifi connection and open the app and look around on the map.

If, for some reason you don't get the map (a server somewhere is down etc) you still should have a blank map or you can use the radar display together with the displayed distance and altitude to tell where your Phantom is.

Cell signal to download a map ..... not needed ...

As L says ... go indoors ... get on the WiFI ... and cache the map .....

THEN disable wifi mode ... go outside and try again ... you will see your map.

Nigel