What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Synthetic runway offset from actual

P

petersb

Well Known Member
GRT HXr synthetic runway display is offset from actual runway position.

The actual GPS location of the airport is offset from the runways, wondering if the EFIS software assumes that the airport GPS location is the point where the two runways intersect.

If so is there anyway to apply a correction so the synthetic runway view lines up with the actual approach
 
It will help GRT and others help you if you post the specific airport you noticed this at.
 
Just a datapoint for you: We've landed at dozens of airports since installing our Horizon HXr, and runways are dead-nuts on, right down to the numbers on the runway. When I am touching down on the center line, the screen is showing me on the center line.

When you set up a synthetic approach, do the boxes lead you to the runway threshold, or is that offset, too?
 
jjhoneck said:
Just a datapoint for you: We've landed at dozens of airports since installing our Horizon HXr, and runways are dead-nuts on, right down to the numbers on the runway. When I am touching down on the center line, the screen is showing me on the center line.

When you set up a synthetic approach, do the boxes lead you to the runway threshold, or is that offset, too?
Click to expand...

Most airports on Ontario canada. Specifically CNC3 Brampton, Collingwood also.

Yes the SAP approach follows the offset and will land you in the adjoining field
 
The same problem happens at the Victoria International airport CYYJ. The SAP takes you on an offset of about 200 feet.
 
You can also add airport runway start and stop lat/Lon's into your user database and build SAPs to runways.
I believe Katie wrote an article on this some time ago.

PS.. I think this is my 1,000th post.

I love this site.
 
Runway Offset

''Canadian airport database issue?''

It has to be because all of the airport I fly to have the runways offset ( CYMX,CYUL,CSE4 ETC...)

The EFIS is a Dynon Skyview SV-700 using PocketFMS

Bruno
[email protected]
 
I had the same issue at KPOU back in 2011 with my AFS4500 getting a GPS feed from my GDU375.

AFS sent me a special update to load on my AFS so they could see exactly what the GPS was doing and they even spoke to Garmin. Of course, even though Garmin has proprietary software they were able to give enough info to send AFS in the right direction. (I was very happy to see competitors work together to help me the customer).

I think it took AFS about half a day at most to reprogram! Once they sent me the update I was on the runway centerline exactly in both actual and SV!

Just send GRT as much info as you can and I am sure they can fix it.
 
Just a story....
In South Africa... pilots were complaining about the runway offset. I told them to be quiet and set their own coordinates. They complained louder. Jepp took them all out of the database. Said they were not surveyed to WGS 84 standards.
Be careful what you ask for....
 
az_gila said:
I thought the US FAA used WGS84 - is that the same thing?
Click to expand...

I am told that the differences between NAD83 and WGS84 in North America are so tiny that they are insignificant for aviation use. For additional confusion, there are different vintages of NAD83 and WGS84, but the changes are miniscule.
 
flightlogic said:
Just a story....
In South Africa... pilots were complaining about the runway offset. I told them to be quiet and set their own coordinates. They complained louder. Jepp took them all out of the database. Said they were not surveyed to WGS 84 standards.
Be careful what you ask for....
Click to expand...

This.

Oh my gawd, my science fiction technology isn't accurate enough! Really!?!? You have, an affordable device, that draws the real world for you on a tv screen! In 3D!!!! And it's not good enough?! :eek::D
 
This is a database issue in which the runway coordinates are not quite right. In Canada, I'm assuming you've got the Jeppesen database-- we see it now and then with the U.S. government database as well. There's really nothing we can do about it from here. You can fix it yourself if it's an airport you go into often by creating a user waypoint virtual airport with correct runway coordinates. It's the same thing you would do if you were adding a private unpublished airstrip to your database. I'll see if I can post the procedure for that on the website this week before we leave for OSH.
 
KatieB said:
This is a database issue in which the runway coordinates are not quite right. In Canada, I'm assuming you've got the Jeppesen database-- we see it now and then with the U.S. government database as well. There's really nothing we can do about it from here. You can fix it yourself if it's an airport you go into often by creating a user waypoint virtual airport with correct runway coordinates. It's the same thing you would do if you were adding a private unpublished airstrip to your database. I'll see if I can post the procedure for that on the website this week before we leave for OSH.
Click to expand...
Something doesn't add up somewhere. I could believe erroneous runway end coordinates provided to Nav Canada by the airport for some airports without LPV approaches. But, some of the airports listed in this thread have LPV approaches, and I know that these approaches are flight checked by Nav Canada - a good friend is one of their flight inspection pilots. If the runway end points used to generate the LPV approach are wrong, that would show up during their flight inspection.

I'll try to get my friend in the loop on this one, to see if we can confirm whether the Nav Canada published data that GRT, Dynon, etc use matches the data used to generate LPV approaches.
 
Please post the airport and runway and we will check the Jeppesen and PocketFMS data.


Rob Hickman
Advanced Flight Systems Inc.
 
Unless the waypoint data comes directly from an IFR-certified GPS, I would always assume that it's VFR-only and that it doesn't have to meet the strict requirements of the LPV database. The Jepp NavData package is available in VFR and IFR versions. If you have the VFR package, it's fair to expect some errors. Our free U.S. nav data is also VFR-only (or not approved for primary IFR navigation).

We can also look up the coordinates of a particular airport and runway in our database package and see what the coordinates look like. There is a way to notify the feds of errors in the U.S. database. I will ask our programmer if there is a contact at Jeppesen for the same.
 
Last edited:
I am using an six year old sport model with the same problems. But it seems to me if my memory is correct that six years ago before any updates to the unit were ever made the approaches to Canadian runways were spot on and I was very impressed with it. Thinking that some day this could be useful having a coupled approach. Again if I remember correctly this all changed a few years ago, mentioned it to GRT at the time and nothing became of it, them saying it was just probably bad data for my local airport.
Maybe if GRT has any of the older data bases or software around from then they could check it out.
 
I think that is why you always want the runway visual environment in actual sight at or above minimums. Just in case a small datum shift to the system has been made since the last time you flew the approach. I still have more comfort and confidence with an ILS than any GPS stand alone approach.
 
Bruno,
I looked at our databases for CYMX, runway 29.

The Jeppesen data lines up with images on google earth exactly. I see no error.

The PFMS databases match the Jeppesen databases exactly as well.

So at least for that airport and runway, I don't see any error.

What would really help is to get a screenshot with your airplane sitting on one of these runways where you see an error, and then a datalog of that too. This way we can see what you are seeing and correlate it to the GPS position we are getting, the database we are using, etc, and see where the error is coming in.
 
I posted a way around this on our Feature of the Week page. The link is here:
http://grtavionics.com/userwp2.html

Before you look at the XML code and totally freak out, realize it's not difficult, just read through the code and you'll figure it out if you know anything about computers. I know nothing about code of any kind and I was able to create an alternate version of my home airport with runways that match reality closer than the database version. I named it something just a character different from the original airport ID and can enter it as my destination in the EFIS flight plan. It appears over top of the published airport on the map. I got my runway coordinates by zooming in close using Google Earth on my computer. One of my customers got spot-on coordinates by recording them off his GPS as he sat on the ends of the runway, and he's now enjoying the SAP into his own grass strip.

It's not perfect, but it can help, particularly if it's your home airport that's messed up. After the OSH dust settles we can dig into the database further.
 
Runway Offset

Hi Dynon Support

I going flying tomorrow and I will try to take a screen shot of Rwy 24 and 29 in CYMX on an approach ( Won't land as there is a landing fee of nearly $30.00 and I have an aversion to landing fees..;-)..

I will also try to take a picture of rwy 28 in CSE4 ( Lachute ) as it displayed the same offset as CYMX and will send them to you probably tomorrow night.

Thanks

Bruno
[email protected]
 
RV-4 said:
Hi Dynon Support

I going flying tomorrow and I will try to take a screen shot of Rwy 24 and 29 in CYMX on an approach ( Won't land as there is a landing fee of nearly $30.00 and I have an aversion to landing fees..;-)..

I will also try to take a picture of rwy 28 in CSE4 ( Lachute ) as it displayed the same offset as CYMX and will send them to you probably tomorrow night.
Click to expand...
Bruno,

It would probably be useful to get screen shots with the aircraft on the ground, on the centre of the runway, pointed down the runway, at some airport that shows the problem. CSE4 would probably work, and that is much cheaper. Having the aircraft actually on the runway would probably simplify the analysis.

How accurate is the heading indication on your EFIS? I suspect that the SVS software must use heading as one of the inputs when it renders the image, and if the heading was in error, that might cause the depicted runway to be displaced to the side of the EFIS when you are on approach.
 
At CEQ3 (14 /32) the heading alignment appears quite parallel, however the synthetic runway is approximately 30 to 50 M east of the actual runway. This is with a new mini X GRT. This is about 1 to 1.5 runway widths east of the actual runway. In the next week I should be able to try some east / west and north / south runways in the area. I suspect it relates to the datum that was used to generate the data base at that point in time.

On the ground, tracking the centerline the gps track is in agreement with the CFS for runway heading.

I will report other locations.
 
Google Earth vs VNC charts

dynonsupport said:
Bruno,
I looked at our databases for CYMX, runway 29.

The Jeppesen data lines up with images on google earth exactly. I see no error.

The PFMS databases match the Jeppesen databases exactly as well.

So at least for that airport and runway, I don't see any error.

What would really help is to get a screenshot with your airplane sitting on one of these runways where you see an error, and then a datalog of that too. This way we can see what you are seeing and correlate it to the GPS position we are getting, the database we are using, etc, and see where the error is coming in.
Click to expand...

It has been awhile since I had a case in front of me, but google earth co-ordinates in Canada do not agree with map co-ordinates or a hand held GPS.

The Map and hand held generally do agree IIRC.

I suppose the upside to this is that if anyone was using google earth to program weapons, they would miss our runways by a few meters.:D

www.ivao.ca/charts/CAP3/CYYC.pdf Here is an address that will get you some approach plates for Calgary. From the ILS 34 plate you can get the threshold Lat and long.

N51 deg 05.80 minutes
W114 deg 01.28 minutes
If you use your google earth cross hairs and place them on the threshold centerline of RWY 34 in Calgary , I think you will find the offset.
Please let me know the difference. I don't have google earth on this computer.
 
Last edited:
i observed similar behaviour depending on which gps source was feeding and looked into details.

turned out to be a floating point issue.
depending on how many digits after the comma the gps is providing position and how many digits after the comma are interpreted by the software (may be truncated either at the sensor or in the interpreter)

database was always right, so it was the own ship position that was not quite right.
 
F1R said:
It has been awhile since I had a case in front of me, but google earth co-ordinates in Canada do not agree with map co-ordinates or a hand held GPS.

The Map and hand held generally do agree IIRC.

I suppose the upside to this is that if anyone was using google earth to program weapons, they would miss our runways by a few meters.:D

www.ivao.ca/charts/CAP3/CYYC.pdf Here is an address that will get you some approach plates for Calgary. From the ILS 34 plate you can get the threshold Lat and long.

N51 deg 05.80 minutes
W114 deg 01.28 minutes
If you use your google earth cross hairs and place them on the threshold centerline of RWY 34 in Calgary , I think you will find the offset.
Please let me know the difference. I don't have google earth on this computer.
Click to expand...

I looked at the threshold of CYYC RWY 34 on maps.google.ca. I zoomed in, and clicked the centre of the threshold, and it claims I clicked at:
51? 05' 48.0"N
114? 01' 17.0"W (or 114? 01' 16.9"W)

Must trusty calculator says this is:
51? 05.80'N
114? 01.28"W

This exactly matches what you got off the approach plate. Two decimal places on the minutes gets us within 60 ft.
 
Runway Offset

Hello Kevin & Dynon Support

I took a couple screen shot today while flying the ILS on CYMX Rwy 24 and it does appears to be fairly close.

From about a mile out ,the runway is slightly to the left but move to the centerline as I was getting closer.Once over the runway and slightly right of centerline as per the localizer, the runway is shown to be a little more to the right.

I then flew down the centerline of RWY 29 and the SV shown me to be flying the right edge of the centerline.


My EFIS heading is dead center normally.

I updated the database this morning.

Will try to take a shot on the ground of the rwys in CSE4 (Lachute ) next time I go there..

Support: Where do you want me to send the pictures??

Bruno
 
Last edited:
Bruno,
Remember that the point of SynVis is to reflect what is out the window. This means that any crosswind will lead to the runway not being in front of you as you crab towards the runway, just like it is off to the side in your real vision.

If you are "just to the right of the center-line," well, some inaccuracy in GPS is just reality. It only takes a few feet to make that error visible when you are on a runway, and a GPS can be off by 10 feet even in normal cases with WAAS.

You can send your pictures to support at dynonavionics.com. Remember to include your data logs as well, or we won't be able to see why there is an error if there is one.
 
A 1? error in heading, or in the variation value used by the EFIS to convert mag heading to true heading, will lead to a 100 ft lateral discrepancy in the runway position when 1 nm back from the threshold. A 5? error would yield a 500 ft lateral discrepancy at 1 nm. Any lateral runway positioning error due to heading or variation will reduce as the aircraft approaches the runway.

I'm wondering what magnetic variation is used by the EFIS, and is it correct? What magnetic variation is the Skyview using at CYMX? The airport diagram suggests the current value is somewhere between 14?W and 15?W.
 
I'll point out that for U.S. civil RNAV procedures, most if not all non RNP (i.e. LNAV, LNAV/VNAV, LP, LPV) final segments are track to fix legs, with the published track magnetic course based upon the true course adjusted by the established airport magnetic variation of record. However, the navigation in a TF leg is not dependent upon that calculated course but is essentially a waypoint to waypoint ground track. Magnetic courses displayed by the FMS or GPS navigator may differ from the published magnetic courses on a procedure depending upon how the FMS or GPS unit calculates local magnetic variation. In most cases, the database used for IFR navigators will properly use the established airport magnetic variation to derive magnetic courses.

Also, the airport's magnetic variation of record may be as much as three degrees (sometimes more) different than the actual magnetic variation at that location. Since the variation drifts, the FAA waits until it drifts three degrees before revising all the SIAPs for the airport.

As an aside, SIAPs using old NAS ground navaids (VOR, ILS, NDB) will use the established magnetic variation of record of the navaid facility providing the final approach course (and not the airport magvar) to calculate the magnetic course from the true course, which is why you will occasionally see the magnetic courses of parallel ILS courses to parallel runways vary by a degree or two, and why RNAV magnetic courses that overlie ILS courses often differ by a degree or two. The intent is that any on airport navaids will utilize the same magnetic variation of record as the airport itself, but that is not always the case.

I am not familiar with how the synthetic vision data is created for display on these avionics systems and how it relates to the underlying RNAV SIAPs, but am curious to find out.
 
Last edited:
Runway Offset

''If you are "just to the right of the center-line," well, some inaccuracy in GPS is just reality. It only takes a few feet to make that error visible when you are on a runway, and a GPS can be off by 10 feet even in normal cases with WAAS''
---------------------------
Today I wasn't trying to be dead on when I flew the ILS 24 in CYMX, I was just trying to see what the SV will give me vs what I had the last time I checked it and it was way better...I don't know if anything has been change but it was more than good enough..

If it is this accurate all the time I'm sure that I will find the runway in just about any weather good enough for an RV-4.I don't plan on testing the CAT 3 capability of an RV-4...I have enough of those at work.

I'll try to remember to get the Datalog next time I go to the hangar.

Thank you for the info guys..

Bruno
 
Hello all.
We have had the same problem with the original supplied database.
After going over to Jeppesen IFR database our GRT Sport SX is showing the runways where they should be.

Best wishes to you all, Bengt, Calgary
 
Kevin Horton said:
I'm wondering what magnetic variation is used by the EFIS, and is it correct? What magnetic variation is the Skyview using at CYMX? The airport diagram suggests the current value is somewhere between 14?W and 15?W.
Click to expand...

Kevin,
SkyView uses the World Magnetic Model from NOAA. This model takes location into account, but also the date and altitude. So there isn't a specific variation it uses at a given location- it's theoretically updating constantly and is always right. As of today at CYMX at 83 meters that calculates to 14.52 degrees.
 
BFoll said:
Hello all.
We have had the same problem with the original supplied database.
After going over to Jeppesen IFR database our GRT Sport SX is showing the runways where they should be.

Best wishes to you all, Bengt, Calgary
Click to expand...

Benget, maybe you can help me. I am
confused about the available third party data base, we have available for the HXr jeppesen and Seattle avionics, are they both maps or airports and frequencies?
 
petersb

The Jeppesen will give you nav data, airspace and airport info, no topo maps.
We can not use Seattle charts in our Sport SX so I have not looked in to that.

Have a nice one,
 
You must log in or register to reply here.
Back
Top