Longitude bug ruined mount

PMTeam@ZWO

Hello, can you upload the log, let me research and development analysis what is the problem

joe3

@asiair@zwo#41291 Hi here are the logs.

autorun-log-2020-10-06-202555.txt

1kB

autorun-log-2020-10-16-214830.txt

5kB

autorun-log-2020-10-17-032658.txt

528B

autorun-log-2020-10-24-225359.txt

15kB

phd2-guidelog-2020-10-06-202830.txt

19kB

PMTeam@ZWO

Sorry, I didn't describe the problem clearly. There is a problem feedback in aisair. Click Submit, fill in your question, and then upload the log

Photon08

I have this issue as well with asiair plus (that I first tried last night) and my Orion dobsonian (alt az) and synscan hand controller (by and firmware 4.x). East and West are swapped (the mount was correctly set to my longitude 122W but asiair thinks that the mount was in 122E). When I touched 'sync to mount' (the phone location derived from the gps is 122W), the mount hand controller is set to 122E (wrong position). If I changed it back to 122W with the synscan hand controller, goto doesn't work at all. The plate solving works fine, but centering after plate solving went wild because the RA and Dec derived from the asiair's longitude and the current alt/az was very different from the plate-solved ra and Dec (although they were very close to each other).

Not syncing to mount doesn't make goto work, either. I submitted the feedback with logs attached. Hopefully you can reproduce and fix the issue.

w7ay

Photon08 East and West are swapped (the mount was correctly set to my longitude 122W but asiair thinks that the mount was in 122E).

This is truly weird, since the SynScan has an explicit flag to indicate East or West. This is from the Synscan protocol manual:

It is almost as if your ASIAIR is ignoring the 'H' flag, or reading it incorrectly. I'll check with my mount simulator when I have some time later. I wrote my own simulator, so I can breakpoint the program to perhaps see what is happening (i.e., if ASIAIR is sending the wrong flag when a sync is executed). I do not have a SynScan mount, else I can just watch the data stream between the mount and ASIAIR (you don't happen to use WireShark, do you?)

Chen

w7ay

Photon08

OK, definitely a ZWO bug.

I connected ASIAIR to my mount simulator, using the SynScan protocol, 9600 baud.

http://www.w7ay.net/site/Applications/MountSim/

My iPad shows 122º West longitude (I am on the left coast of the USA).

Right after I connected ASIAIR Mount Settings to my virtual mount, ASIAIR pushed a Time/Location command to the mount (i.e., my simulator).

My mount simulator has a warning if the Latitude or Longitude signs in a command are not the same as the values I set as default in my program. I had added this warning because ASIAIR has a long standing problem whenever it comes to date, time, location, pier side of the telescope, etc (i.e., anything to do with celestial navigation); I have not seen any other software produce so many bugs in my life.

Well, I immediately got a warning!

I then placed a breakpoint in my code right at where I received the Time/Location command from ASIAIR. Sure enough, when I reconnected ASIAIR to my simulator, the "H" field in the Time/Location command (see my previous post) contained zero. I.e., ASIAIR was pushing an East Longitude to the mount (simulator) even though my tablet says I am located at West longitude.

Send ZWO support a nasty letter to tell them that the ASIAIR in SynScan mode is sending Time/Location with a zero in the H field even though you are located on the western hemisphere.

Interestingly, the Latitude is correct. But probably because I am north of the equator (since the "D" field is 0 for North). If I were in the southern hemisphere, I wouldn't be a bit surprised if it moved me north of the equator.

Nothing wrong with SynScan or your mount. It is ASIAIR's fault.

Chen

Photon08

w7ay Thanks a lot, Chen, for debugging and figuring out the root cause. I Hopefully, ZWO will take care of this long standing issue asa and release an update. I'll also contact zwo support to alert about this bug. Thanks again !

Dspital

w7ay has this ever been sorted out?
I’m pretty sure that my Celestron CG-5 Meridian flip issues are being caused by Asiair sending faulty goto coordinates.

Photon08

@asiair@zwo , I sent a feedback form with @w7ay's findings above. It'd be great if you could take a look at it and fix it soon. I believe that ASIAIR has bugs in both reading and writing the longitude from and to Synscan HC. When reading it from the mount, it appears to ignore 'H' flag and interpret 122W as 122E. When writing the longitude (sync to mount), it fails to set 'H' flag correctly to 1 for the Western hemisphere (e.g. 122W) and set the mount to 122E.

Those with equatorial mounts can work around this bug by connecting ASIAIR and the mount with an EQMod cable bypassing HC, but in my case, that's not an option because I have an alt-az mount and EQMod cable does not work.

katzinaz

I have the same issues…my W latitudes get changed to E when connecting to my Cgem Ii mount. Also have had the same randomness with pa 60 degree rotation, etc.

w7ay

katzinaz I have the same issues…my W latitudes get changed to E when connecting to my Cgem Ii mount. Also have had the same randomness with pa 60 degree rotation, etc.

I don't have a Synscan mount, so I am not sure if the following will work, but it might be worth a try.

Both Celestron and SkyWatcher are manufactured by Synta, so they pretty much share the same protocol.

What I had notice is that the ASIAIR pushes the Longitude to the mount right after you connect the mount to ASIAIR. And the Longitude the ASIAIR sends has the wrong Hemisphere, if you live west of Greenwich.

However, I also don't see ASIAIR push the location coordinates (by itself) to the mount at any other time, as long as you don't tap on that green Sync To Mount button in the Mount Settings window.

Connect the mount as usual, and give ASIAIR 10 seconds or so to finish sending the wrong longitude. Now go to the hand controller, and force your Longitude back to the correct Hemisphere. If you live south of the equator, do a sanity check that your mount still thinks it in the southern hemisphere. Without tricking my tablet, I have no way of testing southern hemisphere coordinates. If it has moved you from south to the north, correct it in the hand controller as well.

Now, see of things start to work properly. But remember not to touch the Sync To Mount button. If you touch that button, you will be teleported to east of Greenwich.

If that does not completely solve the problem, ASIAIR may still send your mount to weird locations with a GOTO, so make sure you have the hand controller ready to abort the GOTO, in case it does that.

Until then, you just have to wait for ZWO to fix the bug. Or, just move on to more advanced systems such as StellarMate or INDIGO Sky, or if you prefer laptops, to N.I.N.A..

With the Meridian Flip problem last year, ZWO's brilliant programmers [/s] could not find the problem until the location of the bug was pinpointed on this forum to them. The same thing may happen with this Hemisphere bug. In all these years, they probably had no clue where the bug resides (this thread was started in 2019, for god's sakes), and now that they are told exactly where the bug is, there is a glimmer of hope that they will finally fix it.

Chen

Photon08

Last night, I tried what Chen wrote with my Synscan HC + Orion XT10G (alt-az) at long last. Unfortunately, it didn't work even though I never touched 'sync to the mount' button, This is what happened.

ASIAir misread the mount position wrong (hemisphere swap for those in the Western hemisphere). I'm in 122W and my scope was set to 122W, but it misread the coordinate from the scope and thinks that it's 122E. This is what we already know. I did NOT touch 'sync to mount' button.
The scope was pointing at Rigel (because I used it as the 2nd star for the 2-star alignment). I used ASIAIr to go to Horsehead nebula. It moved correctly there (this go to worked properly)
I ran 'platesolve' and sync & center. The plate solving found that it's slightly off. So far so good
Weird things began to happen when it tried to 'sync and center'. Apparently,'sync' after platesolving includes sync'ing ASIAir's long/lat to the mount. It tried to set long to 122W but due to a bug found here, it ended up setting it to 122E.
Once the scope position was set incorrectly, 'center' part made a wild move. Note that the scope was only slightly off the target position so that it needed a little bit of slewing. Instead, the scope began to have a wide swing that I had to stop.
I checked the HC and found that indeed the long was incorrectly set to 122E.

In my feedback email, I pointed ZWO to this thread and specifically what Chen found. I also attached Synscan serial communication protocol ( https://inter-static.skywatcher.com/downloads/synscanserialcommunicationprotocol_version33.pdf ) to my email.

With this much information, I hope ZWO engineers can resolve this issue once and for all before long.

Sam@ZWO

Photon08 we will check, did you upload log? can you tell us your upload email?

Sam@ZWO

@Photon08 this bug should be fixed from 1.8.1, which version did you test?
you need to use sync scan beta instead of syncscan

katzinaz

I am experiencing the same issue. I have a CGEM II mount. Running 1.83 firmware 9.1.2. I set the mount type to Celestron AVX/CGEM. It will come up initially with longitude of W111:28:05, or sometimes, and I don't know why it is different sometimes, it will say E248"31"55. These are what the ASIAIR software reads from the mount. The Longitude from the phone (via the software) is correct at W111:28:05. As soon as I slew to an object, the longitude will change to E and the scope will be pointing as if I was in the E hemisphere. For example, I can do a GoTo, and longitude would be E111:28:05 instead of W. I ran some tests yesterday and uploaded those logs, but of course it seemed to be slewing in the right direction then. As soon as the rain clears, I will test again and upload more logs.

If I set the mount to Sync scan beta, then it will initially read the longitude correctly as W111:28:05. However, once a GoTo command is issued, longitude switches to E111:28:05.

Tech@ZWO

katzinaz Hi Our devs got this problem with CGEM and will fix it in the next version.

Photon08

Sam@ZWO#54334 I uploaded the log when I reported the bug for the first time, but I haven't uploaded the log 11 days ago. I use the iPad version of ASIAIR app and it's 1.8.3 (9.1.2).

katzinaz

@tech@zwo#54551 I look forward to trying it out.

Photon08

@Sam@ZWO#54334 I tried once more yesterday and uploaded the log. Hope your team can find out what's going on and fix.

Tech@ZWO

Photon08 Could you tell us your email address so that we can locate your log, and which options did you choose to connect the mount? please try syscan beta (not synscan)

Photon08

My feedback includes the URL to this thread (https://bbs.astronomy-imaging-camera.com/d/10397-longitude-bug-ruined-mount/20 ) at the very beginning so that it'd be easy to find it. (search for "10397-longitude-bug"). I'll try 'Synscan beta' and let you know what happens.