hardware connection issues

[Janne Kurri]

I brought this up in the segfault thread below but since that error is not showing up currently after upgrading os to 16.04 from 14.04, I thought I should start a new one.

So, the issue is that I can detect / connect to usb webcam and synscan mount on rs port when I enter the indi setting window in phd, and while in there I am able to slew the mount too. So it works that far. When I exit the settings window and click to connect devices, I simply get a failed to connect camera / mount errors. I tried using a simulator camera, just to troubleshoot the mount connection for now:

12:00:23.985 00.000 140656712763904 Auto-loading dark library
12:00:23.985 00.000 140656712763904 Error thrown from /build/phd2-fb0NEj/phd2-2.6.1.rev20160407/myframe.cpp:1968->File does not exist
12:00:23.985 00.000 140656712763904 failed to load dark frames from /home/janne/.phd2/darks_defects/PHD2_dark_lib_1.fit
12:00:23.985 00.000 140656712763904 Status Line: Darks not loaded
12:00:23.987 00.002 140656712763904 Status Line: Camera Connected
12:00:23.992 00.005 140656712763904 Connected AO:None
12:00:23.993 00.001 140656712763904 Status Line: Connecting to Mount ...
12:00:24.001 00.008 140656712763904 Resizing image to 772,618


// few hundred lines of that......

12:00:28.994 00.042 140656712763904 Resizing image to 772,618
12:00:29.032 00.038 140656712763904 Resizing image to 772,618
12:00:29.071 00.039 140656712763904 Throw from /build/phd2-fb0NEj/phd2-2.6.1.rev20160407/gear_dialog.cpp:1175->OnButtonConnectScope: connect failed
12:00:29.072 00.001 140656712763904 Status Line: Mount Connect Failed
12:00:29.074 00.002 140656712763904 Resizing image to 772,618
12:00:29.085 00.011 140656712763904 Connected AuxScope:None
12:00:29.086 00.001 140656712763904 Connected Rotator:None

Having launched phd2 from terminal, it returns

:~$ phd2
12.00.02: Debug: Failed to connect to session manager: SESSION_MANAGER environment variable not defined
Dispatch command error(-1): INDI: Could not find property EQUATORIAL_EOD_COORD in SynScan
<setNumberVector device="SynScan" name="EQUATORIAL_EOD_COORD" state="Ok" timeout="60" timestamp="2016-04-17T09:00:13">
    <oneNumber name="RA">
12.176959991455078125
    </oneNumber>
    <oneNumber name="DEC">
90
    </oneNumber>
</setNumberVector>


And, in terminal where I started indi server, only one returned line contains "synscan":

2016-04-17T08:45:58: Driver indi_synscan: Mount does not support setting location

Full debug log: http://pastebin.com/RbRBmMue

Any ideas, what could I try next..?

[Janne Kurri]

Compiled dev version from source, to same end :/ Terminal returns:

./phd2
INDI server localhost/7625 disconnected.

Dispatch command error(-1): INDI: Could not find property EQUATORIAL_EOD_COORD in SynScan

<setNumberVector device="SynScan" name="EQUATORIAL_EOD_COORD" state="Idle" timeout="60" timestamp="2016-04-17T14:11:48">
    <oneNumber name="RA">
21.309784412384033203
    </oneNumber>
    <oneNumber name="DEC">
62.587888240814208984
    </oneNumber>
</setNumberVector>
INDI server localhost/7625 disconnected.

and indi server:

indiserver -p 7625 indi_synscan
2016-04-17T14:03:47: startup: indiserver -p 7625 indi_synscan
2016-04-17T14:03:47: Driver indi_synscan: EnumeratePlugins - found plugin SVD Math Plugin
2016-04-17T14:04:14: Driver indi_synscan: Mount does not support setting location
2016-04-17T14:06:24: Driver indi_synscan: read status bytes didn't get a full read

and the log: http://pastebin.com/1JLSHQVW

[Janne Kurri]

Connecting the webcam as v4l2 imager seems to work - choosing the device as guider instead of imager, pdh won't connect to the camera. Any ideas why this happens - and what difference if any does it make if it's connected to as imager or guider?

Connection from phd to sw eq3-2 synscan mount fails with indi synscan driver but is successfully established on /dev/ttyS0 using indi_eqmod_telescope. KStars/ekos is able to connect with either driver.

Last night was clear enough that I could try this out, and I was able to find a star and start guide calibration - which failed. Apparently the pulses did not actually move the mount at all :( Here's the log file: http://pastebin.com/QyLrRpEe I see in there "Pixel scale = unspecified, Binning = 1, Focal length = unspecified" although the camera was detected and I certainly did enter the focal length information for both "telescope" and "guider"... If the camera is connected as imager, should I manually enter the pixel size somewhere when I'm trying to use it as a guider? Shouldn't the calibration process move the mount by some amount regardless?

[Janne Kurri]

Hmm apparently scope size / pixel size set in indi options does not make it into the log file - any values entered in setup wizard do.

[Patrick Chevalley]

Hi,

It was a long time I not use the synscan with my Heq5, using indi-eqmod is much better and work fine with PHD2.

But I make some inside testing today.
The reason it cannot connect is because the indi-synscan driver do not implement pulse guiding, nor setting the motion rate in milliseconds. Without any of this PHD2 cannot use this driver for guiding.
The only change I can do in PHD2 is to give a clear error message.

Now about the camera.

You must select "Main imager" with a webcam, "Guider" is only for dual chip camera like the ST-7.
I can improve this dialog to show the Guider choice only if two chips are detected.

A v4l2 webcam do not report it's pixel size, so the good option is to set it in PHD2.

Patrick

[Janne Kurri]

On Thursday, April 21, 2016 at 5:40:12 PM UTC+3, Patrick Chevalley wrote:

A v4l2 webcam do not report it's pixel size, so the good option is to set it in PHD2.

Patrick

 
Hi,

Thank you for your response. I'll try again during the weekend if weather permits (sadly does not seem too likely).

I noticed about pixel size, that it does appear to be autodetected - see the attached screenshot.

[Patrick Chevalley]

I make some test with the indi ccd simulator. If I change the pixel size in the indi client (from phd2 or from Kstar), the value is correctly reflected to the phd2 brain.

I remark that your INDI client screenshot use the coma as decimal separator.
Can you check you use the same locale setting to start the indi server and phd2.
You can have lot of problem if one use the coma and the other the dot.

Patrick

[Janne Kurri]

Thank you for taking the time to look into this. Hm. I have no idea why the telescope size and pixel size values would not have appeared on the phd2 end of things in my case... None the less, I have set the values in phd2, and just had the chance to try it outside - now the calibration succeeded and phd started guiding. Yay!

I have another question regarding the synscan/eqmod driver: to connect to the mount with indi_eqmod_telecope, synscan hand controller needs to be in PC direct mode; I was planning to run pec training when phd2 is guiding by selecting the function on the hand controller, but this isn't possible when in PC direct mode. Is it possible to perform pec training from within phd2? I understand that the windows-only eqmod application can do this (but I do not have any windows machines).

[Patrick Chevalley]

You can run a PEC analysis using PecPrep 2.0, it run fine on Linux with Wine and can load your PHD2 guide log. Just guide normally with PHD2 for 4 or 5 time your worm cycle.

This is very useful to better know your mount and if you want to do some hardware tunning, but inidi_eqmod as no option to load a pec file so you cannot use it when guiding. This is not a big problem because I doubt that any small Synta mount can benefit from a PEC correction, there is so much difference between each worm cycle that applying a mean PEC is more harmful than useful.

Patrick

[steve]

El 24/04/2016 a las 09:15, Patrick Chevalley escribió:
> I doubt that any small Synta mount can benefit from a PEC correction

Is an eq6 a 'small Synta mount'?
Cheers,
Steve

[Patrick Chevalley]

Is an eq6 a 'small Synta mount'?

No, we speak about an eq3-2 here.
 

[Janne Kurri]

Too bad, I often do my imaging "in the field" without ac power and I was thinking I could do the polar alignment and pec training before the laptop runs out of juice - certainly it won't last a few hours of guiding in freezing weather. But for now, at least I got the guiding bit to work, now the astronomical dark is over until autumn, perhaps I will have come up with a solution by then :) 

Thanks for the enlightenment!