xthestreams wrote: Amazing! Thank you, I had a feeling it might have been something like that, as you mentioned the base classes have gone through a ton of updates, my next step before seeing this was going to be to recompile kstars/ekos.
Would that fix the problem?
nou wrote: Then there must be some glitch for me because I compiled everything from source.
This is fairly common version incompatibility issue with 3rd-party drivers whenever the base INDI library interfaces change as they are compiled with a certain version of the base INDI library and if the system has a different version installed, there is a good chance it won't work. There have been a lot of changes in the base classes recently so this is expected. Things stabilize in a few days as Launchpad catches up with the packages and everything falls back into sync. To fix this there should be versioned dependency between the packages (so that the 3rd-party packages would depend on the exact version of the base library they were compiled with), but Launchpad doesn't support that in an easy way (all drivers should be recompiled every time the base library is updated). The alternative is to compile the 3rd-party drivers yourself with the base library version installed in your system so that the interfaces match.
It has been merged to git and in nightly PPA since Nov 23.
Frame width / height and offsets use the same SDK functions as ZWO's ASIStudio's ROI and do increase frame rates in similar way. The ADC depth shown in INDI panel is just for information, the driver always saves either 8 or 16 bit data depending on the chosen format. As for the high speed ADC conversion (usually 10 bit) there is control HighSpeedMode in INDI panel among all the others exposed by the SDK, gain, offset, USB bandwidth, and so on. Setting it to 1 enables the faster ADC function.
This seems to be specific to ASI driver as my ATIK camera doesn't start cooling upon connection and the temperature is not saved to the configuration file either. Unfortunately both my ASI cameras are uncooled so they don't have it in the configuration file either. But it seems both the atik_ccd.cpp and asI_ccd.cpp are trying to save the cooler state for some reason in saveConfigItems function, even though the cooler state is defined as write only and temperature and power are read only, which does sound like an unwanted feature.
From the log it seems you have set the flip limit to one minute (Hour Angle "00h 01m 03s" in the log) which is very low and quite possibly due to cone and other pointing errors the mount's step count still hasn't reached the meridian point even if the telescope is already pointing over it so the mount doesn't flip. Unless you have very limited clearance for filter wheel, camera etc. I suggest increasing the limit to at least 5 minutes, maybe even more to make sure the mount itself is also past the meridian when the flip is requested. I have pier in my observatory and no immediate danger of hitting anything so I can quite easily image about an hour on the "wrong" side of meridian in both directions.
I don't see there would be any PI specific issues, the driver is using very basic serial communication like most INDI drivers. So if you can compile INDI from source and connect the PI to the controller, that should be enough.
I have an untested driver for DDW done, just haven't had time to go to our astronomy club's observatory to actually test it. If someone is adventurous enough to try, the code can be found in
(ie. ddw_driver branch in my fork of main indi repo). Any reports if it can even find the controller, let alone do something with it would be welcome. I can't promise it won't break anything, so being present to physically stop things might be a good idea :)
As for the pointing model, KStars/Ekos has a tool for creating mount model in the alignment tab, but I haven't had the need as I just plate solve the targets anyway.
Can't comment on the handset as I haven't really used one, just EQDir cable. LCD screens might get very slow in cold but haven't seen any break so far. As for dew heaters it's most down to pure laziness, I originally didn't install them, noticed I didn't seem to need them in the observatory so to save weight on the HEQ5 I left them off. Now with EQ8 weight isn't an issue anymore, but haven't bothered. With refractor, SCT or similar closed OTA or without shelter the case would probably be different. Then again the worst humidity issues I've had have been not in properly cold weather but just above freezing in the autumn.
I live in Finland and have had an automated remote observatory for three winters now (and one winter I imaged from my backyard) with temperature routinely under -15C. My observatory is not insulated (glass fiber dome and walls) and not heated apart from the equipment itself (only the control computer and IP-power switch are powered all the time, but don't have extra heating in a cabinet) and for the most part everything has worked just fine regardless of the temperature.
The specifically cold related things I've done are:
- using silicone cabling for power and USB (I have only one each going to the OTA with power and USB hubs mounted on top, so most cables don't move) as they don't get stiff in cold
- changing lubrication in my previous HEQ5 mount (though I did run the first two years and my current EQ8 with originals so far)
- tuning mount gear meshing in the cold and not during summer
The observatory is a bit warmer than the environment as it shields everything from wind and so on so dew and frosting hasn't really been an issue at all, though my open RC OTA is easier than SCT or refractor in that regards. I don't even have dew heaters installed even though I have them as haven't had the need.
The problems I've had related to cold are:
- some guiding issues probably from stiff cables at first before changing them, haven't had those since, though I also improved the cable management at the same time
- had some largeish backlash issues with the mount at first, possibly related to the lubrication and worm gear meshing as they disappeared after changing those
- my first remote controllable power switch just couldn't keep its relay on if it was cold enough and just turned everything off
- my dome closed switch stuck to closed position due to some powder snow having found its way just inside the shutter and melting there, dripping over the switch and freezing there
So all in all things work surprisingly well in cold as well with some caveats.